Daniel M. Keller, PhD

November 21, 2012

BOSTON — Bariatric surgery appears to raise the risk for acute liver failure (ALF) from acetaminophen poisoning, according to a retrospective review of adult patients.

The research showed a higher prevalence of ALF among patients who had had bariatric surgery, Edward Holt, MD, from California Pacific Medical Center (CPMC) in San Francisco, reported here at The Liver Meeting 2012: American Association for the Study of Liver Diseases (AASLD) 63rd Annual Meeting. The association was independent of depression, alcohol abuse, the use of combination analgesics, or intent to cause self-harm.

Dr. Holt said that acetaminophen overdose is the leading cause of ALF in the United States and that about half of ALF cases caused by acetaminophen resulted from unintentional overdose, often from the use of combination analgesics containing acetaminophen.

The prevalence of obesity in US adults has reached epidemic proportions, and bariatric surgery can help provide individuals with significant and durable weight loss. Its use has grown during the past 20 years, and now almost 4.5 million US adults aged 18 to 75 years have undergone bariatric surgery.

Such surgery may alter the transit, absorption, metabolism, and pharmacokinetics of some drugs and alcohol. Precursors to glutathione may be deficient after bariatric surgery; glutathione participates in excretion of metabolites of acetaminophen from hepatocytes. In addition, higher rates of suicide have been documented in patients after bariatric surgery.

Because the investigators saw an unexpectedly high proportion of patients with acetaminophen-associated liver failure who had also had bariatric surgery, they studied whether there was an association between them in a retrospective review of all adult patients with ALF at CPMC for the years 2009 to 2011.

Acetaminophen ALF was defined as encephalopathy and coagulopathy with a history of or positive serum acetaminophen level in the absence of known underlying liver disease and competing causes of ALF.

Patients with ALF (n = 101) had an average age of 45 years; 82% of patients were women, and 67% were white. Of the 101 patients, 29% had a history of depression, 24% had a history of alcohol abuse, almost 37% had used combination drugs containing acetaminophen, 18% had had a transplant, 9% had had bariatric surgery, and 19% died.

The investigators determined that in 54 cases (53.5%), acetaminophen was the cause of the ALF. Thirteen patients (12.9%) had an indeterminate cause. Nine patients with ALF (8.9%) had prior bariatric surgery (8 Roux-en-Y gastric bypass, 1 duodenal switch). The mean time from bariatric surgery to ALF was 5.9 years (range, 2 to 9 years).

There were significantly more women and white persons in the group with acetaminophen ALF than in the group (n = 47) with ALF from other causes (women: 88.9% vs 74.5%, respectively, P = .02; white persons: 81% vs 51.1%, respectively, P < .001). There was also significantly more depression (P < .001), alcohol abuse (P < .01), and use of combination drugs containing acetaminophen (P < .001) in the acetaminophen ALF group.

Conversely, the group with acetaminophen ALF had a lower proportion of transplant recipients (3.7% vs 34%; P < .001) and a lower proportion of deaths (14.8% vs 23.4%; P < .001) compared with patients with ALF from other causes.

More Bariatric Surgery Patients in Acetaminophen ALF Group

Of the 54 patients with acetaminophen ALF, 9 (16.7%) had undergone bariatric surgery compared with no patients in the group with ALF from other causes (P = .003). In the bariatric surgery group, 66.7% had used combination drugs vs 33.7% in the group without bariatric surgery (n = 92) (P = .07). There was proportionally more depression and alcohol abuse in the non–bariatric surgery group, but the differences between the non–bariatric surgery and bariatric surgery groups were not significant.

For the groups of patients with acetaminophen ALF with or without bariatric surgery, only alcohol abuse differed between the groups, and it was more prevalent in the group without bariatric surgery: 0 of 9 in the bariatric surgery group vs 16 of 45 in the non–bariatric surgery group (35.6%) (P = .045).

"There was no difference in age, gender, or race between the 2 groups. We found the patients with bariatric surgery did not have more depression, alcohol abuse, or the use of combination drugs, and they did not have a higher intent to cause themselves harm," Dr. Holt said. "Interestingly, we reported a nearly 80% rate of unintentional overdose in the bariatric surgery patients."

"We observed a higher proportion of prior bariatric surgery among patients with [acetaminophen] ALF compared to ALF of other etiologies, yielding a prevalence that was strikingly higher than in the general population," he said during his presentation. The prevalence of bariatric surgery in the acetaminophen ALF group was 16.7% vs 0.66% in the general population, a 25.3-fold higher rate of bariatric surgery in the acetaminophen ALF group compared with the general population.

Limitations of the study were its retrospective nature, that it was performed in a single center, and that it could not address mechanisms of increased risk for acetaminophen ALF in bariatric surgery patients. Whatever mechanisms are at work, the investigators concluded that they appeared to be independent of depression, alcohol abuse, use of combination analgesics, and intent to cause self-harm.

Dr. Holt said the study results have important implications if they are validated in a larger cohort. "Our novel finding may identify a new group of patients at higher risk for [acetaminophen] ALF. If validated, additional warnings for patients with prior bariatric surgery may be warranted, similar to those currently in place for patients who consume 3 or more alcoholic drinks daily."

Session moderator David Mulligan, MD, a transplant surgeon and director of the transplant center at Mayo Clinic in Phoenix, Arizona, called the study very interesting because "it showed an association in a group of patients that's a growing population, that are having morbid obesity surgery.... And now looking at a population of individuals who are having a highly successful bariatric procedure and finding out that there's a commonly used household medication that may put them at increased risk of liver failure is really pivotal."

He said the study warrants further investigation on a larger scale "to see if there's a true risk here and whether in fact the [US Food and Drug Administration] needs to look at adding to the warnings [that] if you've had bariatric surgery, [you should] watch how much [acetaminophen] you take. Maybe you need to take less."

Dr. Mulligan suggested that future studies will need to look at dose effects, the timing of acetaminophen use relative to the time of bariatric surgery, the risk of medications that combine acetaminophen with other drugs, and possibly mechanisms for the observed effects.

Bariatric surgeons "at least who are doing the gastric bypass operation" may need to warn their patients to cut back on acetaminophen, certainly to below the maximum daily dose stated on the product label, he added. What is not known is how much patients would need to reduce their use of acetaminophen and whether the risk is limited to patients who have had gastric bypass or if it extends to gastric sleeve operations, gastric bands, or other bariatric procedures.

Dr. Mulligan also recommended closely watching patients who had had bariatric surgery and have any kind of liver disease and advising them to refrain from acetaminophen use entirely.

There was no commercial support for the study. Dr. Holt and Dr. Mulligan, who was not involved in the study, have disclosed no relevant financial relationships.

The Liver Meeting 2012: American Association for the Study of Liver Diseases (AASLD) 63rd Annual Meeting. Abstract 4. Presented November 11, 2012.