Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is a contagious respiratory illness which can cause serious complications including stroke, kidney failure, and cardiovascular complications.1 Cardiovascular complications are a major risk factor for COVID-19 mortality.2 3 4 The aim of this paper was to review the literature, published by December 2020, in order to elucidate the risk factors, putative causes, diagnosis, and management of cardiovascular complications of COVID-19.

Two early studies on COVID-19 reported that 20% to 28% of patients with COVID-19 had cardiac injury associated with cardiac dysfunction and arrhythmias.2 3 In a cohort of 416 patients hospitalised with confirmed COVID-19, cardiac injury was reported to occur in 19.7%, and was associated with an unexpectedly high risk of mortality during hospitalisation. Symptoms of COVID-19 were more severe when accompanied by cardiac injury; the mortality rate was higher among patients with cardiac injury than among those without (51.2% vs 4.5%).2

Independently, in another cohort of 187 patients, those with cardiac injury were more likely to be male, older and to have more co-morbidities including diabetes, hypertension, coronary artery disease, chronic kidney disease, chronic lung disease, etc. Severe COVID-19 infections were also potentially associated with cardiac arrhythmias and the need for mechanical ventilation. The mortality during hospitalisation was 7.62% for patients without underlying cardiovascular disease and normal cardiac troponin (c-TN) levels, but as high as 69.44% for those with underlying cardiovascular disease and elevated c-TN.3

In another report of 72 314 cases (44 672 confirmed) of COVID-19, the crude mortality rate was 2.3%.4 For octogenarians, the case fatality rate was 14.8%. A history of coronary artery disease was present in 4.2% of all cases, but in 22.7% of fatal cases. Case fatality rates were 10.5% for coronary artery disease, 7.3% for diabetes, and 6% for hypertension. Risk of COVID-19 death is highest among the oldest and lowest among the youngest populations. Compared with those aged 18 to 29 years, people aged 75 to 84 years and those aged ≥85 years have 200-times and 630-times, respectively, higher average death rates.5

In a retrospective study of 393 patients, the prevalence of obesity and male sex also appears to be higher in patients with COVID-19 who developed severe symptoms compared with those who did not.6

Cardiovascular complications of COVID-19 are generally associated with poor prognosis. Therefore, prevention and treatment of COVID-19 should be considered a priority. To that end, an understanding of the possible pathogenetic mechanisms resulting in myocardial injury would be helpful. Putative causes of cardiovascular complications in patients with COVID-19 include: cytokine storm, myocarditis, extreme physical and emotional stress, ischaemic injury caused by cardiac microangiopathy or macrovascular coronary artery disease, hypercoagulopathy, right heart strain, and cor pulmonale associated with adult respiratory distress syndrome.

In addition to direct viral damage, uncontrolled inflammation or ‘cytokine storm’—indicated by high levels of inflammatory markers including C-reactive protein (CRP), ferritin, and D-dimer, and increased levels of inflammatory cytokines and chemokines—has been reported in patients with COVID-19.7 8 However, the exact pathogenetic relevance of cytokine storm has yet to be confirmed.9

Myocardial inflammation (myocarditis) is evidenced by elevated c-TN level in some patients10 and autopsy data show mononuclear infiltrate in the myocardium, with related cardiomyocyte necrosis.11 Although there have been case reports of myocarditis in patients with COVID-19, it is unclear whether myocarditis is caused by direct viral invasion or an uncontrolled inflammatory response.10 12

In a cohort study of 39 autopsy cases of COVID-19, cardiac infection with COVID-19 was frequently found; however, overt myocarditis was not observed in the acute phase.13 In contrast, another study reported on the detection of SARS-CoV-2 genomes in endomyocardial biopsies.14

A cardiac magnetic resonance (MR) imaging study of 100 patients recently recovered from COVID-19 reported cardiac involvement in 78% of them, with evidence of ongoing myocardial inflammation in 60% of them. Such involvement appeared independent of pre-existing conditions, severity, overall course of the acute illness, and the time from diagnosis.13 Of 26 competitive athletes, four (15.4%) had cardiac MR findings suggestive of myocarditis and eight additional athletes (30.8%) exhibited late gadolinium enhancement without T2 elevation suggestive of prior myocardial injury.15

In a study of 145 student athletes with COVID-19 who were either asymptomatic or had mild to moderate symptoms during acute infection, cardiac MR findings (at a median of 15 days after a positive test result for COVID-19) were consistent with myocarditis in only two patients (1.4%), based on updated Lake Louise criteria.16

In contrast, preliminary data based on a small autopsy study of 40 patients showed that cardiac injury results more from clotting than from inflammation; microthrombi were frequent, whereas none of the patients had myocarditis.17 While this observation has implications for thromboprophylaxis, whether COVID-19 can cause a viral myocarditis is yet to be confirmed.

Cases of typical stress cardiomyopathy have also been reported,18 suggesting that both physical and emotional stress may be in part contributory to some cases of cardiovascular complications of COVID-19.

In some patients, ST-segment elevation myocardial infarction (STEMI) may be the first clinical manifestation of COVID-19.19 However, patients with c-TN elevations may not have epicardial coronary artery obstruction at angiography. In a case series of 18 patients with COVID-19 with STEMI, nine patients underwent coronary angiography; six of them (67%) had obstructive disease. A total of 13 patients died in the hospital (4 due to fatal myocardial infarction and 9 due to noncoronary myocardial injury).20 In contrast, patients with COVID-19 with STEMI had more thrombus burden and required more anticoagulation than patients with no COVID-19 infection.21 Very-late stent thrombosis has also been reported with patients with COVID-19 and can be one of the presenting features of COVID-19 in those with a history of coronary stenting.22

Coronavirus disease 2019 is associated with a hypercoagulable state.23 Although the pathogenesis is not completely understood, the following may be observed: elevated fibrinogen and D-dimer; prolongation of both the prothrombin time and activated partial thromboplastin time; and mild thrombocytosis or thrombocytopenia. Major adverse cardiovascular events, and symptomatic thromboembolism, occur frequently in patients with COVID-19, especially among those in the intensive care unit (ICU), even after thromboprophylaxis.24

Unchecked vascular thrombosis may result in neurological complications. In a case series of 214 patients with COVID-19, neurological symptoms were seen in 36.4% of patients and were more common in patients with severe infection.25 A retrospective study of 214 patients reported six patients with acute stroke, of which five were ischaemic stroke.26 Stroke has also been reported in younger patients (aged 33-49 years) with COVID-19.27

Post-mortem studies of 12 patients have reported pulmonary embolism as the direct cause of death in four patients (33%) and deep venous thrombosis in seven patients (58%).28 The risk for venous thromboembolism is markedly elevated with prevalence up to 32%,24 29 30 highest with patients in the ICU.30 In a large study involving 3334 consecutive hospitalised patients with COVID-19, among 829 patients in the ICU, 29.4% had a thrombotic event (13.6% venous and 18.6% arterial).30 Although low-dose anticoagulation has been used for thromboprophylaxis, in a series of 184 critically ill patients with COVID-19, 31% suffered clinically significant thrombotic complications despite low-dose nadroparin.31

A meta-analysis demonstrated thrombocytopenia in patients with severe disease is associated with increased risk of COVID-19 mortality.32 How thrombocytopenia should be factored into the decision to prescribe anticoagulant therapy has yet to be studied.

An echocardiographic study of 110 COVID-19 cases noted right ventricular dilation in 31% of patients.33 Another study demonstrated that when compared with those in the lowest quartile, patients with the highest right ventricular longitudinal strain quartile had an increased risk of elevated D-dimer and CRP levels, acute cardiac injury, acute respiratory distress syndrome, deep vein thrombosis as well as mortality.34 Acute cor pulmonale, right heart strain, and/or pulmonary hypertension should always be considered in critically ill patients with COVID-19.35

Early data suggested an incidence of 16.7% arrhythmias among hospitalised patients with COVID-19 and 44.4% of ICU admissions.36 A multicentre study of 192 patients with COVID-19 reported a prevalence of 12.5% for atrial fibrillation among hospitalised patients with COVID-19.37 Another study evaluating 115 patients with COVID-19 reported atrial tachyarrhythmia in 16.5% of patients, with atrial fibrillation being the most common (63%).38 Those with atrial tachyarrhythmia had higher CRP and D-dimer levels compared with those without atrial tachyarrhythmia. Among 393 patients with COVID-19, atrial arrhythmias were more common among patients on ventilators (18.5% vs 1.9%).6 In another study of 700 patients with COVID-19, nine patients experienced cardiac arrest. All cardiac arrests occurred in patients in the ICU. No patients experienced sustained monomorphic ventricular tachycardia, ventricular fibrillation, or complete heart block. Twenty-five patients had atrial fibrillation, nine had significant bradyarrhythmia, and 10 had non-sustained ventricular tachycardia.39 Among 187 patients with COVID-19, when compared with patients with normal c-TN levels, those with elevated c-TN levels developed more frequent malignant arrhythmias (17.3% vs 1.5%), including ventricular tachycardia/ventricular fibrillation.3

Patients with cardiovascular disease and heart failure are more susceptible to COVID-19 and have a more severe clinical course once infected.40 41 In two studies of patients with COVID-19 hospitalised in Wuhan, heart failure was identified as a complication in about 50% of the fatalities.42 In a retrospective multicentred study, among 8383 patients with heart failure who were hospitalised with COVID-19, nearly one in four died during hospitalisation.43 Evidently, heart failure in patients with COVID-19 may be triggered or aggravated by the acute infection in patients with pre-existing cardiovascular disease or incident acute myocardial insult.

Malignant tachyarrhythmias resulting in cardiac arrest present a dilemma for caregivers. The outcomes of out-of-hospital cardiac arrest were worse during the first weeks of the COVID-19 pandemic in the United States, and this was observed not only in areas with high case-fatality rates but also ones with lower rates.44 In a retrospective study of 136 patients with COVID-19, 119 (87.5%) had a respiratory cause for their cardiac arrest, and the initial rhythm was asystole in 89.7%, pulseless electrical activity in 4.4%, and shockable in 5.9%. The return-of-spontaneous-circulation rate was 13.2% and 30-day survival rate was only 2.9%.45 In another study of 54 patients with COVID-19, the mortality rate following cardiopulmonary resuscitation was even worse (100%). The initial rhythm was non-shockable for 52 patients (96.3%), with pulseless electrical activity being the most common (81.5%). Although the return-of-spontaneous-circulation rate was achieved in 29 patients (53.7%), none survived to be discharged home.46

Prognostic parameters for cardiovascular complications in patients with COVID-19 include c-TN level, D-dimer level, and N-terminal pro B-type natriuretic peptide (NT-proBNP) level.

Increases in c-TN level indicative of myocardial injury is common in patients with COVID-19 and is associated with adverse outcomes such as arrhythmias and death. The risk of cardiac injury, as diagnosed by increased c-TN levels (>99th percentile), was found in up to 22% of patients in the ICU, and in 59% of those that died.36 In another study of 2736 patients with COVID-19, c-TN elevation was observed in 36%, and c-TN elevation (>0.09 ng/dL) appears to triple the mortality risk.47 Other studies of patients with COVID-19 have also demonstrated a poorer prognosis, including mortality, in patients with c-TN elevation.41 48 Both c-TN and NT-proBNP levels were documented to be elevated significantly during the course of hospitalisation among those who eventually died, but no dynamic changes were observed among the survivors.3 Moreover, patients with COVID-19 with myocardial injury who also have transthoracic echocardiography abnormalities had a higher mortality risk.49

Elevated D-dimer levels were higher among patients with COVID-19 and was correlated with a poorer prognosis. Multivariate analysis showed increasing odds of in-hospital death associated with D-dimer value above 1 μg/mL.50 In a study of 343 patients with COVID-19, D-dimer levels ≥2.0 μg/mL had a higher incidence of mortality compared with those with D-dimer levels <2.0 μg/mL (12/67 vs 1/267, P<0.001).51 A markedly elevated D-dimer (>6 times the upper limit of normal) is a consistent predictor of thrombotic events and poor overall prognosis.52 Indeed, the International Society on Thrombosis and Haemostasis has advised that for patients who have markedly raised D-dimers (arbitrarily defined as three- to four-fold increase), admission to hospital should be considered even in the absence of other severe symptoms.53 The importance of D-dimer is emphasised in several other international guidelines.52 53 54 55

As a biomarker of heart failure, NT-proBNP levels are commonly elevated in hospitalised patients with COVID-19, particularly in those with elevated c-TN levels. The report by Shi et al2 showed that NT-proBNP levels were significantly higher in patients with elevated c-TN levels than in those without c-TN elevation (1689 vs 139 pg/mL). A study of 3219 hospitalised patients with COVID-19, elevated c-TN was detected in 6.5%, and an elevated NT-proBNP level in 12.9%.56 The adjusted hazard ratio for 28-day mortality for c-TN was 7.12 and for NT-proBNP 5.11, confirming that elevated NT-proBNP levels also carry prognostic information. Although NT-proBNP provides corroborating laboratory information on heart failure, the caveat is that NT-proBNP levels increase with age and with various other conditions including renal failure, thus compromising its utility in older patients with confounding variables.

The approach to the diagnosis and management of STEMI in patients with COVID-19 is similar to that for patients without (Table 1). The approaches endorsed by the American College of Cardiology are recommended57: their emphasis is on patient selection for the cardiac catheterisation laboratory, resource allocation, and protection of the interventional team and other healthcare workers involved in caring for the COVID-19 patient.

On occasion, it is reasonable to liberalise the use of intravenous thrombolytic therapy relative to primary percutaneous coronary intervention. Intravenous thrombolytic therapy can be considered for a relatively stable patient with STEMI and COVID-19. Obviously, in those STEMI patients who are critically ill with COVID-19, the decision to reperfuse with either primary percutaneous coronary intervention or intravenous thrombolytic therapy should be individualised, and contingent upon hospital resources. In this regard, the consensus statement from the Taiwan Society of Cardiology is both pragmatic and reasonable.58

In the event that primary percutaneous coronary intervention is to be performed, maximum personal protective equipment is essential. Intubation, suction, and cardiopulmonary resuscitation all result in aerosolisation of respiratory secretions and increase the risks to the hospital staff. Patients already intubated pose less of an infectious risk. Hence patients with COVID-19 or suspected COVID-19 requiring intubation should be intubated prior to arrival to the catheterisation suite.

In the treatment of STEMI patients, an early Hong Kong study reported that both the “symptom onset to first medical contact” and the “door-to-device” times pertaining to primary percutaneous coronary intervention were reported to be substantially prolonged.59

Studies from both England60 and the United States61 have confirmed that hospital admissions for acute coronary syndrome declined by 40% to 48% in the early days of COVID-19. It is likely that patients with acute coronary syndromes avoided attending hospital during this period.

Standard indications for use of various agents for treatment of heart failure apply to patients with COVID-19. The coexistence of heart failure and COVID-19 complicates diagnosis and management because of overlapping chest findings; however, there are notable differences in chest computed tomography between heart failure and COVID-19 pneumonia, such as lesion distribution/morphology, and pulmonary vein engorgement, which can all help to differentiate between the two.62

Cardiopulmonary resuscitation poses a very high risk for viral spread, and full personal protective equipment should be provided. Immediate intubation should be prioritised in order to minimise the duration of any aerosolisation. While awaiting intubation, bag/mask ventilation with filter is advised.

Several international guidelines have issued recommendations advocating chemoprophylaxis in all hospitalised patients with COVID-19,63 64 65 66 in the absence of both contra-indications and bleeding complications (Table 2). In the event thromboprophylaxis is deemed indicated, low-molecular-weight heparin is preferred, but unfractionated heparin can be used if low-molecular-weight heparin is unavailable or if kidney function is severely impaired. Low-molecular-weight heparin may be preferred over unfractionated heparin for staff safety reasons.

As for athletes who have recovered from COVID-19 infections, a recent expert consensus article recommended 2-week convalescence followed by no diagnostic cardiac testing if asymptomatic, and an electrocardiogram and transthoracic echocardiogram in mildly symptomatic athletes with COVID-19 to return to participate in competitive sports.67

Cardiovascular complications of COVID-19 are associated with higher fatality rates. Putative causes of cardiac injury include cytokine storm, myocarditis, extreme physical and emotional stress, ischaemic injury, hypercoagulopathy, right heart strain, and cor pulmonale, or combinations thereof. Echocardiography and c-TN, D-dimer, and NT-proBNP levels all provide prognostic information. Aside from percutaneous coronary intervention for STEMI, there is no specific treatment for COVID-19-associated cardiac injury, and management is primarily supportive. Whether antiviral therapies administered early in the course of disease will prevent severe disease and cardiovascular complications associated with COVID-19 remain to be seen.

Concept or design: YSA Lo.
Acquisition of data: YSA Lo, C Jok.
Analysis or interpretation of data: YSA Lo.
Drafting of the manuscript: YSA Lo.
Critical revision of the manuscript for important intellectual content: YSA Lo, HF Tse.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

The authors have no conflicts of interest to disclose.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1. Drake TM, Riad AM, Fairfield CJ, et al. Characterisation of in-hospital complications associated with COVID-19 using the ISARIC WHO Clinical Characterisation Protocol UK: a prospective, multicentre cohort study. Lancet 2021;398:223-7. Crossref

2. Shi S, Qin M, Shen B, et al. Association of cardiac injury with mortality in hospitalized patients with COVID-19 in Wuhan, China. JAMA Cardiol 2020;5:802-10. Crossref

3. Guo T, Fan Y, Chen M, et al. Cardiovascular implications of fatal outcomes of patients with coronavirus disease 2019 (COVID-19). JAMA Cardiol 2020;5:811-8. Crossref

4. Wu Z, McGoogan JM. Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: summary of a report of 72 314 cases from the Chinese Center for Disease Control and Prevention. JAMA 2020;323:1239-42. Crossref

5. Centers for Disease Control and Prevention, US Department of Health & Human Services. COVID-19 hospitalization and death by age. 11 February 2020 (updated 18 August 2020). Available from: https://www.cdc. gov/coronavirus/2019-ncov/covid-data/investigations-discovery/hospitalization-death-by-age.html. Accessed 2 Dec 2020.

6. Goyal P, Choi JJ, Pinheiro LC, et al. Clinical characteristics of Covid-19 in New York city. N Engl J Med 2020;382:2372-4. Crossref

7. Mehta P, McAuley DF, Brown M, et al. COVID-19: consider cytokine storm syndromes and immunosuppression. Lancet 2020;395:1033-4. Crossref

8. Zhang X, Tan Y, Ling Y, et al. Viral and host factors related to the clinical outcome of COVID-19. Nature 2020;583;437-40. Crossref

9. Sinha P, Matthay MA, Calfee CS. Is a “Cytokine Storm” relevant to COVID-19? JAMA Intern Med 2020;180:1152-4. Crossref

10. Zeng J, Liu Y, Yuan J, et al. First case of COVID-19 complicated with fulminant myocarditis: a case report and insights. Infection 2020;48:773-7. Crossref

11. Yao XH, Li TY, He ZC, et al. A pathological report of three COVID-19 cases by minimally invasive autopsies [in Chinese]. Zhonghua Bing Li Xue Za Zhi 2020;49:411-7.

12. Lindner D, Fitzek A, Bräuninger H, et al. Association of cardiac infection with SARS-CoV-2 in confirmed COVID-19 autopsy cases. JAMA Cardiol 2020;5:1281-5. Crossref

13. Puntmann VO, Carerj ML, Wieters I, et al. Outcomes of cardiovascular magnetic resonance imaging in patients recently recovered from coronavirus disease 2019 (COVID-19). JAMA Cardiol 2020;5:1265-73. Crossref

14. Escher F, Pietsch H, Aleshcheva G, et al. Detection of viral SARS-CoV-2 genomes and histopathological changes in endomyocardial biopsies. ESC Heart Fail 2020;7:2440-7. Crossref

15. Rajpal S, Tong MS; Borchers J, et al. Cardiovascular magnetic resonance findings in competitive athletes recovering from COVID-19 Infection. JAMA Cardiol 2021;6:116-8. Crossref

16. Starekova J, Bluemke DA, Bradham WS, et al. Evaluation for myocarditis in competitive student athletes recovering from coronavirus disease 2019 with cardiac magnetic resonance imaging. JAMA Cardiol 2021;6:945-50. Crossref

17. Phend C. COVID heart autopsies point more to clot damage than myocarditis. Available from: https://www.medpagetoday.com/meetingcoverage/tct/89143. Accessed 7 Dec 2020.

18. Jabri A, Kalra A, Kumar A, et al. Incidence of stress cardiomyopathy during the coronavirus disease 2019 pandemic. JAMA Netw Open 2020;3:e2014780. Crossref

19. Stefanini GG, Montorfano M, Trabattoni D, et al. ST-elevation myocardial infarction in patients with COVID-19: clinical and angiographic outcomes. Circulation 2020;141:2113-6. Crossref

20. Bangalore S, Sharma A, Slotwiner A, et al. ST-segment elevation in patients with Covid-19—a case series. N Engl J Med 2020;382:2478-80. Crossref

21. Choudry FA, Hamshere SM, Rathod KS, et al. High thrombus burden in patients with COVID-19 presenting with ST-segment elevation myocardial infarction. J Am Coll Cardiol 2020;76:1168-76. Crossref

22. Prieto-Lobato A, Ramos-Martínez R, Vallejo-Calcerrada N, Corbí-Pascual M, Córdoba-Soriano JG. A case series of stent thrombosis during the COVID-19 pandemic. JACC Case Rep 2020;2:1291-6. Crossref

23. Maier CL, Truong AD, Auld SC, Polly DM, Tanksley CL, Duncan A. COVID-19-associated hyperviscosity: a link between inflammation and thrombophilia? Lancet 2020;395:1758-9. Crossref

24. Cui S, Chen S, Li X, Liu S, Wang F. Prevalence of venous thromboembolism in patients with severe novel coronavirus pneumonia. J Thromb Haemost 2020;18:1421-4. Crossref

25. Mao L, Jin H, Wang M, et al. Neurologic manifestations of hospitalized patients with coronavirus disease 2019 in Wuhan, China. JAMA Neurol 2020;77:683-90. Crossref

26. Li Y, Li M, Wang M, et al. Acute cerebrovascular disease following COVID-19: a single center, retrospective, observational study. Stroke Vasc Neurol 2020;5:279-84. Crossref

27. Oxley TJ, Mocco J, Majidi S, et al. Large-vessel stroke as a presenting feature of Covid-19 in the young. N Engl J Med 2020;382:e60. Crossref

28. Wichmann D, Sperhake JP, Lütgehetmann M, et al. Autopsy findings and venous thromboembolism in patients with COVID-19: a prospective cohort study. Ann Intern Med 2020;173:268-77. Crossref

29. Artifoni M, Danic G, Gautier G, et al. Systematic assessment of venous thromboembolism in COVID-19 patients receiving thromboprophylaxis: incidence and role of D-dimer as predictive factors. J Thromb Thrombolysis 2020;50:211-6. Crossref

30. Bilaloglu S, Aphinyanaphongs Y, Jones S, Iturrate E, Hochman J, Berger JS. Thrombosis in hospitalized patients with COVID-19 in a New York City Health System. JAMA 2020;324:799-801. Crossref

31. Klok FA, Kruip MJ, van der Meer NJ, et al. Incidence of thrombotic complications in critically ill ICU patients with COVID-19. Thromb Res 2020;191:153-5. Crossref

32. Lippi G, Plebani M, Henry BM. Thrombocytopenia is associated with severe coronavirus disease 2019 (COVID-19) infections: a meta-analysis. Clin Chim Acta 2020;506:145-8. Crossref

33. Argulian E, Sud K, Vogel B, et al. Right ventricular dilation in hospitalized patients with COVID-19 infection. JACC Cardiovasc Imaging 2020;13:2459-61. Crossref

34. Li Y, Li H, Zhu S, et al. Prognostic value of right ventricular longitudinal strain in patients with COVID-19. JACC Cardiovasc Imaging 2020;13:2287-99. Crossref

35. Creel-Bulos C, Hockstein M, Amin N, Melhem S, Truong A, Sharifpour M. Acute cor pulmonale in critically Ill patients with Covid-19. N Engl J Med 2020;382:e70. Crossref

36. Wang D, Hu B, Hu C, et al. Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus–infected pneumonia in Wuhan, China. JAMA 2020;323:1061-9. Crossref

37. Russo M, Di Maio M, Attena E, et al. Clinical impact of pre-admission antithrombotic therapy in hospitalized patients with COVID-19: a multicenter observational study. Pharmacol Res 2020;159:104965. Crossref

38. Colon CM, Barrios JG, Chiles JW, et al. Atrial arrhythmias in COVID-19 patients. JACC Clin Electrophysiol 2020;6:1189-90. Crossref

39. Bhatla A, Mayer MM, Adusumalli S, et al. COVID-19 and cardiac arrhythmias. Heart Rhythm 2020;17:1439-44. Crossref

40. Tomasoni D, Italia L, Adamo M, et al. COVID-19 and heart failure: from infection to inflammation and angiotensin II stimulation. Searching for evidence from a new disease. Eur J Heart Fail 2020;22:957-66. Crossref

41. Du RH, Liang LR, Yang CQ, et al. Predictors of mortality for patients with COVID-19 pneumonia caused by SARS-CoV-2: a prospective cohort study. Eur Respir J 2020;55:2000524. Crossref

42. Chen T, Wu D, Chen H, et al. Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study. BMJ 2020;368:m1091. Crossref

43. Bhatt AS, Jering KS, Vaduganathan M, et al. Clinical outcomes in patients with heart failure hospitalized with COVID-19. JACC Heart Fail 2021;9:65-73. Crossref

44. Chan PS, Girotra S, Tang Y, et al. Outcomes for out-of-hospital cardiac arrest in the United States during the coronavirus disease 2019 pandemic. JAMA Cardiol 2021;6:296-303. Crossref

45. Shao F, Xu S, Ma X, et al. In-hospital cardiac arrest outcomes among patients with COVID-19 pneumonia in Wuhan, China. Resuscitation 2020;151:18-23. Crossref

46. Thapa SB, Kakar TS, Mayer C, Khanal D. Clinical outcomes of in-hospital cardiac arrest in COVID-19. JAMA Intern Med 2021;181:279-81. Crossref

47. Lala A, Johnson KW, Januzzi JL, et al. Prevalence and impact of myocardial injury in patients hospitalized with COVID-19 infection. J Am Coll Cardiol 2020;76:533-46. Crossref

48. Lippi G, Lavie CJ, Sanchis-Gomar F. Cardiac troponin I in patients with coronavirus disease 2019 (COVID-19): evidence from a meta-analysis. Prog Cardiovasc Dis 2020;63:390-1. Crossref

49. Giustino G, Croft LB, Stefanini GG, et al. Characterization of myocardial injury in patients with COVID-19. J Am Coll Cardiol 2020;76:2043-55. Crossref

50. Zhou F, Yu T, Du R, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet 2020;395:1054-62. Crossref

51. Zhang L, Yan X, Fan Q, et al. D-dimer levels on admission to predict in-hospital mortality in patients with Covid-19. J Thromb Haemost 2020;18:1324-9. Crossref

52. Spyropoulos AC, Levy JH, Ageno W, et al. Scientific and Standardization Committee communication: clinical guidance on the diagnosis, prevention and treatment of venous thromboembolism in hospitalized patients with COVID-19. J Thromb Haemost 2020;18:1859-65. Crossref

53. Thachil J, Tang N, Gando S, et al. ISTH interim guidance on recognition and management of coagulopathy in COVID- 19. J Thromb Haemost 2020;18:1023-6. Crossref

54. American Society of Hematology. COVID-19 and VTE/ anticoagulation: frequently asked questions. Available from: https://www.hematology.org/covid-19/covid-19-and-vte-anticoagulation. Accessed 23 Jun 2020.

55. Bikdeli B, Madhavan MV, Jimenez D, et al. COVID-19 and thrombotic or thromboembolic disease: implications for prevention, antithrombotic therapy, and follow-up: JACC state-of-the-art review. J Am Coll Cardiol 2020;75:2950-73. Crossref

56. Qin JJ, Cheng X, Zhou F, et al. Redefining cardiac biomarkers in predicting mortality of inpatients with COVID-19. Hypertension 2020;76:1104-12. Crossref

57. Mahmud E, Dauerman HL, Welt FG, et al. Management of acute myocardial infarction during the COVID-19 pandemic: a position statement from the Society for Cardiovascular Angiography and Interventions (SCAI), the American College of Cardiology (ACC), and the American College of Emergency Physicians (ACEP). J Am Coll Cardiol 2020;76:1375-84. Crossref

58. Li YH, Wang MT, Huang WC, Hwang JJ. Management of acute coronary syndrome in patients with suspected or confirmed coronavirus disease 2019: Consensus from Taiwan Society of Cardiology. J Formos Med Assoc 2021;120:78-82. Crossref

59. Tam CC, Cheung KS, Lam S, et al. Impact of coronavirus disease 2019 (COVID-19) outbreak on ST-segment–elevation myocardial infarction care in Hong Kong, China. Circ Cardiovasc Qual Outcomes 2020;13:e006631. Crossref

60. Mafham MM, Spata E, Goldacre R, et al. COVID-19 pandemic and admission rates for and management of acute coronary syndromes in England. Lancet 2020;396:381-9. Crossref

61. Solomon MD, McNulty EJ, Rana JS, et al. The covid-19 pandemic and the incidence of acute myocardial infarction. N Engl J Med 2020;383:691-3. Crossref

62. Zhu ZW, Tang JJ, Chai XP, et al. Comparison of heart failure and COVID-19 in chest CT features and clinical characteristics. Zhonghua Xin Xue Guan Bing Za Zhi 2020;48:467-71.

63. Moores LK, Tritschler T, Brosnahan S, et al. Prevention, diagnosis, and treatment of VTE in patients with coronavirus disease 2019 CHEST guideline and expert panel report. Chest 2020;158:1143-63. Crossref

64. American Society of Hematology. ASH Guidelines on Use of Anticoagulation in Patients with COVID-19. Available from: https://www.hematology.org/education/clinicians/guidelines-and-quality-care/clinical-practice-guidelines/venous-thromboembolism-guidelines/ash-guidelines-on-use-of-anticoagulation-in-patients-with-covid-19. Accessed 7 Dec 2020.

65. Thachil J, Juffermans NP, Ranucci M, et al. ISTH DIC subcommittee communication on anticoagulation in COVID-19. J Thromb Haemost 2020;18:2138-44. Crossref

66. National Institutes of Health. Antithrombotic therapy in patients with COVID-19. Last updated: 12 May 2020. Available from: https://www.covid19treatmentguidelines. nih.gov/adjunctive-therapy/antithrombotic-therapy/. Accessed 7 Dec 2020.

67. Phelan D, Kim JH, Chung EH. A game plan for the resumption of sport and exercise after coronavirus disease 2019 (COVID-19) infection. JAMA Cardiol 2020;5:1085-6. Crossref

The coronavirus disease 2019 (COVID-19) pandemic has spread to 213 countries and territories, posing a severe threat to public healthcare systems worldwide. As of 29 May 2020, the total number of confirmed cases has surged to 5 701 337, with 357 688 deaths recorded worldwide.1 Although multiple pharmacological agents are being evaluated, no beneficial, targeted drug or vaccine has been discovered to date, and the number of cases is rising daily. The causative agent of COVID-19 is severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which is believed to be transmitted through respiratory droplets and person-to-person contact. However, evidence of viral RNA in the faeces of COVID-19 patients also suggests the possibility of faecal-oral transmission.2 3 The disease typically presents with viral pneumonia-like symptoms of fever, dry cough, shortness of breath, and fatigue. Nevertheless, gastrointestinal symptoms like diarrhoea, vomiting, and abdominal pain have also been reported.4

Although coronavirus mainly targets the respiratory system, it also exhibits many extrapulmonary manifestations. Sepsis, acute cardiac injury, multiple organ failure, and alkalosis are some of the critical complications that have been observed in patients who die of COVID-19.5 Several studies have acknowledged the presence of liver injury in patients with COVID-19, mainly indicated by abnormal liver function tests (LFTs).6 7 The exact pathophysiology behind the LFT derangements is unknown. It has been suggested that SARS-CoV-2 causes liver injury either via direct viral insult or through an inflammatory cytokine storm.8 Other potential mechanisms, such as drug-induced hepatotoxicity and hypoxic injury, have also been implicated.

Given the rampant nature of SARS-CoV-2 and its repercussions on human health, the research community has responded expeditiously to the new virus, and studies regarding its systemic involvement are continuously being published. We have conducted a scoping review to summarise all articles published regarding hepatic damage in this setting. In this review, we aim to provide evidence of the incidence, patterns, risk factors, and histopathological findings of liver injury in COVID-19 and its association with the severity of disease. Furthermore, we highlight hepatotoxicity in patients with COVID-19 who are treated with antiviral (lopinavir/ritonavir) or antimalarial (hydroxychloroquine) drugs. We identify the existing gaps in current knowledge regarding the topic and provide recommendations for further research. This will help healthcare providers to identify hepatic complications during the pandemic.

A scoping review was conducted following the methodological framework of Arksey and O’Malley9 by taking the following steps: (a) identification of a definite research objective and search strategy; (b) identification and screening of research articles; (c) selection of research articles according to pre-defined eligibility criteria; (d) extraction and charting of data, and (e) reporting, summarising, and discussing the results.

The reviewed literature was identified by searching five online databases (PubMed, Google Scholar, Scopus, Wiley, and ScienceDirect) without any language restriction from 1 January 2020 to 22 May 2020. Grey literature was also searched in medRxiv and bioRxiv. Moreover, the reference lists of all identified articles were searched for additional sources. A variety of keywords were employed, according to the following search string: “liver injury” OR “hepatic damage” OR “liver functional abnormality” OR “cirrhosis” OR “decompensated liver disease” OR “acute liver failure” OR “chronic liver failure” OR “acute on chronic liver failure” AND “COVID-19” OR “SARS-CoV-2” OR “coronavirus disease”. The full electronic search strategy is provided in the online supplementary Appendix.

We aimed to summarise all of the scientific literature demonstrating liver dysfunction in COVID-19 and to identify the gaps in knowledge regarding hepatic damage in SARS-CoV-2 infection for further research. Three researchers (TBA, SK, and MSS) independently searched through the literature, and all sets of literature were then compared. Disagreements on the inclusion or exclusion of literature were resolved through discussion or, if necessary, by including a fourth researcher (HH) to make the final decision. Articles were screened according to pre-defined eligibility criteria. The inclusion criteria were as follows: (1) study design: retrospective observational study, retrospective cohort study, retrospective descriptive study, prospective observational study, prospective case-cohort, cross-sectional, case report, case series, or meta-analysis; (2) language: studies published in English only; (3) publication status: preprints and published articles; (4) dates considered: studies published from 1 January 2020 to 22 May 2020; and (5) all relevant papers describing functional abnormalities of the liver in COVID-19. The exclusion criteria were as follows: (1) language: articles published in any language other than English; (2) study design: review article, editorial, letter to the editor, correspondence, or commentary; and (3) studies conducted on patients who had undergone organ transplants. Duplicate articles were excluded. Ultimately, 62 articles were included in this review conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for scoping reviews (Fig 1).10

After article selection, data were extracted and recorded on a pre-designed datasheet. The extracted data included the article’s title, study design, study setting, study population, sample size, research domain, and key conclusions.

The articles that assessed liver injury in patients with COVID-19 belong to two categories: (a) studies that employed pre-defined clinical criteria for liver injury in COVID-19, and (b) studies that did not employ any pre-defined criteria and only reported LFT derangements in COVID-19. Based on the primary research objectives, each article was classified into one of the following main research domains: incidence of liver injury, patterns of liver injury, and risk factors for liver injury in COVID-19, associations of liver injury or underlying liver disease (eg, chronic liver disease) with the severity of COVID-19, drug-induced liver injury in COVID-19, and histopathological findings of liver injury in COVID-19. The methodological characteristics (study design, study setting, type of population, and sample size) of all studies were also analysed.

A total of 62 articles were included in this scoping review, among which 10 were preprints, and 52 were published in peer-reviewed journals, including The Lancet and Journal of the American Medical Association. About 23 studies (16 retrospective observational, 2 retrospective cohort, 1 retrospective descriptive, 1 prospective observational, 1 prospective case-cohort, 1 cross-sectional, and 1 meta-analysis) documented the incidence of liver injury in COVID-19.11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Around half of the eligible studies (n=29, 46.8%) showed an association between the severity of COVID-19 and the degree of liver injury.11 12 13 14 15 16 17 19 20 21 22 23 28 29 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Out of the 27 studies assessing liver injury in COVID-19, 44.4% (n=12) had pre-defined clinical criteria for liver injury, whereas 55.6% (n=15) did not have any specific pre-defined criteria. The details of these studies are given below. The studies predominantly depicted significant elevation of aspartate aminotransferase (AST) than of alanine aminotransferase (ALT) in case of liver injury, which was found to be proportional to the severity of COVID-19.

Fewer studies (n=6) mentioned any histopathological findings of liver injury in patients with COVID-19, but the most common findings mentioned were mild sinusoidal dilatation, microvesicular steatosis, and minimal lymphocytic infiltration.36 49 50 51 52 53 Eight studies assessed the impact of drugs on potential liver damage. Half of those studies (n=4, 50%) concluded that the use of lopinavir/ritonavir increases the odds of liver injury. Other drugs described as having the potential to cause hepatotoxicity in COVID-19 included hydroxychloroquine (n=1, 12.5%), tocilizumab (n=2, 25%), and remdesivir (n=1, 12.5%).12 18 54 55 56 57 58 59

The methodological characteristics of the finalised studies were also analysed. The largest number of the studies were retrospective observational studies (n=26, 41.9%), followed by meta-analyses (n=10, 16.1%), case reports (n=9, 14.5%), case series (n=7, 11.3%), prospective observational studies (n=3, 4.8%), and others (Table 1). All studies except for meta-analyses, case reports, and series included a targeted population. Among the 36 articles with a targeted population, more than three-quarters (n=28, 77.8%) were conducted only on living patients with COVID-19, whereas the remainder (n=8, 22.2%) included patients who died. Of the finalised studies, 38.9% (n=14) had sample sizes of 5 to 60. The included studies’ methodological characteristics are given in Table 1.

Several observational studies documenting the clinical characteristics of patients with COVID-19 have reported liver injury.11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 They have mentioned liver enzyme elevation without commenting on the clinical signs of hepatic dysfunction, which include hepatomegaly, ascites, and jaundice.11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 32 The incidence of liver injury has varied widely across studies, from 4.8% to a striking 78%.11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 However, the term ‘liver injury’ has not been defined uniformly. The definitions used in various studies have ranged from slight transaminasaemia to enzyme elevation more than 3 times higher than the upper limit of normal (Table 2 11 12 16 17 18 20 22 23 24 31 32 33). Additionally, several studies did not establish any clinical or laboratory criteria to define liver injury (Table 3 13 14 15 19 21 25 26 27 28 29 34 35 37 46 60). Many studies failed to mention the date on which LFTs were performed, thus creating non-uniformity in their reported values. Case reports have identified the presence of liver injury across the entire age spectrum, ranging from 55 days to 65 years.50 57 58 61 62

The pathogenesis of liver involvement in COVID-19 infection is assumed to be multifactorial. However, none of the available hypotheses provide a complete explanation, and further investigation is required not only to understand the mechanism but also to formulate appropriate management plans. Figure 2 illustrates the possible mechanisms of hepatic dysfunction in COVID-19.

The proposed receptor for the virus, angiotensinconverting enzyme 2 receptor (ACE2R), has been found only sparsely in hepatocytes. Chai et al63 demonstrated ACE2 expression in 2.6% of hepatocytes, whereas up to 59.7% of cholangiocytes expressed ACE2R. Seow et al64 also revealed the presence of ACE2 in liver progenitor cells, especially those destined to become cholangiocytes. These findings imply direct invasion of cholangiocytes and progenitor cells, thus resulting in necrosis and impaired regeneration of cholangiocytes. The tight junctions between cholangiocytes also seem to be altered during COVID-19 infection, which may be responsible for the observed cholestasis in patients.65 Significant necrosis of and rapid viral replication within cholangiocytes has also been observed by Zhao et al65 in a human liver ductal organoid model.

However, Zhou et al66 argued against this proposed mechanism by highlighting that ACE2Rs on cholangiocytes are confined to the apical surface, from where viral invasion is unlikely. Furthermore, the hepatic pattern of LFT elevation fails to explain this possible ductal pathology. Hepatocytes also express the protein furin, which may play a role in liver damage upon entry of the virus into the cells.

Decreased oxygen saturation has been a feature of COVID-19 pneumonia, and this may result in hypoxic injury to multiple organs, including the liver.67

High values of positive end expiratory pressure used during mechanical ventilation in severe patients may result in hepatic congestion by increasing the pressure on the right atrium and thereby impeding venous return. However, the presence of comparable liver functional abnormalities in patients without ventilation renders that assumption inconclusive.68

An inflammatory response to the virus may lead to persistent leukocytic activation and the release of many mediators responsible for cellular injury. The involvement of a cytokine storm in liver damage has been supported by patients’ elevated levels of interleukins 2, 6, and 10, interferon-gamma, serum ferritin, and C-reactive protein.17

Liver dysfunction may occur secondary to vascular pathology, resulting in endotheliitis, coagulopathy, thrombus formation, and ischaemic parenchymal necrosis. Angiotensin-converting enzyme 2 receptors are expressed on the endothelium, making it susceptible to viral invasion, which leads to the recruitment of inflammatory cells and elaboration of inflammatory cytokines. The immune response may also exacerbate the damage.69

A number of drugs currently in use have hepatotoxic potential, which might be further exaggerated in the setting of chronic liver disease (CLD).12 56 58 The mechanisms for individual drugs are not clearly defined.

The patterns of liver injury in COVID-19 patients include both reversible dysfunction and irreversible injury as a component of multiorgan failure in terminally ill patients.40 57 58 59 61 62 However, hepatic dysfunction in COVID-19 cases is usually mild, and deranged LFTs tend to recover within a few days after discharge.48 Predominant elevation of ALT and AST indicates hepatocellular injury.12 The abnormalities in AST have been more severe compared with those of ALT.20 22 23 34 37 44 45 This finding is intriguing, as ALT, being more liver-specific, is the enzyme that is generally expected to be significantly elevated in case of hepatocellular injury. However, a few studies have hypothesised that AST elevation could be secondary to viral-mediated direct liver damage.20 21 The mechanism behind predominant AST elevation in the presence of viral aetiology remains unclear.

Elevation of the ductal enzymes gamma-glutamyl transferase and alkaline phosphatase (ALP) has been reported in some studies.22 23 70 Elevated ALP was also reported alongside elevated AST and ALT in a case of acute hepatitis following COVID-19 infection.61 Cardoso et al71 studied the temporal patterns of liver enzyme levels in critically ill patients and observed that a cholestatic pattern emerged later in the course of illness. Most studies have not mentioned the presence of any liver enzyme abnormalities at the time of liver injury. Hence, the extent of liver damage and the pattern of injury could not be accurately assessed.

Histopathological findings of autopsied liver samples have provided evidence of direct viral invasion and changes secondary to hypoxia, sepsis, and pro-inflammatory and pro-coagulant states. Wang et al11 revealed the presence of hepatic apoptosis, occasional bi- or multi-nucleated hepatocytes, mitochondrial swelling, and decreased glycogen granules. These findings strongly suggest direct cytopathic effects of COVID-19 on the liver. Electron microscopy also showed the presence of viral particles. Other non-specific findings have included varying degrees of steatosis,11 12 49 50 51 52 mild portal lymphocytic infiltration,11 36 50 51 mild sinusoidal dilation,36,51,53 and inflamed cells within the sinusoids.12 However, samples obtained via needle biopsy did not facilitate effective determination of the histology of the ductal epithelium, which carries a higher density of ACE2Rs.51 Ductal pathology was highlighted by Lax et al,52 indicating the presence of canalicular cholestasis and mild nuclear pleomorphism of cholangiocytes. Patterns of both massive and focal patchy necrosis were reported in the periportal and centrilobular areas.51 52 The authors suggested that sepsis and systemic inflammation might be responsible for acute hepatic necrosis. Furthermore, reverse transcription-polymerase chain reaction of one liver sample was positive for COVID-19.51 In that case, an ultrasound-guided autopsy observed centrilobular congestion (which was likely attributable to shock), ischaemic necrosis, portal tract inflammation, and Kupffer cell activation.51

The watery degeneration of some hepatocytes observed by Cai et al12 was likely due to ischaemia and hypoxia. The presence of thrombi within the liver, among other organs, also demonstrates the possibility of COVID-associated coagulopathy.52 Liver involvement with COVID-19 infection may further elaborate the inflammatory cascade and alter the secretion of coagulation factors, thus playing a role in causing widespread thrombosis.52 Endotheliitis, acute and chronic vascular changes, and sinusoidal arterialisation due to pressure elevation observed in the liver further support the involvement of underlying endothelial pathology in causing coagulative derangements.53 69

Studies have consistently shown liver injury to be associated with severe COVID-19.11 12 13 14 15 16 34 42 43 44 Deranged LFTs have also been linked to prolonged hospital stays18 and worse clinical outcomes.19 38 40 45 Disease severity is most likely linked with the elevation of AST rather than ALT.20 21 45 Additionally, hypoproteinaemia and cholestasis in early-stage disease have been shown to increase the risk of death.15 However, the impact of AST on mortality has been controversial.22 46

These findings cannot reliably establish that elevated LFT levels were solely caused by COVID-19 infection, as many studies did not exclude patients with CLD, nor did they consider other possible reasons for liver enzyme elevation. Furthermore, there is still not enough evidence to suggest that mild derangement has a high likelihood of progressing into fulminant liver failure. Yet, patients with deranged LFT patterns of the hepatocellular or mixed types at the time of admission or during hospitalisation were more likely to progress to severe disease,12 47 thus necessitating adequate monitoring.38 44 45 48 Additionally, there is no evidence that liver dysfunction can directly cause mortality in patients with COVID-19.

Studies have reported associations between multiple risk factors and liver injury in the setting of SARS-CoV-2 infection:

1. Abnormal white blood cell parameters, including elevated neutrophils and decreased lymphocytes, have been associated with elevated risk of liver injury.14 15 17 22 The loss of lymphocytes responsible for suppression of the immune response during viral infection may have contributed to the damage.14 Similarly, high levels of C-reactive protein and procalcitonin were associated with increased risk of liver damage.14 17 18 The cytokine storm and systemic inflammation might be implicated, as they result in leukocyte activation and the release of a large quantity of inflammatory mediators that directly or indirectly damage cells.14
2. The use of hepatotoxic drugs, including antivirals, hydroxychloroquine, tocilizumab (discussed below),12 16 18 58 and antifungals for superimposed infections has been established as a risk factor for liver dysfunction.22 Systemic corticosteroids were also associated with an increased risk of AST elevation,22 perhaps due to drug-induced lymphopoenia and alteration of the immune response.
3. A correlation between the severity of lung involvement and the incidence of liver injury has also been noticed.17 As severe lung lesions indicate a robust inflammatory state, the liver might be affected for the same reason (ie, a hyperinflammatory state).17 That study did not indicate the role of hypoxia, which is also a possible contributing factor to hepatic damage.
4. Non-modifiable risk factors such as male sex (odds ratio=1.60; P<0.001) and old age (odds ratio=1.01; P=0.031) have been linked to a higher risk of liver damage.22 23
5. Patients with elevated ALT levels were more likely to have a history of drinking (P=0.032).14 However, that study did not comment on elevation of AST, the dominant enzyme involved in both alcoholic liver disease and COVID-19. Furthermore, that study’s very small sample size necessitates further investigation of this risk factor.
6. Patients with gastrointestinal symptoms were more likely to have liver injury than those without such symptoms (P=0.035).24
7. Diabetes mellitus was a risk factor for cholestasis in patients with COVID-19 (P=0.044),15 which is predicted to be a possible mechanism of liver injury in the setting of viral infection.65
8. Invasive mechanical ventilation increased the risk of LFT elevation42 and acute liver injury.33 Such injury may be caused by hepatic congestion that results from elevated right atrial pressure, which in turn is caused by high levels of positive end expiratory pressure.68

The drugs currently used to manage COVID-19 infection also carry hepatotoxic potential. Muhović et al56 reported a 40-fold rise in transaminases following two doses of tocilizumab, an interleukin-6 receptor antagonist, which regressed 10 days later. Morena et al59 also reported elevated liver enzymes in 29% of patients who were receiving tocilizumab. In addition, Falcão et al58 reported a 10-fold elevation in transaminases following two doses of hydroxychloroquine. Upon withdrawal, the enzyme levels dropped to near normal after 5 days.

Antivirals have also been demonstrated to cause liver toxicity. In one study, people receiving lopinavir/ritonavir had a higher incidence of liver dysfunction compared with those in whom these drugs were not administered (51.8% vs 31.3%, respectively).18 Similarly, Young et al55 reported abnormal LFTs in three out of five patients receiving lopinavir/ritonavir. According to Cai et al,12 the use of lopinavir/ritonavir increased the likelihood of liver injury 4-fold. Durante-Mangoni et al54 reported that remdesivir caused elevation of liver enzymes in three out of four patients, and Weber et al57 suggested that drugs may play a role in precipitating acute liver failure. Administration of lopinavir/ritonavir and interferon was followed by progressive worsening of LFTs. This effect may have been attenuated by the use of Ramipril for arterial hypertension.57 Additionally, Lei et al22 showed that elevated AST and ALP levels were associated with the use of antifungal medications. The above findings are from case reports, retrospective studies, and very small-scale prospective studies. Further large-scale prospective studies, including randomised controlled trials, need to be conducted to establish their efficacy and safety in patients with COVID-19.

The effects of underlying liver disease on the severity of COVID-19 are controversial. Zhou et al35 suggested the presence of CLD as a risk factor for severe COVID-19. However, that study included only eight known cases of CLD. Similarly, Qi et al60 indicated that decompensated liver cirrhosis might be a risk factor for poor outcomes of COVID-19. In contrast, a meta-analysis by Wang et al72 that included five studies concluded that prior liver disease does not impact the severity of COVID-19. Likewise, the presence of pre-existing cirrhosis had no direct prognostic association in the setting of COVID-19.73 Some studies in our review included patients with CLD, which may account for some of the LFT derangements observed in patients. A meta-analysis by Mantovani et al41 estimated that the baseline prevalence of CLD was 3%. This figure is much lower than the proportion of people with liver dysfunction. Hence, the role of CLD in worsening the prognosis of COVID-19 infection seems to be minor, if there is any.

Nevertheless, acute-on-chronic liver failure (ACLF) following COVID-19 infection has been reported. One case was a female patient with decompensated alcoholic cirrhosis (ACLF Grade 2) who developed a mixed hepatic and cholestatic pattern of liver dysfunction following COVID-19 infection. However, her prognosis was good.74 Another patient was an older man with ACLF Grade 1 non-alcoholic cirrhosis. He developed hepatorenal syndrome-type acute kidney injury following COVID-19 infection. His liver failure subsequently progressed to Grade 2 after catheter-associated urinary tract infections and complicated paracentesis.75

Cui et al61 revealed that anal swabs of an infant with liver injury remained positive for COVID-19 even after throat swabs returned to a negative state. However, polymerase chain reaction of stool samples was not performed in most of the articles included in our review. The association of liver dysfunction with the risk of oro-faecal transmission remains to be investigated. If transmission via this route is possible, existing isolation and discharge protocols may need to be revised.

Our review is subject to certain limitations. First, the majority of the included studies did not have any specific pre-defined clinical criteria for diagnosing liver injury in COVID-19. Moreover, the included studies did not distinguish between a history of liver disease (eg, CLD) and liver injury secondary to COVID-19. Hence, our results need to be interpreted cautiously, as they do not accurately describe the level of incidence of liver injury that is caused by COVID-19. Second, we did not include studies published in any language other than English, which might have provided additional insight. Third, the inclusion of a large number of studies prevented us from critically appraising the individual studies’ sources of evidence. There is a need for a comprehensive systematic review or meta-analysis to summarise the statistics and provide a clearer picture of liver injury in SARS-CoV-2 infection. Transplant recipients, a group that is vulnerable to liver injury, were also not reviewed.

Many studies have defined ‘liver injury’ as non-specific elevation of LFTs above the upper limit of normal. Further, many of the investigated studies did not assess the bilirubin levels or coagulation profiles of patients with COVID-19, both of which are important indicators of liver function. Moreover, there have been no reports of liver failure or hepatic cell death secondary to COVID-19 to date. Hence, we recommend the use of scientifically relevant terms ‘liver dysfunction’ or ‘liver enzyme derangement’ to explain non-specific LFT abnormalities until an appropriate definition for liver injury is devised. Furthermore, we propose that pre-defined criteria for liver injury should be set and that mild, non-specific derangements of liver function should not be labelled as liver injury. Given the existing controversy in the literature, we recommend a thorough investigation into the pathogenesis of liver injury, especially the mode of direct viral invasion.

Additional studies are required to investigate whether mild derangement of liver function can cause hepatic failure in COVID-19. The reason for the hepatocellular pattern with predominant AST elevation also needs to be elucidated. Finally, the safety and efficacy of hepatotoxic drugs in COVID-19 should also be established via randomised controlled trials.

Liver injury is a common extrapulmonary feature of COVID-19. However, the absence of standardised clinical criteria for liver injury in this setting needs to be addressed. Derangements of LFT levels are markers of the severity of COVID-19 infection, but the association between LFT derangements and disease progression requires further investigation because to date, liver dysfunction has not been shown to directly cause mortality in patients with COVID-19. The pattern of injury is predominantly hepatocellular, accompanied by greater elevation of AST than of ALT. Possible pathogenetic mechanisms include direct viral invasion, hypoxia, systemic inflammation, endothelial dysfunction, and the use of mechanical ventilation. Histopathological findings in the liver support viral-induced pathology in addition to non-specific changes. Nevertheless, these studies are sparse, and more research is required. Potentially hepatotoxic drugs have also been observed to cause liver injury in patients with COVID-19, and thus, the administration of these drugs necessitates careful monitoring. Large-scale studies are needed to establish their role in the management of COVID-19.

Concept or design: T Bin Arif.
Acquisition of data: T Bin Arif, S Khalid, MS Siddiqui, H Hussain.
Analysis or interpretation of data: H Sohail.
Drafting of the manuscript: S Khalid, MS Siddiqui, H Hussain.
\ Critical revision of the manuscript for important intellectual content: T Bin Arif, H Sohail.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1. World Health Organization. Coronavirus disease (COVID-2019) situation reports–124. Available from: https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200523-covid-19-sitrep-124.pdf?sfvrsn=9626d639_2. Accessed 29 May 2020.

2. Wang W, Xu Y, Gao R, et al. Detection of SARS-CoV-2 in different types of clinical specimens. JAMA 2020;323:1843- 4. Crossref

3. Chen J, Zhu H, Wang D, et al. Clinical features of stool SARS-CoV-2 RNA positive in 137 COVID-19 patients in Taizhou, China. SSRN Electronic J 2020 Mar 23. Available from: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3551383. Accessed 28 May 2020. Crossref

4. Pan L, Mu M, Yang P, et al. Clinical characteristics of COVID-19 patients with digestive symptoms in Hubei, China: a descriptive, cross-sectional, multicenter study. Am J Gastroenterol 2020;115:766-73. Crossref

5. Chen T, Wu D, Chen H, et al. Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study. BMJ 2020;368:m1091. Crossref

6. Chen N, Zhou M, Dong X, et al. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study. Lancet 2020;395:507-13. Crossref

7. Zhang C, Shi L, Wang FS. Liver injury in COVID-19: management and challenges. Lancet Gastroenterol Hepatol 2020;5:428-30. Crossref

8. Mehta P, McAuley DF, Brown M, et al. COVID-19: consider cytokine storm syndromes and immunosuppression. Lancet 2020;395:1033-4. Crossref

9. Arksey H, O’Malley L. Scoping studies: towards a methodological framework. Int J Soc Res Methodol 2005;8:19-32. Crossref

10. Tricco AC, Lillie E, Zarin W, et al. PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation. Ann Inter Med 2018;169:467-73. Crossref

11. Wang Y, Liu S, Liu H, et al. SARS-CoV-2 infection of the liver directly contributes to hepatic impairment in patients with COVID-19. J Hepatol 2020;73:807-16. Crossref

12. Cai Q, Huang D, Yu H, et al. COVID-19: Abnormal liver function tests. J Hepatol 2020;73:566-74. Crossref

13. Ding Q, Lu P, Fan Y, Xia Y, Liu M. The clinical characteristics of pneumonia patients coinfected with 2019 novel coronavirus and influenza virus in Wuhan, China. J Med Virol 2020 Mar 20. Epub ahead of print. Crossref

14. Li L, Li S, Xu M, et al. Risk factors related to hepatic injury in patients with corona virus disease 2019. medRxiv. 2020 Mar 10. Available from: https://www.medrxiv.org/content/10.1101/2020.02.28.20028514v2. Accessed 27 May 2020. Crossref

15. Fu L, Fei J, Xu S, et al. Acute liver injury and its association with death risk of patients with COVID-19: a hospital-based prospective case-cohort study. medRxiv. 2020 Apr 6. Available from: https://www.medrxiv.org/content/10.1101/ 2020.04.02.20050997v1. Accessed 27 May 2020. Crossref

16. Cai Q, Huang D, Ou P, et al. COVID-19 in a designated infectious diseases hospital outside Hubei Province, China. Allergy 2020;75:1742-52. Crossref

17. Xie H, Zhao J, Lian N, Lin S, Xie Q, Zhuo H. Clinical characteristics of non-ICU hospitalized patients with coronavirus disease 2019 and liver injury: a retrospective study. Liver Int 2020;40:1321-6. Crossref

18. Fan Z, Chen L, Li J, et al. Clinical features of COVID-19-related liver functional abnormality. Clin Gastroenterol Hepatol 2020;18:1561-6. Crossref

19. Yang F, Shi S, Zhu J, Shi J, Dai K, Chen X. Analysis of 92 deceased patients with COVID-19. J Med Virol 2020 Apr 15. Epub ahead of print. Crossref

20. Bloom PP, Meyerowitz EA, Reinus Z, et al. Liver biochemistries in hospitalized patients with COVID-19. Hepatology 2020 May 16. Epub ahead of print. Crossref

21. Cholankeril G, Podboy A, Aivaliotis VI, et al. High prevalence of concurrent gastrointestinal manifestations in patients with SARS-CoV-2: early experience from California. Gastroenterology 2020;159:775-7. Crossref

22. Lei F, Liu YM, Zhou F, et al. Longitudinal association between markers of liver injury and mortality in COVID19 in China. Hepatology 2020;72:389-98. Crossref

23. Hajifathalian K, Krisko T, Mehta A, et al. Gastrointestinal and hepatic manifestations of 2019 novel coronavirus disease in a large cohort of infected patients from New York: clinical implications. Gastroenterology 2020;159:1137-40. e2. Crossref

24. Jin X, Lian JS, Hu JH, et al. Epidemiological, clinical and virological characteristics of 74 cases of coronavirus-infected disease 2019 (COVID-19) with gastrointestinal symptoms. Gut 2020;69:1002-9. Crossref

25. Du Y, Tu L, Zhu P, et al. Clinical features of 85 fatal cases of COVID-19 from Wuhan: a retrospective observational study. Am J Respir Crit Care Med 2020;201:1372-9. Crossref

26. Li X, Wang L, Yan S, et al. Clinical characteristics of 25 death cases with COVID-19: a retrospective review of medical records in a single medical center, Wuhan, China. Int J Infect Dis 2020;94:128-32. Crossref

27. Lian J, Jin X, Hao S, et al. Epidemiological, clinical, and virological characteristics of 465 hospitalized cases of coronavirus disease 2019 (COVID-19) from Zhejiang province in China. Influenza Other Respir Viruses 2020;14:564-74. Crossref

28. Yang X, Yu Y, Xu J, et al. Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study. Lancet Respir Med 2020;8:475-81. Crossref

29. Zhang B, Zhou X, Qiu Y, et al. Clinical characteristics of 82 death cases with COVID-19. medRxiv. 2020 Feb 27. Available from: https://www.medrxiv.org/content/10.1101/2020.02.26.20028191v1. Accessed 28 May 2020. Crossref

30. Zhu J, Ji P, Pang J, et al. Clinical characteristics of 3062 COVID-19 patients: a meta-analysis. J Med Virol 2020 Apr 15. Epub ahead of print. Crossref

31. Chen G, Wu D, Guo W, et al. Clinical and immunological features of severe and moderate coronavirus disease 2019. J Clin Invest 2020;130:2620-9. Crossref

32. Qi X, Liu C, Jiang Z, et al. Multicenter analysis of clinical characteristics and outcome in patients with COVID-19 who develop liver injury. J Hepatol 2020;73:455-8. Crossref

33. Zheng Y, Sun L, Xu M, et al. Clinical characteristics of 34 COVID-19 patients admitted to ICU in Hangzhou, China. medRxiv. 2020 Apr 15. Available from: https://www.medrxiv.org/content/10.1101/2020.04.12.20062604v1.Accessed 28 May 2020. Crossref

34. Wan S, Xiang Y, Fang W, et al. Clinical features and treatment of COVID-19 patients in Northeast Chongqing. J Med Virol 2020;92:797-806. Crossref

35. Zhou Y, He Y, Yang H, et al. Development and validation a nomogram for predicting the risk of severe COVID-19: a multi-center study in Sichuan, China. PLoS One 2020;15:e0233328. Crossref

36. Zhang Y, Zheng L, Liu L, Zhao M, Xiao J, Zhao Q. Liver impairment in COVID-19 patients: a retrospective analysis of 115 cases from a single centre in Wuhan City, China. Liver Int 2020;40:2095-103. Crossref

37. Wang L, He W, Yu X, et al. Coronavirus disease 2019 in elderly patients: characteristics and prognostic factors based on 4-week follow-up. J Infect 2020;80:639-45. Crossref

38. Parohan M, Yaghoubi S, Seraj A. Liver injury is associated with severe coronavirus disease 2019 (COVID-19) infection: a systematic review and meta-analysis of retrospective studies. Hepatol Res 2020;50:924-35. Crossref

39. Mao R, Qiu Y, He JS, et al. Manifestations and prognosis of gastrointestinal and liver involvement in patients with COVID-19: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol 2020;5:667-78. Crossref

40. Henry BM, de Oliveira MH, Benoit S, Plebani M, Lippi G. Hematologic, biochemical and immune biomarker abnormalities associated with severe illness and mortality in coronavirus disease 2019 (COVID-19): a meta-analysis. Clin Chem Lab Med 2020;58:1021-8. Crossref

41. Mantovani A, Beatrice G, Dalbeni A. Coronavirus disease 2019 and prevalence of chronic liver disease: a meta-analysis. Liver Int 2020;40:1316-20. Crossref

42. Israelsen SB, Kristiansen KT, Hindsberger B, et al. Characteristics of patients with COVID-19 pneumonia at Hvidovre Hospital, March-April 2020. Dan Med J 2020;67:A05200313.

43. Zhao X, Lei Z. The impact of coronavirus disease 2019 (COVID-19) on liver injury in China: a systematic review and meta-analysis. medRxiv. 2020 May 8. Available from: https://www.medrxiv.org/content/10.1101/2020.05.03.20089557v1. Accessed 28 May 2020. Crossref

44. Dong X, Zeng DY, Cai YY, et al. Liver chemistries in patients with severe or non-severe COVID-19: a metaanalysis. medRxiv. 2020 Apr 29. Available from: https:// www.medrxiv.org/content/10.1101/2020.04.24.20074179v1. Accessed 29 May 2020. Crossref

45. Xing QQ, Dong X, Ren YD, et al. Liver chemistries in COVID-19 patients with survival or death: a metaanalysis. medRxiv. 2020 May 1. Available from: https://www.medrxiv.org/content/10.1101/2020.04.26.20080580v1. Accessed 28 May 2020. Crossref

46. Wu C, Chen X, Cai Y, et al. Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China. JAMA Intern Med 2020;180:1-11. Crossref

47. Xu L, Mao Y, Chen G. Risk factors for severe corona virus disease 2019 (COVID-19) patients: a systematic review and meta analysis. medRxiv. 2020 Apr 1. Available from: https://www.medrxiv.org/content/10.1101/2020.03.30.20047415v1. Accessed 28 May 2020. Crossref

48. Chen C, Jiang J, Xu X, Hu Y, Hu Y, Zhao Y. Dynamic liver function indexes monitoring and clinical characteristics in three types of COVID-19 patients. medRxiv. 2020 May 20. Available from: https://www.medrxiv.org/content/10.1101/2020.05.13.20099614v2. Accessed 28 May 2020. Crossref

49. Nunes Duarte-Neto A, de Almeida Monteiro RA, da Silva LF, et al. Pulmonary and systemic involvement of COVID-19 assessed by ultrasound-guided minimally invasive autopsy. Histopathology 2020 May 22. Epub ahead of print. Crossref

50. Xu Z, Shi L, Wang Y, et al. Pathological findings of COVID-19 associated with acute respiratory distress syndrome. Lancet Respir Med 2020;8:420-2. Crossref

51. Tian S, Xiong Y, Liu H, et al. Pathological study of the 2019 novel coronavirus disease (COVID-19) through postmortem core biopsies. Modern Pathol 2020;33:1007-14. Crossref

52. Lax SF, Skok K, Zechner P, et al. Pulmonary arterial thrombosis in COVID-19 with fatal outcome: results from a prospective, single-center, clinicopathologic case series. Ann Intern Med 2020 May 14. Epub ahead of print. Crossref

53. Sonzogni A. Liver histopathology in COVID 19 infection is suggestive of vascular alteration. 2020 May 11. Available from: https://www.medrxiv.org/content/10.1101/2020.05. 06.20092718v1. Accessed 28 May 2020. Crossref

54. Durante-Mangoni E, Andini R, Bertolino L, et al. Early experience with remdesivir in SARS-CoV-2 pneumonia. Infection 2020;48:779-82. Crossref

55. Young BE, Ong SW, Kalimuddin S, et al. Epidemiologic features and clinical course of patients infected with SARS-CoV- 2 in Singapore. JAMA 2020;323:1488-94. Crossref

56. Muhović D, Bojović J, Bulatović A, et al. First case of drug-induced liver injury associated with the use of tocilizumab in a patient with COVID-19. Liver Int 2020 May 17. Epub ahead of print. Crossref

57. Weber S, Mayerle J, Irlbeck M, Gerbes AL. Severe liver failure during SARS-CoV-2 infection. Gut 2020;69:1365-7. Crossref

58. Falcão MB, Pamplona de Góes Cavalcanti L, Filgueiras Filho NM, Antunes de Brito CA. Case report: hepatotoxicity associated with the use of hydroxychloroquine in a patient with COVID-19. Am J Trop Med Hyg 2020;102:1214-6. Crossref

59. Morena V, Milazzo L, Oreni L, et al. Off-label use of tocilizumab for the treatment of SARS-CoV-2 pneumonia in Milan, Italy. Eur J Intern Med 2020;76:36-42. Crossref

60. Qi X, Wang J, Li X, et al. Clinical course of COVID-19 in patients with pre-existing decompensated cirrhosis: initial report from China. Hepatol Int 2020;14:478-82. Crossref

61. Wander P, Epstein M, Bernstein D. COVID-19 presenting as acute hepatitis. Am J Gastroenterol 2020;115:941-2. Crossref

62. Cui Y, Tian M, Huang D, et al. A 55-day-old female infant infected with 2019 novel coronavirus disease: presenting with pneumonia, liver injury, and heart damage. J Infect Dis 2020;221:1775-81. Crossref

63. Chai X, Hu L, Zhang Y, et al. Specific ACE2 expression in cholangiocytes may cause liver damage after 2019-nCoV infection. bioRxiv. 2020 Feb 4. Available from: https://www.biorxiv.org/content/10.1101/2020.02.03.931766v1. Accessed 28 May 2020. Crossref

64. Seow JJ, Pai R, Mishra A, et al. scRNA-seq reveals ACE2 and TMPRSS2 expression in TROP2+ liver progenitor cells: implications in COVID-19 associated liver dysfunction. bioRxiv. 2020 Mar 25. Available from: https://www.biorxiv.org/content/10.1101/2020.03.23.002832v1. Accessed 28 May 2020.

65. Zhao B, Ni C, Gao R, et al. Recapitulation of SARS-CoV-2 infection and cholangiocyte damage with human liver ductal organoids. Protein Cell 2020;11:771-5. Crossref

66. Zhou L, Niu Z, Jiang X, et al. Systemic analysis of tissue cells potentially vulnerable to SARS-CoV-2 infection by the protein-proofed single-cell RNA profiling of ACE2, TMPRSS2 and Furin proteases. 2020 Apr 18. Available from: https://www.biorxiv.org/content/10.1101/2020.04.06.028522v2. Accessed 28 May 2020. Crossref

67. Waseem N, Chen PH. Hypoxic hepatitis: a review and clinical update. J Clin Transl Hepatol 2016;4:263-8.

68. Kukla M, Skonieczna-Żydecka K, Kotfis K, et al. COVID-19, MERS and SARS with concomitant liver injury—systematic review of the existing literature. J Clin Med 2020;9:1420. Crossref

69. Varga Z, Flammer AJ, Steiger P, et al. Endothelial cell infection and endotheliitis in COVID-19. Lancet 2020;395:1417-8. Crossref

70. Tian S, Zhu X, Sun X, et al. Longitudinal analysis of laboratory findings during the process of recovery for patients with COVID-19. medRxiv. 2020 Apr 7. Available from: https://www.medrxiv.org/content/10.1101/2020.04.04.20053280v1. Accessed 28 May 2020. Crossref

71. Cardoso FS, Pereira R, Germano N. Liver injury in critically ill patients with COVID-19: a case series. Crit Care 2020;24:190. Crossref

72. Wang B, Li R, Lu Z, Huang Y. Does comorbidity increase the risk of patients with COVID-19: evidence from meta-analysis. Aging (Albany NY) 2020;12:6049-57. Crossref

73. Qi X, Liu Y, Wang J, et al. Clinical course and risk factors for mortality of COVID-19 patients with pre-existing cirrhosis: a multicentre cohort study. Gut 2020 May 20. Epub ahead of print. Crossref

74. Qiu H, Wander P, Bernstein D, Satapathy SK. Acute on chronic liver failure from novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Liver Int 2020;40:1590-3. Crossref

75. Große K, Kramer M, Trautwein C, Bruns T. SARS-CoV-2 as an extrahepatic precipitator of acute-on-chronic liver failure. Liver Int 2020;40:1792-3. Crossref

Severe acute respiratory system coronavirus-2 (SARS-CoV-2) outbreak emerged as a series of idiopathic cases of severe pneumonia in early December 2019, with the first report made to the regional office of the World Health Organization on 30 December 2019.1 The epidemiological curve of affected individuals rose steeply worldwide, thus leading to the outbreak being declared as a public health emergency of international concern on 30 January 20201 and subsequently a pandemic on 11 March 2020.2 As of 30 April 2020, 3 090 445 cases have been reported worldwide,3 with a mortality rate of 3.4%,4 culminating in the deaths of 217 769 individuals.3

Like other members of the Coronaviridae family, SARS-CoV-2, which causes coronavirus disease 2019 (COVID-19), is a positive-sense single-stranded RNA virus with an icosahedral capsid that primarily affects the respiratory tract.5 Other members of the family have caused pandemics with similar clinical presentations, such as Middle East respiratory syndrome–related coronavirus (MERS-CoV) and SARS-CoV. However, COVID-19 outnumbers both of the others in terms of cases and deaths, despite its lower mortality rate compared with SARS-CoV and MERS-CoV.6

The COVID-19 outbreak primarily affects adults, and the severity of disease increases in an age-dependent fashion. Additional risk factors include male sex and co-morbidities including diabetes, hypertension, and previous respiratory impairments.7 Children and adolescents comprise a relatively minor proportion of patients who have tested positive for the virus. Moreover, the paediatric body responds to the disease differently from the adult body. This leads to heterogeneous clinical presentation, disease severity, and mortality rates across the age spectrum.

review of 72 314 cases by the Chinese Centre for Disease Control and Prevention reported that the age-groups of 10 to 19 years and under 10 years contributed 1% each to the total disease burden.8 Children aged under 18 years have been reported to comprise 1.2% of the total cases in Italy,9 whereas in the US and Madrid, Spain, the contribution of this age-group has been reported as 1.7%10 and 0.8%11, respectively. Statistics from Pakistan reveal that children aged 10 to 19 years comprise 7.25% of the total number of cases, and this age-group’s mortality has been approximated as 0.52%.12 Although the outbreak appears to be stabilising or declining in certain areas of the world, many regions are still witnessing an upward trend or even a resurgence.

Increased incidence of asymptomatic carriage and milder symptoms may lead to a decreased need for testing in the paediatric population, especially in already burdened healthcare systems. Hence, this age-group may remain as a source of continued transmission, the magnitude of which remains unexplored. Thus, we aimed to review the characteristics and presentation of COVID-19 among children and describe any subtle characteristics that may strengthen the clinical suspicion of infection and prompt further testing. This may help to break the chain of transmission and effectively decrease the global burden of the pandemic.

Considering the date of the earliest confirmed report of COVID-19, we searched Medline for studies published from 1 December 2019 until 9 April 2020, with no language restriction and with a combination of the key search terms “coronavirus” OR “COVID-19” or “2019-nCoV” OR “SARS-CoV-2” AND “baby” OR “babies” OR “pediatric” OR “paediatric” OR “newborn” OR “neonate” OR “adolescent” OR “child” OR “children” OR “infant” OR “boy” OR “girl” OR “teenage”. We used a comprehensive search strategy to identify the relevant studies. The screening process was conducted by two independent reviewers (MJ and MN), and a third reviewer (DN) was consulted in the event of discrepancies. The articles were screened on the basis of title and abstract to assess their relevance to the aims of our study, followed by full-text screening. For each retrieved full-text article, we hand-searched and examined the citation chain for additional studies. The PubMed search identified 325 articles. After excluding 229 irrelevant articles, 96 full-text articles were reviewed, of which only 27 were included in this review.10 11 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

This systematic review was conducted in accordance with the PRISMA statement.38 An inclusive approach to eligibility assessment was taken. Studies were deemed eligible if they provided clinical data on COVID-19 in the paediatric population (ie, aged 0-19 years). Therefore, a few letters that provided original data were also included in the study. We excluded letters, case reports, editorials, opinion pieces, or studies from which no data from paediatric patients could be extracted, focused on other coronaviruses than COVID-19, epidemiological studies that provided no clinical findings, and those in other languages with no English translation. The Figure shows PRISMA chart outlining the search strategy.

The clinical manifestations and outcomes of COVID-19 in the paediatric population are summarised in Table 1.10 11 14 15 16 18 19 20 21 22 23 24 25 26 27 28 29 30 33 39 40 41

Korean data suggested that a 10-year-old female child was the first paediatric case, and the youngest paediatric case of COVID-19 as of 2 March 2020 was a 45-day-old male baby.42 To date, the youngest reported paediatric case was a 36-hour-old baby in China.13

The median age of the paediatric patients with COVID-19 from Madrid, Spain was 1 year (range, 0-15 years). The largest dataset of 2143 paediatric patients from China reported the median age at diagnosis as 7 years.14 However, in the largest-to-date American dataset of 2572 COVID-19 cases in children aged <18 years, the median age was reported as 11 years.10

Comparing the two large studies,10 14 we believe that there is variation in the median age between different regions of the world. However, paediatric data from other parts of the world to confirm this are still lacking. We deduce that children of all ages can be infected, including neonates, infants, and young children.

An observational cohort study of 36 patients identified that 64% of the patients infected with SARS-CoV-2 were male.15 This is consistent with findings from other small studies.16 17 However, a large study by Dong et al14 did not find any statistically significant trend in the sex of paediatric patients, with boys comprising 1213 cases (56.6%). This is in line with another large study from the US describing 2490 paediatric cases: 57% of the patients were male.10

The incubation period of COVID-19 ranged from less than 1 day to as long as 16 days.19 The median incubation period in children reported by Han et al18 and Shen et al19 was 5 days and 7.5 days, respectively.

Infected children were either asymptomatic or reported symptoms like fever and dry cough as the most common symptoms.13 15 19 20 21 22 23 24 25 26 Data reported by the US Centers for Disease Control and Prevention resonate with these findings: fever and cough were reported in 56% and 54% of the paediatric patients, respectively, whereas in another study, 68% of the paediatric cases had no obvious symptoms.10 A case series comprising of six patients reported fever and cough in all of them and vomiting in four of them.27 These results were confirmed by Xia et al,16 who found cough almost equally frequently as fever (65% and 60%, respectively). However, polypnoea has been reported as the most common symptom in severely affected patients, followed by fever and cough.28 A high proportion of patients presented initially with gastrointestinal symptoms such as nausea, vomiting, abdominal pain, constipation, and diarrhoea.10 15 16 18 19 21 22 27 28 29 Other reported symptoms were fatigue, myalgia, headache,10 16 18 22 28 and upper respiratory tract symptoms such as sore throat, nasal discharge, tachypnoea, and expectoration.10 13 15 16 17 19 21 22 23 24 25 28 29

Presentation of COVID-19 in paediatric patients is much milder than that in adults. Therefore, children may be a hidden source of infection.22 Fever, cough, and shortness of breath were more commonly reported among adult patients (93%) than paediatric patients (73%).10 Another study showed a statistically significant difference between symptomatic presentation in children and adults (P<0.0001): fever (36% and 86%, respectively), cough (19% and 62%), pneumonia (53% and 95%), and severe disease type (0% and 23%).15 In contrast, gastrointestinal symptoms were more common in children.10 18 Another difference was observed between the median duration of fever in children (1 day; range, 0-3 days) and adults (4 days; range, 1-10 days).18

Dong et al14 reported that only 34.1% of cases were laboratory-confirmed, whereas the remainder had clinically suspected disease on the basis of their symptoms. However, this does not rule out their chance of other respiratory infections, so we cannot fully rely on clinical suspicion. Children with SARS-CoV, SARS-CoV-2, and H1N1 influenza presented with somewhat similar symptoms, with the most common being fever. However, cough and pharyngeal congestion are not as common in SARS-CoV-2 (7% and 3%, respectively) as in SARS-CoV (64% and 14%) and H1N1 influenza (83% and 95%). These findings indicate that SARS-CoV-2 has little effect on the upper respiratory tract of children.15 24 25

There are many plausible explanations for why the disease’s manifestations are milder in the paediatric population as compared with adults. First, it is very rare for children to have co-morbidities like diabetes, cardiovascular disease, and hypertension. Adults have a higher prevalence of C-reactive protein and longer duration of fever, suggesting a stronger immunological response compared with that in children.15 18 Further, children tend to remain at home and have fewer opportunities for exposure to pathogens or patients. Angiotensin-converting enzyme II is the receptor speculated to be affected by SARS-CoV-2.43 44 Children are less sensitive to COVID-19 because of their lower maturity and binding ability to a different distribution of angiotensin-converting enzyme II receptors.45 Furthermore, children experience respiratory infections in winter more often than adults do, which may result in higher levels of antibodies against other respiratory viruses, providing cross-protection against SARS-CoV-2. Additionally, children are mostly infected by the second or third generation of viral infections; hence, the viruses with which they are infected have weak virulence. Further, children’s immune response is still under development and may show different responses to pathogens than that of adults.23 24 A different yet possible explanation is that like SARS and MERS-CoV infection in children, SARS-CoV-2 infection may follow a much milder and shorter course.21

Among the studies that focused on radiographic findings in paediatric patients (n=224), 147 (65.6%) patients showed positive findings suggestive of pneumonia. Most of the patients presented with abnormalities, such as multiple bilateral, peripheral ground-glass opacities (GGO), and consolidation.15 18 19 20 23 24 27 29 30 In comparison with adults, pulmonary inflammatory changes have been reported to be milder,15 23 24 and nodular changes (75%), revealed as halo sign and air bronchogram sign on computed tomography (CT), are more common in paediatric patients.16 22 23 Therefore, these should be considered as typical signs in paediatric patients.

Chest CT images demonstrated bilateral lung involvement in about 70% of children aged <3 years, and unilateral lesions and normal lungs were reported more frequently in children aged ≥6 years.21 Lesions were mainly distributed in the middle and outer bands of the lungs near the pleura. However, 53% had no obvious abnormalities.22 When another group of 24 asymptomatic carriers was reviewed, 29.2% had normal CT images.30 Feng et al24 examined nine of the 15 confirmed paediatric patients, of whom 54% had no clear symptoms on admission. However, their CT results were typical of SARS-CoV-2 infection. Chest CT scans revealed improvement in children after 3 to 5 days of treatment.17 20 24 29 However, lesions were sometimes still visible on chest CT despite two consecutive negative nucleic acid tests.16

The severity of the lesions was limited in the early stage, but they increased in density as the disease advanced, involving multiple lobes bilaterally.16 22 Earlier in the course of disease, CT showed consolidation with surrounding halo sign, GGO, fine mesh shadow, and tiny nodules in 50%, 60%, 20%, and 15% of cases, respectively.16 In severely ill patients, bilateral multiple patch-like shadows, GGO, ‘white lung’ change accompanied by air bronchogram sign, and pleural thickening was the characteristic picture on CT.16 21 28 In neonates, chest X-ray commonly showed GGO and blurred lung margins followed by bilateral pneumothorax and signs of neonatal respiratory distress syndrome.17

Computed tomography features could play an important role in screening suspected cases radiographically while awaiting confirmation by real-time reverse transcription–polymerase chain reaction (PCR), the results of which could be used to decide the subsequent plan of action. Computed tomography can also be useful in cases that yield multiple false-negative real-time reverse transcription–PCR tests despite being clinically symptomatic.

Analysis of laboratory tests has revealed different laboratory parameters in children compared with adults. In children, the peripheral white blood cell count and absolute lymphocyte count are usually normal or slightly reduced.11 15 19 20 22 24 27 28 29 31 In contrast, laboratory analyses of adults have shown low leukocyte counts and significant reductions in absolute peripheral blood lymphocyte counts. One study revealed leukopenia as a common finding among adults (20%), whereas leukocytosis was more frequent in children (28.6%; P=0.014).18 One study found an elevated lymphocyte count on the initial routine blood test in 66% of paediatric subjects.23 Although the association is not clear, this altered immune response and lack of significant lymphopenia might help to cause the milder presentation in children. Some features differed significantly according to disease severity. Sun et al28 observed normal or mildly raised levels of leukocytes, neutrophils, and lymphocytes in severely ill patients, whereas low counts were observed in critically ill patients with serious complications. Abnormalities in the cytokine spectrum, characterised by increased plasma concentrations of inflammatory cytokines, were seen more frequently in critically ill than severe patients.28

Inflammatory markers like C-reactive protein and erythrocyte sedimentation rate were normal or transiently elevated.19 22 23 38 Raised levels of procalcitonin (PCT) were linked with severe disease in children.11 16 22 The erythrocyte sedimentation rate was raised significantly in adults as compared with children (P=0.047), whereas PCT was elevated in 42.9% of children but no adult patients (P=0.007).18 Elevated PCT in children may indicate bacterial co-infection, and timely administration of antibiotics might prove beneficial.

Another characteristic feature of COVID-19 is that it affects vital organs like the lungs, liver, and heart, indicated by increased levels of myocardial enzymes, aspartate aminotransferase, alanine aminotransferase, and D-dimer. Myocardial zymography revealed a higher frequency of elevated levels of isoenzyme in children than in adults.11 16 18 20 28 The level of creatine kinase, an indicator of myocardial injury, is significantly higher in severely ill patients.28 Moreover, brain natriuretic peptide has been found in a few paediatric patients.28 The presence of creatine kinase and brain natriuretic peptide indicates that SARS-CoV-2 has the potential to cause heart injury. Therefore, attention should be paid to those laboratory results.

In summary, the laboratory findings reported in children with SARS-CoV-2 are inconsistent with those observed in adult cases. Disease progression is characterised by amplified inflammatory response or cytokine storm. Monitoring of laboratory parameters is suggested to identify patients who might show improvement with anti-inflammatory treatments.

The principles of early identification, early isolation, early diagnosis, and early treatment should be stressed. Our review did not identify any treatment protocols or trials specific to the paediatric population. Although most children with mild disease may not have indications for hospitalisation, supervision must be ensured to contain and prevent transmission. As no vaccine is currently available, management plans include bed rest and supportive treatments like maintenance of water electrolyte balance and homeostasis, administration of antipyretics, and administration of broad-spectrum antibiotics because of the probability of co-infecti on.10 16 19 20 21 22 27 29 30 31 32 33 34 38

Given by spray or nebulisation in the early phase of disease, interferons, alone or in combination with other antivirals, have been shown to improve symptoms.10 11 20 22 23 29 30 33 38 Oral lopinavir/ritonavir or ribavirin have been used; however, their efficacy and safety remain to be determined.10 11 19 22 23 27 29 30 33 34 38 Corticosteroids and intravenous immunoglobulin have been used in severe cases only.10 18 19 22 27 33 38 Use of steroids for treatment of SARS-CoV and MERS-CoV resulted in increased rates of secondary bacterial and fungal infections and longer duration of hospital stay. Thus, in addition to suppressing the inflammatory response, steroids also delay viral clearance.46 Because of the lack of evidence regarding efficacy, the World Health Organization’s interim guidance advised against the use of steroids for treatment of novel coronavirus, unless indicated.47 For such cases, it is recommended to use steroids only in the short term, and only as a part of a clinical trial, to efficiently weigh their harms and benefits.48

Patients with COVID-19 should be closely monitored for signs of clinical deterioration, such as rapidly progressive respiratory failure, central cyanosis, coma, convulsion, and sepsis. If respiratory distress develops despite the use of a nasal catheter or mask oxygenation, a heated humidified high-flow nasal cannula and non-invasive ventilation should be used to target SpO2 ≥94%.9 11 17 18 19 27 32 42 Mechanical ventilation with endotracheal intubation should be adopted when no improvement is seen.10 11 38 Increased levels of pro-inflammatory factors have been seen in children.11 16 19 20 22 23 24 27 28 29 31 38 Thus, targeted anti-inflammatory therapies that might help with early control of disease progression are warranted in the future. Monitoring patients’ conditions closely and the application of timely and effective therapeutic protocols through multidisciplinary approaches could serve as the cornerstones of COVID-19 treatment.

Compared with adults, children rarely had co-morbidities.15 18 Zheng et al21 reported two patients with congenital heart disease, one of whom also had malnutrition and metabolic diseases. Among 345 paediatric cases with information on underlying conditions, 23% had at least one underlying condition, with chronic lung disease being the most common, followed by cardiovascular and immunosuppressive diseases.10 Tagarro et al11 reported that 27% of patients had underlying disease. Furthermore, Xia et al16 reported that 35% patients had a history of underlying diseases, which may indicate that such patients are more susceptible to SARS-CoV-2.

Although paediatric patients are susceptible to COVID-19, the case fatality rate of severe paediatric patients is much lower than that of adults (49.0%),15 which indicates that they have favourable outcomes compared with adults.21 Paediatric patients mostly recover in 1 to 3 weeks and are generally discharged after consecutive negative nucleic acid tests.15 16 18 19 20 21 22 24 27 29 30 31 33 Tagarro et al11 reported that 60% of paediatric patients were hospitalised, with only 10% admitted to paediatric intensive unit care. Sun et al28 also reported that most severely ill patients recovered and were discharged. Another two children who received paediatric intensive unit care recovered without any adverse outcomes reported.21

Disease duration is relative to the severity of the disease: the duration is over 10 days across all patients and over 20 days in critically ill patients.28 Qiu et al15 concluded that patients with the moderate clinical type spent more days in hospital compared with those with mild clinical type (P=0.017). The length of hospital stay has varied between different studies: the minimum and maximum averages reported have been 8.327 and 15.3 days,19 respectively. Because this parameter is multifactorial and influenced by isolation policy and the availability of health facilities and laboratory tests in different hospitals, the length of hospital stay does not necessarily predict the prognosis, and detailed analysis is expected on its significance.

Overall, the prognosis in neonates is also good. Most of them have been discharged after consecutive negative nucleic acid test results.13 A few have been kept under observation despite stable condition and negative clinical and radiological findings because of positive COVID-19 pharyngeal swab nucleic acid test results.13 16 17

Almost all studies have reported recovery without any complications. In critically ill patients, septic shock and multiple organ dysfunction syndrome were the most common complications, and intussusception, toxic encephalopathy, status epilepticus, disseminated intravascular coagulation (DIC), hydronephrosis, cardiac insufficiency, coagulopathy, hypoglobulinaemia, and gastroenteritis were also reported.28 One study reported two critical cases with abnormal renal function and coagulapathy.21

In China, the deaths of a 14-year-old boy14 and a 10-month-old baby with intussusception who developed multi-organ failure 4 weeks post-admission were reported.39 Three paediatric deaths were reported in the US.10 None had been reported in Italy or Spain as of 15 March and 8 April 2020, respectively.9 11

The ability of SARS-CoV-2 to infect the gastrointestinal tract is supported by detection of its nucleic acids in stool samples from adults and children.15 20 26 29 32 In spite of negative nucleic acid results from throat swab specimens, children’s stools were still nucleic acid–positive after 10 days of recovery.29 Other studies have reported re-admission of discharged children with positive stool specimens but negative respiratory specimens. Although their prognosis is better than that of adults; the period of PCR positivity is longer in children.15 20 26 Poorer hand hygiene practices causing faecal-oral transmission might be a reason for the delayed clearance of viral RNA in children’s stools. Although positive results cannot confirm that live virus is present in the stool, this still increases the infection risk to the public, so follow-up of specimen collection should be considered. The isolation period for children should be reviewed because of the transmission risk.

The virus is mainly transmitted through respiratory droplets or contact.28 Transmission among the paediatric population mostly occurs by close contact with family members,14 18 19 20 21 22 23 29 a history of exposure to the epidemic area, or both.15 21 According to Wang et al,22 90% of cases were clustered in families. Another study reported that 62.5% cases were associated with familial clustering.28 Neonates and infants are more likely than adults to be infected via close contact with COVID-19-positive family members.16 25

To date, nine studies have reported on neonates born to COVID-19 positive mothers (n=115); only seven of those neonates were SARS-CoV-2-positive.13 16 32 Xia et al,16 Yu et al,13 and Zeng et al32 reported that 3/20 (15%), 1/7 (14%), and 3/33 (9%) neonates were positive, respectively. Despite the placenta and cord blood being negative for SARS-CoV-2, one neonate was diagnosed as positive 36 hours after birth.13 Other studies have also found that the placenta, cord blood, and breast milk were negative for SARS-CoV-2.17 31 34 37 Because of the limited evidence regarding vertical transmission, we speculate that close contact could explain the positive results. We are still not confident about the probability of vertical maternal-fetal transmission, and further studies need to be performed on this subject. Neonates born to infected mothers should be separated immediately after birth and undergo an isolation period.

Although the majority of the studies showed that neonates were negative for SARS-COV-2 and that most neonates had excellent outcomes,36 a few studies revealed that neonates born to COVID-19-positive mothers could develop other complications that could lead to poor neonatal outcomes. The rate of premature birth among newborns born to mothers with confirmed COVID-19 pneumonia (23.5%) was significantly higher than the 2020 and 2019 control rates (5.8% and 5.0%, respectively). Low birth weight was also more frequent in infants of infected mothers (17.6%) than in the control group (2.5%).34 Some incidence of premature birth, fetal distress, premature rupture of membranes, small size for gestational age, and large size for gestational age was observed in neonates born to COVID-19-positive mothers.17 35 Two COVID-19-positive neonates developed DIC, of whom one died on the 9th day secondary to refractory shock, multiple organ dysfunction syndrome, and DIC.17 The neonatal outcomes are summarised in Table 2.13 16 17 31 32 34 35 36 37

Paediatric patients make up a small fraction of COVID-19 cases, and they have a better prognosis than adult patients have. The differences in the mechanisms behind COVID-19’s clinical manifestations between children and adults need to be verified by large, well-designed studies. Children are likely to become a hidden source of infection, which may delay the diagnosis of COVID-19, leading to unfavourable outcomes and causing community transmission. The probability of intrauterine vertical transmission in neonates is low, and close contact is the only plausible explanation for the observed positive results in neonates.

M Jahangir designed the study. All authors contributed to the acquisition and analysis of data, and wrote the manuscript. M Jahangir had critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

The authors have disclosed no conflicts of interest.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1. World Health Organization. Coronavirus disease (COVID-19)–events as they happen. Rolling updates on coronavirus disease (COVID-19). Available from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019/events-as-they-happen. Accessed 30 Apr 2020.

2. World Health Organization. Coronavirus disease 2019 (COVID-19) situation report–51. Available from: https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200311-sitrep-51-covid-19.pdf?sfvrsn=1ba62e57_10. Accessed 30 Apr 2020.

3. World Health Organization. Coronavirus disease 2019 (COVID-19) situation report–101. Available from: https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200430-sitrep-101-covid-19.pdf?sfvrsn=2ba4e093_2. Accessed 30 Apr 2020.

4. World Health Organization. WHO Director-General’s opening remarks at the media briefing on COVID-19–3 March 2020. Available from: https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19---3-march-2020. Accessed 30 Apr 2020.

5. Cascella M, Rajnik M, Cuomo A, Dulebohn SC, Di Napoli R. Features, evaluation and treatment coronavirus (COVID-19). StatPearls Publishing; 2020 Jan. Available from: https://www.statpearls.com/as/infectious/52171/. Accessed 30 Apr 2020.

6. Mahase E. Coronavirus: COVID-19 has killed more people than SARS and MERS combined, despite lower case fatality rate. BMJ 2020;368:m641. Crossref

7. Worldometer. Age, sex, existing conditions of COVID-19 cases and deaths. Available from: https://www.worldometers.info/coronavirus/coronavirus-age-sex-demographics/. Accessed 30 Apr 2020.

8. Wu Z, McGoogan JM. Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China: summary of a report of 72 314 cases from the Chinese Center for Disease Control and Prevention. JAMA 2020;323:1239-42. Crossref

9. Livingston E, Bucher K. Coronavirus disease 2019 (COVID-19) in Italy. JAMA 2020;323:1335. Crossref

10. CDC COVID-19 Response Team. Coronavirus disease 2019 in children–United States, February 12–April 2, 2020. MMWR Morb Mortal Wkly Rep 2020;69:422-6. Crossref

11. Tagarro A, Epalza C, Santos M, et al. Screening and severity of coronavirus disease 2019 (COVID-19) in children in Madrid, Spain. JAMA Pediatr 2020;Apr 8. Epub ahead of print. Crossref

12. Health Advisory Platform by Ministry of National Health Services Regulations and Coordination, Government of Pakistan. COVID-19. Available from: http://covid.gov.pk/stats/pakistan. Accessed 5 May 2020.

13. Yu N, Li W, Kang Q, et al. Clinical features and obstetric and neonatal outcomes of pregnant patients with COVID-19 in Wuhan, China: a retrospective, single-centre, descriptive study. Lancet Infect Dis 2020;20:559-64. Crossref

14. Dong Y, Mo X, Hu Y, et al. Epidemiology of COVID-19 among children in China. Pediatrics 2020;145:e20200702. Crossref

15. Qiu H, Wu J, Hong L, Luo Y, Song Q, Chen D. Clinical and epidemiological features of 36 children with coronavirus disease 2019 (COVID-19) in Zhejiang, China: an observational cohort study. Lancet Infect Dis 2020;20:689- 96. Crossref

16. Xia W, Shao J, Guo Y, Peng X, Li Z, Hu D. Clinical and CT features in pediatric patients with COVID-19 infection: different points from adults. Pediatr Pulmonol 2020;55:1169-74. Crossref

17. Zhu H, Wang L, Fang C, et al. Clinical analysis of 10 neonates born to mothers with 2019-nCoV pneumonia. Transl Pediatr 2020;9:51-60. Crossref

18. Han YN, Feng ZW, Sun LN, et al. A comparative-descriptive analysis of clinical characteristics in 2019-coronavirusinfected children and adults. J Med Virol 2020;Apr 6. Epub ahead of print. Crossref

19. Shen Q, Guo W, Guo T, et al. Novel coronavirus infection in children outside of Wuhan, China. Pediatr Pulmonol 2020;55:1424-9. Crossref

20. Su L, Ma X, Yu H, et al. The different clinical characteristics of corona virus disease cases between children and their families in China—the character of children with COVID-19. Emerg Microbes Infect 2020;9:707-13. Crossref

21. Zheng F, Liao C, Fan QH, et al. Clinical characteristics of children with coronavirus disease 2019 in Hubei, China. Curr Med Sci 2020;40:275-80. Crossref

22. Wang D, Ju XL, Xie F, et al. Clinical analysis of 31 cases of 2019 novel coronavirus infection in children from six provinces (autonomous region) of northern China [in Chinese]. Zhonghua Er Ke Za Zhi 2020;58:269-74.

23. Zhou Y, Yang GD, Feng K, et al. Clinical features and chest CT findings of coronavirus disease 2019 in infants and young children [in Chinese]. Zhongguo Dang Dai Er Ke Za Zhi 2020;22:215-20.

24. Feng K, Yun YX, Wang XF, et al. Analysis of CT features of 15 children with 2019 novel coronavirus infection [in Chinese]. Zhonghua Er Ke Za Zhi 2020;58:E007.

25. Wei M, Yuan J, Liu Y, Fu T, Yu X, Zhang ZJ. Novel coronavirus infection in hospitalized infants under 1 year of age in China. JAMA 2020;323:1313-4. Crossref

26. Ma X, Su L, Zhang Y, Zhang X, Gai Z, Zhang Z. Do children need a longer time to shed SARS-CoV-2 in stool than adults? J Microbiol Immunol Infect 2020;53:373-6. Crossref

27. Liu W, Zhang Q, Chen J, et al. Detection of COVID-19 in children in early January 2020 Wuhan, China. N Engl J Med 2020;382:1370-1. Crossref

28. Sun D, Li H, Lu X, et al. Clinical features of severe pediatric patients with coronavirus disease 2019 in Wuhan: a single center’s observational study. World J Pediatr 2020;16:251-9. Crossref

29. Zhang T, Cui X, Zhao X, et al. Detectable SARS-CoV-2 viral RNA in feces of three children during recovery period of COVID-19 pneumonia. J Med Virol 2020;92:909-14. Crossref

30. Hu Z, Song C, Xu C, et al. Clinical characteristics of 24 asymptomatic infections with COVID-19 screened among close contacts in Nanjing, China. Sci China Life Sci 2020;63:706-11. Crossref

31. Chen H, Guo J, Wang C, et al. Clinical characteristics and intrauterine vertical transmission potential of COVID-19 infection in nine pregnant women: a retrospective review of medical records. Lancet 2020;395:809-15. Crossref

32. Zeng L, Xia S, Yuan W, et al. Neonatal early-onset infection with SARS-CoV-2 in 33 neonates born to mothers with COVID-19 in Wuhan, China. JAMA Pediatr 2020;174:722-5. Crossref

33. Li W, Cui H, Li K, Fang Y, Li S. Chest computed tomography in children with COVID-19 respiratory infection. Pediatr Radiol 2020;50:796-9. Crossref

34. Li N, Han L, Peng M, et al. Maternal and neonatal outcomes of pregnant women with COVID-19 pneumonia: a case-control study. Clin Infect Dis 2020;Mar 30. Epub ahead of print. Crossref

35. Zhang L, Jiang Y, Wei M, et al. Analysis of the pregnancy outcomes in pregnant women with COVID-19 in Hubei Province [in Chinese]. Zhonghua Fu Chan Ke Za Zhi 2020;55:166-71.

36. Chen S, Liao E, Cao D, Gao Y, Sun G, Shao Y. Clinical analysis of pregnant women with 2019 novel coronavirus pneumonia. J Med Virol 2020;Mar 28. Epub ahead of print. Crossref

37. Fan C, Lei D, Fang C, et al. Perinatal transmission of COVID-19 associated SARS-CoV-2: Should we worry? Clin Infect Dis 2020;Mar 17. Epub ahead of print.

38. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA Statement. PLoS Med 2009;6:e1000097. Crossref

39. Lu X, Zhang L, Du H, et al. SARS-CoV-2 infection in children. N Engl J Med 2020;382:1663-5. Crossref

40. Recommendations for the diagnosis, prevention and control of the 2019 novel coronavirus infection in children (first interim edition) [in Chinese]. Zhonghua Er Ke Za Zhi 2020;58(0):E004.

41. National Health Commission of the People’s Republic of China. Diagnosis and treatment scheme for new coronavirus infected pneumonia (fifth interim edition, revised). Available from: http://www.gov.cn/zhengce/zhengceku/2020-02/09/5476407/files/765d1e65b7d1443081053c29ad37fb07.pdf. Accessed 30 Apr 2020.

42. Park JY, Han MS, Park KU, Kim JY, Choi EH. First pediatric case of coronavirus disease 2019 in Korea. J Korean Med Sci 2020;35:e124. Crossref

43. Zhou P, Yang XL, Wang XG, et al. A pneumonia outbreak associated with a new coronavirus of probable bat origin. Nature 2020;579:270-3. Crossref

44. Wrapp D, Wang N, Corbett KS, et al. Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation. Science 2020;367:1260-3. Crossref

45. Fang F, Luo XP. Facing the pandemic of 2019 novel coronavirus infections: the pediatric perspectives [in Chinese]. Zhonghua Er Ke Za Zhi 2020;58:81-5.

46. Russell CD, Millar JE, Baillie JK. Clinical evidence does not support corticosteroid treatment for 2019-nCoV lung injury. Lancet 2020;395:473-5. Crossref

47. World Health Organization. Clinical management of severe acute respiratory infection (SARI) when COVID-19 disease is suspected: interim guidance. Available from: https://www.who.int/publications/i/item/clinical-management-of-covid-19. Accessed 29 May 2020.

48. Shen K, Yang Y, Wang T, et al. Diagnosis, treatment, and prevention of 2019 novel coronavirus infection in children: experts’ consensus statement. World J Pediatr 2020;16:223-31. Crossref

Labour pain is so notoriously painful that opium and its derivatives have been used in childbirth for several thousand years, along with numerous folk medicines and remedies. Nulliparous women suffer greater sensory pain during the early stage of labour compared with multiparous women, for whom the second stage is more intense.1 Labour pain has both visceral and somatic components.2 The first stage of labour pain is caused by contraction of the uterus and gradual dilatation of the cervix. The visceral pain is carried by small unmyelinated C-fibres through sympathetic nerves to the T10 to L1 segments of the dorsal horn of the spinal cord. The pain is often referred to as located in the front and back of the lower abdomen and sacrum. Stretching of the vaginal wall, perineum, and vaginal surface of the cervix in the later stage of labour causes ischaemic pain, which is conducted through thick myelinated A fibres in the pudendal and perineal branches of the posterior cutaneous nerve in the thigh to the S2 to S4 nerve roots, Thus, women who are giving birth feel sharp somatic pain in the perineum.

As well as being unpleasant, labour pain may have harmful effects on the mother and baby,1 3 as pain stimulates catecholamine release, which constricts the uterine blood vessels. Pain also causes maternal hyperventilation, resulting in hypocapnia, which further constricts the uterine vessels and decreases the mother’s ventilatory drive between contractions, thereby causing the left shift of the maternal oxygen dissociation curve. These factors compromise oxygen supply to the fetus and can lead to fetal hypoxaemia and fetal metabolic acidosis (Fig 1). Premature ‘bearing down’ can also lead to birth canal trauma and birth injury. Parenteral opioids can exacerbate maternal respiratory depression, whereas regional analgesia can reduce the adverse effects of labour pain on respiration and the sympathetic nervous system. Therefore, good labour analgesia should aim not only to relieve the pain and suffering of the mother but also to decrease fetal acidosis and make the delivery process safer for both the mother and baby. Traditionally, pain relief methods are classified into non-pharmacological, pharmacological, and regional techniques. In this article, we examine the most recent evidence on the efficacy and safety of the commonly available methods.

Mild labour pain may be reduced by massage, psychological relaxation techniques, transcutaneous electrical nerve stimulation, aromatherapy, hypnosis, sterile water injection, acupuncture, deep breathing, and hydrotherapy. However, most of the evidence on non-drug interventions is based on anecdotal reports from a small number of studies. A Cochrane systematic review reported that immersion and relaxation produced good satisfaction, and both relaxation and acupuncture decreased the use of forceps and ventouse, with acupuncture also decreasing the number of Caesarean sections.4 There was insufficient evidence to judge whether or not hypnosis, biofeedback, sterile water injection, aromatherapy, and transcutaneous electrical nerve stimulation are effective.4

Entonox is a mix of 50% nitrous oxide in oxygen that has been in use for a long time. It has some analgesic efficacy, but many women who used it felt drowsy, nauseous, or were sick.4 Nitrous oxide has detrimental effects on vitamin B12 metabolism, and there are valid concerns about occupational exposure to healthcare professionals in the delivery suite, although the use of a proper scavenging system can help. It has the advantage of being easy to use by self-administration, but around 30% to 40% of patients found pain relief inadequate with Entonox alone.5

Sub-anaesthetic doses (0.8% in oxygen) of sevoflurane have been evaluated as an alternative to Entonox.6 7 In those studies, despite its lack of analgesic effects and increased level of sedation, most women preferred it to Entonox. It also caused less nausea and vomiting than Entonox. However, there are valid concerns about loss of consciousness, fetal toxicity, and air pollution; therefore, it is not popular.

Intramuscular pethidine is widely prescribed. Pethidine is a potent opioid, making the side-effects of somnolence, nausea, vomiting, and respiratory depression common. It is less effective than epidural analgesia4 and cannot be given near the end of the first stage or during the second stage of labour because of its respiratory depressant effects on the baby. It also has a neuroexcitatory metabolite, norpethidine.

Remifentanil, an ultra-short acting opioid with a half-life of about 3 minutes irrespective of the duration of infusion, is usually given intravenously using a patient-controlled analgesic pump. In 2001, we found that the time to first request for rescue analgesia and maternal satisfaction were higher with patient-controlled analgesic remifentanil compared with intramuscular pethidine. There was no sedation, apnoea, or oxygen desaturation in either group, and Apgar scores of the groups were similar.8 In 2018, the RESPITE trial showed that remifentanil halved the proportion of epidural conversions compared with intramuscular pethidine.9 The pooled risk ratio for rescue analgesia of remifentanil relative to pethidine was 0.54. The study also reported that remifentanil posed no excessive risk of respiratory depression to the mothers or babies, thus challenging pethidine’s routine use as a first-line opioid in the management of labour pain. Although its analgesia is not superior to an epidural, remifentanil is an efficacious alternative for patients who have contra-indications to epidural administration, including back problems, coagulopathy, and fixed cardiac output diseases. Many local and overseas centres have incorporated this option into their labour pain management programmes. The RemiPCA SAFE Network has been established to set standards and monitor maternal and fetal outcomes when remifentanil is used for labour analgesia.10

Epidural analgesia, introduced in the 1960s, is still the most effective method of labour pain relief.11 It involves placing a very fine catheter into the epidural space for repeat boluses or continuous infusion of local anaesthetics. This allows for continuous pain relief throughout labour and ‘top-up’ boluses, if required, for operative deliveries. New drugs and technological advancements have improved safety, and our understanding of its effects on obstetric outcomes has been revised (Table 1). Levobupivacaine and ropivacaine are the newest amide local anaesthetics, and they are less cardiotoxic than bupivacaine. Traditionally, a high concentration of local anaesthetic (eg, 0.2%-0.25% bupivacaine) has been used to maintain labour epidural analgesia. Over the years, the adoption of a lower concentration of local anaesthetic (0.0625%-0.1%) and lipophilic opioids (fentanyl or sufentanil) has lessened side-effects such as motor blockage and hypotension.12 These drugs have made it possible for women to walk or move around more easily in bed and retain a mild sensation of uterine contraction and urgency of bearing down, thereby facilitating pushing the baby out in the second stage of labour. In the Comparative Obstetric Mobile Epidural Trial study, the use of low-dose infusion significantly reduced the incidence of assisted vaginal delivery.13 Meta-analysis showed that a lower concentration of local anaesthetic reduces the incidence of assisted vaginal delivery and urinary retention and shortens the second stage compared with a higher concentration.14 A 2018 Cochrane review stated that this type of epidural analgesia has no adverse impact on the proportions of Caesarean section, long-term backache, or neonatal outcomes.11

In the ‘needle-through-needle’ combined spinal-epidural (CSE) technique, a 25- or 27-G pencil point spinal needle with a locking device is inserted through the epidural needle that allows the deposition of a small dose of local anaesthetic, with or without opioids, into the cerebrospinal fluid in the intrathecal space. The onset of analgesia is rapid. An epidural catheter is then threaded through the epidural needle after withdrawing the spinal needle. A review of the complications has concluded that CSE is equally safe to a conventional epidural.15 The use of CSE has increased relative to that of the conventional epidural technique, as it has a quicker onset of analgesia in mothers with severe pain, those in the advanced stage of labour, and those who are multiparous. The technique also improves the success of correct functioning epidural catheter placement by prior verification of placement in the subarachnoid space with the spinal needle.16 Despite the increasingly widespread use of this technique and numerous published investigations, the optimal intrathecal drug regimen has not yet been determined. The disadvantage of CSE is immediate uncertainty about whether the epidural is working because of the initial effects of spinal analgesia. However, a 2016 study refuted this and favoured CSE earlier detection of failed epidural analgesia.17 The use of a 27-G spinal needle is preferred, as its small size is associated with a lower risk of post-dural puncture headache.18 Although there is faster onset of analgesia, the effects on maternal satisfaction are controversial. A systematic review found no differences in maternal satisfaction, mode of delivery, or ambulatory ability between CSE and the conventional epidural technique.19 Subsequently, the choice between conventional epidural and CSE has often been dictated by the clinical situation, institutional protocols, available equipment, and practitioner preference/experience.

In continuous intrathecal labour analgesia, local anaesthetic with or without opioids is directly deposited into the intrathecal space using a 23- to 28-G microcatheter. This technique can provide rapid analgesia or anaesthesia and higher maternal satisfaction with less use of local anaesthetic, but it is also associated with more technical difficulties and catheter failure compared with epidural analgesia. It is theoretically advantageous in the management of morbidly obese patients, patients with significant co-morbidities who cannot tolerate haemodynamic instability, and patients with potentially difficult airways who undergo Caesarean section, as it allows gradual titration and slower onset of subarachnoid blockage.20 This technique is still uncommonly used because of various concerns including post-dural puncture headache and neuraxial infection. Further studies are required to assess whether it can assist in the management of patients with conditions that make neuraxial labour analgesia challenging.

Once an epidural catheter is placed, analgesia can be maintained by intermittent top-ups, continuous infusion, patient-controlled analgesia, or programmed intermittent epidural boluses (PIEB). Continuous infusion technique became popular in the early 1980s. This delivery method reduced the variability of analgesia during labour, especially when high concentrations of local anaesthetics were replaced by low concentrations with the addition of a lipophilic opioid. Unfortunately, this modality does not suit all patients despite many combinations of infusion rate, local anaesthetic concentration, and additives having been investigated. Many patients still require clinician-initiated top-ups or experience unacceptable motor blockage.

Patient-controlled epidural analgesia (PCEA) was first described in 1988.21 Boluses of 4 to 8 mL of epidural mixture are delivered on patient demand with a lockout interval of 10 to 20 minutes. As labour pain has highly variable intensity, and the character of the pain often changes as it progresses, it makes sense that patients may be the best managers of their own pain relief. There is recent evidence that genetic polymorphism may also affect the patient’s labour progress and response to labour analgesia. One example is the Mu opioid receptor gene single-nucleotide polymorphism (OPRM1, A118G), which is believed to be present in 30% of women in labour and may affect the response to neuraxial opioids.22 23 Administration of PCEA allows for some self-titration. Over the past 20 years, PCEA has been widely studied and the technique refined. High-volume, dilute local anaesthetic solutions with a continuous background infusion appear to be the best PCEA regimen.24 The American Society of Anesthesiologists practice guidelines for obstetric anaesthesia advise that basal infusion improves analgesia when provided as part of a PCEA regimen.25 Studies have also shown that PCEA requires less anaesthesia intervention, lower doses of local anaesthetic, and produces less motor blockage than continuous epidural infusion.26 27 Although PCEA delivery devices tend to be more expensive than continuous infusion pumps, the technique may have important benefits. The optimum method of administration requires communication with both the midwife and the patient.

An alternative approach to determining the background infusion rate during PCEA is the use of a computer programme to automatically adjust the background infusion rate according to the amount of local anaesthetic used in the previous hour. A laptop computer is connected to a PCEA pump. In theory, a system that responds to a patient’s analgesic requirements should improve efficacy while minimising the amount of local anaesthetic used for background infusions. Initial studies with this system have been encouraging. In a study comparing demand-only PCEA with computer-integrated background infusion PCEA (CIPCEA), the CIPCEA group had similar local anaesthetic consumption but increased maternal satisfaction.28 Another study found that CIPCEA reduced the incidence of breakthrough pain without increasing drug consumption compared with continuous epidural infusion.29 When CIPCEA was compared with PCEA using fixed-rate continuous infusion, the CIPCEA group had higher maternal satisfaction, whereas local anaesthetic consumption, visual analogue pain scores, and incidence of breakthrough pain were similar between the two groups.30 Therefore, an adjustable background infusion appears to increase maternal satisfaction and may further reduce the incidence of breakthrough pain without increasing local anaesthetic consumption.

Programmed intermittent epidural boluses is a novel technology in which boluses of epidural mixture are delivered at predetermined intervals. Improved analgesia may be offered by PIEB, as the local anaesthetic is administered in boluses under high driving pressure, which can disperse the solution more widely than continuous infusion31 with multi-orifice catheters.32 A system has been developed in which a computer delivers both automated and manual boluses. The authors demonstrated that this ‘programmed intermittent mandatory epidural bolus’ with a PCEA regimen provided advantages over a PCEA plus background infusion regimen: the former used less local anaesthetic dose, but resulted in a higher maternal satisfaction and a longer duration of analgesia. However, there was no difference in the incidence of breakthrough pain between the two groups.33 34 In 2012 and 2014, respectively, Health Canada and the United States Food and Drug Administration approved PIEB combined with PCEA (CADD Solis Epidural Pump, Smiths Medical, St Paul [MN], United States) for clinical use.35 A 2013 systematic review investigating PIEB for maintenance of labour analgesia that included nine randomised controlled trials with 694 patients36 showed that the vast majority of studies associated PIEB with decreased local anaesthetic consumption, improved maternal satisfaction scores, decreased instrumental delivery, and lessened need for anaesthesia intervention. A recent trial confirmed that reduced motor blockage was associated with PIEB,37 although that study could not identify other significant outcomes.

Although ultrasound is widely used in the placement of central venous catheters and peripheral nerve blockage, it is less commonly used in neuraxial analgesia for obstetric patients. It can be used either before the procedure to study the site of needle entry and the depth of the epidural space or for real-time needle guidance (Fig 2). Although the preprocedural use of ultrasound in normal pregnant mothers seems to have limited efficacy among both experienced clinicians38 and trainees,39 some study findings have suggested that it is a useful tool40 to consider in obese patients41 or those with lumbar spine problems. In 2008, the United Kingdom’s National Institute for Health and Care Excellence determined that sufficient evidence had been published to support the routine use of ‘ultrasound to facilitate the catheterisation of the epidural space’.42 In March 2016, the American Society of Regional Anesthesia and Pain Medicine43 published the second evidence-based medicine assessment of ultrasoundguided regional anaesthesia to ‘enable practitioners to make an informed evaluation regarding the role of ultrasound-guided regional anaesthesia in their practice’. A high-quality review article by Arzola outlined the controversies, advantages, and practical applications of preprocedural ultrasound in obstetric patients.44

Neuraxial analgesia is now also safer with the availability of intralipid as an antidote for local anaesthetic toxicity.45 46 Intralipid binds with amide local anaesthetic molecules in the plasma, thereby decreasing the free fraction available to bind with cardiac muscle. It has become widely adopted as part of the resuscitation protocol for local anaesthetic-caused systemic toxicity and should be readily available in all delivery units where neuraxial analgesia is practised. It is given intravenously by boluses followed by continuous infusion according to body weight (Table 2).

Previous concerns that early epidural initiation (when cervical dilatation <4 cm) would increase the rate of instrumental delivery and Caesarean section have been alleviated by more recent research. Wong et al47 found that neuraxial analgesia in early labour did not increase the rate of Caesarean delivery but provided better analgesia and resulted in a shorter duration of labour than systemic analgesia. The latest Cochrane review indicated that there is abundant high-quality evidence that early and late epidural initiation have similar effects on all measured outcomes.48 The American College of Obstetricians and Gynaecologists and the American Society of Anesthesiologists49 have also jointly emphasised that there is no need to wait until cervical dilation has reached 4 to 5 cm and stated that ‘maternal request is a sufficient indication for pain relief in labour’.50 When delivery is imminent, the decision to offer regional anaesthesia should be individualised and depends on various factors including a woman’s parity, fetal condition, and whether a prolonged second stage is expected, such as malposition of the fetus or macrosomia. The Royal College of Anaesthetists recommends that the time from epidural request to the anaesthetist attending should not exceed 30 minutes, after which a second anaesthetist should be available.51

There is insufficient evidence to support the discontinuation of epidural analgesia late in labour as a means to reduce adverse delivery outcomes.52 Doing so also increases the rate of inadequate pain relief in the second stage of labour. A meta-analysis of high-quality studies did not show significant differences in outcomes with immediate and delayed pushing in the second stage of labour.53

The effects of neuraxial analgesia on successful breastfeeding have been evaluated in several studies with controversial results. A recent large, randomised, double-blind, controlled trial showed that epidural solutions containing fentanyl concentrations as high as 2 μg/mL did not affect breastfeeding rates at 6 weeks postpartum.54 The results correlated with those of another study investigating women with previous breastfeeding experience, as both studies showed no difference in the breastfeeding rates at 6 weeks postpartum between groups of women who did and did not receive epidural analgesia.55 Therefore, factors other than epidural and fentanyl administration can affect the successful breastfeeding rate.

The association of maternal fever with epidural analgesia has remained an area of clinical and research interest.56 A 2016 expert panel defined maternal fever as maternal temperature of ≥38°C measured orally for two readings 30 minutes apart.57 Up to one third of mothers may be affected, and the aetiology and prophylactic prevention are still not well understood, although the local anaesthetic used for epidural analgesia is a likely culprit. Sterile inflammation and activation of inflammasomes probably play a role,58 and this is an area of ongoing research.59

Epidural analgesia remains the best method of relieving pain during labour. Advances in technology have made it even safer than before. In the absence of any medical contra-indications, maternal request is a sufficient indication to initiate epidural analgesia, and if it is properly conducted, it can be considered at any stage of labour without affecting the rate of instrumental or Caesarean delivery. Future improvements may lie in preventing breakthrough pain via interaction with various closed-loop feedback drug delivery systems. Remifentanil-based opioid techniques are becoming a popular alternative if epidural is contra-indicated.

All authors contributed to the concept of study, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

As an editor of the journal, MG Irwin was not involved in the peer review process of the article. The other authors have no conflicts of interest to disclose.

The authors thank Prof Ki-jinn Chin, Associate Professor, Department of Anesthesia, Toronto Western Hospital, University of Toronto for permission to use the image in Figure 2.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1. Labor S, Maguire S. The pain of labour. Rev Pain 2008;2:15- 9. Crossref

2. Rowlands S, Permezel M. Physiology of pain in labour. Bailliere Clin Obstet Gynaecol 1998;12:347-62. Crossref

3. Wong CA. Advances in labor analgesia. Int J Womens Health 2009;1:139-54. Crossref

4. Jones L, Othman M, Dowswell T, et al. Pain management for women in labour: an overview of systematic reviews. Cochrane Database Syst Rev 2012;(3):CD009234. Crossref

5. Carstoniu J, Levytam S, Norman P, Daley D, Katz J, Sandler AN. Nitrous oxide in early labor safety and analgesic efficacy assessed by a double-blind, placebo-controlled study. Anesthesiology 1994;80:30-5. Crossref

6. Yeo ST, Holdcroft A, Yentis SM, Stewart A. Analgesia with sevoflurane during labour: i. Determination of the optimum concentration. Br J Anaesth 2007;98:105-9. Crossref

7. Yeo ST, Holdcroft A, Yentis SM, Stewart A, Bassett P. Analgesia with sevoflurane during labour: ii. Sevoflurane compared with Entonox for labour analgesia. Br J Anaesth 2007;98:110-5. Crossref

8. Ng TK, Cheng BC, Chan WS, Lam KK, Chan MT. A double-blind randomised comparison of intravenous patient-controlled remifentanil with intramuscular pethidine for labour analgesia. Anaesthesia 2011;66:796-801. Crossref

9. Wilson MJ, MacArthur C, Hewitt CA, et al. Intravenous remifentanil patient-controlled analgesia versus intramuscular pethidine for pain relief in labour (RESPITE): an open-label, multicentre, randomised controlled trial. Lancet 2018;392:662-72. Crossref

10. Melber AA, Jelting Y, Huber M, et al. Remifentanil patient-controlled analgesia in labour: six-year audit of outcome data of the RemiPCA SAFE Network (2010-2015). Int J Obstet Anesth 2019;39:12-21.Crossref

11. Anim-Somuah M, Smyth RM, Cyna AM, Cuthbert A. Epidural versus non-epidural or no analgesia for pain management in labour. Cochrane Database Syst Rev 2018;(5):CD000331. Crossref

12. Sharma RM, Setlur R, Bhargava AK, Vardhan S. Walking epidural: an effective method of labour pain relief. Med J Armed Forces India 2007;63:44-6. Crossref

13. Comparative Obstetric Mobile Epidural Trial (COMET) Study Group UK. Effect of low-dose mobile versus traditional epidural techniques on mode of delivery: a randomised controlled trial. Lancet 2001;358:19-23. Crossref

14. Sultan P, Murphy C, Halpern S, Carvalho B. The effect of low concentrations versus high concentrations of local anesthetics for labour analgesia on obstetric and anesthetic outcomes: a meta-analysis. Can J Anaesth 2013;60:840-54. Crossref

15. Skupski DW, Abramovitz S, Samuels J, Pressimone V, Kjaer K. Adverse effects of combined spinal-epidural versus traditional epidural analgesia during labor. Int J Gynecol Obstet 2009;106:242-5. Crossref

16. Norris MC, Fogel ST, Conway-Long C. Combined spinalepidural versus epidural labor analgesia. Anesthesiology 2001;95:913-20. Crossref

17. Booth JM, Pan JC, Ross VH, Russell GB, Harris LC, Pan PH. Combined spinal epidural technique for labor analgesia does not delay recognition of epidural catheter failures: a single-center retrospective cohort survival analysis. Anesthesiology 2016;125:516-24. Crossref

18. Landau R, Ciliberto CF, Goodman SR, Kim-Lo SH, Smiley RM. Complications with 25-gauge and 27-gauge Whitacre needles during combined spinal-epidural analgesia in labor. Int J Obstet Anesth 2001;10:168-71. Crossref

19. Simmons SW, Taghizadeh N, Dennis AT, Hughes D, Cyna AM. Combined spinal-epidural versus epidural analgesia in labour. Cochrane Database Syst Rev 2012;(10):CD003401. Crossref

20. Drasner K, Smiley R. Continuous spinal analgesia for labor and delivery: a born-again technique? Anesthesiology 2008;108:184-6. Crossref

21. Gambling DR, Yu P, Cole C, McMorland GH, Palmer L. A comparative study of patient controlled epidural analgesia (PCEA) and continuous infusion epidural analgesia (CIEA) during labour. Can J Anaesth 1988;35:249-54. Crossref

22. Landau R, Cahana A, Smiley RM, Antonarakis SE, Blouin JL. Genetic variability of μ-opioid receptor in an obstetric population. Anesthesiology 2004;100:1030-3. Crossref

23. Sia AT, Lim Y, Lim EC, et al. A118G single nucleotide polymorphism of human μ-opioid receptor gene influences pain perception and patient-controlled intravenous morphine consumption after intrathecal morphine for postcesarean analgesia. Anesthesiology 2008;109:520-6. Crossref

24. Halpern SH, Carvalho B. Patient-controlled epidural analgesia for labor. Anesth Analg 2009;108:921-8. Crossref

25. American Society of Anesthesiologists Task Force on Obstetric Anesthesia. Practice guidelines for obstetric anesthesia: an updated report by the American Society of Anesthesiologists Task Force on Obstetric Anesthesia. Anesthesiology 2007;106:843-63.

26. van der Vyver M, Halpern S, Joseph G. Patient-controlled epidural analgesia versus continuous infusion for labour analgesia: a meta-analysis. Br J Anaesth 2002;89:459-65. Crossref

27. D’Angelo R. New techniques for labor analgesia: PCEA and CSE. Clin Obstet Gynecol 2003;46:623-32. Crossref

28. Lim Y, Sia AT, Ocampo CE. Comparison of computer integrated patient controlled epidural analgesia vs. conventional patient controlled epidural analgesia for pain relief in labour. Anaesthesia 2006;61:339-44. Crossref

29. Sia AT, Lim Y, Ocampo CE. Computer-integrated patientcontrolled epidural analgesia: a preliminary study on a novel approach of providing pain relief in labour. Singapore Med J 2006;47:951-6.

30. Sng BL, Sia AT, Lim Y, Woo D, Ocampo C. Comparison of computer-integrated patient-controlled epidural analgesia and patient-controlled epidural analgesia with a basal infusion for labour and delivery. Anaesth Intensive Care 2009;37:46-53. Crossref

31. Hogan Q. Distribution of solution in the epidural space: examination by cryomicrotome section. Reg Anesth Pain Med 2002;27:150-6. Crossref

32. Duncan LA, Fried MJ, Lee A, Wildsmith JA. Comparison of continuous and intermittent administration of extradural bupivacaine for analgesia after lower abdominal surgery. Br J Anaesth 1998;80:7-10. Crossref

33. Sia AT, Lim Y, Ocampo C. A comparison of a basal infusion with automated mandatory boluses in parturient-controlled epidural analgesia during labor. Anesth Analg 2007;104:673-8. Crossref

34. Leo S, Sia AT. Maintaining labour epidural analgesia: what is the best option? Curr Opin Anaesthesiol 2008;21:263-9. Crossref

35. Carvalho B, George RB, Cobb B, McKenzie C, Riley ET. Implementation of programmed intermittent epidural bolus for the maintenance of labor analgesia. Anesth Analg 2016;123:965-71. Crossref

36. George RB, Allen TK, Habib AS. Intermittent epidural bolus compared with continuous epidural infusions for labor analgesia: a systematic review and meta-analysis. Anesth Analg 2013;116:133-44. Crossref

37. Ojo OA, Mehdiratta JE, Gamez BH, Hunting J, Habib AS. Comparison of programmed intermittent epidural boluses with continuous epidural infusion for the maintenance of labor analgesia: a randomized, controlled, double-blind study. Anesth Analg 2020;130:426-35. Crossref

38. Ansari T, Yousef A, El Gamassy A, Fayez M. Ultrasound-guided spinal anaesthesia in obstetrics: is there an advantage over the landmark technique in patients with easily palpable spines? Int J Obstet Anesth 2014;23:213-6. Crossref

39. Arzola C, Mikhael R, Margarido C, Carvalho JC. Spinal ultrasound versus palpation for epidural catheter insertion in labour: a randomised controlled trial. Eur J Anaesthesiol 2015;32:499-505. Crossref

40. Perlas A, Chaparro LE, Chin KJ. Lumbar neuraxial ultrasound for spinal and epidural anesthesia: a systematic review and meta-analysis. Reg Anesth Pain Med 2016;41:251-60. Crossref

41. Creaney M, Mullane D, Casby C, Tan T. Ultrasound to identify the lumbar space in women with impalpable bony landmarks presenting for elective Caesarean delivery under spinal anaesthesia: a randomised trial. Int J Obstet Anesth 2016;28:12-6. Crossref

42. National Institute for Health and Care Excellence (NICE). Interventional procedures guidance (IPG249): Ultrasound-guided catheterisation of the epidural space. 23 Jan 2008.

43. Neal JM, Brull R, Horn JL, et al. The Second American Society of Regional Anesthesia and Pain Medicine evidence-based medicine assessment of ultrasound-guided regional anesthesia: executive summary. Reg Anesth Pain Med 2016;41:181-94. Crossref

44. Arzola C. Preprocedure ultrasonography before initiating a neuraxial anesthetic procedure. Anesth Analg 2017;124:712-3. Crossref

45. Kosh MC, Miller AD, Michels JE. Intravenous lipid emulsion for treatment of local anesthetic toxicity. Ther Clin Risk Manag 2010;6:449-51. Crossref

46. Weinberg GL. Lipid emulsion infusion: resuscitation for local anesthetic and other drug overdose. Anesthesiology 2012;117:180-7. Crossref

47. Wong CA, Scavone BM, Peaceman AM, et al. The risk of Cesarean delivery with neuraxial analgesia given early versus late in labor. N Engl J Med 2005;352:655-65. Crossref

48. Sng BL, Leong WL, Zeng Y, et al. Early versus late initiation of epidural analgesia for labour. Cochrane Database Syst Rev 2014;(10):CD007238. Crossref

49. Lim G, Levine MD, Mascha EJ, Wasan AD. Labor pain, analgesia, and postpartum depression: are we asking the right questions? Anesth Analg 2020;130:610-4. Crossref

50. Rendon KL, Wheeler V. Epidural analgesia and risk of Cesarean delivery. Am Fam Physician 2018 Nov 1;98:Online.

51. Chapter 9: Guidelines for the provision of anaesthesia services (GPAS). Guidelines for the Provision of Anaesthesia Services for an Obstetric Population. Royal College of Anaesthetists; 2020: 9.

52. Torvaldsen S, Roberts CL, Bell JC, Raynes-Greenow CH. Discontinuation of epidural analgesia late in labour for reducing the adverse delivery outcomes associated with epidural analgesia. Cochrane Database Syst Review 2004;(4):CD004457. Crossref

53. Tuuli MG, Frey HA, Odibo AO, Macones GA, Cahill AG. Immediate compared with delayed pushing in the second stage of labor: a systematic review and meta-analysis. Obstet Gynecol 2012;120:660-8. Crossref

54. Lee AI, McCarthy RJ, Toledo P, Jones MJ, White N, Wong CA. Epidural labor analgesia–fentanyl dose and breastfeeding success: a randomized clinical trial. Anesthesiology 2017;127:614-24. Crossref

55. Orbach-Zinger S, Landau R, Davis A, et al. The effect of labor epidural analgesia on breastfeeding outcomes: a prospective observational cohort study in a mixed-parity cohort. Anesth Analg 2019;129:784-91. Crossref

56. Chan JJ, Dabas R, Han RN, Sng BL. Fever during labour epidural analgesia. Trends in Anaesthesia and Crit Care 2018;20:21-5. Crossref

57. Higgins RD, Saade G, Polin RA, et al. Evaluation and management of women and newborns with a maternal diagnosis of chorioamnionitis: summary of a workshop. Obstet Gynecol 2016;127:426-36. Crossref

58. Sultan P, David AL, Fernando R, Ackland GL. Inflammation and epidural-related maternal fever: proposed mechanisms. Anesth Analg 2016;122:1546-53. Crossref

59. Lim G, Facco FL, Nathan N, Waters JH, Wong CA, Eltzschig HK. A review of the impact of obstetric anesthesia on maternal and neonatal outcomes. Anesthesiology 2018;129:192-215. Crossref

Recently, survival among patients with breast cancer has significantly improved. With better understanding of the disease’s diverse biology and increased availability of treatments, the 5-year survival rate for women diagnosed with breast cancer has increased from 75.2% between 1975 and 1977 to 88.2% between 2001 and 2008, leading to a substantial increase in breast cancer survivors.1 More strikingly, over 10% of breast cancer cases occur in women under age 45 years. Among these premenopausal survivors, 50% or more will live 20 years or longer following diagnosis. Thus, there is a need to address survivorship issues pertaining to long-term toxicity associated with breast cancer treatment.1 A study showed that 20 038 premenopausal women are diagnosed with breast cancer annually in the United States, with an estimated 96% (19 416) of these premenopausal patients at risk of infertility because of chemotherapy or hormonal therapy.1 Following chemotherapy, the reported incidence of amenorrhea varies between 40% and 68%.2 The agents most responsible for inducing amenorrhoea and premature ovarian failure (POF) are alkylating agents such as cyclophosphamide, whereas antimetabolites have a lesser effect.2 A brief summary of the association between infertility risk and type of chemotherapy can be seen in Table 1.3 4

Apart from chemotherapy, hormonal modulation with tamoxifen is beneficial for hormone receptor–positive disease, which accounts for 70% of breast cancers. A 5-year course of tamoxifen reduces recurrence by 47% and mortality by 26%.5 However, tamoxifen is teratogenic, and pregnancy is contra-indicated during treatment; hence, pregnancy is often postponed. Although tamoxifen may not directly damage the ovaries, several studies have reported its association with higher rates of treatment-related amenorrhea, particularly after age 40 years.5 It is likely that tamoxifen’s length of therapy may indirectly contribute to infertility alongside age-related fertility decline.5 Apart from difficulties with conception, premenopausal patients with breast cancer also experience high rates of spontaneous abortion (29%) and premature deliveries with low birth weight (40%).2

Minimising cancer treatment–associated infertility is especially important for premenopausal women who have not yet established a family. Female infertility in premenopausal women can cause these women great distress by preventing them from achieving the important life and social goal of motherhood. A recent study ranks this among the top five concerns among patients with premenopausal breast cancer.6 Moreover, in addition to impacting patients’ mental health, the fear of infertility also impacts treatment compliance.6 Of all patients, 29% do not comply with treatment because of treatment-associated infertility fears, which impacts patients’ prognosis and life expectancy.7 8 9

While these fears are increasingly recognised by doctors, recent literature suggests that 32% of patients do not recall discussing fertility with their doctors. This lack of communication could be caused by a lack of awareness or confidence discussing fertility management in the context of breast cancer. This phenomenon was highlighted by a finding that 37% of oncologists feared delaying chemotherapy for fertility preservation, and 49% of oncologists were concerned about pregnancy safety after breast cancer treatment.8 10 A Hong Kong study showed that only 45.6% of clinicians from diverse fields such as clinical oncology, haematology, obstetrics and gynaecology, paediatrics, and surgery were familiar with fertility preservation.11 Unfortunately, this lack of clinicians’ awareness is detrimental to premenopausal patients with breast cancer. Modern reproductive medicine can preserve female fertility in these patients.12 This review aims to summarise the existing fertility preservation options for patients with breast cancer to enable clinicians to have informed discussions with their patients.

A literature search was conducted using PubMed, Google Scholar, and Science Direct using the terms “breast cancer”, “fertility preservation”, “oocyte and embryonic cryopreservation”, “GnRH-a co-administration”, and “ovarian tissue cryopreservation and transplantation”. Only studies published in English from 1980 to 2019 were included. The focus of the literature review was on identifying the current fertility preservation methods available to premenopausal women with breast cancer and the barriers that impede access.

Chemotherapy accelerates follicular destruction, which reduces synthesis of inhibins and oestrogens. The use of adjuvant gonadotropin-releasing hormone agonist (GnRH-a) to limit chemotherapy-induced ovarian toxicity has been proposed by Glode et al.13 Gonadotropin-releasing hormone agonist has been shown to be protective against ovarian follicular depletion in mice.14 15 A current postulate on the protective mechanism of GnRH-a is the creation of a hypogonadotropic state. Reduced oestrogen and inhibin levels increase follicle-stimulating hormone (FSH) secretion via negative feedback. Consequently, supraphysiological FSH levels accelerate preantral follicle maturation and recruitment, which is vulnerable to chemotherapy. Gonadotropin-releasing hormone agonist induces pituitary desensitisation, preventing any increase in FSH concentration and hence minimising chemotherapy-induced follicle destruction.16 17

The evidence that GnRH-a reduces chemotherapy-associated POF and improves pregnancy rates is increasing.18 The use of GnRH-a to preserve ovarian function and fertility has recently been recommended as a reliable strategy for at least breast cancer.19 A pilot study undertaken by Recchia et al20 reported that patients <40 years with breast cancer who received chemotherapy with GnRH-a co-treatment of 3.6 mg goserelin every 28 days for 1 year resumed cyclic ovarian function. Following a median follow-up of 79 months, Recchia et al20 observed amenorrhea in none of the patients aged <40 years and 49% of patients aged >40 years. Four pregnancies were observed, three ended at term, and one was voluntarily terminated. Additionally, such a procedure did not affect the clinical outcomes of patients with breast cancer: after a median follow-up of 55 months, the disease-free survival and overall survival were 84% and 94%, respectively.20 The limitation of that study was the absence of a parallel control group. However, the observed excellent survival rates lessen the theoretical risk of hormonal manipulation in oestrogen-sensitive cancers. Other studies have also shown reduced onset of POF and no significant disruption of cyclical ovarian function.21 22

Subsequent trials such as PROMISE-GIM6 have built upon the work by Del Mastro et al.23 In the PROMISE-GIM6 trial, GnRH-a triptorelin 3.75 mg was administered intramuscularly at least 1 week prior to chemotherapy and subsequently every 4 weeks for the duration of chemotherapy. Patients were premenopausal women with stage I to III breast cancer who were offered adjuvant or neoadjuvant chemotherapy.23 The early menopause rate was 25.9% in the control group and 8.9% in the triptorelin group: an absolute difference of -17% (95% confidence interval [95% CI]=-26% to -7.9%; P<0.001) was observed.23 Another trial showed that in premenopausal women with either hormone-receptor-positive or hormone-receptor-negative breast cancer, concurrent administration of triptorelin and chemotherapy was associated with a higher long-term probability of recovery of ovarian function compared with chemotherapy alone, without a statistically significant difference in pregnancy rate.24 Another recent prospective, randomised, parallel group study using GnRH-a goserelin administered prior and throughout chemotherapy for patients with stage I-IIIB breast cancer showed a reduction in the rate of POF for women aged <40 years.7 Goserelin reduced the prevalence of amenorrhoea between 12 and 24 months from 38% in the control group to 22% in the treated group. The prevalence of POF was also reduced to 18.5% from 34.8% in the control group.7 Finally, a meta-analysis of randomised control trials involving GnRH-a during chemotherapy in patients with premenopausal breast cancer showed a reduced rate of POF and increased pregnancy rate without negative prognostic consequences.24 The meta-analysis included 12 trials involving 1231 breast cancer patients, and the data showed that GnRH-a was associated with a significantly reduced risk of POF (odds ratio=0.36, 95% CI=0.23-0.57; P<0.001).25 Among the five trials that reported pregnancies, more patients treated with GnRH-a achieved pregnancy (33 vs 19 women; odds ratio=1.83, 95% CI=1.02-3.28; P=0.041). In the three studies that reported disease-free survival, no between-group difference was observed (hazard ratio=1.00, 95% CI=0.49-2.04; P=0.939).25 These results suggest that GnRH-a provides improved fertility preservation for female patients with breast cancer without affecting cancer progression and survival rates.24 However, data regarding live births (which are the primary goal of fertility preservation) following GnRH-a administration have been relatively scant. Currently, it seems that administration of GnRH-a to patients with breast cancer is a potentially beneficial fertility preservation strategy. The ease of GnRH-a co-administration also has the potential to benefit patients outside tertiary and university medical centres.

Embryo cryopreservation is the most established fertility preservation technique and has entered routine clinical practice. Following oocyte harvesting, oocytes can be fertilised in vitro by donor or partner sperm and the embryos cryopreserved. The benefit of this technique is that embryos tend to survive cryopreservation better than oocytes. The improvements in vitrification technology have led to an even higher embryo survival rate.

An alternative to embryo cryopreservation is mature oocyte cryopreservation. After embryo cryopreservation, this is the second-most established technique, and it has entered clinical practice. An advantage to this technique over embryo cryopreservation is that it does not require sperm from a donor or partner, which is more suitable for single women.

Vitrification has enabled mature oocyte cryopreservation by improving oocyte survival rates and clinical outcomes.26 A prospective, randomised study conducted with healthy young oocyte donors showed that a 97% survival rate was obtained through this technology.26 Traditional cryopreservation exposes cells to low temperatures for prolonged periods causing, cytoplasmic ice crystal formation, which compromises cell survival upon thawing.27 Vitrification is solidification that occurs without ice crystallisation but through extreme elevation in viscosity. This phenomenon is achieved using high cooling rates from -15 000 to -30 000°C/min, minimising ice crystal formation. Vitrification has been optimised to reduce cryoprotectant concentration and subsequent cytotoxicity.28

A study on mature oocyte cryopreservation via vitrification for non-oncological patients yielded 693 oocytes, of which 666 (96.1%) survived.26 A total of 487 (73.1%) were then successfully fertilised, leading to 117 embryos transferred to 57 recipients. The pregnancy rate and implantation rate per transfer were 63.2% and 38.5%, respectively, resulting in 28 healthy babies born.26 An overview of the work of several research groups shows that oocyte survival rates using this method have ranged between 91% and 99%, fertilisation rates were between 87% and 91%, and pregnancy rates were between 33% and 57%.29 30 31 Subsequent work was done in cancer patients: 357 cancer patients had their oocytes cryopreserved, and 11 patients returned post-treatment for in vitro fertilisation (IVF).32 The oocyte survival rate was 92.3%, the fertilisation rate was 76.6%, and average number of embryos transferred was 1.8 ± 0.7. Four live births at term were achieved with no malformations.32 Table 2 summarises some outcomes of the recent studies on oocyte cryopreservation for fertility preservation in cancer.32 33

Immature oocyte cryopreservation is another extension of oocyte cryopreservation technology. This technique is far less established, but it enables immature eggs to be removed before chemotherapy and have the eggs matured in vitro. This technique is suitable for single female patients who have oncological emergencies and cannot delay chemotherapy for ovarian stimulation. The benefit of immature oocytes over mature oocytes is their survivability: immature oocytes are relatively cryoresistant because of their smaller size and lack of meiotic spindle. The greatest challenge for clinical translation of immature oocyte cryopreservation technology is the difficulties encountered during in vitro maturation of immature oocytes. Currently, only a single live birth has been reported following the slow freezing of immature oocytes.34

Embryonic or oocyte cryopreservation is a proven approach in fertility preservation, especially in chemotherapy.35 For both types of cryopreservation, a period of 8 to 12 days is needed for ovarian stimulation and subsequent oocyte harvesting. Traditionally but now uncommonly performed, the ovarian stimulation protocol begins with GnRH-a administration during the preceding cycle’s luteal phase to promote ovarian quiescence followed by daily gonadotropin injections. Serial measurements of oestradiol levels and follicular diameter are taken for monitoring. When there are more than three dominant follicles present, human chorionic gonadotropin is administered to trigger ovulation and oocytes collected.36 This protocol has many disadvantages, chiefly delay of the commencement of chemotherapy and high oestrogen exposure (which leads to increased risk of breast cancer progression and recurrence, particularly in hormone-sensitive breast cancer). Therefore, various new stimulation protocols have developed to enable either a shorter stimulation period or a lower oestrogen level during stimulation.

Random start protocols have been proposed to minimise the time for oocyte collection by decreasing total duration of the IVF cycle and have been shown to be equally effective as conventional start protocols in terms of the total number of mature oocytes retrieved, oocyte maturity rate, and fertilisation rate.37 38 39 Other novel stimulation protocols with tamoxifen or aromatase inhibitors have been developed to increase the safety margin of ovarian stimulation in patients with oestrogen-sensitive tumours. Tamoxifen is a selective oestrogen receptor modulator. In addition to its anti-oestrogenic action in breast tissue, it acts as an antagonist in the central nervous system and interferes with the negative feedback exerted by oestrogen on the hypothalamic/pituitary axis, leading to an increase in GnRH secretion from the hypothalamus. A few studies have explored tamoxifen use for ovarian stimulation in patients with breast cancer.40 41 42 A higher number of mature oocytes and subsequent embryos were obtained from the tamoxifen group compared with natural cycle, with at least one embryo generated per tamoxifen patient. Two patients conceived, one miscarried at 8 weeks of pregnancy, but her risk of spontaneous abortion was high at age 42 years. The other patient delivered a healthy set of twins.40 In a more recent study, co-administration of tamoxifen with ovarian stimulation for fertility preservation did not interfere with IVF results. Comparisons were made between those who did and did not receive concurrent tamoxifen. The mean percentages of oocytes retrieved were 12.65% and 10.2%, respectively, and the numbers of embryos cryopreserved were 8.5 and 6.4, respectively.42 Patients co-treated with tamoxifen had marginally higher oestradiol levels, but the difference was not statistically significant. However, co-treatment with tamoxifen was considered to be safe, as the long-term recurrence risk at up to 10 years was not increased.42

Aromatase inhibitors (eg, letrozole) significantly suppress plasma oestrogen levels by competitively inhibiting aromatase enzyme activity. Centrally, it releases the hypothalamic/pituitary axis from oestrogenic negative feedback, increasing secretion of FSH by the pituitary gland while increasing the FSH sensitivity of the ovarian granulosa cells. Combined letrozole-FSH protocols have resulted in oestradiol levels lower than those seen in natural cycles alongside fertility outcomes similar to standard IVF protocols.43 44 In a recent trial, 131 women with stage ≤3 breast cancer underwent ovarian stimulation with concurrent daily letrozole 5 mg prior to cryopreserving embryos and subsequent adjuvant chemotherapy. The overall live birth rate per embryo transfer was similar to the United States national mean among infertile women of a similar age without cancer who underwent IVF–embryo transfer (45.0 vs 38.2; P=0.2).45 Another trial highlighted the safety, feasibility, and utility of two consecutive ovarian stimulation cycles with the use of letrozole-gonadotropin protocol for fertility preservation in patients with breast cancer.46 The mean total number of oocytes harvested (16.1 ± 13.2 vs 9.1 ± 5.2) and embryos generated (6.4 ± 2.9 vs 3.7 ± 3.1) were significantly higher in patients who underwent two cycles compared with those who underwent one cycle.46 There was no significant delay in time interval from surgery to chemotherapy between the two-cycle and single-cycle groups (63.7 ± 7.7 vs 58.0 ± 12.1 days, respectively). The recurrence rate was similar between two-cycle (0 of 17) and single-cycle (2 of 49) patients.46

Cryopreservation and autotransplantation of ovarian tissue are emerging technologies, and women considering such treatments should do so judiciously under specialised expertise in the setting of clinical trials. These techniques rely upon slow freezing technology used to cryopreserve oocytes and embryos, but it is more difficult to optimise the procedure because ovaries contain many different cell types. Upon freezing and thawing, problems occur with fertilisation of maturing oocytes.47 The best follicular survival rate is approximately 70% to 80%, with light microscopy revealing normal follicles. However, electron microscopy has detected ultrastructural changes.48 The benefits of autotransplantation of ovarian tissue include restoration of endocrine and reproductive function; however, greater clinical evaluation of this technology is needed. Autotransplantation of ovarian tissue can be orthotopic or heterotopic. Orthotopic refers to transplantation of ovarian tissue to the original ovary site, while heterotopic refers to transplanting the ovarian tissue to a foreign site. Currently, the most effective site for graft longevity is still unknown. Orthotopic ovarian transplantation allows for natural conception, while heterotopic transplantation in accessible sites minimises repeated invasive procedures and improves ease of oocyte recovery.49

Oktay et al50 reported restoration of hormonal function, follicular growth, and oocyte retrieval after heterotopic transplantation in a patient with breast cancer. They retrieved 20 oocytes from subcutaneously implanted ovarian tissue 6 years after chemotherapy resulting in one fertilisation, but no pregnancy ensued.50 There have since been case reports of spontaneous pregnancy and live births after autotransplantation of cryopreserved human ovarian tissue in patients with cancer.51 52 An early case report described a patient with triple negative medullary breast cancer undergoing ovarian tissue cryopreservation. Following cancer treatment, she underwent an ovarian tissue transplant, and menses occurred 63 days after transplantation. Sixteen mature oocytes were harvested following four sessions of ovarian stimulation. All vitrified oocytes survived thawing, and 77.7% were fertilised. Two day 3 embryos were implanted, and two healthy boys were born at 34 weeks.53

As of 2017, there have been an estimated 86 successful births and nine ongoing pregnancies using cryopreserved ovarian tissue internationally.54 The majority of the patient population that has undergone this procedure has a cancer diagnosis, of which many are haematological malignancies, which require urgent chemotherapy. Furthermore, of the singleton pregnancies, the mean gestational age was 39 weeks, and the mean birth weight was 3168 g, which are within normal standards.54 These early results suggest that ovarian tissue preservation and subsequent transplantation might become a suitable fertility preservation therapy in premenopausal women. A summary of a large case series of live births from ovarian tissue transplantation is shown in Table 3. It focuses on the perinatal outcomes of 40 live births from 32 women.54 The reference also briefly compiles the 86 live births and nine ongoing pregnancies after transplantation of frozen-thawed ovarian cortex.54

The advantage of ovarian cryopreservation and transplantation is that it does not require an ovarian stimulation protocol, which delays cancer treatment. Additionally, this procedure is especially suitable for prepubescent cancer patients, in whom ovarian stimulation is contra-indicated. Moreover, ovarian cryopreservation and transplantation not only restores fertility but also restores gonadal/endocrine function. Finally, because hundreds of immature oocytes can be harvested at once, a huge ovarian reserve can be preserved. The disadvantage of ovarian tissue cryopreservation and transplantation is that it requires at least two surgical operations: one for removal and another for future re-implantation. Following implantation, there is the challenge of minimising ischaemia, which could lead to follicle loss or initiate maturation of the immature oocytes. To minimise follicle loss post-transplant, the entire ovarian cortex is often cryopreserved. Another potential concern of autotransplantation is the risk of cancer cell transmission, which has the highest probability in cases of haematological cancers. Shaw et al55 reported lymphoma in mice with fresh or cryobanked grafted ovarian tissue from donor mice with lymphoma. Clinically, however, ovarian metastases are uncommon in cancers affecting young people.56 Kim et al57 reported in a mouse model that ovarian tissue from patients with high-grade lymphoma appears safe for autotransplantation, as none of the grafted mice tested positive for lymphoma. Further research is required to assess the optimal site for transplantation, improve methods of detecting residual disease in harvested tissue, ascertain the optimal size of ovarian grafts, optimise freezing/thawing techniques, and promote re-vascularisation of the transplanted tissue.

In the context of Asia, the Asian Society for Fertility Preservation (ASFP) was established to promote the science and clinical application of fertility preservation techniques. The members of the ASFP include China, Hong Kong, Taiwan, Singapore, Korea, Japan, Vietnam, India, Thailand, Indonesia, the Philippines, and Pakistan. This review will focus broadly on fertility preservation care services in East Asia (ie, China, Singapore, Korea, and Japan), as seen in Table 4.

In Japan, oocyte, embryo, and ovarian tissue cryopreservation are available. However, despite the availability of the technology, uptake of fertility preservation techniques has been limited, as the majority of cancer hospitals do not provide fertility preservation services.58 This problem is further compounded by the lack of robust coordinated care networks to assist in delivering oncofertility care.58 To overcome this problem, the Japanese have been aggressively experimenting with various referral service models: (1) a reproductive care centre-led model wherein reproductive care centres reach out to cancer centres, and (2) a cancer centre model in which the cancer centre serves as the basis of the referral network.59 To facilitate the running of the referral network, the Japanese have focused on harnessing allied health to drive a psychosocial-based care delivery system.59 They aim to train oncofertility navigators that are able to provide psychosocial care whilst also guiding patients on the technical aspects of their fertility preservation journey.59

The efforts of the Korean Society for Fertility Preservation (KSFP) have produced a well-established referral network that even covers regional healthcare centres, rendering high-quality fertility preservation treatments accessible at various institutions.58 60 Fertility preservation treatments have a multidisciplinary focus incorporating physicians, nurses, mental health professionals, office staff, and laboratory personnel.58 60 To facilitate patient communication regarding fertility preservation under the time pressure of cancer treatment, print material and web resources are distributed to patients during the fertility preservation consultation.58 60 However, despite this concerted effort, uptake of fertility preservation treatments has been limited.60 The main issue raised by the KSFP was access to care: the oncologists noted that there was a lack of discussion of fertility preservation options and referrals to fertility specialists.60 This may be for several reasons: exclusive focus on cancer treatment and its perceived urgency, the perception of limited options for fertility preservation, the perception that fertility is unimportant to patients, and not knowing the referral pathway for patients interested in fertility preservation.60

Like many East Asian countries, China has the technology to perform fertility preservation techniques. However, similarly, the limiting factor for uptake of these technologies is barriers to access. Like in Korea, there is a lack of knowledge and awareness of fertility preservation techniques among healthcare professionals: among obstetrics and gynaecology specialists, despite knowing about cancer therapy being gonadotoxic, about 20% of them were not familiar with fertility preservation techniques.61 Only 50% of obstetrics and gynaecology specialists who were familiar with fertility preservation techniques had been consulted by oncologists about managing a patient’s infertility risk.61 Despite this, 96.6% of obstetrics and gynaecology specialists reported being keen on collaborating with oncologists to preserve the fertility of female patients with cancer.61 This suggests that despite a willingness to collaborate on fertility preservation, there has been limited communication between oncologists and obstetrics and gynaecology specialists.

Finally, Singapore, being a small city-state, has limited resources for fertility preservation care delivery services: in fact, it does not have a fertility preservation society.58 All fertility preservation services are concentrated in one large tertiary hospital in Singapore.58 Reproductive preservation techniques exist in Singapore; however, access to them is again limited by both the patient’s and oncologist’s knowledge and awareness of the technology.58

Fertility preservation is a major concern for premenopausal patients. Hence, at the outset of chemotherapy, its gonadotoxic effects must be discussed by a multidisciplinary team. When discussing fertility preservation options, their nature, success rate, risk, cost, and potential ethical implications should be discussed with the patient.62 Additionally, it is important to set realistic expectations about the fertility preservation treatment with the patient: factors such as patient age, ovarian reserve, presence of a partner, presence of previous live births, financial status, and religious background should be considered.62 Finally, medical factors that may influence the feasibility of fertility preservation, such as severity of the gonadotoxic chemotherapy, time available before commencing chemotherapy, and available expertise and facilities, should also be explained to the patient. A sample decision algorithm for deciding about the suitability of fertility preservation options in both emerging and routine clinical practice is shown in the Figure.

Another common patient concern is cancer recurrence or disease progression due to future pregnancy. This fear may cause patients to delay or abandon future pregnancies. A meta-analysis showed that women who become pregnant following breast cancer treatment have an improved prognosis, reflected by significantly increased overall survival compared with those who did not become pregnant (pooled hazard ratio=0.63, 95% CI=0.51-0.79).63 The meta-analysis also revealed a non-significant increase in disease-free survival for pregnant women.63 The meta-analysis findings, however, may be due to a selection bias termed the “healthy mother” effect, where healthier women are more likely to conceive than those who have relapsed hence skewing the true effect.64 Nonetheless a subsequent multicentre case-control study has supported the conclusion of the meta-analysis regarding pregnancy safety following breast cancer treatment.65 At 7.2 years after pregnancy, no difference in disease-free survival was observed between pregnant and nonpregnant patients.65 Although there was no difference in overall survival for oestrogen receptor–positive pregnant patients, there was an increased overall survival for oestrogen receptor–negative patients.65 Therefore, pregnancy is being considered safe following previous breast cancer diagnosis and maybe associated with an improved prognosis for oestrogen receptor–negative patients.

All of the patient’s fears and concerns regarding fertility preservation should be discussed with a dedicated oncofertility unit. This allows for informed discussion between patient and healthcare practitioner, which can help to allay patients’ fears. Such communication can be facilitated via printouts, brochures, and decision aids, which can help to avoid miscommunication and allow patients to be fully informed of their potential choices in fertility preservation. Research has shown that a dedicated oncofertility unit can improve the frequency and thoroughness of fertility preservation discussions.66 This research is particularly relevant, as oncofertility units allow focus on young cancer patients, including those with breast cancer, who will benefit the most from fertility preservation options.

Fertility preservation is still a major issue for premenopausal patients with breast cancer. Several treatment modalities can now be considered and combined.67 A potential fertility preservation protocol for premenopausal patients with breast cancer could involve an initial oocyte and ovarian tissue harvest and subsequent cryopreservation of oocytes, embryos, and ovarian tissue. Gonadotropin-releasing hormone agonist can be co-administrated alongside chemotherapy to minimise POF.

This review aimed to summarise and evaluate the current clinical status of fertility preservation techniques available to premenopausal patients with breast cancer, thereby raising awareness of fertility preservation techniques among oncologists, fertility specialists, surgeons, nurses, and psychologists who care for premenopausal patients with breast cancer. Hopefully, a multidisciplinary and holistic approach to fertility preservation treatments for premenopausal patients with breast cancer will be possible in East Asia.

Concept or design: SSY Wang.
Acquisition of data: SSY Wang.
Analysis or interpretation of data: All authors.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: All authors.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

As editors of the journal, H Loong and JPW Chung were not involved in the peer review process. Other authors have no conflicts of interest to declare.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1. Trivers KF, Fink AK, Partridge AH, et al. Estimates of young breast cancer survivors at risk for infertility in the U.S. Oncologist 2014;19:814-22. Crossref

2. Kasum M, Beketić-Orešković L, Peddi PF, Orešković S, Johnson RH. Fertility after breast cancer treatment. Eur J Obstet Gynecol Reprod Biol 2014;173:13-8. Crossref

3. Levine J. Gonadotoxicity of cancer therapies in pediatric and reproductive-age females. In: Gracia C, Woodruff TK, editors. Oncofertility Medical Practice. New York: Springer; 2012: 3-14. Crossref

4. Yuksel A, Bildik G, Senbabaoglu F, et al. The magnitude of gonadotoxicity of chemotherapy drugs on ovarian follicles and granulosa cells varies depending upon the category of the drugs and the type of granulosa cells. Human Reprod 2015;30:2926-35. Crossref

5. Llarena NC, Estevez SL, Tucker SL, Jeruss JS. Impact of fertility concerns on tamoxifen initiation and persistence. J Natl Cancer Inst 2015;107(10):djv202. Crossref

6. Howard-Anderson J, Ganz PA, Bower JE, Stanton AL. Quality of life, fertility concerns, and behavioral health outcomes in younger breast cancer survivors: a systematic review. J Natl Cancer Inst 2012;104:386-405. Crossref

7. Paluch-Shimon S, Pagani O, Partridge AH, et al. ESO-ESMO 3rd international consensus guidelines for breast cancer in young women (BCY3). Breast 2017;35:203-17. Crossref

8. Ruddy KJ, Gelber SI, Tamimi RM, et al. Prospective study of fertility concerns and preservation strategies in young women with breast cancer. J Clin Oncol 2014;32:1151-6. Crossref

9. Partridge AH, Gelber S, Peppercorn J, et al. Web-based survey of fertility issues in young women with breast cancer. J Clin Oncol 2004;22:4174-83. Crossref

10. Biglia N, Torrisi R, D’Alonzo M, Codacci Pisanelli G, Rota S, Peccatori FA. Attitudes on fertility issues in breast cancer patients: an Italian survey. Gynecol Endocrinol 2015;31:458-64. Crossref

11. Chung JP, Lao T, Li TC. Evaluation of the awareness of, attitude to, and knowledge about fertility preservation in cancer patients among clinical practitioners in Hong Kong. Hong Kong Med J 2017;23:556-61. Crossref

12. Doyle JO, Richter KS, Lim J, Stillman RJ, Graham JR, Tucker MJ. Successful elective and medically indicated oocyte vitrification and warming for autologous in vitro fertilization, with predicted birth probabilities for fertility preservation according to number of cryopreserved oocytes and age at retrieval. Fertil Steril 2016;105:459-66. e2. Crossref

13. Glode LM, Robinson J, Gould SF. Protection from cyclophosphamide-induced testicular damage with an analogue of gonadotropin-releasing hormone. Lancet 1981;317:1132-4. Crossref

14. Bohlmann MK, von Wolff M, Strowitzki T. Comment on the symposium article “Fertility after treatment for Hodgkin’s disease”, by Z. Blumenfeld, E. Dann, I. Avivi et al. (Ann Oncol 2002;13 Suppl 1:138-147). Ann Oncol 2003;14:499. Crossref

15. Oktay K, Sönmezer M, Öktem Ö, Fox K, Emons G, Bang H. Absence of conclusive evidence for the safety and efficacy of gonadotropin-releasing hormone analogue treatment in protecting against chemotherapy-induced gonadal injury. Oncologist 2007;12:1055-66. Crossref

16. Blumenfeld Z. How to preserve fertility in young women exposed to chemotherapy? The role of GnRH agonist cotreatment in addition to cryopreservation of embrya, oocytes, or ovaries. Oncologist 2007;12:1044-54. Crossref

17. Lobo RA. Potential options for preservation of fertility in women. N Engl J Med 2005;353:64-73. Crossref

18. Blumenfeld Z, von Wolff M. GnRH-analogues and oral contraceptives for fertility preservation in women during chemotherapy. Hum Reprod Update 2008;14:543-52. Crossref

19. Lambertini M, Del Mastro L, Pescio MC, et al. Cancer and fertility preservation: international recommendations from an expert meeting. BMC Med 2016;14:1. Crossref

20. Recchia F, Sica G, De Filippis S, Saggio G, Rosselli M, Rea S. Goserelin as ovarian protection in the adjuvant treatment of premenopausal breast cancer: a phase II pilot study. Anticancer Drugs 2002;13:417-24. Crossref

21. Huser M, Crha I, Ventruba P, et al. Prevention of ovarian function damage by a GnRH analogue during chemotherapy in Hodgkin lymphoma patients. Hum Reprod 2008;23:863-8. Crossref

22. Blumenfeld Z, Avivi I, Eckman A, Epelbaum R, Rowe JM, Dann EJ. Gonadotropin-releasing hormone agonist decreases chemotherapy-induced gonadotoxicity and premature ovarian failure in young female patients with Hodgkin lymphoma. Fertil Steril 2008;89:166-73. Crossref

23. Del Mastro L, Boni L, Michelotti A, et al. Effect of the gonadotropin-releasing hormone analogue triptorelin on the occurrence of chemotherapy-induced early menopause in premenopausal women with breast cancer: A randomized trial. JAMA 2011;306:269-76. Crossref

24. Lambertini M, Boni L, Michelotti A, et al. Ovarian suppression with triptorelin during adjuvant breast cancer chemotherapy and long-term ovarian function, pregnancies, and disease-free survival: A randomized clinical trial. JAMA 2015;314:2632-40. Crossref

25. Lambertini M, Ceppi M, Poggio F, et al. Ovarian suppression using luteinizing hormone-releasing hormone agonists during chemotherapy to preserve ovarian function and fertility of breast cancer patients: a meta-analysis of randomized studies. Ann Oncol 2015;26:2408-19. Crossref

26. Cobo A, Domingo J, Pérez S, Crespo J, Remohí J, Pellicer A. Vitrification: an effective new approach to oocyte banking and preserving fertility in cancer patients. Clin Transl Oncol 2008;10:268-73. Crossref

27. Edgar DH, Gook DA. A critical appraisal of cryopreservation (slow cooling versus vitrification) of human oocytes and embryos. Hum Reprod Update 2012;18:536-54. Crossref

28. Kuwayama M. Highly efficient vitrification for cryopreservation of human oocytes and embryos: The Cryotop method. Theriogenology 2007;67:73-80. Crossref

29. Kuwayama M, Vajta G, Kato O, Leibo SP. Highly efficient vitrification method for cryopreservation of human oocytes. Reprod Biomed Online 2005;11:300-8. Crossref

30. Lucena E, Bernal DP, Lucena C, Rojas A, Moran A, Lucena A. Successful ongoing pregnancies after vitrification of oocytes. Fertil Steril 2006;85:108-11. Crossref

31. Antinori M, Licata E, Dani G, Cerusico F, Versaci C, Antinori S. Cryotop vitrification of human oocytes results in high survival rate and healthy deliveries. Reprod Biomed Online 2007;14:72-9. Crossref

32. Martinez M, Rabadan S, Domingo J, Cobo A, Pellicer A, Garcia-Velasco JA. Obstetric outcome after oocyte vitrification and warming for fertility preservation in women with cancer. Reprod Biomed Online 2014;29:722-8. Crossref

33. Cobo A, García-Velasco J, Domingo J, Pellicer A, Remohí J. Elective and onco-fertility preservation: factors related to IVF outcomes. Hum Reprod 2018;33:2222-31. Crossref

34. Tucker MJ, Wright G, Morton PC, Massey JB. Birth after cryopreservation of immature oocytes with subsequent in vitro maturation. Fertil Steril 1998;70:578-9. Crossref

35. Kim SS. Fertility preservation in female cancer patients: current developments and future directions. Fertil Steril 2006;85:1-11. Crossref

36. Hacker NF, Gambone JC, Hobel CJ. Hacker & Moore’s essentials of obstetrics and gynecology. 6th ed. Philadelphia: Elsevier Health Sciences; 2015.

37. Cakmak H, Rosen MP. Random-start ovarian stimulation in patients with cancer. Curr Opin Obstet Gynecol 2015;27:215-21. Crossref

38. Cakmak H, Katz A, Cedars MI, Rosen MP. Effective method for emergency fertility preservation: random-start controlled ovarian stimulation. Fertil Steril 2013;100:1673- 80. Crossref

39. Kuang Y, Hong Q, Chen Q, et al. Luteal-phase ovarian stimulation is feasible for producing competent oocytes in women undergoing in vitro fertilization/intracytoplasmic sperm injection treatment, with optimal pregnancy outcomes in frozen-thawed embryo transfer cycles. Fertil Steril 2014;101:105-11. Crossref

40. Oktay K, Buyuk E, Davis O, Yermakova I, Veeck L, Rosenwaks Z. Fertility preservation in breast cancer patients: IVF and embryo cryopreservation after ovarian stimulation with tamoxifen. Hum Reprod 2003;18:90-5. Crossref

41. Oktay K, Buyuk E, Libertella N, Akar M, Rosenwaks Z. Fertility preservation in breast cancer patients: a prospective controlled comparison of ovarian stimulation with tamoxifen and letrozole for embryo cryopreservation. J Clin Oncol 2005;23:4347-53. Crossref

42. Meirow D, Raanani H, Maman E, et al. Tamoxifen co-administration during controlled ovarian hyperstimulation for in vitro fertilization in breast cancer patients increases the safety of fertility-preservation treatment strategies. Fertil Steril 2014;102:488-95.e3. Crossref

43. Oktay K, Hourvitz A, Sahin G, et al. Letrozole reduces estrogen and gonadotropin exposure in women with breast cancer undergoing ovarian stimulation before chemotherapy. J Clin Endocrinol Metab 2006;91:3885-90. Crossref

44. Azim AA, Costantini-Ferrando M, Oktay K. Safety of fertility preservation by ovarian stimulation with letrozole and gonadotropins in patients with breast cancer: a prospective controlled study. J Clin Oncol 2008;26:2630-5. Crossref

45. Oktay K, Turan V, Bedoschi G, Pacheco FS, Moy F. Fertility preservation success subsequent to concurrent aromatase inhibitor treatment and ovarian stimulation in women with breast cancer. J Clin Oncol 2015;33:2424-9. Crossref

46. Turan V, Bedoschi G, Moy F, Oktay K. Safety and feasibility of performing two consecutive ovarian stimulation cycles with the use of letrozole-gonadotropin protocol for fertility preservation in breast cancer patients. Fertil Steril 2013;100:1681-5.e1.Crossref

47. Oktay K, Karlikaya G. Ovarian function after transplantation of frozen, banked autologous ovarian tissue. N Engl J Med 2000;342:1919. Crossref

48. Kim SS, Battaglia DE, Soules MR. The future of human ovarian cryopreservation and transplantation: fertility and beyond. Fertil Steril 2001;75:1049-56. Crossref

49. Donnez J, Martinez-Madrid B, Jadoul P, Van Langendonckt A, Demylle D, Dolmans MM. Ovarian tissue cryopreservation and transplantation: a review. Hum Reprod Update 2006;12:519-35. Crossref

50. Oktay K, Buyuk E, Veeck L, et al. Embryo development after heterotopic transplantation of cryopreserved ovarian tissue. Lancet 2004;363:837-40. Crossref

51. Demeestere I, Simon P, Emiliani S, Delbaere A, Englert Y. Fertility preservation: successful transplantation of cryopreserved ovarian tissue in a young patient previously treated for Hodgkin’s disease. Oncologist 2007;12:1437-42. Crossref

52. Donnez J, Dolmans MM, Demylle D, et al. Livebirth after orthotopic transplantation of cryopreserved ovarian tissue. Lancet 2004;364:1405-10. Crossref

53. Sánchez-Serrano M, Crespo J, Mirabet V, et al. Twins born after transplantation of ovarian cortical tissue and oocyte vitrification. Fertil Steril 2010;93:268.e11-3. Crossref

54. Jensen AK, Macklon KT, Fedder J, Ernst E, Humaidan P, Andersen CY. 86 successful births and 9 ongoing pregnancies worldwide in women transplanted with frozen-thawed ovarian tissue: focus on birth and perinatal outcome in 40 of these children. J Assist Reprod Genet 2017;34:325-36. Crossref

55. Shaw J, Bowles J, Koopman P, Wood E, Trounson A. Fresh and cryopreserved ovarian tissue samples from donors with lymphoma transmit the cancer to graft recipients. Hum Reprod 1996;11:1668-73. Crossref

56. Lee SJ, Bae JH, Lee AW, Tong SY, Park YG, Park JS. Clinical characteristics of metastatic tumors to the ovaries. J Korean Med Sci 2009;24:114-9. Crossref

57. Kim SS, Radford J, Harris M, et al. Ovarian tissue harvested from lymphoma patients to preserve fertility may be safe for autotransplantation. Hum Reprod 2001;16:2056-60. Crossref

58. Harzif AK, Santawi VPA, Maidarti M, Wiweko B. Investigation of each society for fertility preservation in Asia. Front Endocrinol (Lausanne) 2019;10:151. Crossref

59. Takai Y. Recent advances in oncofertility care worldwide and in Japan. Reprod Med Biol 2018;17:356-68. Crossref

60. Kim J, Kim SK, Hwang KJ, Kim SH. Fertility preservation during cancer treatment: The Korean Society for Fertility Preservation clinical guidelines. Clin Exp Reprod Med 2017;44:171-4. Crossref

61. Ju K, Kopp M, Wang Y, et al. A survey study of attitude and knowledge regarding female fertility preservation among reproductive health professionals in Fujian, China. J Adolesc Young Adult Oncol 2018;8:67-73. Crossref

62. Tomasi-Cont N, Lambertini M, Hulsbosch S, Peccatori AF, Amant F. Strategies for fertility preservation in young early breast cancer patients. Breast 2014;23:503-10. Crossref

63. Hartman EK, Eslick GD. The prognosis of women diagnosed with breast cancer before, during and after pregnancy: a meta-analysis. Breast Cancer Res Treat 2016;160:347-60. Crossref

64. Sankila R, Heinävaara S, Hakulinen T. Survival of breast cancer patients after subsequent term pregnancy: “healthy mother effect”. Am J Obstet Gynecol 1994;170:818-23. Crossref

65. Lambertini M, Kroman N, Ameye L, et al. Long-term safety of pregnancy following breast cancer according to estrogen receptor status. J Natl Cancer Inst 2018;110:426- 9. Crossref

66. Lewin J, Ma JMZ, Mitchell L, et al. The positive effect of a dedicated adolescent and young adult fertility program on the rates of documentation of therapy-associated infertility risk and fertility preservation options. Support Care Cancer 2017;25:1915-22. Crossref

67. Blumenfeld Z, Katz G, Evron A. ‘An ounce of prevention is worth a pound of cure’: the case for and against GnRHagonist for fertility preservation. Ann Oncol 2014;25:1719- 28. Crossref