Fulminant Hepatitis A in a Patient with Severe Hyperthyroidism: Rapid Recovery from Hepatic Coma after Plasmapheresis and Total Thyroidectomy
Michael Enghofer,
Klaus Badenhoop,
Stefan Zeuzem,
Andreas Schmidt-Matthiesen,
Christoph Betz,
Albrecht Encke and
Klaus Henning Usadel
Departments of Medicine I (M.E., K.B., K.H.U.), II (S.Z.), and IV
(C.B.), and Department of Surgery (A.S.-M., A.E.), Johann Wolfgang
Goethe University, D-60590 Frankfurt am Main, Germany
Address correspondence and requests for reprints to: Michael Enghofer, M.D., Department of Medicine I, Johann Wolfgang Goethe University, Theodor-Stern-Kai 7, D-60590 Frankfurt am Main, Germany.
 |
Introduction
|
---|
A 62-yr-old Chinese man was transferred to the
medical intensive care unit of our university hospital because
of hepatic coma and severe hyperthyroidism. The software engineer, who
had lived in Germany for 10 yr, had been well until 3 days before
admission, when he experienced acute onset of nausea, vomiting, and
diarrhea. The next day, he started passing black, tarry stools.
Jaundice, low-grade fever, and pronounced malaise developed. Three days
after onset of symptoms he was admitted to another hospital, where
gastroscopy showed diffuse gastric bleeding due to erosive gastritis.
He was treated with ranitidine and metoclopramide iv. The laboratory
investigations at admission (Table 1
)
showed markedly elevated liver enzymes, a prolonged prothrombin time,
and decreased serum levels of antithrombin III and total protein.
Glucose, electrolytes, blood urea nitrogen, creatinine, creatine
phosphokinase, amylase, lipase, uric acid, complete blood count (with
the exception of a reduced platelet count), and erythrocyte
sedimentation rate were within the normal range. Serologic tests for
antibodies against hepatitis A (IgG and IgM) were positive, and the
patient showed marked hyperthyroidism (Table 1
). Thyroid scintiscanning
revealed a diffusely increased thyroid
99mTc-pertechnetate uptake without evidence for
hyperfunctioning nodules. Diffuse goiter with an irregular and slightly
hypoechoic structure was diagnosed by thyroid sonography. The patient
received 40 mg methimazole per day iv. The coagulopathy was treated
with 10 mg phytonadione (vitamine K1) per day iv
and by infusion of a total of 400 mL fresh-frozen plasma. The patient
had been weak and drowsy on admission to the other hospital, but was
able to communicate. He was oriented to time, place, and self. However,
his mental status rapidly deteriorated. Hepatic encephalopathy with
confusion developed, and he showed increased psychomotor activity. No
focal neurologic signs were noted. A cerebral computed tomography scan
(without administration of iodine containing contrast medium) was
normal. A central line was inserted, and the patient received glucose,
an amino acid solution enriched with branched amino acids, electrolyte
solutions, and ornithine aspartate iv. For sedation, midazolam,
flunitrazepam, and morphine sulfate were administered iv. The following
day, the patient was in a deep coma, and he was transferred to our
university hospital.
The patients wife reported a history of mild Graves disease for 5
yr. The patient had refused radioiodine therapy or thyroidectomy and
had been taking oral methimazole irregularly for 2 yr, although
repeatedly serum T4 levels were markedly elevated
and thyrotropin was undetectable in serum. His general practitioner had
recommended a dose of 20 mg methimazole per day. Due to compliance
problems, he also did not follow the recommendation to minimize his
iodine intake, but continued to eat seafood regularly. However,
administration of iodine containing x-ray contrast medium in the
previous months could be ruled out as a predisposing factor for the
development of his thyrotoxicosis. Furthermore, atrial fibrillation was
present for 5 yr before admission. The patient repeatedly refused
digoxin treatment, cardioversion, or anticoagulation for this
condition. However, he was able to work hard up to 12 h a day and
had just arrived from a 4-week business trip to southern China and
Beijing 1 month before the onset of his current illness. His German
business partner, who accompanied the patient during the whole trip to
China, as well as his wife and daughter, stayed well. There was no
known history of contact with persons infected with hepatitis A. The
patient drank one to two bottles of beer a day. He did not take
acetaminophen or other potentially hepatotoxic drugs,
apart from methimazole. He had never had symptoms suggestive of
hepatitis or other liver diseases and had never been hospitalized
before. Three months before admission, liver function tests had been
normal, but hyperthyroidism was present.
On arrival in our intensive care unit, the jaundiced and dehydrated
patient was deeply comatose (Glasgow coma scale, 3/15) and did not show
any reaction to painful stimuli. He was breathing spontaneously, and
his respirations were 28/min. His rectal temperature was 36.3 C (96.7
F), the blood pressure was 100/70 mm Hg, and the pulse was arrhythmic.
Atrial fibrillation with a ventricular response of 100120 bpm was
present. There were no murmurs or rubs. The lungs were clear. The lips
and oral mucosa showed old encrusted blood, but no active bleeding.
Apart from one isolated spider naevus in the left clavicular region, no
peripheral signs of chronic liver disease were seen. The thyroid was
diffusely enlarged. Neither endocrine ophthalmopathy nor dermopathy was
present. The liver and spleen were not palpable, and there were no
clinical signs indicating the presence of ascites.
 |
Differential Diagnosis
|
---|
This 62-yr-old Chinese man had no history of previous liver
disease. He presented with rapidly deteriorating hepatic
encephalopathy. Graves hyperthyroidism was known for 5 yr. He had
been on methimazole for more than 2 yr when he developed fulminant
hepatitis. It seemed highly unlikely that in this case acute hepatitis
was caused by hepatotoxic adverse effects of antithyroid drugs alone.
He had been taking methimazole more or less regularly for more than 2
yr before the rapid onset of acute severe liver disease, but even 3
months before admission there was no laboratory evidence for adverse
hepatotoxicity of this treatment. A latency period of more than 2 yr
between the start of antithyroid therapy and the induction of
associated hepatotoxicity has not been reported in the literature yet.
Fatal methimazole-induced hepatitis has been described in a patient
with micronodular cirrhosis and persistent hepatitis B antigenemia, who
had no evidence for hepatitis B reactivation or chronic hepatitis B
(1). Our patient had acute hepatitis A, as shown by demonstration of
IgM antibodies against the hepatitis A virus. A fulminant course with
the development of acute liver dystrophy is a very rare complication of
hepatitis A (2). Hence, preexisting hyperthyroidism and/or chronic
treatment with antithyroid drugs apparently predisposed this patient
for hepatic failure induced by the hepatitis A infection. Other
infectious hepatitis types could be ruled out serologically and by PCRs
(Table 3
). There was no evidence for toxic liver damage by drugs or
other toxins. The amount of alcohol consumed was less than 40 g
per day. Thrombotic occlusion of hepatic veins, the Budd-Chiari
syndrome, was considered unlikely because the liver was not enlarged on
ultrasonography and there was no ascites. The absence of splenomegaly
and ascitic fluid was evidence against portal vein thrombosis. Because
chronic atrial fibrillation was present and the patient had refused
oral anticoagulation with warfarin for years, thromboembolism to the
hepatic artery might have occurred. Liver failure induced by
thromboembolic impairment of hepatic arterial blood supply, however,
has not been reported so far. There was no history of acute hypotension
due to any type of shock in this patient. The absence of leukocytosis,
of leukopenia, and of fever more than 38 C, as well as the almost
normal levels of acute phase proteins, was evidence against septic
liver failure.
 |
Diagnostic Workup
|
---|
Laboratory testing in our hospital (Table 2
) confirmed the presence of severe
hepatitis and Graves hyperthyroidism. Thyrotropin receptor
autoantibodies showed markedly elevated levels (Table 2
). Glucose,
electrolytes, blood urea nitrogen, creatinine, creatine phosphokinase,
amylase, and the complete blood count (with the exception of a reduced
platelet count; Table 2
) were within the normal range. Acute hepatitis
A was diagnosed serologically (Table 3
).
There was no history of exposure to hepatotoxins apart from the
methimazole the patient was taking irregularly for 2 yr.
Thyroid sonography showed diffuse goiter with a very patchy and
irregular hypoechogenic texture. The thyroid volume was 105 mL (normal,
<25 mL). Abdominal sonography excluded the presence of ascites or
splenomegaly and demonstrated a homogenous and slightly hypoechogenic
liver of normal shape and size. There was no other pathology in this
abdominal ultrasound study. The condition of the comatose patient was
critical. Liver function tests deteriorated rapidly (Table 2
). Free
T4 levels in serum were extremely high. A liver
biopsy was not performed initially because of the high risk of bleeding
and the impossibility to obtain informed consent from the patient.
 |
Clinical Course
|
---|
The ventricular rate was slowed by continuous infusion of esmolol,
and 100 mg prednisone were administered iv. Parenteral nutrition was
instituted. It was postulated that the high levels of thyroid hormones
resulted in increased metabolic stress of hepatocytes, thereby
potentiating the hepatotoxic effect of hepatitis A infection. Continued
administration of antithyroid agents would have carried the risk of
additional drug-induced hepatotoxicity. Therefore, we strongly
recommended immediate thyroidectomy. Informed consent was obtained from
the patients wife. One plasmapheresis treatment was performed to
rapidly lower thyroid hormone levels before surgery (Table 4
). Fifty milliliters of plasma/kg body
weight were exchanged. The removed plasma (3094 mL) was replaced by
fresh-frozen plasma from matched blood donors. Before and 10 min after
plasma separation, blood was collected to quantify the reduction of
circulating levels of T4 and thyronine (Table 4
).
After plasmapheresis, the patient was given 1000 units antithrombin III
and 2000 units prothrombin complex iv before surgery. Two hours after
plasmapheresis the patient was transferred to the operating room, where
total thyroidectomy was performed in general anesthesia without
complications.
After thyroidectomy the patient was treated in the medical
intensive care unit for 10 days. Following surgery he was sedated
and mechanically ventilated for 28 h. Fresh-frozen plasma had to
be administered repeatedly. Lactulose was given via a gastric tube, and
ornithine aspartate 40 g/24 h was infused iv to lower ammonia levels.
Normocaloric parenteral nutrition, including an amino acid solution
enriched with branched chain amino acids and with a low content in
aromatic amino acids, was resumed on postoperative day 2. The patient
could be extubated in the morning of the 3rd day of his hospital stay
in our institution. His mental status slowly improved. Although ammonia
levels and liver enzymes rapidly normalized in the early postoperative
course, the patient developed severe direct hyperbilirubinemia (Fig. 1
) and continued to require infusions of
fresh-frozen plasma to keep international normalized ratio values below
2.0 and antithrombin III levels above 30%. Oral levothyroxine
replacement therapy was started on the 6th day. On the 10th hospital
day, the patient was transferred to a medical ward. The patient refused
liver biopsy. Bilirubin levels peaked at 42.3 mg/dL, and there was no
tendency for a decline when the patient decided to leave the hospital
against our advice on the 30th hospital day. He traveled to China to be
treated there according to the rules of traditional Chinese medicine.
However, he had to be admitted to an intensive care unit of a major
hospital in Beijing soon after his arrival due to extreme
hyperbilirubinemia and coagulopathy. The patient was treated there for
4 months before he was transferred to a traditional Chinese medical
facility for rehabilitation. Meanwhile, he has returned to Germany and
has started to work again. His mental status, liver function, and most
other laboratory results have returned to normal 12 months after the
acute onset of hepatic coma (Table 2
).

View larger version (25K):
[in this window]
[in a new window]
|
Figure 1. Liver function tests before and after
plasmapheresis and thyroidectomy performed on day 1
(arrow).
|
|
 |
Discussion
|
---|
Disturbances of hepatic function are known complications of
antithyroid drug treatment for hyperthyroidism (3). The spectrum of
observed changes ranges from mild and asymptomatic elevation of
cholestatic and cytosolic liver enzymes (3) to fulminant and fatal
necrotizing hepatitis (4, 5, 6). Namely, propylthiouracil has been
described to be able to cause severe toxic hepatopathies (5, 6, 7),
whereas cholestatic jaundice, without evidence of hepatic necrosis on
liver biopsy, has been typically associated with methimazole (8). In
one study, the incidence of toxic hepatitis due to hepatotoxic adverse
reactions against propylthiouracil was 6% (9). Others, however, found
a much lower incidence of severe propylthiouracil-induced hepatitis,
whereas asymptomatic elevation of liver enzymes under this therapy
occurs in up to 28% of patients (3). Nevertheless, there are
occasional reports on severe and even fatal cases of drug-induced
hepatitis under treatment of hyperthyroidism with methimazole (10) and
carbimazole (4). It is also well known that hyperthyroidism itself may
cause elevations of liver enzymes (11).
We assume that the chronic hyperthyroidism, which was present in our
patient for at least 5 yr before admission, as well as the necessary
treatment with antithyroid drugs, imposed such a significant metabolic
and possibly toxic stress on the liver that it predisposed him for a
fulminant course of viral hepatitis. We, therefore, recommended
plasmapheresis, followed by immediate total thyroidectomy, to
definitively eliminate the need for further antithyroid treatment and
to rapidly lower the concentration of circulating free
T4 and T3. In a German survey (12),
early thyroidectomy has been reported to reduce the mortality of
thyrotoxicosis with coma from 2040% under conservative treatment to
less than 10%.
Plasmapheresis has been used successfully to treat patients with very
severe thyrotoxicosis for more than 25 yr (13). In our patient, plasma
exchange significantly reduced the levels of circulating free thyroid
hormones (Table 4
). The small increase of total
T3 and TSH immediately after plasmapheresis was
the result of the administration of donor plasma containing TSH and
albumin that binds free T3. Furthermore, plasma
exchange not only resulted in a rapid and lasting decline of alanine
aminotransferase and aspartate aminotransferase levels,
but also reduced hyperammonemia (Fig. 1
). If the infection with
hepatitis A had been the only relevant pathogenetic factor in this
case, a much more prolonged period of liberation of cytosolic liver
enzymes from necrotic hepatocytes would have been expected. Apparently,
the course of fulminant hepatitis A was significantly attenuated by a
single plasmapheresis treatment, followed by thyroidectomy, in the case
under discussion. However, progressive and severe hyperbilirubinemia
could not be prevented. Cholestatic hepatitis is a rare, albeit well
recognized, variant of hepatitis A. Protracted cholestatic jaundice may
last several months and is typically observed in extraordinarily severe
hepatitis A, like in our patient. There was no evidence for underlying
hemolytic anemia because lactate dehydrogenase levels rapidly
normalized and the hematocrit was stable. One year after surgery,
bilirubin levels were measured within the normal range and the patient
had fully recovered.
In conclusion, preexisting hyperthyroidism and its treatment with
thionamides might predispose patients for acute liver failure after
infection with the hepatitis A virus. Due to possible hepatotoxic side
effects, treatment with antithyroid drugs should be avoided in
hyperthyroid patients with viral hepatitis. We recommend immediate
institution of plasmapheresis, followed by thyroidectomy, in such
patients because it is the most rapid and reliable way to lower
circulating levels of thyroid hormones.
Received October 13, 1999.
Revised December 1, 1999.
Accepted December 6, 1999.
 |
References
|
---|
-
Kang H, Choi JD, Jung IG, et al. 1990 A case
of methimazole-induced acute hepatic failure in a patient with chronic
hepatitis B carrier status. Korean J Intern Med. 5:6973.[Medline]
-
Ghoshal UC, Mukherjee S, Dey BK, Das P, Sarkar S,
Mazumder DN. 1997 Subacute hepatic failure in hepatitis A. Indian
J Gastroenterol. 16:109.
-
Liaw YF, Huang MJ, Fan KD, Li KL, Wu SS, Chen TJ. 1993 Hepatic injury during propylthiouracil therapy in patients
with hyperthyroidism. A cohort study. Ann Intern Med. 118:424428.[Abstract/Free Full Text]
-
Binder C, Lang W. 1993 Necrotizing hepatitis with
a fatal outcome after carbimazole therapy. Dtsch Med Wochenschr. 118:15151519.[Medline]
-
Hanson JS. 1984 Propylthiouracil and hepatitis.
Two cases and a review of the literature. Arch Intern Med. 144:994996.[Abstract]
-
Limaye A, Ruffolo PR. 1987 Propylthiouracil-induced fatal hepatic necrosis. Am J
Gastroenterol. 82:152154.[Medline]
-
Weiss M, Hassin D, Bank H. 1980 Propylthiouracil-induced hepatic damage. Arch Intern Med. 140:11841185.[Abstract]
-
Fischer MG, Nayer HR, Miller A. 1973 Methimazole-induced jaundice. J Am Med Assoc. 223:10281029.[CrossRef][Medline]
-
Romaldini JH, Bromberg N, Werner RS, et al. 1983 Comparison of effects of high and low dosage regimens of antithyroid
drugs in the management of Graves hyperthyroidism. J Clin
Endocrinol Metab. 57:563570.[Abstract]
-
Baker B, Shapiro B, Fig LM, Woodbury D, Sisson JC,
Beierwaltes WH. 1989 Unusual complications of antithyroid drug
therapy: four case reports and review of literature. Thyroidology. 1:1726.[Medline]
-
Gurlek A, Cobankara V, Bayraktar M. 1997 Liver
tests in hyperthyroidism: effect of antithyroid therapy. J Clin
Gastroenterol. 24:180183.[CrossRef][Medline]
-
Hintze G, Köbberling J, Usadel KH. 1989 Schwere jodinduzierte Hyperthyreose und Thyreoidektomie. Innere
Medizin. 16:161164.
-
Ashkar FS, Katims RB, Smoak WM, Gilson AJ. 1970 Thyroid storm treatment with blood exchange and plasmapheresis. J Am Med Assoc. 214:12751279.[CrossRef][Medline]