In January 2020, a 32-year-old Filipino woman was admitted to the medical unit of our hospital with subacute onset of a facial malar rash, digital vasculitis purpura and alopecia. Systemic lupus erythematosus (SLE) had been newly diagnosed according to the Systemic Lupus International Collaborating Clinics criteria. Initial blood tests revealed an elevated serum anti-ds DNA antibody level, positive anti-SM antibody, low C3 and C4 as well as pancytopenia. Fasting glucose and lipid profile were unremarkable. She was first treated with oral hydroxychloroquine 200 mg daily and oral prednisolone 20 mg daily. One week later, she presented to our ophthalmology clinic with acute-onset painless blurring of vision in her right eye upon waking that morning. Physical examination revealed a best-corrected visual acuity of finger counting at 30 cm in her right eye and 6/6 in her left eye with a right relative afferent pupillary defect. Slit-lamp examination was unremarkable with no anterior chamber inflammation. Fundus examination revealed numerous Purtscher flecken, which are polygonal patches of retinal whitening with distinct border and normal retina in between, across the posterior pole. A pseudo cherry-red spot and confluent retinal whitening could be seen in the right eye secondary to proximal occlusion of the retinal arteries. A few flame-shaped retinal haemorrhages were observed inferior and temporal to the optic disc in the right eye. The right optic disc was mildly pale with sharp margins and the retinal vessels were not tortuous, while the left optic disc is pink (Fig 1).

Urgent optical coherence tomography of the right eye showed hyper-reflectivity in the retinal nerve fibre layer and areas of retinal thickening. Fundus fluorescein angiography (FFA) of the patient’s right eye showed capillary non-perfusion corresponding to Purtscher flecken, superior branch retinal arterial occlusion, and peripapillary staining (Fig 2). Optical coherence tomography and FFA of the patient’s left eye were normal. Given the typical fundus appearance and FFA findings, our provisional diagnosis was Purtscher-like retinopathy with branch retinal artery occlusion related to SLE. We advised her medical team to escalate her systemic treatment immediately and she was given intravenous methylprednisolone 500 mg daily for 3 days, followed by oral cyclophosphamide 500 mg for 1 day before switching to oral prednisolone 40 mg daily. Oral hydroxychloroquine 200 mg daily was continued. On high-dose steroids, she developed features of psychosis that subsequently resolved. Magnetic resonance imaging of the brain showed no features typical of central nervous system lupus such as ischaemia and vasculitis. We closely monitored her condition and visual acuity had not improved 1 month after systemic treatment was started, remaining at finger counting at 30 cm. Fundus examination of the right eye after treatment revealed macula oedema and pseudo cherry-red spot, with no signs of optic atrophy. Cotton wool spots in the left eye had resolved. The patient chose to seek further medical attention overseas.

At least one third of patients with SLE have ophthalmological involvement. Retinal and choroidal pathology are sinister ocular complications of SLE that can lead to permanently impaired visual acuity. Purtscher-like retinopathy is a type of vaso-occlusive retinopathy associated with multiple conditions including connective tissue disorders, pancreatic disease, renal disease, and haematological disease.1

We report this rare case of Purtscher-like retinopathy in a patient with SLE which had a devastating visual outcome despite systemic treatment. In an observational case series of 5688 patients with SLE, eight cases of Purtscher-like retinopathy were diagnosed.1 All patients had received treatment but most had optic atrophy and persistent low visual acuity,1 consistent with the outcome for our patient. Poor visual acuity has also been attributed to presentation with a pseudo cherry-red spot. The prognosis of Purtscher-like retinopathy is better in patients with other underlying diseases. For example, Shahlaee et al2 reported a case of postviral Purtscher-like retinopathy presenting with finger count visual acuity in both eyes. No treatment was given but the patient’s final visual acuity improved to 20/20 with bilateral scotoma on visual field assessment. Another case of Purtscher-like retinopathy in a patient with adult-onset Still’s Disease was reported by Yachoui,3 in which initial visual acuity was 20/20 in the right eye and 20/25 in the left. Oral prednisolone 60 mg daily was initiated and the patient’s final visual acuity was 20/25 in both eyes with evidence of established bilateral optic atrophy.

In a retrospective case control study, Gao et al4 reported a 0.66% prevalence of retinal vasculopathy among SLE patients with signs ranging from cotton wool spots, retinal vascular attenuation to retinal haemorrhages. Lupus retinopathy can result in severe complications such as neovascularisation, macula oedema and retinal vessel occlusion resulting in vision loss.4 It is known that patients with high Systemic Lupus Erythematosus Activity Disease Index score are at higher risk of Purtscher-like retinopathy, central nervous system lupus1 and vaso-occlusive retinopathy, reflecting severe systemic microangiopathy.1 4

Because patients with active lupus are at risk of Purtscher-like retinopathy that can lead to severe irreversible visual loss,1 it is important to raise awareness in patients with active lupus and visual symptoms. A high index of suspicion for retinopathy is needed for those with very active disease and prompt referral to an ophthalmologist is warranted in the presence of visual symptoms. Despite escalation of treatment after the onset of visual symptoms, our patient’s vision did not recover. Early aggressive treatment to control SLE disease activity is essential to prevent this blinding complication.

1. Wu C, Dai R, Dong F, Wang Q. Purtscher-like retinopathy in systemic lupus erythematous. Am J Ophthalmol 2014;158:1335-1341.e1. Crossref

2. Shahlaee A, Sridhar J, Rahimy E, Shieh WS, Ho AC. Paracentral acute middle maculopathy associated with post viral Purtshcer-like retinopathy. Retinal Cases Brief Rep 2019;13:50-3. Crossref

3. Yachoui R. Purtscher-like retinopathy associated with adult-onset still disease. Retin Cases Brief Rep 2018;12:379-81. Crossref

4. Gao N, Li MT, Li YH, et al. Retinal vasculopathy in patients with systemic lupus erythematous. Lupus 2017;26:1182-9. Crossref

In July 2020, a 50-year-old man presented to a public hospital in Hong Kong with a 3-week history of abdominal pain localising to the right upper quadrant, associated with low-grade fever, chills, and fatigue. He denied any diarrhoea or vomiting. His medical history and family history were unremarkable. He was a non-smoker and consumed alcohol infrequently. He had recently returned from Southern California, United States, where he was bitten by a horse 1 month before the onset of his symptoms. Despite sustaining deep wounds to his left neck (Fig 1), he did not seek medical attention at the time.

On admission, a contrast-enhanced computed tomography (CT) scan of the abdomen revealed three liver abscesses. He was treated initially with intravenous antibiotic (amoxicillin/clavulanic acid for 5 days before switching to piperacillin/tazobactam) but his clinical condition showed no improvement. A new CT scan 5 days later showed enlargement of the abscesses. Percutaneous drainage of the abscesses was performed under ultrasound guidance and the patient was transferred to our private hospital for further management.

On arrival, the patient was haemodynamically stable but clinically dehydrated, jaundiced, and delirious. The abdomen was soft and non-tender, with three abdominal drainage tubes in situ. A new CT scan confirmed that the liver abscesses and the drainage tubes were blocked, and they had to be replaced (Fig 2). Drainage fluid afterwards was noted to have the appearance of anchovy sauce and amoebic infection was suspected.

Blood tests on arrival showed leucocytosis (white cell count 31 × 10⁹/L) with markedly elevated inflammatory markers (C-reactive protein 357 mg/L), and deranged liver function (alkaline phosphatase 231 U/L, alanine aminotransferase 64 U/L, bilirubin 56 μmol/L and albumin 26.5 g/L). Blood cultures were negative. No abnormalities were found on colonoscopy and stool ova, cysts and parasite microscopy were negative. Fluid drained from the abscesses was negative on microscopy and culture, but positive for Entamoeba histolytica DNA on polymerase chain reaction. Initial amoebic serology was negative for E histolytica immunoglobulin G antibodies. Intravenous metronidazole was commenced empirically in addition to piperacillin/ tazobactam. The patient subsequently made a steady recovery. New amoebic serology demonstrated a significant titre of E histolytica immunoglobulin G antibodies 9 days later, indicating recent amoebic infection. Repeated blood tests showed continued normalisation of initial results and follow-up CT scans showed progressive resolution of abscesses. The patient was discharged from the hospital after 2 weeks.

Amoebiasis is a leading parasitic infection in terms of morbidity and mortality worldwide. It is most prevalent in developing countries and in tropical regions. It is caused by the protozoa, E histolytica, transmitted via the faecal-oral route by ingestion of contaminated food or water containing amoebic trophozoites or cysts. Acute infection typically presents as amoebic colitis. The most common symptoms associated with amoebiasis are abdominal pain (98% of cases), fever (74%), and dysentery (30%).1 2 Although extraintestinal complications of invasive infection are rare (<1% of cases), liver abscesses are the most common secondary manifestation, occurring when trophozoites invade the colonic mucosa and penetrate mesenteric venules to enter the portal circulation.3

This case was an unusual presentation of amoebic liver abscess, confirmed by polymerase chain reaction testing of aspirates and serology results that suggested an acute infection. However, the patient presented with no gastrointestinal symptoms typical of amoebic colitis and stool investigations and colonoscopy were normal. The typical period of incubation for amoebiasis in those with liver abscesses has been reported to be 8 to 20 weeks.4 In this case, the patient presented with symptoms of invasive disease 4 weeks following his horse bite. The patient denied recent travel to an endemic area or any sick contacts. Southern California is known to have many migrants from South America where amoeba is prevalent. There was little evidence from the patient’s clinical history or investigations to support a faecal-oral route of transmission. As such, the possibility of liver abscess from a cutaneous source was considered.

The patient was initially treated for pyogenic liver abscess. Drainage of the abscesses and the addition of metronidazole dramatically improved his condition. It is conceivable that the horse bite harboured amoebic trophozoites or otherwise facilitated their invasion from contaminated environmental water, soil, or vegetation. Just as invasive trophozoites are known to reach the liver by hematogenous dissemination to form abscesses, in our patient they may have entered the circulation via the bite wound to invade the liver. The short incubation period was also consistent with direct inoculation. Although cutaneous bacterial infection leading to liver pyogenic abscesses is well reported in the literature, the development of amoebic liver abscesses from a similar source has not previously been described. This may be the first described case of amoebic liver abscesses of cutaneous origin and warrants further study.

Concept or design: AKK Chui.
Acquisition of data: AKK Chui.
Analysis or interpretation of data: AKK Chui.
Drafting of the manuscript: Both authors.
Critical revision of the manuscript for important intellectual content: Both authors.

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

As an editor of the Journal, AKK Chui was not involved in the peer review process for this article. The other author has disclosed no conflicts of interest.

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

The patient was treated in accordance with the tenets of the Declaration of Helsinki. Patient consent was obtained.

1. Akgün Y, Taçyιlιdιz ÏH, Çelik Y. Amoebic liver abscess: changing trends over 20 years. World J Surg 1999;23:102-6. Crossref

2. Donovon AJ, Yellin AE, Ralls PW. Hepatic abscess. World J Surg 1991;15:162-9. Crossref

3. Swaminathan V, O’Rourke J, Gupta R, Kiire CF. An unusual presentation of an amoebic liver abscess: the story of an unwanted souvenir. BMJ Case Rep 2013;2013:bcr2012006964. Crossref

4. Li E, Stanley SL Jr. Protozoa. Amebiasis. Gastroenterol Clin North Am 1996;25:471-92. Crossref

A 64-year-old man presented to the emergency department with acute onset abdominal pain. His medical history was unremarkable. Initial vital signs showed blood pressure of 87/43 mmHg and pulse rate of 110 beats per minute. His abdomen was distended, with generalised tenderness and guarding. The patient was stabilised with 1 L of crystalloid. Bedside ultrasonographic examination revealed intraperitoneal free fluid. Subsequent computed tomography of the abdomen demonstrated gross haemoperitoneum and multiple hepatic lesions highly suspicious of hepatocellular carcinoma (HCC). Active contrast extravasation was noted at the posterior aspect of segment 2/3 lesion, compatible with the diagnosis of ruptured HCC. The initial haemoglobin level was 8.7 g/dL. Transarterial embolisation was urgently arranged but the patient went into cardiac arrest. Spontaneous circulation returned after 8 minutes of cardiopulmonary resuscitation. Haemoglobin level fell to 5.5 g/dL and platelet count was normal. Prothrombin time and activated partial thromboplastin time were prolonged to 20.3 s and 62.1 s, respectively. International normalised ratio was 1.9. He was transfused with 2 units of pack cells and 4 units of plasma. Transarterial embolisation was successfully performed. Active contrast extravasation over a branch of the left hepatic artery was demonstrated but controlled by Gelfoam injection.

The patient was then transferred to the intensive care unit. Rotational thromboelastometry (ROTEM^®; Tem International GmbH, Munich, Germany) was performed to guide transfusion strategy (Fig a). Maximum lysis was shown to be 100%, indicating abnormally accelerated clot lysis. Maximum lysis over 15% is diagnostic of hyperfibrinolysis. The lysis index at 30 minutes was 5%, indicating almost complete clot dissolution 30 minutes after initial formation. No dysfunction in coagulation activation or clot propagation was otherwise detected. Tranexamic acid 1 g was administrated according to the interpretation of ROTEM results. A follow-up ROTEM analysis after antifibrinolytic treatment demonstrated restoration of normal fibrinolysis (Fig b). Haemostasis was achieved and no further blood transfusion was needed.

Testing for hepatitis B surface antigen was later revealed to be reactive. This patient probably had liver cirrhosis and HCC consequent to chronic hepatitis B infection. The occurrence of ruptured HCC tipped the balance in this vulnerable patient. He developed abdominal compartment syndrome and hepatic failure. The sequential organ failure assessment score was 13 with an estimated mortality of over 90%. The surgical team advised optimum medical supportive treatment in view of the extremely high operative risk. The patient succumbed 16 hours after hospital admission.

Haemocoagulation is a complex interaction of procoagulants, anticoagulants, fibrinolytic proteins, and cellular components. Platelets are activated in response to vascular injury, leading to primary haemostasis. Activated platelets adhere to damaged endothelium and aggregate to create a temporary platelet plug. Secondary haemostasis involves activation of the coagulation cascade, resulting in fibrin formation. The platelet plug is strengthened by the cross-linked fibrin to form a stable clot. Fibrinolysis serves as the final stage of coagulation to regulate the extent of clot formation and to maintain vascular patency. Cross-linked fibrin is broken down by plasmin and normal blood flow is restored.

Hyperfibrinolysis refers to excessive fibrinolytic activity that threatens clot integrity and leads to defective haemostasis. It is common in patients with chronic liver disease, major trauma, and obstetric complications. However, the incidence is not well studied and it has often been underdiagnosed due to an absence of appropriate tests.1 Conventional coagulation tests, such as prothrombin time, activated partial thromboplastin time, and thrombin time, are ineffective in detecting hyperfibrinolysis. Viscoelastic haemostatic assays are the only tests for rapid detection and quantification of hyperfibrinolysis.2

Rotational thromboelastometry is a form of viscoelastic assay. It provides a global haemostatic assessment from initial platelet activation, through platelet aggregation, clot strengthening by cross-linked fibrin, to clot dissolution. The degree of clot lysis is expressed numerically as maximum lysis. Hyperfibrinolysis is diagnosed when maximum lysis exceeds 15% within a 60-minute ROTEM analysis. Schöchl et al3 further quantified this condition based on the time course of clot dissolution, as reflected by the lysis index at 30 minutes and at 60 minutes. They categorised complete clot lysis within 30 minutes as fulminant hyperfibrinolysis; intermediate when complete clot lysis was within 30 to 60 minutes; and late when complete clot lysis exceeded 60 minutes. Prompt recognition of hyperfibrinolysis by ROTEM analysis guides appropriate antifibrinolytic therapy. Untreated hyperfibrinolysis has been shown to be associated with refractory bleeding and increased mortality.1

Rotational thromboelastometry has multiple advantages over conventional coagulation tests. In contrast to conventional coagulation tests that use plasma, ROTEM uses whole blood and can determine the contribution of both cellular and plasma components of haemostasis. Conventional coagulation tests provide only a quantitative assessment of individual procoagulation factors; results are not necessarily a good indication of the in vivo haemostatic process, especially in patients with chronic liver disease.4 Rotational thromboelastometry also provides real-time functional assessment of the haemostatic process, from clot formation to its lysis, and is useful for identifying hyperfibrinolysis. There is evidence that use of ROTEM-guided haemostatic strategy reduces transfusion requirement in different clinical settings.5 6 7

Understanding its limitations is equally important in the interpretation of ROTEM analysis. It is insensitive to the effect of platelet inhibitors such as aspirin and clopidogrel. Thrombin, the strongest activator of platelets, is produced in large amounts during ROTEM analysis, masking the inhibitory effects of antiplatelet agents.8 It is also poor at detecting conditions that affect platelet adhesion and aggregation, with Von Willebrand’s disease being a well-known example.9 The ROTEM analysis is undertaken at 37°C to mimic physiological conditions so the negative effect of hypothermia on coagulation is not reflected by the analysis.9

In conclusion, owing to the limitations of conventional coagulation tests, hyperfibrinolysis is frequently underdiagnosed. With its ability of real-time functional haemostatic assessment from clot formation to lysis, ROTEM analysis allows better insight into the complex haemocoagulation process. The rapid detection of hyperfibrinolysis by ROTEM can guide prompt antifibrinolytic treatment, possibly reducing the transfusion requirement in bleeding patients.

Concept or design: KM Kwok, KL Lee, SY Lam, T Liong, KI Law.
Acquisition of data: KM Kwok.
Analysis or interpretation of data: KM Kwok, KL Lee, SY Lam, T Liong, KI Law.
Drafting of the manuscript: KM Kwok.
Critical revision of the manuscript for important intellectual content: All authors.

The authors have no conflicts of interest to disclose.

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

The patient was treated in accordance with the Declaration of Helsinki. Patient was incapable to provide informed consent for publication.

1. Theusinger OM, Wanner GA, Emmert MY, et al. Hyperfibrinolysis diagnosed by rotational thromboelastometry (ROTEM) is associated with higher mortality in patients with severe trauma. Anesth Analg 2011;113:1003-12. Crossref

2. Yeung MC, Tong SY, Tong PY, Cheung BH, Ng JY, Leung GK. Use of viscoelastic haemostatic assay in emergency and elective surgery. Hong Kong Med J 2015;21:45-51. Crossref

3. Schöchl H, Frietsch T, Pavelka M, Jámbor C. Hyperfibrinolysis after major trauma: differential diagnosis of lysis patterns and prognostic value of thrombelastometry. J Trauma 2009;67:125-31. Crossref

4. Tripodi A, Mannucci PM. The coagulopathy of chronic liver disease. N Engl J Med 2011;365:147-56. Crossref

5. De Pietri L, Bianchini M, Montalti R, et al. Thrombelastography-guided blood product use before invasive procedures in cirrhosis with severe coagulopathy: a randomized, controlled trial. Hepatology 2016;63:566-73. Crossref

6. Veigas PV, Callum J, Rizoli S, Nascimento B, da Luz LT. A systematic review on the rotational thrombelastometry (ROTEM®) values for the diagnosis of coagulopathy, prediction and guidance of blood transfusion and prediction of mortality in trauma patients. Scand J Trauma Resusc Emerg Med 2016;24:114. Crossref

7. Vymazal T, Astraverkhava M, Durila M. Rotational thromboelastometry helps to reduce blood product consumption in critically ill patients during small surgical procedures at the intensive care unit—a retrospective clinical analysis and literature search. Transfus Med Hemother 2018;45:385-7. Crossref

8. Lang T, von Depka M. Possibilities and limitations of thrombelastometry/-graphy [in German]. Hamostaseologie 2006;26(3 Suppl 1):S20-9. Crossref

9. Srivastava A, Kelleher A. Point-of-care coagulation testing. Continuing Educ Anaesthesia Crit Care Pain 2013;13:12-6. Crossref

A 79-year-old man presented with left-sided weakness, drowsiness and dysphasia following left atrial appendage occlusion implantation in a cardiac intensive care unit. A sudden drop in systolic blood pressure to 40 mmHg was noted. Echocardiogram and angiogram revealed no obvious coronary artery disease. Computed tomography (CT) of the brain showed no major vessel occlusion or gas pocket. A new CT of the brain showed more established infarct and oedema. Four sessions of hyperbaric oxygen therapy (HBOT) were conducted in accordance with United States Navy Treatment Tables (USNTT)1: two USNTT6 and two USNTT5.

A 40-year-old woman was admitted to a surgical unit for suspected acute appendicitis. Urgent CT of the abdomen with contrast was performed, but approximately 60 mL of air was injected intravenously during the procedure and gas was evident in the right ventricle. Luckily, she was asymptomatic as the gas emboli had entered only the venous system and remained in the heart. One session of HBOT was conducted with USNTT6.

A 57-year-old woman underwent fine needle aspiration cytology under CT guidance for lung mass. The patient developed sudden onset bilateral anopia and left upper limb weakness during the procedure. Computed tomography brain and thorax identified a small gas embolism in the pulmonary vein with pulmonary haemorrhage. No obvious gas bubble was noted in the brain. A total of four sessions of HBOT were conducted with two USNTT6 and two USNTT5.

An 84-year-old man underwent CT-guided lung biopsy for right lung mass but experienced an acute deterioration in Glasgow Coma Scale (GCS) score. Urgent CT of the brain showed multiple acute gas emboli and confirmed the diagnosis of cerebral arterial gas embolism (AGE). On arrival at the HBOT centre, he had GCS of 13/15 and right limb weakness with power of 1/5 and 3/5, respectively. Two sessions of HBOT were conducted with USNTT6.

A 66-year-old man with biliary pancreatitis underwent urgent endoscopic retrograde cholangiopancreatography. After the procedure he developed left-sided weakness with limb power of 0/5 after 5 minutes. Urgent CT of the brain revealed branching gas densities at the sulcal space of his right frontal and parietal lobes (Fig 1). Two sessions of HBOT with USNTT6 were conducted and the patient’s symptoms considerably improved.

A 72-year-old man underwent CT-guided fine needle aspiration cytology for a lung lesion. The patient developed transient hypotension on needle removal and CT of the thorax revealed gas embolism in his left ventricle with no pneumothorax (Fig 2). The patient remained asymptomatic with no neurological deficit. Gas embolism had been introduced through the pulmonary circulation but remained in the heart, avoiding the complication of cerebral arterial occlusion. One session of HBOT was conducted with USNTT6.

Gas embolism is a serious and life-threatening condition. It is usually iatrogenic and should be treated promptly to prevent serious complications such as cerebral AGE. It presents with sudden deterioration in neurological and haemodynamic status following risky medical procedures, most notably lung biopsy and cardiac interventions. This case report revealed six patients with gas embolism: five with AGE and one with venous gas embolism. Four patients with AGE manifested as cerebral AGE with neurological deficits such as limb weakness and anopia (Table).

As soon as AGE is suspected, the patient should start receiving 100% high-flow oxygen and be placed in the right lateral decubitus position. The definitive treatment for AGE is HBOT with 100% oxygen, although no randomised controlled trials have been conducted to confirm its efficacy in this disease. Hyperbaric oxygen therapy “crushes” the bubbles with pressure, accelerates gas diffusion and bubble resolution with oxygen, oxygenates ischaemic tissue, reduces cerebral oedema, and decreases neutrophil adhesion to the endothelium.1 2 A retrospective study in France showed that the initial neurological symptoms were impaired consciousness (70%), focal motor deficits (60%), seizures (11%), visual disturbance (10%), dysarthria/aphasia (5%) and vertigo (5%). In all, 26% of patients were haemodynamically unstable and subject to hypotension, circulatory collapse, and cardiac arrest. Respiratory disturbances such as dyspnoea, cough or chest pain were present in 23% of patients.3

In a study of 45 patients with AGE, good neurological outcome was achieved in 27 (60%) of them.4 Time to receipt of HBOT was the only statistically significant factor predictive of a good outcome with mean delay 8.8 hours.4 Although probability of a good outcome is highest when HBOT is administered as soon as possible, a good response can still be obtained if treatment is delayed for longer than 24 hours.3 5 There is a tendency for patients with AGE to deteriorate after their initial apparent recovery; thus, early HBOT is still recommended for patients with a seemingly spontaneous recovery.6 Although identification of gas bubbles on CT of the brain is not a prerequisite to HBOT, early treatment can also attenuate leucocyte adherence to damaged endothelium and secondary inflammation that in turn facilitate the return of blood flow. However, gas bubbles can persist for several days, and many case reports have demonstrated the benefit of HBOT after delays ranging from hours to days. Complete recovery has been reported in one patient in whom treatment was delayed for 28 hours. In 2017, another patient who was deeply comatose with AGE recovered and was able to lead a functional life after HBOT had been delayed for 6 days.7 8 In our cases, all the patients were treated within 7 hours onset of symptoms. Most fully recovered except for Patient 1. The aetiology of neurological deterioration in Patient 1 was likely due to a combination of gas embolism, clots, and hypotension.

According to the United States Navy Diving Manual,9 USNTT6 is the standard initial compression therapy for AGE. It is a table with maximum pressure up to 2.8 atmospheres absolute (ATA) for 1 hour 15 minutes with slow depressurisation to 1.9 ATA lasting for 2.5 hours. Total treatment time is approximately 4 hours 45 minutes. There is no conclusive evidence that use of pressure higher than 2.8 ATA has any advantage. Use of pressure up to 6 ATA poses further risks to the patient and attending staff, including gas toxicity and decompression sickness. Typically, one to three hyperbaric treatments are required to treat AGE; the clinical condition should be monitored until there is no further stepwise improvement. There are reports that up to 10 treatments have been required to reach a stable condition.10 11

The USNTT5, which is a shorter treatment used to treat mild decompression sickness, may be used together with USNTT6 for AGE when prolonged treatment is required. This treatment table advises a maximum pressure up to 2.8 ATA for 45 minutes with slow depressurisation to 1.9 ATA for 30 minutes. Total treatment time is approximately 2 hours 15 minutes. According to the Royal Navy Diving manual,12 which is commonly used in Europe, RN62 Table is the standard to treat AGE. This treatment is similar to USNTT6 with a maximum pressure of 2.8 ATA.

Arterial gas embolism presents in highly variable ways with no pathognomonic signs or symptoms. It is uncommon to achieve complete consensus on the diagnosis, and it is common practice to err on the side of caution. Hyperbaric oxygen therapy should be initiated as early as possible although delayed treatment is also effective. Failure to administer treatment can result in long-term disability. Hyperbaric oxygen therapy should be considered for any suspected AGE.

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

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

All authors have disclosed no conflicts of interest.

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

The patients were treated in accordance with the Declaration of Helsinki. The patients provided informed consent for the treatment/procedures, and verbal consent for publication.

1. United States Navy. Diagnosis and treatment of decompression sickness and arterial gas embolism. In: US Navy Diving Manual. Revision 7. Vol 5. Available from: https://www.navsea.navy.mil/Portals/103/Documents/SUPSALV/Diving/US%20DIVING%20MANUAL_REV7.pdf?ver=2017-01-11-102354-393. Accessed 30 Sep 2020.

2. Grim PS, Gottlieb LJ, Boddie A, Batson E. Hyperbaric oxygen therapy. JAMA 1990;263:2216-20. Crossref

3. Blanc P, Boussuges A, Henriette K, Sainty JM, Deleflie M. Iatrogenic cerebral air embolism: importance of an early hyperbaric oxygenation. Intensive Care Med 2002;28:559-63.Crossref

4. Beevor H, Frawley G. Iatrogenic cerebral gas embolism: analysis of the presentation, management and outcomes of patients referred to The Alfred Hospital Hyperbaric Unit. Diving Hyperb Med 2016;46:15-21.

5. Massey EW, Moon RE, Shelton D, Camporesi EM. Hyperbaric oxygen therapy of iatrogenic air embolism. J Hyperb Med 1990;5:15-21.

6. Pearson RR, Goad RF. Delayed cerebral edema complicating cerebral arterial gas embolism: case histories. Undersea Biomed Res 1982;9:283-96.

7. Benson J, Adkinson C, Colier R. Hyperbaric oxygen therapy of iatrogenic cerebral arterial gas embolism. Undersea Hyperb Med 2003;30:117-26.

8. Perez MF, Ongkeko Perez JV, Serrano AR, Andal MP, Aldover MC. Delayed hyperbaric intervention in lifethreatening decompression illness. Diving Hyperb Med 2017;47:257-9 Crossref

9. United States Navy. US Navy Diving Manual. Revision 7. Vol 5: Diving medicine and recompression chamber operations. 2016. Available from: https://www.navsea.navy.mil/Portals/103/Documents/SUPSALV/Diving/US%20DIVING%20MANUAL_REV7.pdf?ver=2017-01-11-102354-393. Accessed 30 Sep 2020.

10. Vann RD, Butler FK, Mitchell SJ, Moon RE. Decompression illness. Lancet 2011;377:153-64.Crossref

11. Undersea & Hyperbaric Medical Society. UHMA Best Practice Guidelines: Prevention and Treatment of Decompression sickness and arterial gas embolism. Available from: https://www.uhms.org/images/DCS-AGE-Committee/dcsandage_prevandmgt_uhms-fi.pdf. Accessed 30 Sep 2020.

12. Royal Navy, Ministry of Defence. Diving Manual. HM Stationery Office, London; 1972.

In July 2017, a 28-month-old boy presented to a private outpatient clinic with a 2-day history of fever and coryzal symptoms (Table). He had enjoyed good past health and his family history was unremarkable. On clinical examination, he was noted to have respiratory distress and tachycardia. Plain radiograph of the chest (CXR) showed perihilar haziness but no consolidation. He was transferred to Queen Mary Hospital, Hong Kong, to exclude myocarditis in view of elevated serum troponin in blood taken in the private clinic. On admission, he was noted to have diffuse crepitations and wheeze suggestive of pneumonitis. Nebulised salbutamol, hypertonic saline, and intravenous cefotaxime were administered. Complete blood count revealed neutrophilia, normal liver and renal function, and normal creatine kinase level. Venous blood gas showed no acidosis. Troponin was high initially but then gradually normalised. Echocardiogram showed no features of myocarditis. Nasopharyngeal aspirate and throat swab test results were positive for enterovirus (EV)/rhinovirus ribonucleic acid (RNA) using reverse transcriptase-polymerase chain reaction (RT-PCR) detection method. EV71 RNA was not evident. Nasopharyngeal aspirate for mycoplasma, throat swab and blood culture were all negative. He then developed progressive respiratory distress with increased cough and high fever up to 40°C on day 7 of illness and CXR showed worsening of bilateral perihilar haziness. In view of the progressive respiratory failure, the child was admitted to the paediatric intensive care unit 2 days later (day 9).

On admission to paediatric intensive care unit, he was noted to have generalised muscle weakness with a weak voice and an inability to sit up. Gag reflex and jerks were preserved. He was then intubated and ventilated under sedation. On reassessment, chest auscultation revealed decreased air entry over the left side, corresponding to the left lower zone collapse evident on CXR. He was prescribed piperacillin-tazobactam and levofloxacin. On day 10, throat swab culture grew only commensals while endotracheal aspirate revealed scanty growth of alpha-haemolytic streptococci with negative fungal smear and culture, and RT-PCR identified EV/rhinovirus. Broncho-alveolar lavage was also performed but results were unremarkable: Pneumocystis jiroveci (carinii), smear and culture for bacteria, fungus and acid-fast bacilli, and RT-PCR for cytomegalovirus, herpes simplex virus, Mycoplasma, Legionella, were all negative. Urine culture was negative for Legionella antigen.

On day 11, the patient exhibited paradoxical breathing on weaning of sedation along with hypotonia and paralysis of four limbs. Urgent sagittal T2-weighted magnetic resonance imaging of spine (day 11) showed mild T2 hyperintensity with mild expansion within the central portion of the cervical cord from C3 to C6 (Fig). No intraspinal mass or collection could be seen. Neurology examination was performed on day 12. Creatine kinase was normal and anti-acetylcholine receptor and anti-aquaporin-4 were negative. Eye examination the following day showed no evidence of optic neuritis. Immunoglobulin G and immunoglobulin M of anti-gangliosides were all negative. Lumbar puncture revealed normal cerebrospinal fluid level of glucose and protein. Total cell count was 100 × 10⁶/L with predominantly (80%) neutrophils. Cerebrospinal fluid culture was negative for viruses, including EV, herpes simplex virus or varicella-zoster virus. Cerebrospinal fluid levels of oligoclonal protein and immunoglobulin G were unremarkable. The working diagnosis was transverse myelitis affecting cervical cord C3 to C6. He was prescribed pulse methylprednisolone 30 mg/kg for 5 days (day 12 to 16) followed by a tapering oral dose of prednisolone together with intravenous immunoglobulin 2 g/kg over 2 days (day 16 and 17). He was also treated with therapeutic plasma exchange with 1.5-times plasma volume for five courses over 2 weeks (day 22, 24, 26, 30, 32). His condition gradually improved and he was extubated (day 47) after 38 days of invasive ventilation. The child was successfully discharged from the paediatric intensive care unit after 2 months (day 64) and achieved full neurological recovery with intensive training by physiotherapists. Subsequent review by a microbiologist using gene sequencing of initial specimens obtained on day 3 of admission to Queen Mary Hospital revealed EV-D68 RNA in nasopharyngeal aspirate, endotracheal aspirate and stool samples. Samples remained positive for 4 weeks (day 30, during acute deterioration warranting intensive care unit admission) and were negative after 6 weeks (day 55). The definitive diagnosis was EV-D68-associated acute flaccid paralysis (AFP) although EV-D68 was not present in the cerebrospinal fluid and plasma/serum samples were not tested for EV-D68 by RT-PCR.

Acute flaccid paralysis is defined by the World Health Organization as a clinical syndrome of diverse aetiology characterised by acute-onset limb weakness or paralysis with varying degrees of autonomic and somatic nervous system dysfunction that reaches maximum severity over a period of days or weeks in a child younger than 15 years of age.1 It is a diagnosis of exclusion. In 1962, a new strain of EV, EV-D68, was identified in Berkeley, California. In 2014, EV-D68 outbreaks were reported in 20 countries including the United States, Canada, Europe, and Asia with a total of over 2000 cases. This corresponded to an increased global incidence of AFP.2 A casual association between EV-D68 and AFP is supported by Bradford Hill criteria.3 Despite public health attempts in 1988 to eliminate AFP through the Global Polio Eradication Initiative4 and roll-out of the oral polio vaccine5 to prevent vaccine-associated poliomyelitis, the emergence of EV-D68-associated AFP has become a significant cause of neurological deficits in children since 2014. Owing to its impact on the healthcare system, a comprehensive literature review and further detailed studies are warranted. This is the first case encountered in our department. It is important for clinicians in Hong Kong to be alert for the disease.

As a newly emerging disease manifestation, a high index of suspicion and clinical awareness is advocated to facilitate earlier recognition and diagnosis through appropriate investigations, and presumably, improved clinical outcomes. The optimum treatment strategy has yet to be defined and preventive strategies are still being developed. Local and international notification systems as well as comprehensive surveillance are suggested since disease outbreaks may occur at any time and may have a serious impact on affected children.

Acquisition of data: WYK Chan, SHY Chim, DML Tse.
Analysis or interpretation of data: WYK Chan, DML Tse, PL Ho.
Drafting of the manuscript: WYK Chan.
Critical revision of the manuscript for important intellectual content: All authors.

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

All authors have disclosed no conflicts of interest.

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

The parents of the patient gave consent for publication.

1. Bitnun A, Yeh EA. Acute flaccid paralysis and enteroviral infections. Curr Infect Dis Rep 2018;20:34. Crossref

2. Holm-Hansen CC, Midgley SE, Fischer TK. Global emergence of enterovirus D68: a systematic review. Lancet Infect Dis 2016;16:e64-75. Crossref

3. Messacar K, Asturias EJ, Hixon AM, et al. Enterovirus D68 and acute flaccid myelitis-evaluating the evidence for causality. Lancet Infect Dis 2018;18:e239-47. Crossref

4. World Health Organization. Global Polio Eradication Initiative. 2018. Available from: http://polioeradication.org/where-we-work/polio-endemic-countries/. Accessed 24 Mar 2018.

5. World Health Organization. Global Polio Eradication Initiative. Circulating vaccine-derived poliovirus. 2018. Available from: http://polioeradication.org/polio-today/polio-now/this-week/circulating-vaccine-derived-poliovirus/. Accessed 24 Mar 2018.