Abstract
To the Editor:
Atomoxetine, a selective norepinephrine reuptake inhibitor, is the first nonstimulant approved by The Food and Drug Administration for the treatment of ADHD in children, adolescents, and adults (Pliszka 2007). This drug was considered to be safe, effective, and associated with relatively few adverse drug reactions: dyspepsia, nausea, vomiting, fatigue, rash, decreased appetite, and weight loss (Caballero and Nahata 2003). The Food and Drug Administration had recently reported two cases of markedly elevated hepatic enzymes and bilirubin, in the absence of other obvious explanatory factors, out of >2 million patients during the first 2 years of postmarketing experience (Lim et al. 2006; Stojanovski et al. 2007). Although to date there has been no recommendation for clinicians to do routine monitoring of liver function tests (LFT) during atomoxetine treatment, it has been recommended that parents and clinicians should be informed of the evidence of liver problems during treatment with this agent (Lilly, Product Information, July 2008).
Studies of safety and effectiveness of atomoxetine conducted on large samples have found no clinically significant abnormalities of LFT with treatment. However, adverse effects of psychotropic agents on the pediatric population can take a long period of time to emerge. Therefore, it may be argued that the effects of atomoxetine on liver function of children and adolescents are currently unclear. In this case report, we present a case of serious liver injury that needed liver transplantation after 5 days of atomoxetine treatment.
Case Report
The patient was a 10-year-old boy who was first seen at age 9 at a child and adolescent psychiatry outpatient clinic in Zonguldak, Turkey. He was brought in by his biological mother with complaints of hyperactivity, attention difficulties, and forgetfulness since infancy and school failure since beginning of the school period. After detailed psychiatric evaluation, he was diagnosed with ADHD according to Diagnostic and Statistical Manual of Mental Disorders, 4th edition criteria (American Psychiatric Association 1994). His developmental history was unremarkable and his intellectual capacity was within subaverage IQ level. He was initially treated with imipramine HCl at a dosage of 25–50 mg/day for 3 months in another hospital for ADHD. During treatment with this agent, patient showed minimal improvement of ADHD symptoms. All laboratory tests on admission including complete blood cell count with differential, serum electrolytes, urinalysis, thyroid function tests, serum iron, ferritin, copper, ceruloplasmin, vitamin B12, folate levels, and LFT were within normal limits. At our center, he was initially treated with methylphenidate immediate-release tablets and methylphenidate extended-release tablets. Initial methylphenidate dose was 20 mg/day bid and was gradually increased to 30 mg tid. There was no concurrent use of any other medication. During the treatment with methylphenidate, alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma glutamyl transpeptidase, and serum bilirubin levels were within normal limits.
As the patient showed minimal improvement of ADHD symptoms with the methylphenidate IR treatment, methylphenidate IR was discontinued and he was started on atomoxetine after 1 week of methylphenidate discontinuation. His height was 160.2 cm (>97th percentile) and weight was 60.0 kg (>97th percentile). The patient was prescribed atomoxetine 40 mg/day (0.75 mg/kg) for 2 days followed by an increased to 60 mg/day (1 mg/kg). There was no concurrent use of any other medication and his family history was negative for any liver problems.
During the second day of atomoxetine treatment, the patient started complaining of fatigue, fever, headache, drowsiness, and poor appetite. He also had abdominal pain and vomiting. He was examined by a pediatrician in the fourth day of these symptoms. On physical examination, he was not jaundiced, and his liver and spleen were not palpable. The rest of his physical examination was also unremarkable. Laboratory tests showed AST of 913 U/L (normal range [NR]: 25–45), ALT of 942 U/L (NR: 0–50), total bilirubin of 5.2 mg/dL (NR: 0–1), and direct bilirubin of 4.9 mg/dL (NR: <0.4). One day later, his parents noticed icterus on his scleras. He was admitted to Pediatrics Department of Zonguldak Karaelmas University Hospital. Physical examination revealed jaundice. No erythematous rash was detected. Hemolytic anemia or eosinophilia was not demonstrated on his peripheral blood film. Laboratory tests showed AST of 4,040 U/L (25–45), ALT of 2,832 U/L (0–50), total bilirubin of 6.5 mg/dL (NR: 0–1), direct bilirubin of 6.2 mg/dL (NR: <0.4), alkaline phosphatase (ALP) of 400 U/L, gamma glutamyl transpeptidase of 192 U/L, and LDH of 1,872 U/L. His prothrombin time was 52 seconds and International Normalized Ratio level was 4.15 (NR: 1–1.2).
The patient was diagnosed with acute liver failure (ALF). Serology was negative for hepatitis A, B, and C and HIV. Serology for acute infection with cytomegalovirus, herpes simplex, and Epstein-Barr viruses were also negative. Antinuclear antibody, liver–kidney microsomal antibody, antimitochondrial antibody, smooth muscle antibody, and perinuclear antineutrophilic cytoplasmic antibody tests were negative as well. Alpha-1 antitrypsin level was normal (150 mg/dL, NR: 110–280). Ceruloplasmin level was 22 mg/dL. There were no Kayser Fleischer rings on slit lamp examination. Twenty-four-hour urine copper level was high (336 μg/dL, NR: 5–50 μg per 24 hours). Serum copper level was normal (138 μg/dL). There was no history of ethanol or drug abuse. He had no family history of liver disease and no exposure to known hepatotoxins or hepatitis viruses. Atomoxetine was suspected as the causative agent of acute liver disease and was discontinued.
Plasmapheresis was performed for 3 hours, using an intermittent aphaeresis device, MCS plus (Haemonetics). To reduce extracorporeal volume, a 125-mL special pediatric bowel was used, and ∼3,600 mL plasma volume was removed in each plasmapheresis session. The drawing and returning flow rate were performed at 50 mL/minute. Fresh-frozen plasma was used as replacement fluid. Despite daily plasmapheresis for 4 days, hepatic encephalopathy emerged. He was referred to the Department of Pediatric Gastroenterology and General Surgery of Baskent University Medical Faculty for liver transplantation.
Upon his admission to this hospital, he had combative–aggressive behavior indicating grade-2 hepatic encephalopathy. His AST and ALT levels were 913 and 487 U/L, respectively. Total bilirubin and direct bilirubin levels were found to be 23 and 14 mg/dL. Serum ammonia level was high (95 μmol/L, NR: 10–55). Factor VII level was low (10%, NR is 50%–200%). Ceruloplasmin level was 25 mg/dL and 24-hour urine copper level was 95 μg/day. Abdominal computerized tomography demonstrated a geographical liver with hypo-hyper perfusion areas indicating necrosis and necrosis-spared parts of his liver. At the second day of his admission, hepatic encephalopathy reached grade 3. He was stuporous but could be awakened by painful stimuli. He underwent living-related left lobe liver transplantation from his mother on April 1, 2009. Pathological examination revealed massive hemorrhagic necrosis, ductular proliferation, and inflammation in partially preserved area, ballooned hepatocytes, and minimal hepato-canalicular cholestasis. There was neither steatosis nor signs of chronic liver disease; findings were consistent with severe acute hepatic injury. Since then, he is alive and taking tacrolimus and mycophenolate mofetil as immunosuppressive treatment.
Discussion
This patient's clinical features, laboratory and radiologic findings, and the histology of liver explants indicated a probable positive relationship between atomoxetine exposure and ALF. Reported causes of ALF are viral hepatitis, toxins, serious vascular events (e.g., ischemic hepatitis, hepatic venous outflow obstruction), hepatic malign infiltration, metabolic disorders (e.g., Wilson's disease [WD]), and autoimmune hepatitis.
Our patient's high urinary copper level that suggested WD was a confusing finding at first sight. It has been well known that patients with WD can present with ALF. It was reported that conventional WD testing utilizing serum ceruloplasmin and/or serum copper levels is less sensitive and specific in identifying patients with WD presenting with ALF than other available tests (Korman et al. 2008). An ALP-to-total bilirubin ratio of <4 yielded a sensitivity of 94% and a specificity of 96% for diagnosing fulminant WD. In addition, an AST:ALT ratio of >2.2 yielded a sensitivity of 94% and a specificity of 86% for diagnosing fulminant WD. Combining the tests provided a diagnostic sensitivity and specificity of 100% (Korman et al. 2008).
Plasmapheresis is known to affect ALT, AST, ALP, bilirubin, and serum copper levels. For this reason, we calculated our patient's ALP-to-bilirubin ratio and AST/ALT ratio from the laboratory tests before plasmapheresis during his admission to Karaelmas University. His AST/ALT ratio was 1.43 and ALP/total bilirubin ratio was 67. Both of these ratios were not indicating a case of fulminant WD. Our patient's ceruloplasmin levels were found to be normal twice. There were no Kayser Fleischer rings. His high urinary copper excretion can be explained by massive hepatic necrosis and copper release from necrotic hepatocytes. WD is an inborn disease of metabolism and most patients with WD presented with ALF had cirrhosis and hepatosteatosis indicating chronicity (Cope-Yokoyama et al. 2010). Our patient's liver histology findings were not in favor of WD. This patient had ALF with massive liver necrosis without cirrhosis and hepatosteatosis findings. According to all findings, this patient did not have WD.
Efficacy of atomoxetine in the treatment of ADHD has been shown. In literature, there have been rare reports of hepatic abnormalities associated with atomoxetine, and most of these abnormalities have been reported to be transient and do not pose a serious risk. However, our case suggests that it may have a considerable potential to cause liver function abnormalities. In accordance with our case, the labeling of atomoxetine was recently modified (late 2004) to include severe liver injury among adverse events, based on recent case reports of hepatotoxicity with atomoxetine (Lim et al. 2006; Stojanovski et al. 2007). First, Stojanovski et al. (2007) reported hepatotoxicity in a child with atomoxetine. This patient's AST, ALT, and bilirubin levels were markedly elevated in the first month of atomoxetine therapy and returned to normal level after discontinuation of atomoxetine. Later, Lim et al. (2006) reported two children who presented with acute hepatitis after starting therapy with atomoxetine. In one child, hepatotoxicity symptoms started after 3 weeks of atomoxetine therapy, no competing diagnosis could be identified, and liver injury resolved completely on withdrawal of the medication. In the second child, hepatotoxicity symptoms started after 4 months of atomoxetine therapy, and the evaluation was suggestive of type 1 autoimmune hepatitis; she subsequently improved with removal of atomoxetine and concomitant immunosuppressive therapy. Our patient had hepatic injury leading to ALF in the first week of atomoxetine therapy. Concordant with our finding, hepatotoxicity occurred in the first month of atomoxetine treatment in two reported cases.
In an investigation of case reports identified by a computerized search that contained potential hepatic events, of the 7,962 pediatric and adult patients treated with atomoxetine in clinical trials, 41 were identified as requiring further analysis. Of those 41 cases, none progressed to liver failure, and most of these events were mild increases in ALT and AST levels (Bangs et al. 2008). During the 4 years since the market launch of atomoxetine, 351 cases of liver injury were related to the drug treatment for ADHD. Of those 351 cases, 69 had explanations unrelated to the use of the drug, 146 presented insufficient information to assess the cause, 133 contained confounding factors and were labeled as possibly related to drug use, and the remaining 3 cases reported liver injury probably related to atomoxetine use (Bangs et al. 2008).
The etiology of drug-induced liver injury with atomoxetine may be metabolic idiosyncrasy or drug-induced autoimmune hepatitis. Hepatotoxicity with most drugs is idiosyncratic. Drug-induced liver injury develops in only a small proportion of subjects exposed to a drug in therapeutic doses, and the risk of ALF associated with idiosyncratic hepatotoxins is usually <1 per 10,000 exposed patients (Russmann et al. 2010). Although ALF is rare, 13%–17% of all ALF cases are attributed to idiosyncratic drug reactions (Hussaini and Farrington 2007). Patients who show an increase in liver enzymes should be checked for hallmarks of hypersensitivity reactions such as gastrointestinal complaints, fever with chills, rash, malaise, fatigue, muscle aches, and eosinophilia, which point to an immunoallergical basis of injury. Exclusion of other causes such as chronic viral hepatitis B and C, alcoholic liver disease, hemochromatosis, and nonalcoholic fatty liver disease is mandatory as well. When recognized and the offending agent is discontinued, most instances of acute hepatocellular drug-induced liver disease are reversible and heal without chronic sequelae (Lewis 2000).
Clinically significant LFT elevations are probably uncommon; however, based upon our case and other reports, baseline as well as periodic monitoring of LFTs in atomoxetine-treated patients is advised. Consistent with the present case and other reported cases, monitoring of LFTs should be made at the onset of atomoxetine therapy.
Footnotes
Disclosures
Drs. Ayten Erdogan, Mehmet Calik, Ethem Piskin, Mehmet Haberal, Mehmet Goksin Karaman, Figen Ozcay, Banu Bilezikci, and Ihsak Tekin have no conflicts of interest or financial ties to disclose.