Drug-induced liver injury resulting in fulminant liver failure is a well-known condition, and many drugs have been documented in the literature as possible etiologies. However, hydralazine has seldom been reported as the offending agent. Our case report is about one such rare scenario of fulminant liver failure due to hydralazine use as an antihypertensive. A 65-year-old female patient presented with signs of fulminant liver failure 2 months after starting hydralazine for hypertension. She underwent extensive workup for the cause of acute liver failure. Other possible medications were ruled out, and workup for autoimmune and other etiologies were also negative. The patient underwent a deceased donor liver transplant and has been doing well since then. Her liver was found to be atrophic, with microscopically confirmed drug-induced liver injury. Hydralazine is used orally to treat essential hypertension and intravenously to emergently lower blood pressure. Hydralazine-induced acute liver failure is extremely rare. However, in this rare case where hydralazine-related drug-induced liver injury worsened to the extent of requiring liver transplant, we felt obliged to document and highlight this complication as a form of reminder to our colleagues of this serious outcome.
Key words : Acute liver failure, Hypertension, Liver toxicity, Liver transplantation
Drug-induced liver injury (DILI) resulting in fulminant hepatic failure is a well-known condition. Acetaminophen, isoniazid, propylthiouracil, phenytoin, and valproate account for 15% of cases requiring liver transplant in the United States (US).1 Hydralazine, a direct vasodilator, is a commonly prescribed antihypertensive medication, with more than 7.5 million prescriptions in the US in 2018.2 It was first used in the 1950s and officially approved for use in the US in 1984. Hydralazine-induced hepatotoxicity is an uncommon condition with first reports dating back to the 1970s and 1980s.3,4 The acute liver injury mechanism associated with hydralazine appears to be due to its metabolism to an immunologic product that can cause both an immune-allergic hepatitis or a more delayed lupus- and/or an autoimmune hepatitis-like syndrome.5,6 In 2 large studies evaluating DILI (n = 1198 and n = 313), there was only 1 case found to be associated with use of hydralazine.7,8
Here, we have described a rare patient case of acute liver injury within 2 months of hydralazine treatment. With no improvement after medication cessation and progression to fulminant liver failure with severe encephalopathy, the patient required a liver transplant.
A 65-year-old woman without significant past medical or surgical history presented to an outside facility with a 4-day history of progressive fatigue. On admission, she was noted to have acute mental status changes with slurred speech, increasing lethargy, constipation, and scleral icterus. Two months before presentation, the patient was diagnosed with hypertension and was started on hydralazine by her primary care provider, with the dose up-titrated 1 month later to 50 mg twice per day. Prior to initiation of hydralazine, her hepatic function panel was within normal limits. She had not used any over-the-counter medications, antibiotics, herbal supplements, or acetaminophen during this period, and she did not report use of alcohol or illicit substances.
At initial evaluation at an outside facility, she was found to be severely hypoglycemic (32 mg/dL), and her laboratory work-up was significant for serum creatinine level of 2.5 mg/dL, white blood cell count of 15 900/mm3, alanine aminotransferase (ALT) of 2065 U/L (normal range, 14-38 U/L), aspartate aminotransferase (AST) of 2055 U/L (normal range of <35 U/L), total bilirubin of 15.2 mg/dL (normal range of <1.2 mg/dL), international normalized ratio (INR) of 10.05, and lactate of 15.1 mmol/L. Alcohol, salicylate, and acetaminophen levels were within normal limits. A head computed tomography scan was negative, and no intra-abdominal abnormality was seen with the exception of a small amount of free fluid in the pelvis.
Because of the severity of her symptoms, she was transferred to the medical intensive care unit at our institution. She underwent extensive work-up for her acute liver failure, including viral serologies for hepatitis (positive hepatitis A immunoglobulin G [IgG]), varicella zoster virus, cytomegalovirus (positive IgG), Epstein-Barr virus (positive IgG; negative viral load), herpes virus (positive herpes simplex virus type 1 and type 2), human immunodeficiency virus, as well as autoimmune markers (antinuclear antibody, antimitochondrial antibody, antineutrophil cytoplasmic antibodies, myeloperoxidase antibodies, and anti-smooth muscle antibody; positive 1:80) and total IgG. N-acetylcysteine was continued (had received 1 dose at outside facility before her transfer) for 24 hours per protocol. A liver ultrasonography examination showed a small liver with no evidence of thrombus within the portal vasculature, hepatic veins, or arteries.
Over the next 2 days, her INR and bilirubin worsened (Table 1), and her mental status severely deteriorated to the extent of requiring intubation for airway protection. A head computed tomography did not show any cerebral edema. On 2-dimensional echo examination of her heart, she was found to have preserved biventricular function with no significant valvular disease; however, a small pericardial effusion persisted on serial examination.
The patient was actively listed for liver transplant on day 4 of hospitalization as a status 1a candidate. She received orthotopic liver transplant the next day from a brain dead donor. Intraoperatively, her liver was observed to be completely collapsed and shrunken. In addition to liver transplant, the patient underwent a pericardial window and pericardial drain placement due to her persistent pericardial effusion. Surgical pathology of the explanted liver showed extensive hepatocyte necrosis and collapse affecting approximately 90% of the hepatic parenchyma, with accompanying congestion, cholestasis, and acute and chronic inflammation. Postoperatively, her hepatic function panel normalized; however, she eventually required intermittent hemodialysis due to rising creatinine and non-anion gap metabolic acidosis. Because of her poor mental status and concerns with airway protection, despite passing spontaneous breathing trials, she underwent tracheostomy on postoperative day 10 (POD10). Her mental status and overall condition gradually improved, and, on POD13, she was transferred to our intermediate care unit with subsequent transfer to a regular transplant service floor bed on POD21. She was decannulated on POD24. On POD45, she was discharged to an acute rehabilitation facility with presentation of tolerating an oral diet, voiding spontaneously, and ambulating with assistance. At over 10 months posttransplant, the patient remains well.
Hydralazine-induced acute liver failure is extremely rare. Frequently, the injury is reversed within days of hydralazine discontinuation, with clinical laboratory normalization, and recurs when the medication is resumed.5 In hydralazine-induced hepatitis, liver damage is variable; it may present as hypersensitivity reaction, mixed hepatocellular injury, cholestatic jaundice, and centrilobular necrosis.9,10 There are 2 different types of injury described with hydralazine: 1 with a short latency period within 1 to 2 months of initiation of therapy and the other with latency periods ranging from 2 months to 1 year.9,11 In the first, patients have symptoms of fever, rash, and eosinophilia; in the latter, the findings are of chronic hepatitis and fibrosis. However, fulminant hepatic failure as observed in our case is unique, and it is important to reiterate that, in 2 large recent studies evaluating DILI with more than 1500 patients combined, only 1 case was found to be associated with hydralazine.7,8
Our patient had positive anti-smooth muscle antibody, which could represent autoimmune hepatitis; however, microscopy of the liver explant was suggestive of DILI rather than autoimmune pathology. On microscopy, her liver was found to have extensive hepatocyte necrosis and drop out, affecting approximately 90% of the hepatic parenchyma, with accompanying congestion, cholestasis, and acute and chronic inflammation. This extensive necrosis with major hepatocyte loss along with portal infiltrates and cholestasis supported a drug-induced pathology rather than autoimmune pathology.12 Reticulin stain showed dense condensation, consistent with extensive parenchymal collapse. Trichrome stain showed increased fibrosis around the portal tracts and condensation similar to the reticulin stain, again consistent with massive hepatocyte loss. A thorough search for alternate etiologies was made, including a review of pharmacy records in search for alternative medications.13
Intraoperatively, the liver was seen to be shrunken, measuring 14.4 × 12.2 × 4.1 cm and weighing 380 g. The liver appeared red-tan and micronodular, with a nutmeg pattern favoring acute liver failure rather than autoimmune pathology.
Hydralazine is used orally to treat essential hypertension and intravenously to emergently lower blood pressure. Since the introduction of angiotensin-converting enzyme inhibitors and beta-blockers, hydralazine as a first-line antihypertensive agent indication is limited to African American individuals for the treatment of congestive heart failure, along with isosorbide dinitrate.14,15 Despite these adverse events and the possibility of DILI, hydralazine has a good safety profile, has been marketed for decades, and should continue to be used when appropriately indicated. Because hydralazine DILI to the extent of requiring liver transplant is extremely rare, our goal was to document and highlight this complication as a form of reminding and alerting our colleagues of the serious outcomes that can be produced.
DOI : 10.6002/ect.2021.0446
From the 1Division of Abdominal Transplant, the 2Department of Surgery, the 3Department of Pharmacy, and the 4Divison of Gastroenterology and Hepatology, University of North Carolina, Chapel Hill, North Carolina, USA
Acknowledgements: The authors have not received any funding or grants in support of the presented research or for the preparation of this work and have no declarations of potential conflicts of interest.
Corresponding author: Chirag S. Desai, Abdominal Transplant Division, Department of Surgery, 4001 Burnett-Womack Building, CB #7050, Chapel Hill, NC 27599, USA
Table 1. Laboratory Results in Patient Over Time