Deep vein thrombosis (DVT) is a major cause of morbidity and mortality, as it can lead to post-thrombotic syndrome (PTS) and pulmonary embolism. According to the American College of Chest Physicians treatment guidelines, DVT has conventionally been treated with low-molecular-weight heparin, unfractionated heparin, or fondaparinux followed by vitamin K antagonists for at least 3 months.1 This recommended regimen is adequate for prevention of thrombus extension, but its effect on clot lysis is minimal. The reported 6-month venous patency rate in patients treated with anticoagulation alone was only 47.4%. Eventually, up to 55.6% of patients with iliofemoral DVT developed PTS as a result of valvular dysfunction.2 Up to 5% to 10% of patients had severe PTS in the form of venous ulceration, which caused significant morbidity and socio-economic cost.3

In view of the suboptimal treatment outcomes of anticoagulation, aggressive means have been developed to achieve early restoration of venous patency and thus preservation of valvular function. A Cochrane review suggested that early thrombus removal by means of systemic thrombolysis can prevent venous dysfunction and PTS. However, its use was limited by its significantly increased risk of bleeding.4

Endovascular modalities including catheter-directed thrombolysis (CDT) and percutaneous mechanical thrombectomy (PMT) were developed to achieve accelerated thrombolysis with less bleeding risk. Catheter-directed thrombolysis was shown to be superior to anticoagulation alone in terms of higher thrombolysis rate and lower rates of recurrence and PTS.2 It features loco-regional delivery of thrombolytic agent over the DVT site via a transluminal catheter. The dosage of the thrombolytic agent can be reduced compared with that of systemic thrombolysis, and thus, a reduction in bleeding complications can be achieved. Its benefits have been validated by a number of randomised controlled trials and meta-analyses, but its application rate remains low because of its substantial bleeding risk and cost.5

Percutaneous mechanical thrombectomy is another form of endovascular treatment, in which thrombectomy devices are passed to the site of DVT and blood clots are removed by different mechanical means. It can also be used as an adjunctive device to CDT or pharmacomechanical thrombectomy. When these two devices are used in combination, the dosage of thrombolytic agents can be lowered further and the duration of procedure can be shortened.6 According to the American College of Chest Physicians guidelines, PMT provides the greatest benefits for young and functionally active patients with acute presentation (<14 days, or presence of phlegmasia cerulea dolens) of extensive proximal DVT.1 Percutaneous mechanical thrombectomy has provided promising results in various studies, while high-level evidence to guide its implementation is still lacking. Against this background, this article aimed to review the evidence about PMT regarding its procedural outcomes and safety profile in the treatment of DVT.

A systematic review was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analysis) statement (http://www.prisma-statement.org/). An electronic search was performed using the PubMed, EMBASE, and Cochrane Databases from January 2006 to December 2016. The medical subject heading (MeSH) terms used were “mechanical thrombectomy” and “venous thrombosis” or “deep vein thrombosis”.

The inclusion criteria were as follows: DVT of the lower extremities; human study; study population aged ≥18 years; and articles published in English. Reviews and case reports were excluded. All studies of interest were obtained as full-text articles and assessed by two authors independently. Final decisions on inclusion in the study were made by the entire research team.

Relevant data were extracted with the following items recorded: author, title, year of publication, number and age of patients, co-morbidities, duration of follow-up, onset of symptoms, location of DVT, type of thrombectomy device, and adjunctive modalities. Efficacy was measured in terms of rates of venous patency, recurrence, and PTS. Complications including bleeding, pulmonary embolism, and mortality were recorded. Secondary outcomes included dosage of thrombolytic agents, cost, and duration of procedure.

Statistical meta-analysis was not performed because of the heterogeneity of the original data. Therefore, descriptive data were summarised and presented in tables to provide a comprehensive overview of different clinical aspects of the studies.

Our initial search yielded a total of 369 articles, including 260 articles from PubMed, 98 articles from EMBASE, and 11 studies from the Cochrane Library. Thirty-one duplicated records and 283 irrelevant studies were excluded upon screening of titles and abstracts, leaving 55 potentially eligible studies. A further 39 articles were excluded after full-text articles were assessed: 28 review articles; one study on the patients with inferior vena cava (IVC) filter; one study on the effect of clot age; seven studies without full text; and two non-English studies. Fifteen retrospective studies and one prospective registry were included into our analysis, in which seven articles reported comparative evidence of PMT versus CDT. There were no published randomised trials available.

Baseline patient demographics and characteristics of the studies are summarised in Table 1.6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 A total of 1170 patients were included (range, 16-329 patients) with a mean age of 53.5 years (range, 16 to 88 years). The mean follow-up time was 12.3 months (range, 1-82 months).

Four different categories of thrombectomy devices were used among the included studies: rheolytic devices,6 7 8 9 10 11 aspiration devices,12 13 14 15 16 rotational devices,16 17 18 19 20 21 and ultrasound-enhanced thrombolysis devices.14 They aimed to achieve transcatheter removal of thrombi via different mechanical means.

The AngioJet system (Possis Medical, Minneapolis [MN], US) is a rheolytic device that generates high-velocity saline jets at the side of catheter, which create a localised low-pressure zone and thus result in maceration and aspiration of the thrombus.

During aspiration thrombectomy, the thrombus was aspirated out through the percutaneous catheter as the catheter was gradually pulled out. The process was repeated until complete removal of the thrombus or at least 90% disappearance of thrombi. Two of the aspiration systems were Aspirex (Straub Medical, Wangs, Switzerland) and the Trellis infusion system (Covidien, Mansfield [MA], US), which is a sophisticated system that contains an oscillation drive unit that mixes the thrombus with thrombolytic agents between two occlusion balloons.

Rotational devices feature high-frequency revolution of a helix that is controlled by a foot pedal. At least four different types of rotational devices were included, including the Amplatz thrombectomy device (Microvena, White Bear Lake [MN], US), Rotarex (Straub Medical, Wangs, Switzerland), Trerotola (Arrow International, Redding [PA], US), and Cleaner (Rex Medical, Fort Worth [TX], US and Argon Medical Devices, Inc, Plano [TX], US). They all consisted of a motor-driven fragmentation helix or basket that was rotated in the thrombosed vein. The thrombus was then aspirated out via the catheter.

An ultrasound-enhanced thrombolysis device (EKOS Corporation, Bothell [WA], US) was selectively used in one study14 for patients with inadequate thrombus removal despite the use of a rotational device. A high-frequency ultrasound wave was emitted through transducers inside the catheter to achieve maceration of the thrombus and mix it with thrombolytic agents.

Nine non-comparative and seven comparative studies were included in our analysis. Efficacy in terms of rates of venous patency, PTS, and recurrent thrombosis is shown in Table 2.6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Venous patency was measured most frequently by Duplex ultrasound (n=9) followed by computed tomographic (CT) venography (n=4) and contrast venography (n=2). Imaging modalities were not mentioned in three studies. Venous patency was further quantified according to a 3-tier system in five studies: Grade I (<50% clot lysis), Grade II (50%-99% clot lysis), and Grade III (100% clot lysis).22 Venous patency was measured at 6 months in four studies and at 1 year in the other 12 studies. Venous patency rates ranged from 75% to 100%.

Rates of PTS were reported in terms of Villalta score (n=2)23 or Venous Clinical Severity Score (VCSS) [n=2].14 Four studies reported the rates of valvular incompetence from 8% to 24%. The rates of DVT recurrence, reported in eight studies, ranged from 0% to 17%.

The major complications of thrombectomy are shown in Table 3.6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Six studies reported rates of pulmonary embolism ranging from 0.3% to 17%, but none of them was clinically significant. No patient had pulmonary embolism in the remaining 11 studies. Garcia et al6 reported major bleeding complications in 3.6% of patients, including intracranial bleeding, gastrointestinal bleeding secondary to gastritis or gastric cancer, retroperitoneal bleeding, and haemolytic anaemia requiring transfusion. Minor bleeding complications were reported at frequencies of up to 28%, most of which were access site bleeding. Blood transfusion and surgical intervention were seldom required. No operative mortality was reported in 12 studies, while only two cases of fatal intracranial haemorrhage were noted in two separate studies. Another mortality was reported by Garcia et al,6 with an unknown cause of death. The overall mortality in this series was 0.26%.

Seven studies reported comparative evidence about PMT versus CDT (Table 46 9 10 11 15 18). Huang et al7 showed that PMT significantly reduced PTS at 1 year, with lower Villalta scores in the PMT group (2.1±3.0) than in the CDT group (5.1±4.1; Wilcoxon rank-sum test, P=0.03). However, no statistical difference was shown in Villalta scores in another retrospective study conducted by Park et al.18

Lin et al9 compared bleeding complications between the two groups in terms of the number of units of packed red blood cells transfused. There was a significant reduction of blood transfusion from 1.2±0.7 units in the CDT group to 0.2±0.3 units in the PMT group (Pearson Chi squared, P≤0.05).9

The dosage of thrombolytic infusion and average procedural time were significantly reduced in the CDT with adjunctive PMT group compared with the CDT alone group, as reported in at least four different retrospective studies.10 11 15 18

From the economic perspective, two retrospective studies performed cost analysis, and PMT was found to be associated with 44% to 49% reduction in total hospital costs.9 10 It was also consistent with shorter hospital and intensive care unit (ICU) stays in the PMT group (4.6±1.3 days of hospital stay and 0.6±0.3 days of ICU stay in the PMT group vs 8.4±2.3 days of hospital stay and 2.4±1.2 days of ICU stay in the CDT group; Pearson Chi squared, P<0.02 to 0.04).

No statistically significant differences in venous patency or symptom improvement between the two groups were reported in this series of comparative studies.

Three studies compared outcomes of different thrombectomy devices. Garcia et al6 created the first prospective multi-centre (PEARL) registry to document the use of the AngioJet rheolytic device. A total of 329 patients were stratified into four treatment subgroups: (1) rheolytic thrombectomy (RT) alone; (2) RT plus CDT; (3) pharmacomechanical CDT (PCDT), and (4) PCDT combined with CDT. Each of these subgroups differed in terms of the presence, timing, and delivery means of thrombolytic agents. Rheolytic thrombectomy was given before or after CDT in subgroup 2, and PCDT was defined as delivery of lytic agent through an AngioJet catheter. This registry demonstrated no statistical difference in venous patency rate between the subgroups, while a significant reduction in procedural time in non-CDT group was observed (Table 4). The investigators concluded that RT was effective and safe, and therefore, the needs for concomitant CDT and intensive care could potentially be reduced.

Shi et al20 and Arko et al8 compared the outcomes of Amplatz versus Rotarex and Trellis versus AngioJet devices, which again showed no significant differences in clinical outcomes between the two groups.

Inferior vena cava filter placement, angioplasty, and stenting were the most commonly performed adjunctive treatments in addition to thrombectomy. Inferior vena cava filters were used in 46% to 100% of patients among 11 studies. The majority of the filters were removed shortly after the procedure without major complications. Lee et al15 reported that 6 of 37 patients had thrombus entrapment in prophylactic IVC filters in the thrombectomy group compared with 0 of 9 patients in the CDT alone group. Arko et al8 reported that 17% of patients showed asymptomatic pulmonary embolism on CT after thrombectomy, in which all patients did not receive IVC filters. This showed that prophylactic IVC filtration could be a useful measure for prevention of pulmonary embolism, especially in patients who undergo aggressive thrombectomy.

Angioplasty with or without stenting was performed in 15 studies, ranging from 14% to 80% of patients. The two main indications were iliac vein compression syndrome (May-Thurner syndrome) and residual thrombus after thrombectomy. One study reported a significantly improved iliac vein patency rate in the group with stents (28.95%) than without stents (11.29%; log rank test, P=0.026).15

Catheter-directed thrombolysis and PMT are both emerging techniques for treatment of acute DVT of the lower extremities that have the advantage of early restoration of venous patency and thus reduction of post-thrombotic complications. A 2015 meta-analysis compared the efficacy of CDT plus anticoagulation versus that of anticoagulation alone in the treatment of proximal DVT. It showed that additional CDT was associated with significantly improved 6-month venous patency and PTS rates. However, there was a two-fold increase in bleeding complications in the CDT group, and concomitant close monitoring under intensive care setting has had a substantial economic burden.5 These two main reasons have precluded the incorporation of CDT into the standard treatment recommendation despite encouraging procedural outcomes.

As compared with CDT, PMT is another endovascular option that has provided promising clinical outcomes with better controlled bleeding risk. This review has served as a comprehensive overview of clinical and safety outcomes across different categories of thrombectomy devices. It demonstrated well that the procedural outcomes of both PMT alone and that with pharmacomechanical devices were non-inferior to that of CDT in treatment of acute DVT in the lower extremities. The rates of PTS, bleeding complications, and hospital costs of PMT were all shown to be favourable to those of CDT alone. In addition, the mortality risk of PMT was minimal and comparable to that of patients treated with anticoagulation alone: 0.4% recurrent fatal venous thromboembolism and 0.2% fatal major bleeding events.24 As illustrated in this review, the balanced risks and benefits of PMT provide a basis for the future initiation of randomised controlled trials on its use.

In addition, PMT is potentially superior to CDT especially in patients in whom thrombolysis therapy is contra-indicated. According to the Society of Interventional Radiology recommendations, CDT is absolutely contra-indicated in patients with recent cerebrovascular events, neurosurgery or intracranial trauma, active internal bleeding, and those with absolute contra-indications to anticoagulation. Other strong relative contra-indications are listed in the Standard of Practice25: recent major surgery, obstetrical delivery or major trauma within 10 days, etc. These patients are prone to the development of DVT, and they have been conventionally treated with anticoagulation or IVC filters. Percutaneous mechanical thrombectomy is another option in this clinically challenging situation. Further studies on this particular group of high-risk patients are necessary to investigate the efficacy and safety of this novel technique.

Nevertheless, this review has several limitations. The studies were heterogeneous in terms of outcome measurements and the use of thrombectomy devices. Post-thrombotic syndrome was measured in terms of Villalta score, VCSS, or valvular incompetence rate. Although these systems were well-defined objective scales for monitoring and documentation of PTS, it was difficult to compare efficacy across studies. Similarly, the important index of venous patency rate was variously measured by Duplex ultrasound, CT venogram, or venography. Inaccuracies during direct comparison between studies were unavoidable.

Other adjunctive modalities in addition to the principal thrombectomy devices including iliac vein angioplasty, stenting, and prophylactic IVC filter were used in a major proportion of the studies. No standardised criteria were outlined for the usage of these devices, and they created a confounding factor during data analysis. With inadequate information on sub-categorisation of the study populations, analysis specific to each type of adjunctive devices was not feasible. Most of the studies were retrospective, and no randomised trials were available for quantitative analysis.

Percutaneous mechanical thrombectomy is a safe and effective treatment for acute iliofemoral DVT in terms of restoration of venous patency, prevention of DVT recurrence, PTS, and pulmonary embolism. The overall clinical outcomes of PMT are superior to those with anticoagulation alone. Compared with CDT alone, adjunctive PMT has a lower risk of PTS and bleeding complications. Randomised studies to demonstrate the efficacy of PMT versus anticoagulation and CDT and compare the efficacy of different types of PMT devices would be most beneficial to guide future strategies for treatment of acute proximal DVT.

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.

Concept or design: All authors.
Acquisition of data: All authors.
Analysis or interpretation of data: PC Wong, YC Chan, Y Law.
Drafting of the manuscript: PC Wong, YC Chan, Y Law.
Critical revision: All authors.

The authors declare no conflicts of interest.

The paper was presented as an abstract in the 21st Asian Congress of Surgery by the Asian Surgical Association, 22-23 November 2017, Tokyo, Japan.

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

1. Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008;133(6 Suppl):454S-545S. Crossref

2. Enden T, Haig Y, Kløw NE, et al. Long-term outcome after additional catheter-directed thrombolysis versus standard treatment for acute iliofemoral deep vein thrombosis (the CaVenT study): a randomised controlled trial. Lancet 2012;379:31-8. Crossref

3. Baldwin MJ, Moore HM, Rudarakanchana N, Gohel M, Davies AH. Post-thrombotic syndrome: a clinical review. J Thromb Haemost 2013;11:795-805. Crossref

4. Watson LI, Armon MP. Thrombolysis for acute deep vein thrombosis. Cochrane Database Syst Rev 2004;(4):CD002783. Crossref

5. Du GC, Zhang MC, Zhao JC. Catheter-directed thrombolysis plus anticoagulation versus anticoagulation alone in the treatment of proximal deep vein thrombosis—a meta-analysis. Vasa 2015;44:195-202. Crossref

6. Garcia MJ, Lookstein R, Malhotra R, et al. Endovascular management of deep vein thrombosis with rheolytic thrombectomy: final report of the prospective multicenter PEARL (Peripheral Use of AngioJet Rheolytic Thrombectomy with a Variety of Catheter Lengths) registry. J Vasc Interv Radiol 2015;26:777-85. Crossref

7. Huang CY, Hsu HL, Kuo TT, Lee CY, Hsu CP. Percutaneous pharmacomechanical thrombectomy offers lower risk of post-thrombotic syndrome than catheter-directed thrombolysis in patients with acute deep vein thrombosis of the lower limb. Ann Vasc Surg 2015;29:995-1002. Crossref

8. Arko FR, Davis CM 3rd, Murphy EH, et al. Aggressive percutaneous mechanical thrombectomy of deep venous thrombosis: early clinical results. Arch Surg 2007;142:513-9. Crossref

9. Lin PH, Zhou W, Dardik A, et al. Catheter-direct thrombolysis versus pharmacomechanical thrombectomy for treatment of symptomatic lower extremity deep venous thrombosis. Am J Surg 2006;192:782-8. Crossref

10. Kim HS, Patra A, Paxton BE, Khan J, Streiff MB. Catheter-directed thrombolysis with percutaneous rheolytic thrombectomy versus thrombolysis alone in upper and lower extremity deep vein thrombosis. Cardiovasc Intervent Radiol 2006;29:1003-7. Crossref

11. Kim HS, Patra A, Paxton BE, Khan J, Streiff MB. Adjunctive percutaneous mechanical thrombectomy for lower-extremity deep vein thrombosis: clinical and economic outcomes. J Vasc Interv Radiol 2006;17:1099-104. Crossref

12. Ozpak B, Ilhan G, Ozcem B, Kara H. Our short-term results with percutaneous mechanical thrombectomy for treatment of acute deep vein thrombosis. Thorac Cardiovasc Surg 2016;64:316-22. Crossref

13. Jia Z, Tu J, Zhao J, et al. Aspiration thrombectomy using a large-size catheter for acute lower extremity deep vein thrombosis. J Vasc Surg Venous Lymphat Disord 2016;4:167-71. Crossref

14. Gagne P, Khoury T, Zadeh BJ, Rajasinghe HA. A multicenter, retrospective study of the effectiveness of the trellis-8 system in the treatment of proximal lower-extremity deep vein thrombosis. Ann Vasc Surg 2015;29:1633-41. Crossref

15. Lee JH, Kwun WH, Suh BY. The results of aspiration thrombectomy in the endovascular treatment for iliofemoral deep vein thrombosis. J Korean Surg Soc 2013;84:292-7. Crossref

16. Karahan O, Kutas HB, Gurbuz O, et al. Pharmacomechanical thrombolysis with a rotator thrombolysis device in iliofemoral deep venous thrombosis. Vascular 2016;24:481-6. Crossref

17. Bozkurt A, Kırbaş İ, Kösehan D, Demirçelik B, Nazlı Y. Pharmacomechanical thrombectomy in the management of deep vein thrombosis using the cleaner device: an initial single-center experience. Ann Vasc Surg 2015;29:670-4. Crossref

18. Park KM, Moon IS, Kim JI, et al. Mechanical thrombectomy with Trerotola compared with catheter-directed thrombolysis for treatment of acute iliofemoral deep vein thrombosis. Ann Vasc Surg 2014;28:1853-61. Crossref

19. Shi HJ, Huang YH, Shen T, Xu Q. Percutaneous mechanical thrombectomy for acute massive lower extremity deep venous thrombosis. Surg Laparosc Endosc Percutan Tech 2011;21:50-3. Crossref

20. Shi HJ, Huang YH, Shen T, Xu Q. Percutaneous mechanical thrombectomy combined with catheter-directed thrombolysis in the treatment of symptomatic lower extremity deep venous thrombosis. Eur J Radiol 2009;71:350-5. Crossref

21. Lee KH, Han H, Lee KJ, et al. Mechanical thrombectomy of acute iliofemoral deep vein thrombosis with use of an Arrow-Trerotola percutaneous thrombectomy device. J Vasc Interv Radio 2006;17:487-95. Crossref

22. Porter JM, Moneta GL. Reporting standards in venous disease: an update. International Consensus Committee on Chronic Venous Disease. J Vasc Surg 1995;21:635-45. Crossref

23. Kahn SR, Partsch H, Vedantham S. Definition of post-thrombotic syndrome of the leg for use in clinical investigations: a recommendation for standardization. J Thromb Haemost 2009;7:879-83. Crossref

24. Carrier M, Le Gal G, Wells PS, Rodger MA. Systematic review: case-fatality rates of recurrent venous thromboembolism and major bleeding events among patients treated for venous thromboembolism. Ann Intern Med 2010;152:578-89. Crossref

25. Vedantham S, Thorpe PE, Cardella JF, et al. Quality improvement guidelines for the treatment of lower extremity deep vein thrombosis with use of endovascular thrombus removal. J Vasc Interv Radiol 2009;20(7 Suppl):S227-39. Crossref

Patients with epilepsy whose seizures do not respond successfully to antiepileptic drug therapy are considered to have drug-resistant epilepsy (DRE). Prior equivalent terms include medically intractable epilepsy or pharmacoresistant epilepsy. Among patients with epilepsy, those with DRE have the greatest burden of epilepsy-related disabilities and of health care expenses.

In 2010, the task force of the International League Against Epilepsy Commission on Therapeutic Strategies proposed a framework of two levels for defining DRE.1 The diagnosis of DRE usually requires failure of two adequate trials of appropriately chosen and administered antiepileptic drugs (sequential monotherapy or combined polytherapy).1 2 It is also important to consider the effect of seizure factors (frequency, severity, associated behavioural problem) on individual psychosocial well-being. This effect will influence the physicians’ decision on drug options and the urgency of considering non-pharmacological therapy.

The important aspects of the clinical assessment of DRE are discussed in the sections below.

A 2004 study estimated the prevalence of epilepsy among Hong Kong Chinese residents aged ≥15 years at about 3 to 5.7 per 1000 population, or approximately 40 000 Hong Kong residents in total.3 Approximately one third of these patients with epilepsy were expected to have DRE.3 A study in 2008 estimated the crude prevalence of active epilepsy and seizure disorder in Hong Kong to be 8.49 per 1000 population; therefore, the number of people in Hong Kong at that time with seizure disorder was approximately 62 000.4 A meta-analysis of international studies showed the point prevalence of epilepsy was 6.38 per 1000 population.5 Furthermore, a recent study regarding prevalence and incidence of epilepsy in the United States population showed that the overall age-adjusted prevalence of epilepsy was 8.5 per 1000 population.5 Previous studies have found that there is a treatment or referral gap of 20 years in the United States for patients with DRE.5 6

Prospective studies of patients with chronic epilepsy suggested that 70% to 80% of patients retain their status as either having intractable seizures or being in remission.7 In other words, around 20% of patients with seizure initially diagnosed as intractable will achieve seizure freedom in the long term. One proposed mechanism that might lead to intractability involves glial proliferation and dendritic sprouting with synaptic recognition in mesial temporal sclerosis.8 Paroxysmal depolarization shift refers to situations where the usual refractory period does not follow rapidly repeating action potentials at a cellular level, resulting in local high-frequency oscillations.9 10

Another compelling theory is the build-up of epileptic “neuronal network”, via alternation in neuronal circuitry.11 A well-defined neuronal network example is the limbic network with sequential propagation path via the hippocampus, amygdala, lateral temporal neocortex, entorhinal cortex, medial thalamus, and frontal inferior lobes. The interest on neuronal network analysis in epilepsy had gained strength with the use of high-resolution recording techniques. Different brain functional regions are interconnected with specific neuronal networks; therefore, a seizure in one part will spread quickly in a typical oscillatory manner.11

A recent study, which adopted the International League Against Epilepsy DRE criteria, observed different patterns of disease progression in an incident cohort of DRE.12 The authors of that study observed that 30% of patients eventually developed DRE. They also found a long delay from disease onset to failure of second antiepileptic drug. This finding might give insights into the pathogenesis of DLE mentioned above.

In general, the mortality and morbidity of DRE is higher than that of seizure-free patients or patients with good seizure control.13 14 15 Even when seizures are infrequent, daily life and subjective well-being are impaired in patients with DLE, to various extents.16

Refractory epilepsy can be subdivided into temporal lobe epilepsy (TLE) and non-temporal lobe epilepsy. Depending on the clinical and radiological manifestations, TLE can be divided further into two distinct groups: mesial TLE and neocortical TLE. Both mesial TLE and neocortical TLE share similar pathological substrates; the primary difference is that mesial temporal sclerosis is found only in mesial TLE.17 Patients with TLE usually present with complex partial seizures, with or without generalised seizures, depending on the neuronal network involved. A minority of patients with TLE will became seizure free after repeated drug trials.18 However, most patients with TLE require surgical intervention.18

For patients with non–temporal lobe epilepsy, the clinical and radiological features are diverse and also depend on the underlying aetiologies or pathological substrates. In general, the seizure semiology is poorly defined and variability in the associated magnetic resonance imaging (MRI) abnormalities makes seizure focus localisation challenging. Usually, a concerted effort and multimodality investigations (in phase 1 of the presurgical evaluation) are required.19

Before patients with DRE are referred to a tertiary centre for preparation for surgical intervention, potential factors for treatment failure should be considered. For example, failure to recognise a generalised epilepsy syndrome can result in an inappropriate choice of first-line antiepileptic drug (eg, carbamazepine) that will aggravate seizures. Poor drug compliance and other lifestyle-related factors can also contribute to seizure recurrence. These factors are often considered as pseudoresistance.20 21 22

The longer the delay between the onset of DRE and surgery intervention, the lower the chance of a seizure-free and improved psychosocial postoperative outcome.5 Therefore, timely referral is crucial for quality care of patients with DRE.

Before recruiting the patient, the first step is to identify the criteria that indicate an appropriate candidate. Only patients who meet all three of these eligibility criteria should be considered for inclusion:
1. The patient and their family understand and accept the surgical treatment and any potential risks;
2. The seizures are disabling despite adequate and appropriate drug trials; and
3. The available imaging and neurophysiological data are consistent with surgically remediable epileptic syndrome.

The first objective of presurgical evaluation is to identify the epileptogenic zone (EZ) by invasive or non-invasive modalities of investigations. More sophisticated or invasive approaches might be required, depending on the clarity of structural identifiable pathologies on neuroimaging, and on the link with the clinical semiology.

The second objective, after successful identification and location of the EZ, is to develop strategies to ensure that the lesion can be safely resected without significant physical or cognitive sequelae.

It is also pragmatic to interview the patient and their family and friends, in order to obtain all relevant history and risk factors or aetiological factors. The latter will also give insight into the prognosis of the epileptic disorder after surgical treatment. For example, a case of post-encephalitic epilepsy would be less suitable for surgical intervention.23

A multidisciplinary evaluation team is required, and investigations should include neurophysiological evaluation, structural neuroimaging, neuropsychological assessment, and psychiatric assessment.

The neurophysiological evaluation includes interictal and ictal electroencephalography (EEG) sampling, which can be attained by non-invasive or invasive means in a long-term recording manner.

Interictal EEG provides important information on the lateralisation or localisation of the EZ. This is particularly true in cases of TLE, in which a solely unilateral anterior temporal spike is a strong predictor of a seizure-free postoperative outcome.24 Unilateral mesial temporal sclerosis with bitemporal interictal epileptiform discharges may also be found.25 Short bursts of low-voltage, high-frequency oscillations are valuable in localising focal cortical dysplasia.26

Video EEG recordings capture habitual seizures and ictal EEG discharges. The lateralisation and localisation of the ictal onset zone can be deduced from analysis of an adequate number of these captured events. After combined analysis of ictal and interictal EEG data, the irritative and ictal onset zones can be estimated.27

Invasive recording is indicated when there is a hypothesised EZ that is not fully supported by the results of non-invasive diagnostic evaluations. These difficult scenarios are more common in cases of non-lesional epilepsy.28 29

Brain MRI is the fundamental yet most important investigation for presurgical evaluation. This is particularly true for certain epileptic disorders, such as TLE in which mesial temporal sclerosis is the pathological substrate. The MRI features of hippocampal sclerosis include hippocampal atrophy on T1-weighted MRI images, increased signal on T2-weighted MRI images or fluid-attenuated inversion recovery MRI sequences, and decreased signal on inversion recovery MRI sequences.30 31 Detection of these abnormalities requires optimised imaging techniques, including angulated coronal sections obtained perpendicular to the long axis of the hippocampal structures.

For extratemporal substrates, MRI can also define hemimegaloencephaly, schizencephaly, and focal subcortical heterotopia. Focal cortical dysplasia is the most common developmental pathology in children with extratemporal lobe seizures, and there is an international classification to define the underlying histopathology and foretell the outlook of surgical success.32

Recently, 3T MRI systems have gained in importance for pre-surgical workup. Studies have shown that, for initially non-lesional cases scanned by a 1.5T system with standard MRI brain protocol, more than half had new findings after rescanning by a 3T MRI system with multichannel phased-array coils.33 34

Diffusion tensor imaging and tractography can be used for fibre tracking and non-invasive structural network mapping and is an optional imaging sequence to aid presurgical trajectory planning. A recent study reported using diffusion tensor imaging for successful identification of significant diffusion abnormalities of tract sections in the ipsilateral dorsal fornix and in the contralateral parahippocampal white matter bundle in patients with poor postoperative seizure control.35 Although more studies are required to draw a definitive conclusion, these results may help to understand the mechanisms of postoperative persistent seizure. Diffusion tensor imaging has potential prognostic value in predicting operation outcome.

Despite the strengths of MRI, there are pathological substrates that go beyond the detection ability of MRI analysis. As a result, multi-modality imaging of the brain will come into play. Functional neuroimaging modalities, such as positron emission tomography, single photon emission computed tomography, functional MRI and magnetoencephalography, can be co-registered with MRI to give a more detailed structural-functional correlated imaging analysis. These imaging modalities can aid the localisation of EZ, but the sensitivity largely depends on the epileptic syndrome. Magnetoencephalography is indicated in both localization of EZ and delineation of the relationship between the index lesion and the surroundings. Magnetoencephalography-guided excisional surgery provides more precise excision in subtle or MRI-negative cases, and in patients with multiple intra-cerebral lesions, such as cavernomata.36 37 38

Neuropsychological tests rely on functional neuroanatomy, in which certain cognitive functions are ascribed to physical areas of the brain. The temporal lobe is associated with memory and language, with the left side representing verbal memory, and the right side representing visual memory. The frontal lobe is associated with executive actions and behaviour, and the posterior of the brain with perception and higher sensory faculties. It is controversial to state the prediction of postoperative cognitive outcome should be based on the side that was to be resected or the side that would remain following surgery.

The Wada test, first introduced in 1949, is used to determine cerebral language dominance.39 The Wada test is used to evaluate the risks of postoperative amnesia and task-specific memory deficits, to lateralise hemispheric dysfunction, and to predict postoperative seizure outcome.39 However, one study involving 145 patients showed that Wada test results were not predictive of outcome.40 Functional MRI is an alternate non-invasive method to determine cerebral language dominance. Functional MRI has been shown as eligible to replace Wada tests in the majority of patients with clearly lateralised language localisation; however, in patients who are agitated or mentally impaired and have bilateral functional MRI activations, Wada tests still provide additional information.41

Lower mental reserve and higher functional adequacy of the resected tissue preclude surgical feasibility.42 43

It is recommended that the presurgical evaluation should include a thorough psychiatric assessment.29 Psychiatric disorders are prevalent in epilepsy patients, and psychopathology is common in patients with TLE. Appropriate assessment can also help to anticipate acute anxiety, delusions, and other symptoms that might be aggravated in some temporal epilepsy cases, especially in the perioperative period. In addition, a history of psychiatric disorder is associated with the worst postoperative seizure outcome, although the existence of stable psychiatric disorder does not preclude epilepsy surgery.44 Psychiatric assessment should include four domains: behavioural, psychiatric, self-esteem profile, and quality of life.

The decision to proceed with surgical intervention is usually made by consensus agreement among the investigating clinicians. The decision is based on a rational estimation of the precision of the EZ (thus the success rate of seizure cure) and the risk-benefit analysis of the potential postoperative outcomes.

In general, outcomes are more favourable for lesional epileptic syndrome with concordance of investigation results and neuropsychological proof of “absence” of important cognitive function within the areas to be resected. In contrast, lack of concordance or evidence of important function in the pathological substrate will preclude surgical feasibility. In addition to the disease factor, patient factors such as seizure frequency, duration of illness, and co-morbidity govern the outcome.45

Conventionally, outcomes after epilepsy surgery are categorised into four classes as described by Engel.46 Epilepsy surgery is typically considered either curative, palliative, or modulatory.

Curative resective surgery involves temporal lobe surgery and extratemporal lobe surgery. Among the different epileptic syndromes, mesial temporal sclerosis has the most favourable outcomes. After curative resective surgery, outcomes in 70% of patients are reported as Engel Class I.47 48

Palliative disconnection surgery includes corpus callosotomy, hemispherectomy (anatomical/ functional), hemispherotomy, or multiple subpial transections. All of these procedures are commonly performed in paediatric patients with epilepsy and have been shown to reduce seizure frequency by 40% to 50%.49

Modulatory surgery includes gamma knife radiosurgery, vague nerve stimulation, and deep brain stimulation. Gamma knife radiosurgery has been shown to be effective in mesial temporal sclerosis patients. Seizure control is typically achieved about 10 months after radiation. Studies have shown the effectiveness of gamma knife radiosurgery in seizure control in patients with central region cavernomas and hypothalamic haemartomas.50 Vagal nerve stimulation is indicated for adults and adolescents aged >12 years with medically intractable partial seizures who are not candidates for potentially curative surgical resections. About 30% to 40% of patients have seizure reduction of ≥50%.51 Deep brain stimulation targets include the anterior nucleus and the centromedian nucleus of the thalamus, the subthalamic nucleus, the caudate nucleus, the hippocampus and the cerebellum.51 By 2 years, a median 56% reduction in seizure frequency was observed after stimulation of anterior nucleus of the thalamus. Among the patient study group, 54% had at least a 50% reduction in seizures, and 14 patients had seizure-free status for at least 6 months.52

Newer treatment entities include responsive or closed-loop cortical stimulation for patients with bitemporal lobe epilepsy or epilepsy originating from the eloquent cortex. The seizure reduction rate for these treatments is >50% at 12 weeks in approximately 30% of patients.51

Epilepsy surgery for TLE is usually recommended because of the promising results. One study from Wiebe et al showed that, for TLE patients, surgical treatment resulted in significantly higher seizure freedom (58%) than did medical treatment (8%) at 1-year follow-up.47 The Early Randomized Surgical Epilepsy Trial in 38 patients showed that 11 of 15 patients who received surgical treatment were seizure free at 2-year follow-up compared with 0 of 23 patients who received medical treatment.53 Another study including more than 3000 patients from Germany concluded that there is an increasing trend the number of patients with non-lesional epilepsy requiring intracranial recordings.54

There is always difficulty in identification of the EZ in non-lesional neocortical epilepsy. Seizure-free outcomes have been reported in 55% of patients with non-lesional TLE and in 43% of patients with non-lesional extratemporal lobe epilepsy.55 Concordance with two or more presurgical evaluations, including interictal EEG, ictal EEG, 18-fluorodeoxyglucose positron emission tomography, or ictal single photon emission computed tomography, is significantly related to a seizure-free outcome.56 Another study showed that 9 out of 24 patients with non-lesional extratemporal epilepsy (38%) had Engel class I outcome at a mean follow-up time of 9 years.57 In patients with TLE with MRI-negative and positron emission tomography–positive findings, surgery could achieve Engel Class I surgical outcomes at postoperative 2 years in about 79.2%.58

Epilepsy surgery failure may be caused by incorrect localisation of the EZ, very widespread EZs, or very limited resection of the suspected EZ.

After mesial temporal resection, patients may experience seizures arising from neocortical regions instead of from the residual hippocampal structure. This may imply the existence of regional epileptogenicity. The hippocampus represents the area of cortex with the lowest threshold for seizure generation and any surrounding neocortical tissue also exhibiting epileptogenicity then becomes the site of ictal onset. About 25% of patients with seizure relapse after mesial temporal sclerosis may have seizure onset in the contralateral temporal region.59

Extensive re-evaluation of patients after epilepsy surgery failure is recommended, for consideration of reoperation if the EZ can be localised.

Current applications of EEG recordings are not limited to scalp EEG and intracranial EEG with subdural electrodes and depth electrodes. Minimally invasive intracranial endovascular EEG monitoring by means of nanowire, catheter, and stent-electrode recordings is evolving.60

Stereo EEG is gaining popularity owing to its ability to make precise recordings from deep cortical areas in bilateral and multiple lobes without subjecting the patient to bilateral large craniotomies. Stereo EEG has been shown to be a useful and relatively safe tool to localise the EZ, with a procedure-related morbidity as low as 5.6%.61 Other centres have incorporated stereo EEG into a neurorobotic surgical system with comparable results.62 63

High-frequency oscillations are a potential marker for epileptogenicity and a predictive factor for positive epilepsy surgery outcomes.64 A meta-analysis has shown a small but significant relationship between the removal of the high-frequency oscillation-generating brain region and favourable outcomes.65

The prerequisites of seizure origin in a well-circumscribed area of the brain and precise localisation of the EZ make modern epilepsy surgery a promising treatment modality for refractory epilepsy.

The presurgical assessment, which includes multiple disciplines, however, should be focused on two important aspects:
1. Data concordance: The individual seizure pattern is ascribed to the hypothetical brain lesions, as suggested by neurophysiological and radiological data.
2. Functional reserve: The brain pathological region, if being resected, will not leave the patient with significant morbidities.

The broad range of available diagnostic tests and surgical techniques has widened the applicability of surgical treatment. The success rates of these surgical interventions range depend strongly on different scenarios, ranging from seizure reductions of 10% to 20% to seizure-free outcomes in >70% of patients.

Epilepsy surgery for DRE involves close collaboration and cooperation by a multidisciplinary team. Epilepsy surgery can be performed in different epilepsy centres. Patients should be referred early in their refractory disease course to a higher-level epilepsy centre for evaluation of the complex surgical options.

All authors have made substantial contributions to the concept of this study; acquisition of data; analysis or interpretation of data; drafting of the article; and critical revision for important intellectual content.

This project was supported in part by an unrestricted grant of the Hong Kong Epilepsy Society.

All authors have disclosed no conflicts of interest. 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.

1. Kwan P, Arzimanoglou A, Berg AT, et al. Definition of drug resistant epilepsy: consensus proposal by the ad hoc task force of the ILAE Commission on Therapeutic Strategies. Epilepsia 2010;51:1069-77. Crossref

2. Kwan P, Brodie MJ. Definition of refractory epilepsy: defining the indefinable? Lancet Neurology 2010;9:27-9. Crossref

3. Hui AC, Kwan P. Epilepsy in Hong Kong: a literature review. Hong Kong Med J 2004;10:185-9.

4. Fong GC, Kwan P, Hui AC, Lui CH, Fong JK, Wong V. An epidemiological study of epilepsy in Hong Kong SAR, China. Seizure 2008;17:457-64. Crossref

5. Fiest KM, Sauro KM, Wiebe S, Patten SB, Kwon S, Dykeman J, Pringsheim T, Lorenzetti DL, Jette N. Prevalence and incidence of epilepsy: A systemic review and meta-analysis of international studies. Neurology 2017;88:296-303. Crossref

6. Helmers SL, Thurman DJ, Durgin TL, Pai AK, Faught E. Descriptive epidemiology of epilepsy in the U.S. population: A different approach. Epilepsia 2015;56:942-8. Crossref

7. Brodie MJ, Kwan P. Staged approach to epilepsy management. Neurology 2002;58(8 Suppl 5):S2-8. Crossref

8. Pitkänen A, Sutula TP. Is epilepsy a progressive disorder? Prospects for new therapeutic approaches in temporal-lobe epilepsy. Lancet Neurol 2002;1:173-81. Crossref

9. Holmes GL, Ben-Ari Y. Seizing hold of seizures. Nat Med 2003;9:994-6. Crossref

10. Bragin A, Engel J Jr, Wilson CL, Fried I, Mathern GW. Hippocampal and entorhinal cortex high-frequency oscillations (100-500 Hz) in human epileptic brain and in kainic acid–treated rats with chronic seizures. Epilepsia 1990;40:127-37. Crossref

11. Stefan H, Lopes da Silva FH. Epileptic neuronal networks: methods of identification and clinical relevance. Front Neurol 2013;4:8. Crossref

12. Choi H, Hayat MJ, Zhang R, et al. Drug-resistant epilepsy in adults: outcome trajectories after failure of two medications. Epilepsia 2016;57:1152-60. Crossref

13. Tomson T. Mortality in epilepsy. J Neurol 2000;247:15-21. Crossref

14. Sperling MR, Feldman H, Kinman J, Liporace JD, O’Connor MJ. Seizure control and mortality in epilepsy. Ann Neurol 1999;46:45-50. Crossref

15. Shackleton DP, Westendorp RG, Kasteleijn-Nolst Trenité DG, de Craen AJ, Vandenbroucke JP. Survival of patients with epilepsy: an estimate of the mortality risk. Epilepsia 2002;43:445-50. Crossref

16. Luoni C, Bisuli F, Canevini MP, et al. Determinants of health-related quality of life in pharmacoresistant epilepsy: results from a large multicenter study of consecutively enrolled patients using validated quantitative assessment. Epilepsia 2011;52:2181-91. Crossref

17. Asadi-Pooya AA, Tajvarpour M, Vedadinezhad B, Emami M. Comparison of temporal lobe epilepsy with hippocampal sclerosis and temporal lobe epilepsies due to other etiologies. Med J Islamic Repub Iran 2015;29:263.

18. Spencer SS. When should temporal-lobe epilepsy be treated surgically? Lancet Neurol 2002;1:375-82. Crossref

19. Cascino GD. From the American Epilepsy Society 2009 Annual Course Non substrate-directed epilepsy and surgery: PRO and CON. Epilepsy Behav 2011;20:190-3. Crossref

20. Smith D, Defalla BA, Chadwick DW. The misdiagnosis of epilepsy and the management of refractory epilepsy in a specialist clinic. QJM 1999;92:15-23. Crossref

21. Perucca E, Gram L, Avanzini G, Dulac O. Antiepileptic drugs as a cause of worsening seizures. Epilepsia 1998;39:5-17. Crossref

22. Cramer JA, Glassman M, Rienzi V. The relationship between poor medication compliance and seizures. Epilepsy Behav 2002;3:338-42. Crossref

23. Trinka E, Dubeau F, Andermann F, et al. Successful epilepsy surgery in catastrophic postencephalitic epilepsy. Neurology 2000;54:2170-3. Crossref

24. Tonini C, Beghi E, Berg AT, et al. Predictors of epilepsy surgery outcome: a meta-analysis. Epilepsy Res 2004;62:75-87. Crossref

25. Ergene E, Shih JJ, Blum DE, So NK. Frequency of bilateral independent interictal epileptiform discharges in temporal lobe epilepsy. Epilepsia 2000;41:213-8. Crossref

26. Gambardella A, Palmini A, Andermann F, et al. Usefulness of focal rhythmic discharges on scalp EEG of patients with focal cortical dysplasia and intractable epilepsy. Electroencephalogr Clin Neurophysiol 1996;98:243-9. Crossref

27. Rosenow F, Lüders H. Presurgical evaluation of epilepsy. Brain 2001;124:1683-700. Crossref

28. Macrodimitris S, Sherman EM, Forde S, et al. Psychiatric outcomes of epilepsy surgery: a systemic review. Epilepsia 2011;52:880-90. Crossref

29. Jayakar P. Invasive EEG monitoring in children; when, where and what? J Clin Neurophysiol 1999;16:408-18. Crossref

30. Cascino GD, Jack CR Jr, Parisi JE, et al. Magnetic resonance imaging–based volume studies in temporal lobe epilepsy: pathological correlations. Ann Neurol 1991;30:31-6. Crossref

31. Cendes F, Andermann F, Glorr P, et al. MRI volumetric measurement of amygdala and hippocampus in temporal lobe epilepsy. Neurology 1993;43:719-25. Crossref

32. Bae YS, Kang HC, Kim HD, Kim SH. New classification of focal cortical dysplasia: application to practical diagnosis. J Epilepsy Res 2012;2:38-42. Crossref

33. Knake S, Triantafyllou C, Wald LL, et al. 3T phased array MRI improves the presurgical evaluation in focal epilepsies: a prospective study. Neurology 2005;65:1026-31. Crossref

34. Hiratsuka Y, Miki H, Kikuchi K, et al. Sensitivity of an eight-element phased array coil in 3 Tesla MR imaging: a basic analysis. Magn Reson Med Sci 2007;6:177-81. Crossref

35. Keller SS, Glenn GR, Weber B, et al. Preoperative automated fibre quantification predicts postoperative seizure outcome in temporal lobe epilepsy. Brain 2017;140:68-82. Crossref

36. Mamelak AN, Lopez N, Akhtari M, Sutherling WW. Magnetoencephalography-directed surgery in patients with neocortical epilepsy. J Neurosurg 2002;97:865-73. Crossref

37. Papanicolaou AC, Simos PG, Castillo EM, et al. Magnetocephalography: a noninvasive alternative to the Wada procedure. J Neurosurg 2004;100:867-76. Crossref

38. Mäkelä JP, Forss N, Jääskeläinen J, Kirveskari E, Korvenoja A, Paetau R. Magnetoencephalography in neurosurgery. Neurosurgery 2006;59:493-511. Crossref

39. Meador KJ, Loring DW. The Wada test: controversies, concerns, and insights. Neurology 1999;52:1535-6. Crossref

40. Chan S, Andrew A, Roberts D, et al. Wada memory performance does not predict memory and seizure outcome after epileptic surgery. Neurol 2017;88(16 Suppl):229.

41. Wagner K, Hader C, Metternich B, Buschmann F, Schwarzwald R, Schulze-Bonhage A. Who needs a Wada test? Present clinical indications for amobarbital procedures. J Neurol Neurosurg Psychiatry 2012;83:503-9. Crossref

42. Luders HO. Textbook of Epilepsy Surgery. London: Informa; 2008.

43. Schwartz TH, Jeha LE, Tanner A, Bingaman W, Sperling MR. Late seizures in patients initially seizure free after epilepsy surgery. Epilepsia 2006;47:567-73. Crossref

44. Kanner AM, Byrne R, Chicharro A, Wuu J, Frey M. A lifetime psychiatric history predicts a worse seizure outcome following temporal lobectomy. Neurology 2009;72:793-9. Crossref

45. Téllez-Zenteno JF, Dhar R, Wiebe S. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain 2005;128:1188-98. Crossref

46. Engel J Jr. Surgical Treatment of the Epilepsies. New York: Raven Press; 1993.

47. Wiebe S. Blume WT, Girvin JP, et al. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311-8. Crossref

48. Engel J Jr. The timing of surgical intervention for mesial temporal lobe epilepsy: a plan for a randomized clinical trial. Arch Neurol 1999;56:1338-41. Crossref

49. Malaysian Society of Neurosciences. Consensus guidelines on the management of epilepsy 2010. Available from: http://www.neuro.org.my/MSN_GUIDELINE/MSN_GUIDELINE_Consensus%20Guidelines%20on%20the%20Management%20of%20Epilepsy%202010.pdf Accessed 15 Apr 2018.

50. Noacthar S, Borggraefe I. Epilepsy surgery: a critical review. Epilepsy Behav 2009;15:66-72. Crossref

51. Rolston JD, Englot DJ, Wang DD, Shih T, Chang EF. Comparison of seizure control outcomes and the safety of vagus nerve, thalamic deep brain, and responsive neurostimulation: evidence from randomized controlled trials. Neurosurg Focus 2012;32:E14. Crossref

52. Fisher R, Salanova V, Witt T, et al. Electrical stimulation of the anterior nucleus of thalamus for treatment of refractory epilepsy. Epilepsia 2010;51:899-908. Crossref

53. Engel J Jr, McDermott MP, Wiebe S, et al. Early surgical therapy for drug-resistant temporal lobe epilepsy: a randomized trial. JAMA 2012;307:922-30. Crossref

54. Cloppenborg T, May TW, Blümcke I, et al. Trends in epilepsy surgery: stable surgical numbers despite increasing presurgical volumes. J Neurol Neurosurg Psychiatry 2016;87:1322-9. Crossref

55. Lee SK. Surgical approaches in nonlesional neocortical epilepsy. J Epilepsy Res 2011;1:47-51. Crossref

56. Lee SK, Lee SY, Kim KK et al. Surgical outcome and prognostic factors of cryptogenic neocortical epilepsy. Ann Neurol 2005;58:525-32. Crossref

57. Noe K, Sulc V, Wong-Kisiel L, et al. Long-term outcomes after nonlesional extratemporal lobe epilepsy surgery. JAMA Neurol 2013;70:1003-8. Crossref

58. Capraz IY, Kurt G, Akdemir Ö, et al. Surgical outcome in patients with MRI-negative, PET-positive temporal lobe epilepsy. Seizure 2015;29:63-8. Crossref

59. Hennessy MJ, Elwes RD, Binnie CD, Polkey CE. Failed surgery for epilepsy: A study of persistence and recurrence of seizures following temporal resection. Brain 2000;123:2445-66. Crossref

60. Sefcik RK, Opie NL, John SE, Kellner CP, Mocco J, Oxley TJ. The evolution of endovascular electroencephalography: historical perspective and future applications. Neurosurg Focus 2016;40:E7. Crossref

61. Cossu M, Cardinale F, Castana L, et al. Stereoelectroencephalography in the presurgical evaluation of focal epilepsy: a retrospective analysis of 215 procedures. Neurosurgery 2005;57:706-18. Crossref

62. González-Martínez J, Bulacio J, Thompson S, et al. Technique, results, and complications related to robot-assisted stereoelectroencephalography. Neurosurgery 2016;78:169-80. Crossref

63. Cardinale F, Rizzi M, d’Orio P, et al. A new tool for touch-free patient registration for robot-assisted intracranial surgery: application accuracy from a phantom study and a retrospective surgical series. Neurosurg Focus 2017;42:E8. Crossref

64. Gloss D, Nevitt SJ, Staba R. The role of high-frequency oscillations in epilepsy surgery planning. Cochrane Database Syst Rev 2017;(10):CD010235. Crossref

65. Holler Y, Kutil R, Klaffenböck L, et al. High-frequency oscillations in epilepsy and surgical outcome. A meta-analysis. Front Hum Neurosci 2015;9:574. Crossref

Group A β-haemolytic Streptococcus (GABHS) is a gram-positive, non-motile, non-spore-forming coccus that tends to grow in chains. It produces a large variety of extracellular enzymes and toxins and has β-haemolytic properties.1 Streptococcus pyogenes is the most important species of GABHS, causing many human diseases. The terms GABHS and S pyogenes are often used synonymously in the literature.2 Infections with GABHS typically begin in the throat or skin. The spectrum of infections ranges from mild pharyngitis (strep throat) and localised skin infections (impetigo) to moderate-to-severe manifestations in the forms of scarlet fever, pneumonia, bacteraemia, erysipelas, cellulitis, and life-threatening conditions such as necrotising fasciitis and toxic shock syndrome. Immune-mediated clinical conditions linked to GABHS infection include rheumatic fever, post-streptococcal glomerulonephritis, arthritis, and paediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS). The manifestations of acute GABHS infection mostly affect schoolchildren. There has been a surge of GABHS in Hong Kong and worldwide in recent years, and its epidemiology and clinical manifestations in Hong Kong children are reviewed in this article. A PubMed search for references was performed using the MeSH terms “Streptococcus pyogenes” or “group A Streptococcus” and “Hong Kong”, limited to ‘human’, with no filters for article type and publication time. This discussion is based on, but not limited to, the search results.

Infections related to GABHS have long been described throughout history, but Streptococcus was not discovered to be the causative organism of erysipelas and wound infection until 1847. It was further named Streptococcus pyogenes (pyo = pus, genes = forming) in 1884 after the bacteria were found in suppurative conditions.3 Scarlet fever was first identified in the medical literature in 1675 and named “scarlatina”. Before the era of antibiotics, epidemics of scarlet fever leading to high mortality had been described for centuries.3 Despite significant improvements to socioeconomic conditions over the past 50 years, S pyogenes still remains one of the top ten infective causes contributing to global childhood mortality.4 5 A review published by the World Health Organization (WHO) in 2005 estimated that approximately 18.1 million people had serious GABHS-related disease, including rheumatic heart diseases, post-streptococcal glomerulonephritis, and over 650 000 cases of invasive diseases. The WHO review identified 1.78 million new cases of serious disease and over 500 000 deaths each year. In addition, the review indicated over 111 million new cases of streptococcal pyoderma and 616 million cases of GABHS pharyngitis each year.6

The prevalence of disease varies with age, time, season, and geographic location. Group A β- haemolytic Streptococcus infections predominantly affect children aged 5 to 15 years.7 8 With the wide range of diseases caused by GABHS and the lack of standard surveillance criteria, accurate data on its associated disease burden are limited. Over the years, reporting of scarlet fever and invasive GABHS diseases has become a statutory requirement in many countries.4 Scarlet fever has been a statutorily notifiable disease in Hong Kong since 1940, with its associated disease activity closely monitored by the Centre for Health Protection. However, reporting of invasive GABHS diseases not presenting as scarlet fever by individuals and institutions is voluntary.9

The incidence of scarlet fever has remained low in developed countries, in which health care and socioeconomic conditions have improved since the late 20th century. However, in 2011, a surge of scarlet fever was observed in East Asia, including in Hong Kong, China, South Korea, and Vietnam.2 10 11 12 Recently, a similar upsurge of reports has occurred in the United States, England, and European countries.13 In Hong Kong, there was an outbreak of 1535 total cases (21.7 cases per 100 000 population) of scarlet fever in 2011. This was almost 10 times the average annual incidence throughout the previous two decades, during which annual incidence ranged from 0.0351 to 3.37 cases per 100 000 population.14 15 During the outbreak, the age distribution was similar to that in past reports, with the majority of cases in children aged <10 years and a median age of 6 years.15 Two deaths were reported. The annual incidence has since remained high, with 1100 to 1500 cases reported in subsequent years, and there was a further increase to 2353 cases reported in 2017 (Fig).16 Epidemiological analysis of reported cases in Hong Kong from 2005 to 2015 indicated that the surge was more apparent among children aged <5 years, with the highest incidence at age 3 to 5 years. There was a 5-fold increase of annual incidence in children aged <5 years, from 3.3 per 10 000 population in 2005 to 2010 to 18.1 per 10 000 population in 2012 to 2015.17

The reason for the worldwide surge of scarlet fever is still mysterious. In the 2011 Hong Kong outbreak, the emergence of scarlet fever S pyogenes emm12 clones was associated with toxin acquisition and multidrug resistance, which could have contributed to the outbreak.18 19 Environmental alterations and host immunity are other possible factors. Similar seasonal trends were observed in the pre-surge (2005-2010) and post-surge (2011-2015) periods, with troughs in early September and rises after school holidays, with peaks in January. There was a slight bimodal elevation in June during the post-surge period. Reductions by 30% to 40% of cases after school holiday weeks were observed during both periods.17 Meteorological factors including humidity, rainfall, and temperature were found to be associated with scarlet fever, but these factors sometimes had opposite valences at different periods and in different places.20

Clinical presentations of streptococcal infections vary according to bacterial virulence factors and individual host response. In 1993, an informal group of microbiologists, clinicians, and epidemiologists formed a working group on severe streptococcal infections and classified GABHS infections into five groups21: streptococcal toxic shock syndrome, invasive infections, scarlet fever, non-invasive infections (throat and skin infections), and non-suppurative sequelae.

Streptococcal toxic shock syndrome is a severe form of streptococcal infection due to toxin-mediated acute illness. Patients present with fever, rash, hypotension, and organ failure, and it can be life-threatening.9 Endotoxins produced by S pyogenes are associated with disease invasiveness. Streptococcal pyrogenic exotoxin A and streptococcal superantigen, are two of the superantigens that stimulate T-cell response and induce cytokine-mediated inflammatory reactions, thereby leading to shock and organ dysfunction. Some exotoxins can also induce shock through non-cytokine-mediated mechanisms. Streptococcal pyrogenic exotoxin B has been shown to release bradykinins, causing vasodilation, while streptolysin O can lead to dysfunction of cardiomyocytes.22 Mortality of streptococcal toxic shock syndrome in children has been reported to be 5% to 10%.23

Streptococcal toxic shock syndrome was defined by the working group on severe streptococcal infections according to the following criteria: (1) identification of GABHS in sterile sites as definite cases or in non-sterile sites as probable cases and (2) the presence of features of shock and organ dysfunction.21 Streptococcal toxic shock syndrome is characterised by the co-existence of shock and organ dysfunction in the early course of the disease. In certain cases, organ dysfunction can precede hypotension. Septic shock leading to organ dysfunction differentiates it from other invasive GABHS infections. Streptococcal toxic shock syndrome is rarely associated with pharyngitis; rather, it more commonly follows skin infections. In 50% of cases, no obvious bacterial portal of entry can be identified.22 Diagnosis is often delayed, as patients present with non-specific symptoms of fever, vomiting, and abdominal pain without localising signs in early stage of illness. The clinical course can progress rapidly to shock and organ dysfunction, and the typical signs of deep-seated infection or necrotising fasciitis usually become obvious in later stages. Streptococcal toxic shock syndrome can occur with coinfections of viral infections such as varicella and influenza B infections. Use of nonsteroidal anti-inflammatory drugs has been reported to be associated with severe necrotising fasciitis, possibly because they mask the signs and symptoms of inflammation.22 23 24 25 The clinical course of streptococcal toxic shock syndrome can be rapidly fatal, and early recognition is very important.25 Management is mainly supportive and consists of maintenance of haemodynamic stability, with administration of fluids and inotropic medications, prompt initiation of antibiotics for infection control, and surgical intervention as indicated in cases of necrotising fasciitis. Intravenous immunoglobulin may be considered as an adjunctive therapy to neutralise superantigens, although clinical evidence regarding mortality outcomes remains controversial.22 23

Invasive GABHS diseases are usually defined as clinical diseases associated with identification of S pyogenes in sterile sites including blood, cerebrospinal or pleural fluids, deep wounds, and muscles. Before the introduction of antibiotics, bacteraemia was not uncommon, especially at extremes of age. In children, upper respiratory infections, varicella infections, and skin wounds are predisposing factors for invasive GABHS infections. Pneumonia usually starts as streptococcal pharyngitis, with 40% to 50% of cases complicated by empyema. Mortality is generally low with the appropriate use of antibiotics and management of empyema.

Scarlet fever is a clinical syndrome associated with S pyogenes pharyngitis or, less commonly, skin and soft tissue infections (the latter is known as “surgical” scarlet fever). It typically presents as abrupt onset of fever, sore throat, beefy red pharynx, and enlarged and erythematous tonsils (with or without exudates) followed by the development of sandpaper-like erythematous skin rashes on the first to second day of the fever.1 Skin rashes often start from the trunk and spread to the limbs, being more prominent over flexures, axilla, and the groin (Pastia lines) while sparing the palms and soles. The cheeks become flushed, leaving the perioral region looking pale (circumoral pallor). The papillae of the tongue become swollen, leading to a strawberry-like appearance of the tongue. The rashes usually resolve in about 1 week, followed by desquamation of the hands and feet.26

The clinical manifestations of scarlet fever are caused by streptococcal pyrogenic exotoxins produced by certain strains of S pyogenes. No single toxin but rather a combination of 11 identified exotoxins are implicated. Strains of S pyogenes producing streptococcal pyrogenic exotoxins A and C and streptococcal superantigen have been associated with a few outbreaks.26

Group A β-haemolytic Streptococcus infection is the most common bacterial cause of pharyngitis and tonsillitis both in children and adults. The typical symptoms include fever, sore throat, exudative tonsils or pharynx, and cervical lymphadenopathy. However, none of these can differentiate bacterial from viral infections. According to population studies in high-resource settings, about 4% to 10% of adults and 15% of schoolchildren had episodes of symptomatic GABHS pharyngitis each year. The rates were 5 to 10 times higher in low-resource settings.5 A meta-analysis involving studies on patients aged <18 years presenting with sore throat in out-patient settings showed that the pooled prevalence of positive throat swabs for GABHS was 37%. The prevalence in children aged <5 years was lower (24%), and it was only 5% to 10% in adults.7 In a 2000 study, patients presenting to the accident and emergency department of a Hong Kong public hospital complaining of sore throat or suspected to have acute pharyngitis had throat swabs taken for culture.8 The prevalence of GABHS pharyngitis in patients aged ≤14 years was 38.2%, which was similar to that in other countries, but the prevalence in patients aged >14 years was only 2.7%. No patient was aged <3 years, and only one patient was aged >60 years.8 The findings concurred with those of previous studies, which indicated that GABHS infections are uncommon at extremes of age.

Carrier status of S pyogenes in the throat is classically defined as isolation of the bacteria in subjects without clinical features of acute pharyngitis.27 Rates of S pyogenes carrier status among asymptomatic schoolchildren have been reported as up to 15% to 20% in some studies.5 7 In Hong Kong, the earliest epidemiological study of GABHS in a general practice setting was reported in 1968.26 It showed carrier rates of 6.1% in the general population, 7.6% to 8.3% in schoolchildren, and 27.5% in a residential children’s home. Approximately 50% of skin infections and 29% of tonsillitis cases were due to GABHS infection.28 With a high carrier rate of GABHS in children, it is sometimes difficult to determine in a child presenting with fever, sore throat, and a positive GABHS culture whether the symptoms are caused by a concurrent viral infection in a carrier or genuine GABHS pharyngitis. Clues in favour of GABHS pharyngitis include epidemiologic factors or clinical findings suggestive of GABHS pharyngitis, a marked clinical response to antimicrobial therapy, negative throat swab cultures between episodes of pharyngitis, and a serological response to GABHS extracellular antigens (eg, anti-streptolysin O, anti-DNase, and antihyaluronidase).2 3 In children with recurrent pharyngitis, it may be warranted to repeat throat cultures when they are asymptomatic after treatment with antibiotics to determine their carrier status and avoid indiscriminate use of antibiotics. Eradication of bacteria by use of antibiotics in asymptomatic carriers is generally not recommended, as the risks of development of complications and transmission to others are both low in these subjects.27 The American Academy of Pediatrics Committee on Infectious Diseases suggested the following indications for treatment of carriers: (1) family history of rheumatic fever or rheumatic heart disease; (2) parental anxiety or parents considering tonsillectomy solely because of the bacterial carriage; and (3) community outbreaks of S pyogenes acute pharyngitis.

Streptococcus pyogenes can cause infections in all layers of the skin and underlying soft tissues. Superficial infections include impetigo with localised infections of the keratin layer of the epidermis. Erysipelas are characterised by multiplication and lateral spread of S pyogenes in the superficial epidermis. They can spread through the lymphatic system, causing more widespread inflammation. In contrast, cellulitis results from deeper infections of subcutaneous tissues. Streptococcus pyogenes invasion and multiplication in the fascia can lead to necrotising fasciitis, which commonly involves all layers of the skin. Streptococcus pyogenes can also cause muscle infections, leading to myositis and myonecrosis.29

Impetigo is prevalent in developing countries and commonly affects young children aged 2 to 5 years. Infections are usually localised, rarely causing systemic spread. Post-streptococcal glomerulonephritis may follow impetigo caused by nephrogenic strains of GABHS, but rheumatic fever has not been found to be associated with impetigo.29 30 31

Cellulitis may result from infections of wounds or adjacent infections such as lymphadenitis or peritonsillar abscesses.32 Such infections may spread to involve large areas of subcutaneous tissue and be complicated by bacteraemia, toxic shock syndrome, and scarlet fever. Necrotising fasciitis is rare but has the potential to be rapidly fatal with high mortality.33 34 35 36 It may be preceded by only trivial injury or varicella infection in children.37 Early recognition with a low threshold for surgical debridement is of paramount importance in its management.34 36

Streptococcus pyogenes can also cause post-infectious non-suppurative diseases. These immune-mediated complications follow a small percentage of infections and include acute rheumatic fever, post-streptococcal reactive arthritis, acute post-streptococcal glomerulonephritis, guttate psoriasis, and Henoch-Schönlein purpura.30 38 Different strains of GABHS have been found to be related to acute rheumatic fever (rheumatogenic types) and acute post-streptococcal glomerulonephritis (nephrogenic types). The incidence rates of these conditions have declined dramatically in the previous half-century with the use of antibiotics. These conditions appear several weeks following initial untreated or partially treated streptococcal infections, but it is relatively common for the preceding symptoms of infection not to be recognised. The Jones diagnostic criteria for acute rheumatic fever were first established in 1944, with recent modifications by the American Heart Association (AHA). The 1992 version stated that two out of the five major criteria for GABHS infection (carditis, arthritis, Sydenham chorea, subcutaneous nodules, and erythema marginatum) or one major with two minor criteria indicate a high probability of acute rheumatic fever.39 In 2015, the AHA incorporated the use of serial echocardiography with Doppler studies to evaluate and monitor for clinical or subclinical carditis in any confirmed or suspected cases of acute rheumatic fever. Subclinical carditis refers to mitral or aortic valvulitis revealed by echocardiography/Doppler studies without the classic auscultatory findings of valvar dysfunction.40

Acute post-streptococcal glomerulonephritis is caused by immune complexes deposited in the glomerulus. The typical clinical presentation includes acute nephritic syndrome, or less commonly, nephrotic syndrome approximately 7 to 10 days after streptococcal pharyngitis and 2 to 4 weeks after skin infections.41 Prognosis in children is usually excellent, and progressive renal failure rarely occurs. The clinical condition PANDAS was described by Swedo et al in 1998.42 The diagnostic criteria for PANDAS include: (1) the presence of obsessive–compulsive disorder and/or any other tic disorders, (2) pre-pubertal onset, (3) abrupt onset and relapsing remitting symptom course, (4) a distinct association with GABHS infection, and (5) association with neurological abnormalities, such as motoric hyperactivity and choreiform movements, during exacerbations. The pathogenesis of PANDAS is uncertain but likely linked to a post-infectious autoimmune phenomenon. Treatment is mainly symptomatic, and the efficacy of immunomodulatory therapies including plasmapheresis and intravenous immunoglobulin needs further evaluation.43

Laboratory diagnosis of GABHS infection is classically based on bacterial culture of clinical specimens. Bacterial colonies of Streptococcus produce rings of β-haemolysis on agar plates and appear as Gram-positive cocci in chains on microscopic examination.1 Group A β-haemolytic Streptococcus can be identified by detection of Lancefield group A antigens on the bacterial surface. Further differentiation of S pyogenes from other GABHS requires pyrrolidonyl arylamidase colorimetric tests and tests of bacitracin susceptibility. Recently, more advanced automated laboratory systems based on molecular techniques have enhanced the identification of specific bacteria in blood cultures.44 The role of rapid antigen detection tests (RADTs) will be discussed in more detail below. Serology tests including anti-streptolysin O titre and anti-DNase are useful for diagnosis of immune-mediated late complications of GABHS infections.

Group A β-haemolytic Streptococcus is serotyped according to the streptococcal M protein coded by the emm gene. Since the 1990s, emm sequence typing has replaced classical M protein serotyping to define GABHS strains. The epidemiology of emm strains varies geographically and temporally. A meta-analysis identified 205 emm types worldwide, with the most common emm types being emm1 and emm12. There was a greater diversity of emm types in Africa and the Pacific region than in high-income countries.45 Local and overseas studies have shown a lack of association between GABHS emm type and invasiveness or disease severity.46 47 48 Presumably, an increase in the prevalence of strains in the general population, host immunity, and environmental factors rather than the strains’ virulence contributed to the surge of GABHS infections and severity of disease.

Pharyngitis in children is commonly caused by viral infections, and GABHS is the most important bacterial cause. Throat swab culture is the gold standard for diagnosis, but the turnaround time is 2 to 3 days, limiting its utility for deciding whether antibiotics should be used for pharyngitis. The Centor score (0-4) is a 4-point clinical scale that was first reported in 1981 and is used for differentiation of GABHS pharyngitis from viral pharyngitis to decide about the use of antibiotics in adults presenting to emergency departments with sore throat. One point is scored for each of the following: fever >38°C, absence of coughing, presence of tonsillar exudates, and swollen and tender anterior cervical nodes.49 The modified Centor score or McIsaac score includes an age score applicable to children by adding one point for age 3 years to <15 years and subtracting one point for age ≥45 years while keeping the minimum score as 0 (0 or -1) and the maximum score as 4 (4 or 5).50 However, a large-scale validation study showed a positive predictive value of only about 35% to 55% for scores 3 and 4 and a negative predictive value of 80% for scores ≤2.51

Since the 1980s, commercially available RADT for detection of GABHS in throat swabs has become more widely available. Various techniques including latex agglutination, enzyme immunoassays, optical immunoassays, and molecular tests have been developed.52 In a Cochrane Review of RADT using enzyme immunoassay and optical immunoassay methods compared with throat swab cultures for diagnosis of streptococcal pharyngitis in children, the pooled sensitivity was 85.6% (95% confidence interval [CI]=83.3-87.6%), and the specificity was 95.5% (95% CI=94.5-96.2%).53 In another meta-analysis identified by the review that included molecular tests, the molecular technique category had higher sensitivity and specificity than the other methods had.54 Molecular tests are not truly point-of-care tests and require specialised equipment and personnel, limiting their application in primary care settings. International guidelines on management of acute pharyngitis for adults and children make varying recommendations about the use of the Centor or McIsaac scores and microbiological tests for diagnosis and management of acute pharyngitis.55 The United States guidelines recommend performing RADT only in patients with suspicion of GABHS pharyngitis based on clinical signs and symptoms or according to the Centor score, while the United Kingdom and some guidelines from Europe do not suggest the use of RADT.55 Practice recommendations published in the Hong Kong College of Paediatricians’ guidelines for management of acute pharyngitis do not recommend the use of any clinical scores for making management decisions. The guidelines suggest performing either a throat swab culture or RADT if clinical and epidemiological factors strongly suggest GABHS. Negative RADT is recommended to be confirmed by throat swab culture in these cases.56

Group A β-haemolytic Streptococcus is generally sensitive to penicillin and other members of the β-lactam group of antibiotics. Failure of treatment with penicillin is generally attributed to other local commensal organisms producing β-lactamase or failure to reach adequate tissue levels in the pharynx.57 In Hong Kong, macrolide-resistant strains of GABHS are common.48 According to surveillance data from the University of Hong Kong and the Public Health Laboratory Centre of the Centre for Health Protection, GABHS isolates from 2011 were all sensitive to penicillin, while more than 50% were resistant to macrolides, clindamycin, and tetracyclines.18 58 International and local management guidelines recommend either a 10-day course of penicillin or amoxicillin as first-line antibiotics for acute pharyngitis or scarlet fever due to GABHS. For patients allergic to penicillin, a narrow-spectrum cephalosporin (cephalexin, cefadroxil) is indicated, while macrolides such as azithromycin or clarithromycin are second-line options55 56 58 59 Oral penicillin V is highly sensitive, and its narrow spectrum makes it the drug of choice over amoxicillin for GABHS infection. Amoxicillin is often preferred in children because of the better taste of the suspension and its availability as chewable tablets.55 56 The Centre for Health Protection of Hong Kong recommends a shorter course of antibiotics of 5 to 7 days for streptococcal pharyngitis based on recent studies showing that shorter courses of antibiotics (mostly cephalosporins) are equally effective as a 10-day course of penicillins.60 61 However, for patients with positive GABHS cultures, scarlet fever, and in areas with high prevalence of rheumatic heart disease, a full 10-day course of treatment may still be needed.59 60 Effective treatment for GABHS infection is important for control of acute infection and prevention of late immunological manifestations including rheumatic heart disease, glomerulonephritis, and PANDAS.

Despite the availability of antibiotics, GABHS remains an important bacterial pathogen that causes a wide variety of diseases in children. Clinicians should be familiar with the clinical features of GABHS infections to decide on the appropriate use of microbiological tests and judicious use of empirical antibiotics. Early recognition of symptoms and signs of invasive infections and serious toxin-mediated conditions is the key to preventing mortality. The reasons for the recent surge of scarlet fever in Hong Kong and other parts of the world are still unclear. In addition to mandatory reporting of scarlet fever and laboratory surveillance of GABHS isolates according to current practice in Hong Kong, enhanced monitoring of severe acute GABHS infections and late immunological manifestations are important to provide a complete picture of the impact of the re-emergence of this pathogen. Further investigations on microbiological, environmental, and host factors are urgently needed to control the upsurge of GABHS infections.

All authors have made substantial contributions to the concept of this study, drafting of the article, and critical revision for important intellectual content.

As an editor of the journal, KL Hon was not involved in the peer review process of the article. The authors have no conflicts of interest to disclose. 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. This paper has not been presented, published or posted before, in whole or in part.

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

1. Leung AK, Kellner JD. Group A beta-hemolytic streptococcal pharyngitis in children. Adv Ther 2004;21:277-87. Crossref

2. Wong SS, Yuen KY. Streptococcus pyogenes and re-emergence of scarlet fever as a public health problem. Emerg Microbes Infect 2012;1:e2. Crossref

3. Ferretti J, Köhler W. History of streptococcal research. In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK333430/. Accessed 14 Jan 2018.

4. Efstratiou A, Lamagni T. Epidemiology of Streptococcus pyogenes. In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK343616/. Accessed 14 Jan 2018.

5. Sims Sanyahumbi A, Colquhoun S, Wyber R, Carapetis JR. Global disease burden of group A Streptococcus. In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK333415/. Accessed 14 Jan 2018.

6. Carapetis JR, Steer AC, Mulholland EK, Weber M. The global burden of group A streptococcal diseases. Lancet Infect Dis 2005;5:685-94. Crossref

7. Shaikh N, Leonard E, Martin JM. Prevalence of streptococcal pharyngitis and streptococcal carriage in children: a meta-analysis. Pediatrics 2010;126:e557-64. Crossref

8. Wong MC, Chung CH. Group A streptococcal infection in patients presenting with a sore throat at an accident and emergency department: prospective observational study. Hong Kong Med J 2002;8:92-8.

9. Hon KL, Fu A, Leung TF, et al. Cardiopulmonary morbidity of streptococcal infections in a PICU. Clin Respir J 2015;9:45-52. Crossref

10. Zhang Q, Liu W, Ma W, et al. Spatiotemporal epidemiology of scarlet fever in Jiangsu Province, China, 2005-2015. BMC Infect Dis 2017;17:596. Crossref

11. Mahara G, Chhetri JK, Guo X. Increasing prevalence of scarlet fever in China. BMJ 2016;353:i2689. Crossref

12. Park DW, Kim SH, Park JW, et al. Incidence and characteristics of scarlet fever, South Korea, 2008-2015. Emerg Infect Dis 2017;23:658-61. Crossref

13. Lamagni T, Guy R, Chand M, et al. Resurgence of scarlet fever in England, 2014-16: a population-based surveillance study. Lancet Infect Dis 2018;18:180-7. Crossref

14. Luk EY, Lo JY, Li AZ, et al. Scarlet fever epidemic, Hong Kong, 2011. Emerg Infect Dis 2012;18:1658-61. Crossref

15. Lau EH, Nishiura H, Cowling BJ, Ip DK, Wu JT. Scarlet fever outbreak, Hong Kong, 2011. Emerg Infect Dis 2012;18:1700-2. Crossref

16. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Statistics on communicable diseases. Available from: https://www.chp.gov.hk/en/statistics/submenu/26/index.html. Accessed 25 Oct 2018.

17. Lee CF, Cowling BJ, Lau EH. Epidemiology of reemerging scarlet fever, Hong Kong, 2005-2015. Emerg Infect Dis 2017;23:1707-10. Crossref

18. Davies MR, Holden MT, Coupland P, et al. Emergence of scarlet fever Streptococcus pyogenes emm12 clones in Hong Kong is associated with toxin acquisition and multidrug resistance. Nat Genet 2015;47:84-7. Crossref

19. Tse H, Bao JY, Davies MR, et al. Molecular characterization of the 2011 Hong Kong scarlet fever outbreak. J Infect Dis 2012;206:341-51. Crossref

20. Duan Y, Yang LJ, Zhang YJ, Huang XL, Pan GX, Wang J. Effects of meteorological factors on incidence of scarlet fever during different periods in different districts of China. Sci Total Environ 2017;581-2:19-24. Crossref

21. The Working Group on Severe Streptococcal Infections. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. JAMA 1993;269:390-1. Crossref

22. Stevens DL, Bryant AE. Severe group A streptococcal infections. In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK333425/. Accessed 14 Jan 2018.

23. Chuang YY, Huang YC, Lin TY. Toxic shock syndrome in children: epidemiology, pathogenesis, and management. Paediatr Drugs 2005;7:11-25. Crossref

24. Lam KW, Sin KC, Au SY, Yung SK. Uncommon cause of severe pneumonia: co-infection of influenza B and Streptococcus. Hong Kong Med J 2013;19:545-8. Crossref

25. Lau SK, Woo PC, Yuen KY. Toxic scarlet fever complicating cellulitis: early clinical diagnosis is crucial to prevent a fatal outcome. New Microbiol 2004;27:203-6.

26. Wessels MR. Pharyngitis and scarlet fever. In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK333418/. Accessed 14 Jan 2018.

27. Martin J. The Streptococcus pyogenes carrier state. In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK374206/. Accessed 14 Jan 2018.

28. Fischbacher E. Streptococcus pyogenes in general practice in Hong Kong. J R Coll Gen Pract 1968;16:345-52.

29. Stevens DL, Bryant AE. Impetigo, erysipelas and cellulitis. In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK333408/. Accessed 14 Jan 2018.

30. Robson WL, Leung AK. Post-streptococcal glomerulonephritis: recent increase of incidence in southern Alberta. Can Fam Physician 1992;38:2883-7.

31. Robson WL, Leung AK. Post-streptococcal glomerulonephritis. Clin Nephrol 1995;43:139.

32. Leung AK, Robson WL. Childhood cervical lymphadenopathy. J Pediatr Health Care 2004;18:3-7. Crossref

33. Donaldson PM, Naylor B, Lowe JW, Gouldesbrough DR. Rapidly fatal necrotising fasciitis caused by Streptococcus pyogenes. J Clin Pathol 1993;46:617-20. Crossref

34. Leung AK, Eneli I, Davies HD. Necrotizing fasciitis in children. Pediatr Ann 2008;37:704-10. Crossref

35. Hon KL, Leung E, Burd DA, Leung AK. Necrotizing fasciitis and gangrene associated with topical herbs in an infant. Adv Ther 2007;24:921-5. Crossref

36. Cheung JP, Fung B, Tang WM, Ip WY. A review of necrotising fasciitis in the extremities. Hong Kong Med J 2009;15:44-52.

37. de Benedictis FM, Osimani P. Necrotising fasciitis complicating varicella. Arch Dis Child 2008;93:619. Crossref

38. Robson WL, Leung AK. Acute rheumatic fever and Henoch-Schönlein purpura. Acta Paediatr 2003;92:513. Crossref

39. Special Writing Group of the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease of the Council on Cardiovascular Disease in the Young of the American Heart Association. Guidelines for the diagnosis of rheumatic fever. Jones Criteria, 1992 update. JAMA 1992;268:2069-73.

40. Gewitz MH, Baltimore RS, Tani LY, et al. Revision of the Jones Criteria for the diagnosis of acute rheumatic fever in the era of Doppler echocardiography: a scientific statement from the American Heart Association. Circulation 2015;131:1806-18. Crossref

41. Rodriguez-Iturbe B, Haas M. Post-streptococcal glomerulonephritis. In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK333429/. Accessed 14 Jan 2018.

42. Swedo SE, Leonard HL, Garvey M, et al. Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: clinical description of the first 50 cases. Am J Psychiatry 1998;155:264-71.

43. Esposito S, Bianchini S, Baggi E, Fattizzo M, Rigante D. Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: an overview. Eur J Clin Microbiol Infect Dis 2014;33:2105-9. Crossref

44. Spellerberg B, Brandt C. Laboratory diagnosis of Streptococcus pyogenes (group A streptococci). In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK343617/. Accessed 14 Jan 2018.

45. Steer AC, Law I, Matatolu L, Beall BW, Carapetis JR. Global emm type distribution of group A streptococci: systematic review and implications for vaccine development. Lancet Infect Dis 2009;9:611-6. Crossref

46. Rogers S, Commons R, Danchin MH, et al. Strain prevalence, rather than innate virulence potential, is the major factor responsible for an increase in serious group A Streptococcus infections. J Infect Dis 2007;195:1625-33. Crossref

47. Ho PL, Johnson DR, Yue AW, et al. Epidemiologic analysis of invasive and noninvasive group A streptococcal isolates in Hong Kong. J Clin Microbiol 2003;41:937-42. Crossref

48. Chan JC, Chu YW, Chu MY, Cheung TK, Lo JY. Epidemiological analysis of Streptococcus pyogenes infections in Hong Kong. Pathology 2009;41:681-6. Crossref

49. Centor RM, Witherspoon JM, Dalton HP, Brody CE, Link K. The diagnosis of strep throat in adults in the emergency room. Med Decis Making 1981;1:239-46. Crossref

50. McIsaac WJ, White D, Tannenbaum D, Low DE. A clinical score to reduce unnecessary antibiotic use in patients with sore throat. CMAJ 1998;158:75-83.

51. Fine AM, Nizet V, Mandl KD. Large-scale validation of the Centor and McIsaac scores to predict group A streptococcal pharyngitis. Arch Intern Med 2012;172:847-52. Crossref

52. Leung AK, Newman R, Kumar A, Davies HD. Rapid antigen detection testing in diagnosing group A beta-hemolytic streptococcal pharyngitis. Expert Rev Mol Diagn 2006;6:761-6. Crossref

53. Cohen JF, Bertille N, Cohen R, Chalumeau M. Rapid antigen detection test for group A Streptococcus in children with pharyngitis. Cochrane Database Syst Rev 2016;(7):CD010502. Crossref

54. Lean WL, Arnup S, Danchin M, Steer AC. Rapid diagnostic tests for group A streptococcal pharyngitis: a meta-analysis. Pediatrics 2014;134:771-81. Crossref

55. Chiappini E, Regoli M, Bonsignori F, et al. Analysis of different recommendations from international guidelines for the management of acute pharyngitis in adults and children. Clin Ther 2011;33:48-58. Crossref

56. Chan JY, Yau F, Cheng F, Chan D, Chan B, Kwan M. Practice recommendation for the management of acute pharyngitis. Hong Kong J Paediatr 2015;20:156-62.

57. Cattoir V. Mechanisms of antibiotic resistance. In: Ferretti JJ, Stevens DL, Fischetti VA, editors. Streptococcus pyogenes: Basic Biology to Clinical Manifestations. Oklahoma City, OK: University of Oklahoma Health Sciences Center; 2016. Available from: https://www.ncbi.nlm.nih.gov/books/NBK333414/. Accessed 14 Jan 2018.

58. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Letter to doctor 13 June 2011: empirical antibiotic treatment of scarlet fever due to group A Streptococcus. Available from: https://www.chp.gov.hk/files/pdf/ltd_treatment_of_gas_20110613.pdf. Accessed 14 Feb 2018.

59. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Acute pharyngitis. 2017. Available from: https://www.chp.gov.hk/files/pdf/guidance_notes_ acute_pharynitis_full.pdf. Accessed 13 May 2018.

60. Altamimi S, Khalil A, Khalaiwi KA, Milner RA, Pusic MV, Al Othman MA. Short-term late-generation antibiotics versus longer term penicillin for acute streptococcal pharyngitis in children. Cochrane Database Syst Rev 2012;(8):CD004872. Crossref

61. Falagas ME, Vouloumanou EK, Matthaiou DK, Kapaskelis AM, Karageorgopoulos DE. Effectiveness and safety of short-course vs long-course antibiotic therapy for group a beta hemolytic streptococcal tonsillopharyngitis: a meta-analysis of randomized trials. Mayo Clin Proc 2008;83:880-9. Crossref

Systemic lupus erythematosus (SLE) is a prototypical multi-systemic autoimmune disease that predominantly affects women of childbearing age. The disease has considerable clinical and immunological heterogeneity; no two patients with SLE are exactly alike. The pathogenesis of SLE remains obscure, with multiple genetic, epigenetic, hormonal, and immunopathological pathways being involved.1 The course of SLE is largely unpredictable and characterised by periods of disease exacerbation and remission that lead to progressive organ damage and dysfunction.2 Compared with the age- and sex-matched general population, SLE is associated with at least a five-fold increase in mortality.3 Patients with SLE have reduced quality of life because of multiple factors, such as organ damage, anxiety, and depression.4 5

In Hong Kong, the prevalence and annual incidence of SLE are estimated to be 0.1% and 6.7 per 100 000 population, respectively.6 The 15-year cumulative survival of local Chinese patients with SLE managed in non-academic hospitals is 86%.7 Infections, cardiovascular events, and malignancies are their most common causes of death. Renal and musculoskeletal complications (eg, avascular bone necrosis and osteoporotic fracture) are the most important contributors to disease and treatment-related organ damage, respectively. One-third of such patients lose their ability to work within 5 years after disease onset; this is mainly attributed to musculoskeletal pain, fatigue, anxiety/depression symptoms, and memory deterioration.8

Strategies for management of SLE by family physicians should target early recognition and diagnosis, treatment and monitoring of mild disease, and referral to specialists to formulate an individualised plan based on age, disease severity, organ function, and other medical co-morbidities.9 In this article, the latest criteria for classification of SLE, use of autoantibodies for diagnosis and assessment, the role of family physicians, disease monitoring, advice on various SLE-related health issues from a local perspective, and pharmacological treatment of the disease will be reviewed.

The manifestations of SLE are protean, and any body system can be involved during the course of the disease. In our experience with 803 Hong Kong Chinese patients with SLE, the most common features are arthritis, glomerulonephritis, facial rash, and haematological disease (Fig).7 In primary care practice, the most frequently encountered early symptoms of SLE include systemic upset (fatigue, fever, weight loss, loss of appetite, and prolonged influenza-like illness), arthralgia or arthritis, facial rash, photosensitivity, mouth sores, pleuritic chest pain, and Raynaud’s phenomenon.6 In hospital practice, more serious manifestations of SLE such as rapidly progressive glomerulonephritis, pulmonary haemorrhage, cardiac tamponade, severe cytopenia, and neuropsychiatric symptoms may be encountered.

The American College of Rheumatology (ACR) classification criteria for SLE were established in 1982,10 and this set of criteria was revised in 1997,11 with the deletion of the LE cell phenomenon and the addition of antiphospholipid antibodies as a criterion. A classification of SLE is made when four or more of the 11 clinical or serological criteria are fulfilled serially or simultaneously. However, in real life practice, many patients with autoimmune cytopenia or hypocomplementaemia are treated as having SLE, even though they do not fulfil the 1997 criteria.11 Moreover, dermatological manifestations other than malar rash and discoid lesions and neuropsychiatric manifestations other than psychosis and seizure are not included in the criteria. Owing to these limitations, the Systemic Lupus International Collaborating Clinics (SLICC) group revised and validated a new set of classification criteria in 2012.12 A patient is classified as having SLE when at least four of the 17 SLICC/ACR criteria are fulfilled. A comparison of the 1997 ACR and 2012 SLICC criteria is shown in Table 1.

The SLICC group emphasises the absolute requirement of at least one clinical or immunologic criterion for a classification of SLE. Lupus malar rash and photosensitivity are no longer separated into two criteria. There is no need to demonstrate the absence of radiological erosion in lupus arthritis. A number of types of subacute and chronic lupus skin lesions are included, and diffuse non-scarring alopecia (excluding alopecia areata or other causes) is also regarded as a criterion for SLE classification. Haemolytic anaemia, leukopenia/lymphopenia, and thrombocytopenia are separated into three criteria, and more neuropsychiatric features are included in addition to psychosis and seizure. For the renal criteria, the dipstick test for urine protein is replaced by either the protein-to-creatinine ratio (spot urine test) or 24-hour urine protein quantification. Finally, an entity called “stand-alone” lupus nephritis is introduced, in which a patient has typical renal biopsy features of lupus nephritis and a positive ANA or anti-dsDNA antibody test in the absence of other features of SLE. The features in the SLICC criteria must be related to active SLE instead of other causes or differential diagnoses. Validation of the SLICC criteria has demonstrated higher sensitivity (97% vs 83%) but lower specificity (84% vs 96%) for SLE than the 1997 ACR criteria.12

To further improve the sensitivity and specificity of SLE classification, the ACR/EULAR (European League Against Rheumatism) is currently validating a new set of criteria consisting of an entry criterion and 10 domains (seven clinical and three immunologic). More items with different weighted scores are included. Applications on mobile devices and desktop computers will be devised to facilitate the calculation of summed scores for classification purposes.

These classification criteria for SLE are being developed to facilitate research and comparison among different cohorts of patients. Although they generally have good specificity to aid diagnosis, false positivity and negativity are bound to occur. The final diagnosis of SLE still requires the meticulous clinical judgement of attending physicians.

Antinuclear antibody (ANA) is the hallmark of SLE. Although this antibody shows extreme sensitivity for SLE (>98%), it has low specificity. As many as 20% to 23% of normal healthy individuals test positive for ANA, particularly older subjects.13 Other autoimmune and non-immune chronic illnesses also generate positivity for ANA, making it grossly unsuitable as a sole diagnostic test. However, ANA is an excellent screening test for SLE, and a negative result by indirect immunofluorescence assay (IIFA) may virtually exclude the diagnosis.

Antinuclear antibody is conventionally detected by the IIFA method, which involves initial screening, serial serum dilution, and determination of the distinct ANA staining patterns on human epithelial cell (HEp-2) slides.14 This is the most sensitive method of ANA detection, but it is labour-intensive and subject to inter-observer reading variability. Although IIFA remains the gold standard of ANA detection, automated and less laborious quantitative methods are often used by service laboratories. Enzyme-linked immunosorbent assay is commonly used to detect serum autoantibodies directed against antigens coated onto plates.15 As the antigens used may be derived from animal tissues or recombinant techniques, the specificity and sensitivity of the results for assessment of SLE vary among different commercial kits adopted by different laboratories. In general, higher ANA titres result in more specific predictions for SLE and related disorders. Therefore, ANA should be interpreted in the clinical context, and a diagnosis of SLE should not be based on a positive ANA result alone.

The dense fine speckle (DSF) pattern of ANA in IIFA is related to autoantibodies against a 70-kDa protein (DSF70).16 Interestingly, this anti-DSF70 antibody is present in around one-third16 17 18 19 of ANA-positive healthy subjects, in contrast to less than 1% of ANA-positive patients with SLE and other autoimmune diseases.20 Anti-DSF70 is becoming a part of the standard report along with the ANA result in public hospitals. Positive results for both ANA and anti-DSF70, in the absence of other autoantibodies such as anti-dsDNA and anti–extractable nuclear antigen (anti-ENA), may virtually exclude SLE or an ANA-related autoimmune disorder.

Table 2 shows a list of pointers that should alert family physicians to consider the possibility of SLE.21 When SLE is suspected, ANA should be included in the screening blood tests. If the patient is positive for ANA, more specific tests such as those for anti-dsDNA, anti-ENA, antiphospholipid antibodies (eg, anticardiolipin antibodies), and complements are needed to confirm the diagnosis. Other relevant investigations are also needed, such as urine analysis, cell counts, renal and liver function tests, and tests for inflammatory markers such as erythrocyte sedimentation rate or C-reactive protein (CRP). Diagnosis of SLE is based on a combination of compatible clinical features and the presence of relevant immunological abnormalities. Therefore, SLE should never be diagnosed by abnormal antibody tests alone.

The ANA titre is not useful for monitoring of SLE activity. The anti-dsDNA titre and complement levels (C3/4) are the standard serological tests for disease activity evaluation (“lupus serology”). The ENAs include a number of soluble cytoplasmic and nuclear antigens. The six main antigens used to detect anti-ENA antibodies are Ro, La, Sm, RNP, Scl-70, and Jo-1. The detected antibodies are associated with certain manifestations of SLE (eg, anti-Ro with cutaneous lupus and photosensitivity) and are relevant in pregnancy (eg, anti-Ro with congenital heart blockage and neonatal lupus). Anti-Sm is specific to SLE and is a criterion for its classification.11 12 As anti-ENA antibodies seldom sero-convert over time, repeating the tests during routine follow-up is not necessary. Table 3 summarises the assessment and monitoring of patients with SLE and includes general advice about various health-related issues.

According to our experience with Chinese patients with SLE, mood disorders are its most frequent psychiatric manifestations.22 The major self-reported symptoms that lead to impaired quality of life are problems with memory and concentration and symptoms of anxiety and depression.23 Therefore, patients with SLE should receive education about the disease, psychological counselling, and support in the primary care setting.

In addition to understanding the clinical presentation of SLE for early diagnosis, trained family physicians are able to treat and monitor mild SLE, which comprises the following characteristics: (1) diagnosis clearly established; (2) clinically stable; (3) absence of life-threatening manifestations; (4) stable function of organ systems; and (5) absence of significant complications related to disease activity or treatment. Patients with stable SLE should be followed at intervals of 3 to 6 months. Referral to specialists is indicated for worsening disease activity, involvement of major organs such as the kidneys, haematological and central nervous system complications, development of disease or treatmentrelated complications, antiphospholipid syndrome, and advice about pregnancy, surgery, and other special circumstances.21

Family physicians may help to monitor disease activity and the adverse effects of drug therapies. A complete blood count, renal function, SLE serology (anti-dsDNA and complements), and urinary protein analysis should be performed every 3 to 6 months for patients with stable disease. As patients with SLE are more prone to accelerated atherosclerosis as a result of disease activity and treatment,24 surveillance for vascular risk factors such as body mass index, blood pressure, fasting lipid profile, and glucose should be done at regular intervals. Erythrocyte sedimentation rate and CRP have little role in the monitoring of SLE activity and should only be performed in patients with active synovitis undergoing specific therapy.

The ultraviolet (UV) light spectrum can be divided into UVC (100-290 nm), UVB (290-320 nm), and UVA (320-400 nm) wavelengths. The superficial layers of the epidermis mainly absorb UVB irradiation, but longer-wavelength UVA can also penetrate the deeper dermis. Ultraviolet light may trigger a complicated process that includes the activation of keratinocytes to release pro-inflammatory cytokines, chemokines, and interferons, which may exacerbate local and systemic autoimmunity.25

Photosensitivity was poorly described in the ACR criteria as skin rash resulting from an unusual reaction to sunlight, as reported in patients’ history or physicians’ observation.10 11 Some clinicians regard photosensitivity as induction or exacerbation of skin lesions after extensive sun exposure, which also includes sunburn. Because of the broad definition of photosensitivity, its incidence in SLE ranges widely (27%-100% in different studies).25 The latency period between UV exposure and skin eruptions can range from several days to 3 weeks. In addition to UV exposure, photosensitivity in SLE can be caused by photosensitising medications and co-existing photodermatosis.

Avoidance of excessive sunshine, particularly during midday hours, is often advised to patients with SLE. Hats, protective clothing, and umbrellas are effective at blocking UV light. Ultraviolet-protective sunglasses and lip balms may also help. Topical sunscreen is a common means of reducing UV light penetration. A sunscreen with sun protection factor 30 absorbs/reflects 97% of UV light.26 Patients with SLE should apply sunscreen (sun protection factor ≥30) 30 minutes before going out into the sun to all exposed body parts and re-apply it after 1 to 2 hours if exposure is to continue. Patients should be reminded that sunscreen does not provide 100% protection from UV light or offer skin support for repair of photodamage. Therefore, avoidance of unnecessary sun exposure remains the most important behavioural modification.

Vitamin D deficiency has recently been postulated to be an environmental trigger for autoimmune diseases, including SLE.27 Compared with age- and sex-matched healthy subjects, patients with SLE have significantly lower serum vitamin D levels, which correlate inversely with disease activity.28 29 30 Vitamin D insufficiency in SLE has multiple contributing factors, which include avoidance of UV exposure by using sunscreen, chronic kidney disease, long-term use of medications that hamper absorption or metabolism of vitamin D, and anti-vitamin D antibodies that may enhance plasma clearance of vitamin D.27 Although there is conflicting evidence regarding the efficacy of vitamin D supplementation at alleviating clinical SLE activity, such supplementation is recommended for prevention and treatment of glucocorticoid-induced osteoporosis.31 According to the updated ACR recommendations, patients receiving ≥3 months of prednisolone (≥2.5 mg/day) should receive elemental calcium (1000-1200 mg/day) and cholecalciferol (600-800 IU/day) along with lifestyle modification (weight bearing exercise, cessation of smoking, balanced diet, and maintaining optimal body weight).31 In patients with SLE aged >40 years, who have a moderate to high risk of a major osteoporotic (>10%) or hip fracture (>1%) within 10 years (as assessed by the fracture risk assessment tool), oral bisphosphonates are recommended. When oral bisphosphonates are inappropriate (eg, owing to intolerance or contra-indication), intravenous bisphosphonates (eg, zoledronate) are the next alternatives to be considered. Other treatment options include teriparatide (which is costly and inconvenient to inject daily), denosumab, and raloxifene (which has a lack of efficacy data regarding fractures). There is a general paucity of efficacy data of these agents in younger patients aged <40 years. The ACR recommends treatment for moderate–to-high-risk younger patients, defined as having a previous osteoporotic fracture; bone mineral density Z-score of <−3.0 at the hip or spine; or rapid loss of ≥10% bone mineral density over 1 year and continuous prednisolone treatment (≥7.5 mg/day for ≥6 months).31 The choice of drugs is the same as that for older patients, except for raloxifene, which is not indicated in premenopausal women or male patients.

Patients with SLE are prone to infections because of the underlying immune aberrations and therapies with immunosuppressive regimens.1 Vaccination offers the most cost-effective method of reducing infection risk in patients with SLE. Non-live vaccines such as influenza and pneumococcal vaccines are generally well tolerated in SLE, although they are less immunogenic than in age-matched individuals.32 Although there is conflicting evidence on whether influenza vaccine exacerbates SLE activity,33 seasonal influenza vaccination according to national guidelines is recommended.34 35 Influenza and pneumococcal vaccination is particularly recommended for patients with SLE before rituximab therapy. Additional vaccinations against Haemophilus influenzae and Neisseria meningitidis are suggested for patients with functional asplenia, splenectomy, or persistently very low complement levels.34 35 Hepatitis B vaccination can be safely administered to patients with SLE who are at risk of infection if it was not given at birth. Female patients with SLE are more prone to persistent genital human papillomavirus (HPV) infection, which predisposes them to cervical cancers. The HPV vaccine is recommended for patients with SLE, preferably prior to the beginning of sexual activity. There is no evidence of increased SLE flares after administration of the quadrivalent or bivalent HPV vaccines.36 In Hong Kong, the quadrivalent and nonavalent HPV vaccine is licensed for female and male patients aged ≥9 years.37 Non-live vaccines should be given to patients with SLE during periods of disease quiescence and minimal immunosuppression.

Live attenuated vaccines are generally not recommended for individuals who are heavily immunocompromised because of the risk of disseminated infections. Of relevance is the live attenuated herpes zoster vaccine, which has been licensed for patients aged >50 years. Patients with SLE are particularly prone to herpes zoster reactivation, with a pooled relative risk of 2.10 compared with the age- and sex-matched general population.38 According to the United States Advisory Committee on Immunization Practices, herpes zoster vaccine should not be given to individuals who are receiving heavy immunosuppressive therapies, such as prednisolone (>20 mg/day for ≥2 weeks), methotrexate (≥0.4 mg/kg/week), and azathioprine (≥3.0 mg/kg/day).39 However, in view of the high incidence of herpes zoster in patients with SLE, herpes zoster vaccine should be considered in those who have stable and remitted disease that does not require intense immunosuppression.34 35 The herpes zoster vaccine has been administered safely to SLE patients without subsequent development of herpetiform lesions or disease flares.40

The fertility of patients with SLE is preserved, unless they develop chronic kidney disease or have been treated with cyclophosphamide. Patients with SLE should not be discouraged regarding pregnancy, provided that their disease has been under good control for at least 6 to 12 months.41 The outcomes of pregnancies have improved for patients with SLE in the past few decades as a result of better risk stratification, pre-conception counselling, and close multidisciplinary surveillance. However, the rates of pregnancy loss, preterm birth, pre-eclampsia, and intrauterine growth retardation remain higher in pregnancies of patients with SLE than in those of patients without.41 The main risk factors for poor maternal and fetal outcomes in pregnancies of patients with SLE are active disease at conception (particularly nephritis), the presence of strongly positive antiphospholipid antibodies (or a history of obstetric antiphospholipid syndrome), and a history of lupus nephritis.42 Some medications such as cyclophosphamide, mycophenolate mofetil, leflunomide, and angiotensin-converting enzyme inhibitors/angiotensin receptor blockers are teratogenic. High-dose glucocorticoid treatment may lead to intrauterine growth retardation and premature delivery. The risk of congenital heart blockage in anti-Ro-positive mothers with SLE is approximately 1% to 2%.42 Close liaison with obstetricians and paediatricians for monitoring of the cardiovascular status of the fetus during pregnancy and assessment of neonatal lupus syndrome is needed. In general, SLE patients with ≥6 months of disease remission who are in good general health may consider conception. Referral to specialists for adjustment of medications and prophylactic heparin/aspirin (in case of obstetric antiphospholipid syndrome) is needed.

The use of assisted reproductive technology is increasing. Despite increases in disease flares and thrombosis after hormonal ovulation stimulation,43 the current recommendation is to individualise the risk of these procedures in patients with SLE.44 Assisted reproductive technology procedures should be discouraged in female SLE patients who have active disease, severe renal insufficiency, serious valvulopathy or coronary heart disease, poorly controlled hypertension, history of major thrombotic events, or antiphospholipid syndrome.44 Counselling should also be given about other serious adverse effects of assisted reproductive technology procedures, such as ovarian hyperstimulation. In patients with SLE who are positive for antiphospholipid antibodies and have no history of thrombosis, aspirin and heparin prophylaxis is recommended during these procedures.45 Similar to naturally achieved pregnancies, the SLE of candidates for assisted reproductive technology should have been quiescent for ≥6 months.46

Patients with SLE should be counselled about contraception methods. Barrier methods are generally safe. Oestrogen-containing oral contraceptive pills were discouraged in the past. However, in a randomised double-blind placebo-controlled trial, a combination of oral contraceptive pills was not shown to increase SLE disease flares or thrombosis after 12 months’ administration as compared with placebo in patients with stable SLE and no antiphospholipid antibodies.47 Another randomised controlled trial did not reveal a difference in disease flares or adverse events in 12 months among patients with SLE who were assigned to receive combined oral contraceptive pill, intrauterine device, and progestogen-only pills for contraception.48 Thus, patients with stable SLE and no antiphospholipid antibodies or other contra-indications may use low-dose oestrogen oral contraceptive pills if they want to adopt a more reliable contraceptive method. When oral contraceptive pills are not appropriate, progestogens and intrauterine devices can be offered to patients with SLE as alternatives.44 Progestogen-impregnated intrauterine devices have the advantage of reducing the incidence of dysmenorrhoea and irregular vaginal bleeding.44

Hydroxychloroquine is an antimalarial drug that exhibits immune-modulatory properties in addition to antithrombotic and lipid- and glucose-lowering properties.49 Hydroxychloroquine is mainly indicated for skin, joint, and serosal manifestations of SLE and has a glucocorticoid-sparing effect. The drug is compatible with pregnancy and breastfeeding and is relatively safe to be prescribed and monitored by trained family physicians. Allergy and acute ocular and neuromuscular toxicity are rare adverse drug reactions. Chronic use of hydroxychloroquine may lead to retinopathy, with the main risk factors being older age, pre-existing liver and renal dysfunction, higher daily dose, and longer duration of therapy.50 51 Early recognition of this adverse drug reaction is essential to minimise damage to vision. Referral to an ophthalmologist for baseline examination and regular retinopathy surveillance is recommended.50

In a recent study, 2361 patients received hydroxychloroquine for >5 years. In that study, the risk of retinopathy was <1% in the first 5 years and <2% in 10 years when the daily dose was <5 mg/ kg of real body weight.51 The risk of retinopathy increased sharply to 20% after 20 years. The daily dose of hydroxychloroquine was the most critical factor for the retinopathy risk, which correlated better with real rather than ideal body weight. The American Academy of Ophthalmology recommends a maximum daily hydroxychloroquine dose of <5.0 mg/kg of real weight to minimise retinal toxicity.50 A baseline ophthalmologic examination within the first year of commencement of drug administration is recommended, and annual screening should start after 5 years of exposure in patients using a lower dosage and without major risk factors. Patients with major risk factors for retinopathy (older age, renal or liver dysfunction, or pre-existing macular or retinal disease) should be screened annually if not more frequently.50

Short courses of non-steroidal anti-inflammatory drugs (NSAIDs) are indicated for control of SLE symptoms such as arthritis, myalgia, serositis, and fever. The risk of allergic and skin reactions, aseptic meningitis, and renal and liver toxicity is increased in SLE patients, despite their younger age. Ovulation may be affected by NSAIDs, and they should be used cautiously during pregnancy. Patients with SLE who have renal insufficiency, bleeding tendency, and pre-existing coronary heart disease should avoid NSAIDs. Except for their lower risk of gastrointestinal toxicity, selective Cox II inhibitors share similar renal, hepatological, and neurological adverse effects with non-selective Cox inhibitors.52 Among the NSAIDs, naproxen appears to be associated with the lowest risk of cardiovascular events and is the preferred NSAID for patients with multiple cardiovascular risk factors.53 Diclofenac is associated with the highest risk and should be avoided in these patients. A recent randomised controlled trial reported that celecoxib was non-inferior to ibuprofen or naproxen with regard to cardiovascular safety in patients with rheumatoid arthritis and osteoarthritis.54 The lowest effective dose of NSAIDs should be used, and their indications should be periodically reviewed. Monitoring of fluid status, kidney function, liver transaminases, and blood pressure is necessary.

Glucocorticoids and a number of non-glucocorticoid immunosuppressive agents are often used to treat more serious organ manifestations of SLE. Systemic glucocorticoids are a major cause of treatment-related organ damage in patients with SLE and contribute significantly to mortality and co-morbidities.7 55 Therefore, the use of systemic glucocorticoids in SLE has to be fully justified, judicious, and closely monitored. Other treatment modalities used for severe SLE include intravenous immunoglobulin and plasmapheresis. A biological agent called belimumab has recently been approved for mild to moderate SLE manifestations that are refractory to standard therapies.56 Although rituximab has not been proven to be more effective than placebo in randomised controlled trials, it is often used off-label for refractory lupus manifestations.1 Many other biological and targeted synthetic agents are being tested in patients with SLE. While it is outside the scope of this review to describe these therapies in detail, they are summarised in Table 4 for quick reference.

Systemic lupus erythematosus is a prototypical autoimmune disease that affects primarily young women of reproductive age. The new SLICC classification has expanded the clinical and serological criteria for its classification. Systemic lupus erythematosus should never be diagnosed based solely on positive test results for antibodies, particularly ANA, which is highly non-specific and should be interpreted in conjunction with clinical signs and symptoms. In view of the disease’s multisystemic involvement, holistic care is necessary to formulate treatment plans for individual patients. Family physicians play an important role in establishing an early diagnosis, treatment and monitoring of mild disease, and making referrals to specialists when appropriate. Education, counselling, and psychological support are equally important to improve treatment adherence and alleviate mood symptoms. General advice about photoprotection, vaccination, prevention of osteoporosis, and reproductive issues may be given in the primary care setting. Hydroxychloroquine is a relatively safe drug that can be commenced and monitored by family physicians. For patients with stable SLE, screening for cardiovascular risk factors and osteoporosis may also be performed periodically in family clinics.

The author has made substantial contributions to the concept or design, acquisition of data, analysis or interpretation of data, drafting of the article, and critical revision for important intellectual content.

The author has disclosed no conflicts of interest. The author had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

1. Mok CC. Biological and targeted therapies of systemic lupus erythematosus: evidence and the state of the art. Expert Rev Clin Immunol 2017;13:677-92. Crossref

2. Mok CC. Epidemiology and survival of systemic lupus erythematosus in Hong Kong Chinese. Lupus 2011;20:767-71. Crossref

3. Mok CC, Kwok CL, Ho LY, Chan PT, Yip SF. Life expectancy, standardized mortality ratios, and causes of death in six rheumatic diseases in Hong Kong, China. Arthritis Rheum 2011;63:1182-9. Crossref

4. Mok CC, Kosinski M, Ho LY, Chan KL, Jolly M. Validation of the LupusPRO in Chinese patients from Hong Kong with systemic lupus erythematosus. Arthritis Care Res (Hoboken) 2015;67:297-304. Crossref

5. Mok CC, Ho LY, Cheung MY, Yu KL, To CH. Effect of disease activity and damage on quality of life in patients with systemic lupus erythematosus: a 2-year prospective study. Scand J Rheumatol 2009;38:121-7. Crossref

6. Mok CC. Emerging biological therapies for systemic lupus erythematosus. Expert Opin Emerg Drugs 2014;19:303-22. Crossref

7. Mok CC, Tse SM, Chan KL, Ho LY. Effect of immunosuppressive therapies on survival of systemic lupus erythematosus: a propensity score analysis of a longitudinal cohort. Lupus 2018;27:722-7. Crossref

8. Mok CC, Cheung MY, Ho LY, Yu KL, To CH. Risk and predictors of work disability in Chinese patients with systemic lupus erythematosus. Lupus 2008;17:1103-7. Crossref

9. Lam NC, Ghetu MV, Bieniek ML. Systemic lupus erythematosus: primary care approach to diagnosis and management. Am Fam Physician 2016;94:284-94.

10. Tan EM, Cohen AS, Fries JF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271-7. Crossref

11. Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1997;40:1725. Crossref

12. Petri M, Orbai AM, Alarcón GS, et al. Derivation and validation of the Systemic Lupus International Collaborating Clinics classification criteria for systemic lupus erythematosus. Arthritis Rheum 2012;64:2677-86. Crossref

13. Pisetsky DS. Antinuclear antibody testing—misunderstood or misbegotten? Nat Rev Rheumatol 2017;13:495-502. Crossref

14. Rigon A, Infantino M, Merone M, et al. The inter-observer reading variability in anti-nuclear antibodies indirect (ANA) immunofluorescence test: A multicenter evaluation and a review of the literature. Autoimmun Rev 2017;16:1224-9.Crossref

15. Emlen W, O’Neill L. Clinical significance of antinuclear antibodies: comparison of detection with immunofluorescence and enzyme-linked immunosorbent assays. Arthritis Rheum 1997;40:1612-8. Crossref

16. Dellavance A, Viana VS, Leon EP, Bonfa ES, Andrade LE, Leser PG. The clinical spectrum of antinuclear antibodies associated with the nuclear dense fine speckled immunofluorescence pattern. J Rheumatol 2005;32:2144-9.

17. Infantino M, Meacci F, Grossi V, et al. The clinical impact of anti-DFS70 antibodies in undifferentiated connective tissue disease: case reports and a review of the literature. Immunol Res 2017;65:293-5. Crossref

18. Mahler M, Hanly JG, Fritzler MJ. Importance of the dense fine speckled pattern on HEp-2 cells and anti-DFS70 antibodies for the diagnosis of systemic autoimmune diseases. Autoimmun Rev 2012;11:642-5. Crossref

19. Mariz HA, Sato EI, Barbosa SH, Rodrigues SH, Dellavance A, Andrade LE. Pattern on the antinuclear antibody-HEp-2 test is a critical parameter for discriminating antinuclear antibody-positive healthy individuals and patients with autoimmune rheumatic diseases. Arthritis Rheum 2011;63:191-200. Crossref

20. Seelig CA, Bauer O, Seelig HP. Autoantibodies against DFS70/LEDGF exclusion markers for systemic autoimmune rheumatic diseases (SARD). Clin Lab 2016;62:499-517. Crossref

21. Mok CC. Systemic lupus erythematosus: when to refer? HK Pract 2002;24:444-9.

22. Mok CC, To CH, Mak A. Neuropsychiatric damage in Southern Chinese patients with systemic lupus erythematosus. Medicine (Baltimore) 2006;85:221-8. Crossref

23. Mok CC, Ho LY, Tse SM, Chan KL. Prevalence of remission and its effect on damage and quality of life in Chinese patients with systemic lupus erythematosus. Ann Rheum Dis 2017;76:1420-5. Crossref

24. Mok CC. Accelerated atherosclerosis, arterial thromboembolism, and preventive strategies in systemic lupus erythematosus. Scand J Rheumatol 2006;35:85-95. Crossref

25. Kuhn A, Wenzel J, Bijl M. Lupus erythematosus revisited. Semin Immunopathol 2016;38:97-112. Crossref

26. Obermoser G, Zelger B. Triple need for photoprotection in lupus erythematosus. Lupus 2008;17:525-7. Crossref

27. Mok CC. Vitamin D and systemic lupus erythematosus: an update. Expert Rev Clin Immunol 2013;9:453-63. Crossref

28. Shahin D, El-Farahaty RM, Houssen ME, et al. Serum 25-OH vitamin D level in treatment-naïve systemic lupus erythematosus patients: relation to disease activity, IL-23 and IL-17. Lupus 2017;26:917-26. Crossref

29. Mok CC, Birmingham DJ, Leung HW, Hebert LA, Song H, Rovin BH. Vitamin D levels in Chinese patients with systemic lupus erythematosus: relationship with disease activity, vascular risk factors and atherosclerosis. Rheumatology (Oxford) 2012;51:644-52. Crossref

30. Amital H, Szekanecz Z, Szücs G, et al. Serum concentrations of 25-OH vitamin D in patients with systemic lupus erythematosus (SLE) are inversely related to disease activity: is it time to routinely supplement patients with SLE with vitamin D? Ann Rheum Dis 2010;69:1155-7. Crossref

31. Buckley L, Guyatt G, Fink HA, et al. 2017 American College of Rheumatology Guideline for the prevention and treatment of glucocorticoid-induced osteoporosis. Arthritis Rheumatol 2017;69:1521-37. Crossref

32. Pugès M, Biscay P, Barnetche T, et al. Immunogenicity and impact on disease activity of influenza and pneumococcal vaccines in systemic lupus erythematosus: a systematic literature review and meta-analysis. Rheumatology (Oxford) 2016;55:1664-72. Crossref

33. Murdaca G, Orsi A, Spanò F, et al. Vaccine-preventable infections in systemic lupus erythematosus. Hum Vaccin Immunother 2016;12:632-43. Crossref

34. van Assen S, Agmon-Levin N, Elkayam O, et al. EULAR recommendations for vaccination in adult patients with autoimmune inflammatory rheumatic diseases. Ann Rheum Dis 2011;70:414-22. Crossref

35. Heijstek MW, Ott de Bruin LM, Bijl M, et al. EULAR recommendations for vaccination in paediatric patients with rheumatic diseases. Ann Rheum Dis 2011;70:1704-12. Crossref

36. Mok CC, Ho LY, Fong LS, To CH. Immunogenicity and safety of a quadrivalent human papillomavirus vaccine in patients with systemic lupus erythematosus: a case-control study. Ann Rheum Dis 2013;72:659-64. Crossref

37. Cervical screening programme, Department of Health, Hong Kong SAR Government. Available from: https://www.cervicalscreening.gov.hk/english/hum/hum_ccv.html#4. Accessed 1 May 2018.

38. Kawai K, Yawn BP. Risk factors for herpes zoster: a systematic review and meta-analysis. Mayo Clin Proc 2017;92:1806-21. Crossref

39. Harpaz R, Ortega-Sanchez IR, Seward JF; Advisory Committee on Immunization Practices (ACIP), Centers for Disease Control and Prevention. Prevention of herpes zoster: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2008;57:1-30.

40. Guthridge JM, Cogman A, Merrill JT, et al. Herpes zoster vaccination in SLE: a pilot study of immunogenicity. J Rheumatol 2013;40:1875-80. Crossref

41. Fischer-Betz R, Specker C. Pregnancy in systemic lupus erythematosus and antiphospholipid syndrome. Best Pract Res Clin Rheumatol 2017;31:397-414. Crossref

42. Peart E, Clowse ME. Systemic lupus erythematosus and pregnancy outcomes: an update and review of the literature. Curr Opin Rheumatol 2014;26:118-23. Crossref

43. Orquevaux P, Masseau A, Le Guern V, et al. In vitro fertilization in 37 women with systemic lupus erythematosus or antiphospholipid syndrome: a series of 97 procedures. J Rheumatol 2017;44:613-8. Crossref

44. Andreoli L, Crisafulli F, Tincani A. Pregnancy and reproductive aspects of systemic lupus erythematosus. Curr Opin Rheumatol 2017;29:473-9. Crossref

45. Andreoli L, Bertsias GK, Agmon-Levin N, et al. EULAR recommendations for women’s health and the management of family planning, assisted reproduction, pregnancy and menopause in patients with systemic lupus erythematosus and/or antiphospholipid syndrome. Ann Rheum Dis 2017;76:476-85. Crossref

46. Levine AB, Lockshin MD. Assisted reproductive technology in SLE and APS. Lupus 2014;23:1239-41. Crossref

47. Petri M, Kim MY, Kalunian KC, et al. Combined oral contraceptives in women with systemic lupus erythematosus. N Engl J Med 2005;353:2550-8. Crossref

48. Sánchez-Guerrero J, Uribe AG, Jiménez-Santana L, et al. A trial of contraceptive methods in women with systemic lupus erythematosus. N Engl J Med 2005;353:2539-49. Crossref

49. Ponticelli C, Moroni G. Hydroxychloroquine in systemic lupus erythematosus (SLE). Expert Opin Drug Saf 2017;16:411-9. Crossref

50. Marmor MF, Kellner U, Lai TY, Melles RB, Mieler WF; American Academy of Ophthalmology. Recommendations on screening for chloroquine and hydroxychloroquine retinopathy (2016 revision). Ophthalmology 2016;123:1386-94. Crossref

51. Melles RB, Marmor MF. The risk of toxic retinopathy in patients on long-term hydroxychloroquine therapy. JAMA Ophthalmol 2014;132:1453-60. Crossref

52. Ostensen M, Villiger PM. Nonsteroidal anti-inflammatory drugs in systemic lupus erythematosus. Lupus 2000;9:566-72. Crossref

53. Pelletier JP, Martel-Pelletier J, Rannou F, Cooper C. Efficacy and safety of oral NSAIDs and analgesics in the management of osteoarthritis: evidence from real-life setting trials and surveys. Semin Arthritis Rheum 2016;45(4 Suppl):S22-7. Crossref

54. Nissen SE, Yeomans ND, Solomon DH, et al. Cardiovascular safety of celecoxib, naproxen, or ibuprofen for arthritis. N Engl J Med 2016;375:2519-29. Crossref

55. Tarr T, Papp G, Nagy N, Cserép E, Zeher M. Chronic high-dose glucocorticoid therapy triggers the development of chronic organ damage and worsens disease outcome in systemic lupus erythematosus. Clin Rheumatol 2017;36:327-33. Crossref

56. Hahn BH. Belimumab for systemic lupus erythematosus. N Engl J Med 2013;368:1528-35. Crossref

Tuberculosis (TB) is a major global health burden that ranks with human immunodeficiency virus (HIV)/acquired immune deficiency syndrome (AIDS) as a leading cause of death worldwide. The World Health Organization (WHO) estimated that 9.6 million people were sickened by TB and 1.5 million died as a result in 2014, with 58% of global TB cases occurring in the South-East Asia and Western Pacific regions.1 2 As a part of the global response to TB, the sixth Millennium Development Goal (MDG) set out to halve TB prevalence and mortality rates by 2015 compared with the 1990 baseline.3 Following significant declines in TB mortality and prevalence rates, in 2015, the third Sustainable Development Goals contained targets to end the epidemics of AIDS, TB, malaria, and neglected tropical disease by 2030.4 The TB target for 2030 is to reduce the number of TB deaths by 90% compared with 2015 numbers. The WHO established the End TB Strategy in 2014, aiming to reduce the TB burden by 2030 and eliminate TB entirely by 2050.5 6 Advanced economies such as the US and Australia7 typically have low TB incidence, and TB is commonly known as a disease of poverty that more heavily affects developing countries.8 The Global Fund is conducting country case studies on HIV/AIDS, TB, and malaria in several developing countries, including Haiti, Pakistan, and the Philippines.9 No country case studies have yet been conducted in developed Asian countries/regions such as Hong Kong, Japan, Singapore, Taiwan, or South Korea, which have good health infrastructure and stable economic growth, but where intermediate levels of TB incidence persist.10 11

Reaching the WHO targets in Asia will require strategies specific to TB epidemiology in this setting. However, information to adequately inform strategies is lacking. The last report of comparative data between Asian countries was published 10 years ago by the WHO.5 The reasons for the gap between the TB burden in Asian countries and that in their equally developed counterparts in other regions need to be understood. The TB burden in low-incidence countries is attributable mostly to immigrants.12 In contrast, the stagnant intermediate incidence in developed Asian countries is ascribed mainly to latent TB infection in ageing populations.13

Compared with that of Singapore, Japan, or Western countries with similar gross domestic products, the notification rate of TB in Hong Kong is relatively high (60 per 100 000 population in 2016).11 14 Presently, TB is the second most common notifiable disease in Hong Kong, following chickenpox.15 The TB notification rate in Hong Kong has declined slowly since 1995, although the notification rate only dropped below 100 per 100 000 population in 2002, and it took until 2011 for the notification rate to decline below 70 per 100 000 population.16

The present case study of the TB situation in Hong Kong highlights successful policies intended to achieve WHO goals and identifies areas for further research or intervention in gaps that could prevent attainment of these targets. This could facilitate useful comparisons with the situation in other developed Asian countries.

A document review including both policy and literature was conducted. Statistics on TB notification in Hong Kong were obtained from the official website of the Tuberculosis and Chest Service, Department of Health of Hong Kong SAR Government.14 The TB notification rates were analysed in terms of immigrant status, age-group, and gender and presented in line graphs. The notification trend was extrapolated to 2030 by using Microsoft Excel’s FORECAST function on the trend in the past 10 years (2005-2015).

Existing documents from the Tuberculosis and Chest Service, Department of Health of Hong Kong SAR Government, such as the TB manual (2006),17 TB annual reports (2007-2013),18 19 20 21 22 23 24 information and guidelines (2006-2015),25 26 27 28 29 30 31 and other recommendations were obtained. Reports and strategies regarding TB control and elimination from the WHO were also reviewed to analyse how the strategy had been operationalised, how this may affect implementation of local programmes, and to identify the policy gap between the Hong Kong government’s and WHO’s strategies.

Two electronic databases, PubMed and Google Scholar, were searched to identify articles related to TB control and elimination in Hong Kong. The key words ‘tuberculosis’ or ‘TB’ in combination with the terms ‘Hong Kong’, ‘epidemiology’, ‘risk factors’, ‘prevention’, ‘treatment’, ‘Latent TB’, ‘MDR-TB’, or ‘XDR-TB’ were used to search for relevant articles.

Selected publications included studies (a) carried out in Hong Kong; (b) published in the past 10 years; (c) related to TB prevalence, at-risk populations, and TB control measures/interventions in Hong Kong; (d) with full-text articles in English; (e) with no overlapping data; and (f) qualitative studies with sufficient sample size, significant results (P<0.05), and specified outcomes/outputs.

Since 1947, a downward trend in total TB notification in Hong Kong has been observed. From 1970 to 1977, TB notification rapidly declined but remained stagnant thereafter (Fig 1). The oldest age-group (≥75 years) had much higher TB notification rates (Fig 2). Between 1995 and 2015, reductions in notification rates occurred in younger age-groups but increased sharply at the turn of the millennium in the oldest group, whose notification rate had only gradually decreased by 2015. Notification rates in both genders showed downward trends, although men had a higher notification rate than women (data not shown).

Tuberculosis notification rates dropped rapidly after the Bacille Calmette–Guérin (BCG) vaccine was introduced in 1952, with a further decline after the introduction of directly observed treatment short course (DOTS) [Fig 1]. The incidence in 2015 had almost halved compared with that in 1990. Extrapolated trends showed that at the current rate, Hong Kong would be unable to meet the WHO target of a 90% drop in incidence rate by 2030. By then, Hong Kong’s TB notification rate is predicted to drop by only 60.2%, compared with that in 2015. The analysis shows that Hong Kong could become a low-incidence country (10 cases per 100 000 population) by 2036.

A comparison between the WHO’s and Hong Kong’s TB policies is shown in Table 1.4 6 17 30 32 33 In 2015, the Sustainable Development Goal 3 included a target to end the TB epidemic by 2030,4 and the End TB Strategy aims to achieve a 90% drop in TB incidence rate and up to 95% reduction in number of TB-related deaths by 2035 compared with those in 2015.6 Yet, the Hong Kong government has not set a clear strategy and timeline for specific goals in TB control and elimination.

The WHO guidelines for management of latent TB infection (LTBI) strongly recommended that high-income and upper-middle income countries with TB incidence less than 100 per 100 000 population per year perform systematic testing and treatment of LTBI in specific groups, and Hong Kong was listed among these.32 Hong Kong follows the WHO recommendations for LTBI screening in high-risk groups. However, conditional recommendations for a number of target populations to be included in active case finding are not included in the local Hong Kong policy documents. According to the Hong Kong TB Manual, active case finding in high-risk groups was not very effective, as only 1% of active TB was found in household contacts in 2004.17

We reviewed the TB literature about studies conducted in Hong Kong published in the past 10 years (Table 2).34 35 36 37 38 39 40 41 42 43 44 45 46 Thirteen published studies were included: two on older adults in old age homes, one on migrant populations, two on drug-resistant TB, two on HIV-related TB, two on primary school children, three on TB treatment outcomes, and one on TB prevalence in Hong Kong.

Among the included studies, three indicated that Hong Kong’s TB prevalence rate is stagnating because of high TB prevalence in older adults and a high risk of TB reactivation34 35 caused by high prevalence of latent infection among older adults in old age homes.36 Some immigrants come from countries with higher TB incidence and drug resistance rates, particularly mainland China. These migrants may also be at increased risk of TB reactivation.37 However, TB in the migrant population is likely to decrease as migration from China is reduced and living conditions for those entering the city improve.38

Multidrug-resistant TB (MDR-TB) is a threat that is more likely in patients diagnosed with TB at younger ages.39 Extensively drug-resistant TB (XDR-TB) significantly increases household TB transmission, demonstrating a need for prolonged household surveillance.40 Treatment of LTBI is recommended to control TB, especially among people with HIV. Two studies reported outcomes of treating LTBI in patients with HIV in Hong Kong, one confirming the usefulness of LTBI treatment,41 while the other doubted the utility of LTBI tests in annual screening of patients with HIV because of discordant results between different tests.42 Identification of children with LTBI is also useful: in a study that described the use of tuberculin tests to screen primary school children, strong tuberculin reactions (>15 mm) predicted TB in adolescence.43 44 Diagnosis of TB is still problematic, and new methods are needed to prevent delayed diagnosis and treatment,35 as mortality of hospitalised TB patients is still high.45 However, one Hong Kong study demonstrated that although early diagnosis and treatment are recommended, TB therapy carried a high risk of side-effects in the study population.36 Directly observed treatment short course has significantly decreased TB incidence,38 46 although not all patients in Hong Kong completed the first 2 months of treatment, with failure to complete treatment predicting poorer outcomes than undergoing the full course.46

In Hong Kong’s older adult population, TB accounts for the majority of the city’s high burden from the disease. In Hong Kong, those aged >75 years showed an especially high TB incidence rate. Migrants and people with HIV also have higher TB prevalence but contribute significantly less to the burden than do older adults. Children with a strong purified protein derivative reaction indicating infection were more prone to develop TB in adolescence. Also, MDR-TB and XDR-TB pose a relatively rare but important threat in Hong Kong. Late or underdiagnosis results in high TB-related mortality in those who present symptoms late and require hospitalisation.

High rates of LTBI in Hong Kong have been documented in other Asian countries with low and intermediate TB burden.47 The BCG vaccine was introduced to Hong Kong in April 195217; therefore, by 1995, 2005 and 2015, those aged >43, >53 and >63 years, respectively, would not have been vaccinated in infancy. The higher prevalence of LTBI and active TB in old age homes compared with that in older adults living in the community is a trend shared with other countries, including low-burden countries such as the US.48 49 50 Despite the higher prevalence of LTBI in institutionalised older adults in Hong Kong (68.6%)36 compared with their American counterparts (5.5%),51 52 53 54 55 56 Hong Kong has not followed the US policy of LTBI testing in this population.57 Further research is needed to explore the feasibility and cost-effectiveness of screening and providing prophylaxis to older adults and other populations.

In contrast to countries with low TB burden, where infections in migrants primarily contribute to the burden,58 the infection rate in Hong Kong’s migrant population is declining.16 However, MDR-TB rates are higher in migrants and younger age-groups in Hong Kong and countries with low TB burden.59 60 A systematic review also concurred with a Hong Kong study’s findings that patients with HIV had a higher risk of MDR-TB.41 61 Meanwhile, the findings on transmission of XDR-TB in Hong Kong differ from those in Peru, where household contacts reported a very high prevalence of XDR-TB.62 It has been postulated that in Hong Kong, XDR-TB is mainly transmitted outside the household setting because of the high population density.40 The Peru study’s different findings may support this idea, as the population density of Hong Kong is more than double that of Lima.63 64

The WHO has called for improved tests to diagnose LTBI, as the current ones lack accuracy.65 This was echoed by findings in the study of patients with HIV by Leung et al.42 The finding that a strong tuberculin reaction in 6-to-10-year-old schoolchildren in Hong Kong predicted TB in adolescence was reinforced by a similar study in Singapore, which is also a developed city with an intermediate TB burden.43 66 67 However, Hong Kong schoolchildren are not routinely screened for LTBI.30 It may be advisable to extend LTBI testing to cover schoolchildren.

The high mortality of hospitalised patients with TB in Hong Kong is also seen in many other countries,45 68 emphasising the need for early detection and treatment. The DOTS strategy is an important cornerstone of TB treatment; however, there is room for improvement in compliance with DOTS in Hong Kong.46 Other developed Asian countries have similar DOTS treatment success rates to Hong Kong.69 Without improvement in medication adherence, treatment success rates are unlikely to rise.

Hong Kong reached the MDG target of reducing TB incidence, with a declining notification rate. However, according to the extrapolated trend, if improvements are not instituted, there will likely be only a 60% reduction in TB notification by 2030 compared with the 2015 baseline. To achieve the goal of 80% reduction in TB incidence proposed by the End TB Strategy,70 an improved supportive protocol targeting older adults with a clear timeline is needed. In addition, the Hong Kong government should consider screening high-risk groups included in the WHO’s conditional recommendations. More research needs to be done to explore whether screening these groups would be beneficial.

This study has several limitations. First, some key literature and important policies or strategies may have been missed, as no systematic review was conducted. This may have imposed error on the screening and article selection. Second, some patients that did not seek health care may have been missed by the system. Despite these limitations, this research has provided helpful suggestions and valuable insights for future research and implementation of TB-related policy.

The TB incidence rate is currently under control in Hong Kong, but further actions are warranted if the elimination targets are to be achieved. More accurate diagnostic tools are required, and policies targeting LTBI in older adults and children should be implemented to achieve the WHO goal by 2030.

Concept or design: G Tam.
Acquisition of data: H Yang.
Analysis or interpretation of data: H Yang.
Drafting of the article: All authors.
Critical revision for important intellectual content: G Tam, T Meyers.

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

All authors have disclosed no conflicts of interest. 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. Abstract of this article was presented at Infection 2016 (13th Annual Scientific Meeting), 22 June 2016, The Chinese University of Hong Kong, Hong Kong.

1. World Health Organization. Tuberculosis, Fact sheet N104. Available from: http://www.who.int/mediacentre/factsheets/fs104/en/. Accessed 11 Jul 2018.

2. World Health Organization. Global tuberculosis report 2015. Available from: http://www.who.int/tb/publications/global_report/gtbr15_main_text.pdf. Accessed 11 Jul 2018.

3. World Health Organization. Global tuberculosis report 2014. Available from: http://www.who.int/tb/publications/global_report/gtbr14_main_text.pdf. Accessed 11 Jul 2018.

4. World Health Organization. Health in 2015: from MDGs to SDGs. WHO report 2015. Available from: http://www.who.int/gho/publications/mdgs-sdgs/en/. Accessed 11 Jul 2018.

5. World Health Organization. Framework towards TB elimination in low-incidence countries. 2014. Available from: http://www.who.int/tb/publications/Towards_TB_Eliminationfactsheet.pdf. Accessed 11 Jul 2018.

6. World Health Organization. The End TB strategy; global strategy and targets for tuberculosis prevention, care and control after 2015. 2016. Available from: http://www.who.int/tb/post2015_TBstrategy.pdf. Accessed 11 Jul 2018.

7. International Monetary Fund. World economic outlook: adjusting to lower commodity prices. 2015. Available from: http://www.imf.org/external/pubs/ft/weo/2015/02/pdf/text.pdf. Accessed 11 Jul 2018.

8. World Health Organization. Addressing poverty in TB control. Options for national TB control programmes. Available from: http://www.who.int/tb/areas-of-work/population-groups/poverty/en/ Accessed 11 Jul 2018.

9. The Global Fund. Where we invest, The Global Fund. Available from: http://www.theglobalfund.org/en/overview/. Accessed 11 Jul 2018.

10. The World Bank. Incidence of tuberculosis (per 100,000 people). Available from: http://data.worldbank.org/indicator/SH.TBS.INCD. Accessed 11 Jul 2018.

11. World Health Organization. Tuberculosis (TB), tuberculosis country profiles. Available from: http://www.who.int/tb/country/data/profiles/en/. Accessed 11 Jul 2018.

12. Pareek M, Abubakar I, White PJ, Garnett GP, Lalvani A. Tuberculosis screening of migrants to low-burden nations: insights from evaluation of UK practice. Eur Respir J 2011;37:1175-82. Crossref

13. World Health Organization. Tuberculosis control in the South-East Asia and Western Pacific regions 2005: a biregional report. Available from: http://iris.wpro.who.int/bitstream/handle/10665.1/5519/9290611960_eng.pdf.Accessed 11 Jul 2018.

14. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Statistics on tuberculosis, 2015. Available from: http://www.info.gov.hk/tb_chest/index_2.htm. Accessed 11 Jul 2018.

15. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Number of notifiable infectious disease by month in 2015, notifiable infectious disease, Statistics on communicable diseases. Available from: http://www.chp.gov.hk/en/data/1/10/26/43/3829.html. Accessed 11 Jul 2018.

16. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Notification and death rate of tuberculosis (all forms), 1947-2015. Available from: http://www.chp.gov.hk/en/data/4/10/26/43/88.html. Accessed 11 Jul 2018.

17. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Tuberculosis manual 2006. 2006. Available from: http://www.info.gov.hk/tb_chest/doc/Tuberculosis_Manual2006.pdf. Accessed 11 Jul 2018.

18. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Annual report 2007. 2007. Available from: http://www.info.gov.hk/tb_chest/doc/Annual Report 2007.pdf. Accessed 11 Jul 2018.

19. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Annual report 2008. 2008. Available from: http://www.info.gov.hk/tb_chest/doc/Annual_Report_2008.pdf. Accessed 11 Jul 2018.

20. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Annual report 2009. 2009. Available from: http://www.info.gov.hk/tb_chest/doc/AnnualReport2009.pdf. Accessed 11 Jul 2018.

21. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Annual report 2010. 2010. Available from: http://www.info.gov.hk/tb_chest/doc/AnnualReport2010.pdf. Accessed 11 Jul 2018.

22. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Annual report 2011. 2011. Available from: http://www.info.gov.hk/tb_chest/doc/AnnualReport2011.pdf. Accessed 11 Jul 2018.

23. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Annual report 2012. 2012. Available from: http://www.info.gov.hk/tb_chest/doc/AnnualReport2012.pdf. Accessed 11 Jul 2018.

24. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Annual report 2013. 2013. Available from: http://www.info.gov.hk/tb_chest/doc/AnnualReport2013.pdf. Accessed 11 Jul 2018.

25. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Recommendations on the management of human immunodeficiency virus and tuberculosis coinfection. 2008. Available from: http://www.chp.gov.hk/files/pdf/recommendations_on_management_of_human_immunodeficiency_virus_and_tuberculosis_coinfection_r.pdf. Accessed 11 Jul 2018.

26. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Guidelines on the management of multidrug-resistant and extensively drugresistant tuberculosis in Hong Kong. 2008. Available from: http://www.info.gov.hk/tb_chest/doc/MDRTBguideline_0812.pdf. Accessed 11 Jul 2018.

27. Centre for Health Protection, Department of Health, Hong Kong SAR Government. The use of BCG vaccine in HIV infected patients. 2009. Available from: http://www.chp.gov.hk/files/pdf/the_use_of_bcg_vaccine_in_hiv_infected_patients_r.pdf. Accessed 11 Jul 2018

28. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Ambulatory treatment and public health measures for a patient with uncomplicated pulmonary tuberculosis—an information paper. 2013. Available from: http://www.info.gov.hk/tb_chest/doc/Information_paper_ambulatory_tb_2013.pdf. Accessed 11 Jul 2018.

29. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Guidance on logistical arrangement for targeted screening and treatment of latent tuberculosis infection among immunocompetent household contacts of smear-positive pulmonary tuberculosis patients in TB and Chest Service. 2013. Available from: http://www.info.gov.hk/tb_chest/doc/Guidance_Notes_Contact_TBCS_upated 1 NOV 2013.pdf. Accessed 11 Jul 2018.

30. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Guidelines on targeted tuberculin testing and treatment of latent tuberculosis infection. 2015. Available from: http://www.info.gov.hk/tb_chest/doc/LTBI_guide_TBCS_2012_update 1 Nov2013_ADD_31March2015.pdf. Accessed 11 Jul 2018.

31. Tuberculosis and Chest Service, Department of Health, Hong Kong SAR Government. Scheme for investigation of close contacts (household) in the Tuberculosis & Chest Service, Department of Health. 2015. Available from: http://www.info.gov.hk/tb_chest/doc/TST_contacts of smear-negative sources.pdf. Accessed 11 Jul 2018.

32. Getahun H, Matteelli A, Abubakar I, et al. Management of latent mycobacterium tuberculosis infection: WHO guidelines for low tuberculosis burden countries. Eur Respir J 2015;46:1563-76. Crossref

33. World Health Organization. Weekly epidemiological record. 2004. Available from: http://www.who.int/wer/2004/en/wer7904.pdf?ua=1. Accessed 11 Jul 2018.

34. Chan-Yeung M, Cheung AH, Dai DL, et al. Prevalence and determinants of positive tuberculin reactions of residents in old age homes in Hong Kong. Int J Tuberc Lung Dis 2006;10:892-8.

35. Vynnycky E, Borgdorff MW, Leung CC, Tam CM, Fine PE. Limited impact of tuberculosis control in Hong Kong: attributable to high risks of reactivation disease. Epidemiol Infect 2008;136:943-52. Crossref

36. Chan-Yeung M, Chan FH, Cheung AH, et al. Prevalence of tuberculous infection and active tuberculosis in old age homes in Hong Kong. J Am Geriatr Soc 2006;54:1334-40. Crossref

37. Leung CC, Chan CK, Chang KC, et al. Immigrants and tuberculosis in Hong Kong. Hong Kong Med J 2015;21:318-26. Crossref

38. Wu P, Cowling BJ, Schooling CM, et al. Age-period-cohort analysis of tuberculosis notifications in Hong Kong from 1961 to 2005. Thorax 2008;63:312-6. Crossref

39. Law WS, Yew WW, Chiu Leung C, et al. Risk factors for multidrug-resistant tuberculosis in Hong Kong. Int J Tuberc Lung Dis 2008;12:1065-70.

40. Leung EC, Leung CC, Kam KM, et al. Transmission of multidrug-resistant and extensively drug-resistant tuberculosis in a metropolitan city. Eur Respir J 2013;41:901-8. Crossref

41. Chan CK, Alvarez Bognar F, Wong KH, et al. The epidemiology and clinical manifestations of human immunodeficiency virus–associated tuberculosis in Hong Kong. Hong Kong Med J 2010;16:192-8.

42. Leung CC, Chan K, Yam WC, et al. Poor agreement between diagnostic tests for latent tuberculosis infection among HIV-infected persons in Hong Kong. Respirology 2016;21:1322-9. Crossref

43. Leung CC, Yew WW, Au KF, et al. A strong tuberculin reaction in primary school children predicts tuberculosis in adolescence. Pediatr Infect Dis J 2012;31:150-3. Crossref

44. Leung CC, Yew WW, Chang KC, et al. Risk of active tuberculosis among schoolchildren in Hong Kong. Arch Pediatr Adolesc Med 2006;160:247-51. Crossref

45. Lui G, Wong RY, Li F, et al. High mortality in adults hospitalized for active tuberculosis in a low HIV prevalence setting. PLoS One 2014;9:e92077. Crossref

46. Wong MY, Leung CC, Tam CM, Lee SN. Directly observed treatment of tuberculosis in Hong Kong. Int J Tuberc Lung Dis 2005;9:443-9.

47. Mori T, Leung CC. Tuberculosis in the global aging population. Infect Dis Clin North Am 2010;24:751-68. Crossref

48. Stead WW, Dutt AK. Tuberculosis in elderly persons. Annu Rev Med 1991;42:267-76. Crossref

49. Stead WW, Lofgren JP, Warren E, Thomas C. Tuberculosis as an endemic and nosocomial infection among the elderly in nursing homes. N Engl J Med 1985;312:1483-7.Crossref

50. Macarthur C, Enarson DA, Fanning EA, Hessel PA, Newman S. Tuberculosis among institutionalized elderly in Alberta, Canada. Int J Epidemiol 1992;21:1175-9. Crossref

51. Shea KM, Kammerer JS, Winston CA, Navin TR, Horsburgh CR Jr. Estimated rate of reactivation of latent tuberculosis infection in the United States, overall and by population subgroup. Am J Epidemiol 2014;179:216-25. Crossref

52. Welty C, Burstin S, Muspratt S, Tager IB. Epidemiology of tuberculous infection in a chronic care population. Am Rev Respir Dis 1985;132:133-6.

53. Dorken E, Grzybowski S, Allen EA. Significance of the tuberculin test in the elderly. Chest 1987;92:237-40. Crossref

54. Stead WW, To T. The significance of the tuberculin skin test in elderly persons. Ann Intern Med 1987;107:837-42. Crossref

55. Nisar M, Williams CS, Ashby D, Davies PD. Tuberculin testing in residential homes for the elderly. Thorax 1993;48:1257-60. Crossref

56. Creditor MC, Smith EC, Gallai JB, Baumann M, Nelson KE. Tuberculosis, tuberculin reactivity, and delayed cutaneous hypersensitivity in nursing home residents. J Gerontol 1988;43:M97-100. Crossref

57. Centers for Disease Control and Prevention. Latent tuberculosis infection: a guide for primary health care providers. 2013. Available from: http://www.cdc.gov/tb/publications/ltbi/targetedtesting.htm. Accessed 11 Jul 2018.

58. Pareek M, Greenaway C, Noori T, Munoz J, Zenner D. The impact of migration on tuberculosis epidemiology and control in high-income countries: a review. BMC Med 2016;14:48. Crossref

59. Hargreaves S, Lönnroth K, Nellums LB, et al. Multidrugresistant tuberculosis and migration to Europe. Clin Microbiol Infect 2017;23:141-6. Crossref

60. Faustini A, Hall AJ, Perucci CA. Risk factors for multidrug resistant tuberculosis in Europe: a systematic review. Thorax 2006;61:158-63.Crossref

61. Mesfin YM, Hailemariam D, Biadgilign S, Kibret KT. Association between HIV/AIDS and multi-drug resistance tuberculosis: a systematic review and meta-analysis. PLoS One 2014;9:e82235. Crossref

62. Becerra MC, Appleton SC, Franke MF, et al. Tuberculosis burden in households of patients with multidrug-resistant and extensively drug-resistant tuberculosis: a retrospective cohort study. Lancet 2011;377:147-52. Crossref

63. World population review. Lima population 2017. 2018. Available from: http://worldpopulationreview.com/worldcities/lima-population/. Accessed 11 Jul 2018.

64. World population review. Hong Kong population 2017. 2018. Available from: http://worldpopulationreview.com/countries/hong-kong-population/#. Accessed 11 Jul 2018.

65. World Health Organization. WHO guidelines on the management of latent tuberculosis infection launched today. Available from: http://www.who.int/tb/features_archive/LTBI/en/. Accessed 11 Jul 2018.

66. Chee CB, Soh CH, Boudville IC, Chor SS, Wang YT. Interpretation of the tuberculin skin test in Mycobacterium bovis BCG-vaccinated Singaporean schoolchildren. Am J Respir Crit Care Med 2001;164:958-61. Crossref

67. World Health Organization. Tuberculosis. 2007. Available from: http://www.wpro.who.int/mediacentre/factsheets/fs_20060829/en/. Accessed 11 Jul 2018.

68. Waitt CJ, Squire SB. A systematic review of risk factors for death in adults during and after tuberculosis treatment. Int J Tuberc Lung Dis 2011;15:871-85. Crossref

69. Tam CM. The DOTS strategy in Hong Kong. Med Bull 2006;11:3-4.

70. World Health Organization. The End TB Strategy. Available from: http://www.who.int/tb/End_TB_brochure.pdf?ua=1. Accessed 11 Jul 2018.