Gabapentin, Pregabalin Tied to Higher Risks Early in Treatment

People prescribed gabapentinoids — gabapentin or pregabalin — had a higher risk of drug poisoning soon after they started treatment, especially if they took benzodiazepines or opioids concurrently, a study of 17,000 U.K. primary care patients showed.

Compared with reference periods, the risk of hospitalization for drug poisoning rose during the first 4 weeks of gabapentinoid treatment, with an adjusted incidence rate ratio (aIRR) of 1.81 (95% CI 1.66-1.99, P<0.001), reported Kenneth Man, PhD, of the University College London, and co-authors.

People taking both gabapentinoids and benzodiazepines were four times more likely to be hospitalized for drug poisoning in the 4 weeks after starting gabapentinoid treatment (aIRR 3.95, 95% CI 3.07-5.07, P<0.001), compared with people taking neither drug, Man and colleagues said.

Gabapentinoids combined with opioids doubled the poisoning risk in the first 4 weeks compared with taking neither drug (aIRR 2.14, 95% CI 1.77-2.58, P<0.001), they wrote in PLoS Medicine.

“Prescription rates for gabapentinoids have been increasing rapidly in recent years, as they are seen as a safe alternative to opioids,” Man said in a statement. “While they can be effective for pain relief and do have better perceived safety profiles than opioids, there are still substantial risks that clinicians and patients should be mindful of.”

Despite limited indications, gabapentin and pregabalin are widely prescribed off-label, especially for chronic pain. In 2024, gabapentin was the fifth most dispensed product in U.S. retail pharmacies, according to the CDC.

In 2019, the FDA issued a warning about respiratory problems that can occur if gabapentinoids are used alongside opioids or other central nervous system depressants like benzodiazepines. Prescribing information now includes this risk.

Man and colleagues assessed patient information from 2010 to 2020 in the U.K. Clinical Practice Research Datalink Aurum database, excluding people with epilepsy or cancer. Over the study period, 1.35 million people had at least one gabapentinoid prescription; of those, 16,827 people had at least one drug poisoning hospitalization.

Using a self-controlled case series approach, the researchers evaluated drug poisoning hospitalizations before and after gabapentinoids were prescribed. The mean age of the study sample was 47, and the mean follow-up time was 8.18 years.

During the observation period, 45.4% of participants were prescribed gabapentin only, 34.7% were prescribed pregabalin only, and 19.9% were prescribed both drugs. Most participants (83.7%) had neuropathic pain or chronic pain, and 76.2% had a mental health condition.

The analysis included intentional and accidental poisonings and did not exclude people who were taking more than prescribed dosages or misusing medications. At some point in the study period, 89.0% of participants took gabapentinoids with prescribed opioids, and 54.7% took them with prescribed benzodiazepines.

The incidence of drug poisoning fell after the first 4 weeks of treatment. For patients on gabapentinoids alone, it eventually reached an aIRR of 1.11 (95% CI 1.05-1.17, P<0.001), where it remained for the rest of the treatment period.

Notably, the risk of drug poisoning was doubled during the 90 days before treatment started (aIRR 2.09, 95% CI 1.98-2.21, P<0.001), compared with nontreatment periods, the researchers pointed out.

“The findings suggest that gabapentinoids are often started during periods of already heightened vulnerability to drug poisoning, and that risk remains modestly elevated during treatment, and further elevated when opioids or benzodiazepines are co-prescribed,” Man and co-authors wrote.

“Clinicians should be aware of this increased vulnerability, monitor patients closely around treatment initiation, and avoid or minimize concurrent opioid or benzodiazepine use where possible to support safer prescribing,” they added.

Self-controlled case series cannot fully rule out the influence of unmeasured time-varying confounders like major life events or illicit drug use, the researchers acknowledged. However, a negative control analysis did not indicate substantial residual confounding, they said.

The database did not include adherence information and captured prescriptions from general practitioners only. In addition, prescribing records were not linked to diagnostic codes, so the researchers could not tie prescriptions to specific indications at each treatment episode.