Fulminant varicella hepatitis: a rare but lethal cause of abdominal pain
- Christopher Fang ,
- Junice Wong and
- Wei Wen Ang
- Department of General Surgery, Tan Tock Seng Hospital, National Healthcare Group, Singapore
- Correspondence to Dr Christopher Fang; Christopher.fang@mohh.com.sg
Abstract
An 81-year-old woman with no history of immunocompromise presented with 2 days of upper abdominal pain associated with nausea. On arrival, her physical examination was unremarkable apart from mild epigastric and right hypochondriac tenderness, and laboratory investigations were unremarkable apart from mild thrombocytopenia and transaminitis. A CT scan performed on the day of admission revealed a tiny 0.3 cm stone in the common bile duct, with no upstream dilatation. On day 2 of admission, she developed a vesicular rash and with acutely worsening transaminitis. She deteriorated rapidly and demised from complications of acute liver failure within the next 24 hours. The diagnosis of varicella was confirmed with antibody testing. Fulminant varicella hepatitis is an extremely rare and lethal condition with only a handful of reported cases in the current literature. We aim to share our clinical experience and summarise the salient points from existing case reports.
Background
Fulminant varicella hepatitis is an extremely rare and lethal complication of disseminated varicella and usually occurs in immunocompromised patients.
Early diagnosis of fulminant varicella hepatitis is extremely difficult due to the fact that most patients present with non-specific mild transaminitis and the onset of abdominal pain can precede the onset of rash. Once worsening transaminitis occurs, this is usually followed by rapid clinical deterioration and subsequent mortality.
We report a case of fulminant varicella hepatitis treated in our institution which closely mimics the clinical presentation and natural progression reported in existing case reports and concurrently provide a thorough review of the existing literature.
Case presentation
Our patient is an active 81-year-old Chinese woman who was premorbidly independent in all activities of daily living. She had a medical history of hypertension, hyperlipidaemia, paroxysmal atrial fibrillation and Hashimoto’s disease for which she underwent a total thyroidectomy in 2017. During history taking, she confirmed she was not on any regular steroids, immunosuppressants or traditional medications/supplements. Clinically, she was well-nourished with a body mass index of 24.8, and did not have any history of alcohol intake, intravenous recreational drug use or previous tattoos. Significantly, she had not been previously infected with varicella, and did not receive any previous vaccination for varicella. She reported no familial history of hepatic disease.
She first presented to our institution’s accident and emergency department with a history of 2 days of epigastric pain which radiated to the right hypochondrial region and nausea without any episodes of vomiting. She did not report feeling febrile or any symptoms of obstructive jaundice.
Her vital signs were stable on presentation with a low-grade temperature of 37.7°C. On physical examination, she appeared alert and comfortable and was not jaundiced. Other than right upper quadrant abdominal tenderness, the rest of the physical examination was unremarkable. Of note, she had no rashes.
Investigations
Blood tests, including a full blood count, renal panel, liver function test, amylase, lipase, coagulation panel and arterial blood gas were performed to workup potential causes of upper abdominal pain. These were grossly unremarkable apart from mild thrombocytopenia [(136×109/L) and mild transaminitis (Alanine transaminase (ALT) 58 U/L, Aspartate transaminase (AST) 141 U/L).
A CT scan of her abdomen and pelvis was also performed which revealed a tiny 0.3 cm stone in the common bile duct with no upstream common bile duct dilatation, and no evidence of pancreatitis, hepatitis or cholecystitis. Significantly, her liver appeared grossly normal on CT, and did not reveal any findings suggestive of hepatic steatosis or chronic liver disease.
A formal 12 lead ECG performed on arrival did not reveal any atrial fibrillation or ischaemic changes.
Treatment
The patient was managed as for possible passed gallstone/biliary colic. Other than one episode of vomiting, she had no other significant events on day 1 of admission.
The next morning, she reported persistent abdominal pain which was transiently relieved with analgesia. Laboratory tests repeated approximately 24 hours after the first set of investigations revealed worsening transaminitis (ALT 58>380 U/L, AST 141>1105 U/L) and thrombocytopenia (136 to 75×109/L).
The patient was reviewed by the gastroenterology team who ordered a hepatitis screen (which was unremarkable) and a magnetic resonance cholangiopancreatography scan to workup the potential cause of transaminitis. The utilisation of any hepatotoxic drugs was also excluded. In lieu of the worsening thrombocytopenia, intravenous antibiotics was also started empirically to treat potentially developing intra-abdominal sepsis although the patient remained clinically afebrile.
Later that evening (approximately 36 hours postpresentation) the patient developed a cluster of painless vesicular rashes over her back which subsequently progressed to her neck and anterior trunk by the next morning. On further questioning, she reported a significant history of exposure to varicella from her husband who had contracted the virus 1 week ago. She was promptly isolated and started on intravenous acyclovir for varicella zoster.
Unfortunately, the patient’s clinical condition rapidly deteriorated on her third day of admission, soon after the clinical diagnosis of varicella zoster. At mid-day that day (approximately 52 hours post-presentation), she developed fulminant liver failure, with a sharp rise in her ALT (380>2794 U/L) and AST (1105>9977 U/L) (table 1). Approximately 4 hours later, she suffered a syncopal episode after an episode of per rectal bleeding and turned hypotensive requiring both aggressive fluid resuscitation and inotropic support. Laboratory tests showed multi-organ failure with severe coagulopathy (Activated partial thromboplastin time (APTT) >120 s, International normalised ratio (INR) >10, Prothrombin time (PT) >100 s), hyperkalaemia and metabolic acidosis. During this time, a formal 12 lead ECG was repeated for the patient which again showed no evidence of atrial fibrillation.
Liver function test (LFT) trend
| LFTs | 0 hours | 24 hours | 52 hours | 60 hours |
| ALP (U/L) normal range: 40–120 | 52 | 70 | 171 | 190 |
| ALT (U/L) normal range: 10–55 | 58 | 380 | 2794 | 3756 |
| AST (U/L) normal range: 20–45 | 141 | 1105 | 9977 | 12 788 |
| GGT (U/L) normal range: 15–90 | 22 | 42 | 145 | 133 |
-
ALP, Alkaline Phosphatase; ALT, Alanine Transaminase; AST, Aspartate Aminotransferase; GGT, Gamma-Glutamyl Transferase.
In lieu of the above, the patient was intubated and transferred to the surgical intensive care unit (ICU). Unfortunately, despite aggressive supportive treatment, she demised from cardiac arrest shortly after. The diagnosis of acute varicella infection was confirmed via plasma serology which was reported postmortem.
Outcome and follow-up
In summary, our patient is an 81-year-old woman who contracted varicella zoster and unfortunately demised as a result of rapid and fulminant liver failure resulting in multi-organ failure.
Discussion
Epidemiology
Varicella zoster rarely results in disseminated disease, which is characterised by a diffuse rash outside the primary or adjacent dermatomes, along with other potential complications such as encephalitis, hepatitis and pneumonitis. Mortality rates in patients with disseminated varicella have been reported to be up to 55%.1
To date, our literature review revealed a total of 18 cases of patients with varicella hepatitis with an overall mortality rate of 77.8% (14 out of 18 patients) (table 2).2–19 The majority of patients (12 out of 18) were immunocompromised. Of the remaining six patients, three had recently been started on low dose oral steroids,6 11 12 which some clinicians believe can result in a temporary immunocompromised state.
Patient demographics and outcomes
| Author | Age/sex | Medical history | Immunocompromise | Mortality |
| Lechiche et al 2 | 35/M | HIV | Yes | No |
| Alvite-Canosa et al 3 | 43/M | Status post-heart transplant | Yes | No |
| Drebber et al 4 | 49/M | Squamous cell carcinoma of neck | Yes | Yes |
| Saitoh et al 5 | 47/M | Multiple myeloma | Yes | Yes |
| Maggi et al 6 | 49/M | Low dose prednisolone for pharyngitis | No | Yes |
| Natoli et al 7 | 15/M | Nil | No | Yes |
| Plisek et al 8 | 26/F | Multiple sclerosis on methylprednisolone | Yes | Yes |
| Ross et al 9 | 64/F | Recent oesophago-gastrectomy and splenectomy | Yes | Yes |
| Brewer and Hunter10 | 66/F | Dermatomyositis on long-term steroids | Yes | Yes |
| Toffaha et al 11 | 26/M | Low dose oral prednisolone for sciatica | No | Yes |
| Pishvaian et al 12 | 38/M | Low dose oral prednisolone for asthma exacerbation (patient had childhood varicella) | No | Yes |
| Morales et al 13 | 51/M | Renal transplant patient | Yes | No |
| Patti et al 14 | 29/M | Renal transplant patient, previous splenectomy | Yes | Yes |
| Soriano et al 15 | 27/M | HIV | Yes | Yes |
| Anderson et al 16 | 26/F | Nil | No | Yes |
| Hsing et al 17 | 21/F | Renal transplant patient | Yes | No |
| Dits et al 18 | 30/M | Renal transplant patient | Yes | Yes |
| Vartian19 | 20/M | Nil | No | Yes |
As illustrated above, the main risk factor for developing fulminant hepatic failure from varicella zoster is a history of immunocompromise which can either be due to iatrogenic or acquired causes. These include patients who underwent either organ transplantation or splenectomy, or patients on steroids or other immunosuppressive agents for conditions such as asthma, inflammatory bowel disease or AIDS. Our case is particularly peculiar as our patient did have any of the above risk factors.
In addition to immunocompromised patients, the WHO Vaccine-Preventable Diseases Surveillance Standards,20 as well as the Centers for Disease Control and Prevention (CDC) Vaccine Preventable Disease Surveillance Manual guideline,21 both highlight pregnant women, children younger than 1 year of age and adults as individuals at higher risk of severe complications of primary zoster infection. Interestingly in our literature review, all three patients who suffered fulminant varicella hepatitis without any preceding history of immunocompromise or steroid use were all below 30 years of age. Ultimately, due to the extreme rarity of the condition, it is difficult to meaningfully assess these purported independent risk factors for statistical significance.
Presentation and diagnosis
Given the rapid progression and lethality of varicella hepatitis, early diagnosis and initiation of treatment is essential. However, making an early diagnosis of acute varicella, which hinges clinically on the presence of its characteristic rash in different stages, is often delayed due to several reasons.
First, these patients usually present with non-specific and at times atypical symptoms (table 3). In our review, although the majority of patients presented with expected symptoms such as fever, abdominal pain, nausea, vomiting and rash, some patients presented atypically such as two patients with an isolated complaint of chest pain,2 6 and five patients without any abdominal pain at presentation.2 6 8 9 17 All five patients without abdominal pain at presentation had either normal or mild transaminitis like the majority of patients reviewed in our literature review (10 out 14 cases where laboratory results on arrival were clearly documented).
Presenting symptoms
| Presenting complaint | Fever (8) Abdominal pain (12) Nausea/vomiting (3) Myalgia (2) |
| 18 patients reviewed | Chest pain (2) Rash (10) |
Second, the onset of abdominal pain can precede the appearance of the characteristic varicella rash such as in our patient. We noted this trend in 5 out of 18 other patients in our literature review who developed rashes 1–4 days after the onset of abdominal pain.2 4 6 14 19
Given the rarity of the condition, without the presence of a rash, it is highly unlikely that the treating clinician would suspect varicella to be the cause of hepatitis and abdominal pain and therefore test for varicella. In fact, in our review, two patients did not develop any rash up to the point of demise, and a diagnosis was only made postmortem.5 9
In the event of atypical presentations of varicella zoster, laboratory investigations can be used to aid in diagnosis. According to the CDC guidelines,21 PCR testing from swabs of unroofed vesicular lesions, scabs from crusted lesions or saliva should be used as first line as it can be conducted rapidly and has good sensitivity. Other options not recommended include direct fluorescent antibody testing (as it is less sensitive than PCR) and Tzanck smears (cannot differentiate between herpes simplex and varicella zoster). Serological testing can also be performed but has a disadvantage of a long turnaround time. Ultimately, if varicella is suspected, treatment should not be delayed pending these laboratory results.
When considering a diagnosis of varicella hepatitis, it is essential more common aetiologies of fulminant hepatitis are ruled out. Differentials to rule out include more common viral aetiologies (such as infectious hepatitis), drug-related causes, metabolic diseases, vascular aetiologies (such as Budd-Chiari syndrome or veno-occlusive disease) and other miscellaneous causes (such as autoimmune hepatitis, malignant infiltration, hyperthermia and ischaemia).
Our patient had an unremarkable hepatitis screen and was not on any hepatotoxic drugs. Furthermore, she did not have any history suggestive of any of the above aetiologies. An ischaemic aetiology was highly unlikely as prior to the development of severe uncorrectable transaminitis the patient was never hypotensive. A cardioembolic event was also considered (in lieu of the patient’s history of paroxysmal atrial fibrillation), but this was unlikely as two separate electrocardiograms (once on arrival and just prior to ICU transfer) and continuous monitoring in ICU did not reveal any recurrence of atrial fibrillation.
Natural history and progression
Varicella hepatitis is associated with rapid and sudden clinical deterioration, with all patients invariably requiring intensive care management after worsening hepatitis. In our review, the mean time from presentation to mortality was 5.2 days (range 1–10), and the median time was 4 days among 11 patients. Three studies were excluded as the time from presentation to mortality was not clearly documented.
In almost all cases reviewed, rapid clinical deterioration occurred concurrently with acutely worsening transaminitis with some patients experiencing up 50-fold increases in liver enzymes within a short span of time.6 Among 10 patients with normal liver function tests or mild transaminitis on arrival, this rapid deterioration in liver enzymes was noted to have occurred at a mean of 2.4 days (range 1–4) postpresentation, with a median of 2 days. Once this occurred, all patients suffered severe hypotension and complications of liver failure such as coagulopathy requiring ICU stay.
Treatment
Given the rarity of fulminant varicella hepatitis and the limited survivorship, there are no existing guidelines specifically for treating the condition, and no high-powered studies comparing different treatments and outcome measures.
In our review of the literature, intravenous acyclovir was started for all patients except two patients where the diagnosis was initially missed.4 5 Only one patient was reported to have been given oral acyclovir before escalation to intravenous acyclovir when the patient deteriorated.8
Intravenous immune globulin (IVIG) was also given in 3 out 18 patients.3 6 17 Interestingly, two out of three patients given IVIG survived.3 17 Although both the CDC and WHO guidelines recommend IVIG mainly only for post-exposure prophylaxis in immune compromised individuals or pregnant non-immune patients,20 21 there has been some anecdotal evidence to suggest that IVIG as an adjuvant together with intravenous acyclovir may help to reduce viral load in immunocompromised individuals with delayed antibody responses.17 However, its use should be carefully considered as worsening renal impairment is not uncommon with IVIG infusions.
In our review, all patients required intensive care support for aggressive supportive treatment. Molecular Absorbent Recirculating System therapy was attempted for one patient while in ICU while awaiting liver transplant, but the patient unfortunately demised.7
In terms surgical treatment, one patient underwent an emergency liver transplant on day 8 of presentation and fortunately survived,3 while another underwent a total hepatectomy and portal caval anastomosis while awaiting urgent liver transplantation after a prior biopsy showed extensive hepatic necrosis.6
Learning points
-
We report a rare case of fulminant varicella hepatitis which closely mirrors existing reports.
-
Although more predominant among immunocompromised patients, fulminant varicella hepatitis can also affect immunocompetent patients with short-term oral steroids a possible risk factor.
-
Making a prompt diagnosis in such cases is challenging as symptoms can often precede the onset of dermatological findings.
-
Diagnosis should be clinical, and treatment should not be delayed pending investigations.
Ethics statements
Patient consent for publication
Footnotes
-
Contributors First author CF was involved in the data collection and preparation of the manuscript. Second and third authors JW and WWA are senior authors whose advice, expertise and guidance were invaluable in the preparation and final editing of this manuscript.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Competing interests None declared.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2021. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer