Gabapentin May Not Raise the Risk of Falling After All

New use of gabapentin (Neurontin) was not associated with greater risk of falls in older adults with neuropathy or fibromyalgia compared with new use of duloxetine (Cymbalta), a commercial claims analysis suggested.

At 30, 90, and 180 days, the weighted cumulative incidence of a fall-related visit was 103.60, 90.44, and 84.44 per 1,000 person-years for new gabapentin users and 203.43, 177.73, and 158.21 for new duloxetine users, reported Alexander Chaitoff, MD, MPH, of the University of Michigan in Ann Arbor, and co-authors.

At 6-month follow-up, new gabapentin users had lower hazard of falls (HR 0.52, 95% CI 0.43-0.64), Chaitoff and colleagues wrote in Annals of Internal Medicine. However, there was no significant difference between gabapentin and duloxetine in the hazards of experiencing a severe fall with hip fracture (HR 0.78, 95% CI 0.32-1.86) or hospitalization associated with a fall (HR 0.68, CI 0.42-1.11).

“Gabapentin has been labeled a ‘bad’ medication and has gotten a lot of negative press in the past couple of years,” Chaitoff told MedPage Today. “But all medications can be ‘good’ or ‘bad’ depending on whether they are given to the right person, for the right indication, at the right time, in the right form.”

The use of gabapentinoids — gabapentin and pregabalin (Lyrica) — has climbed in the U.S. since the drugs were first approved. Common side effects of gabapentin include somnolence, dizziness, and ataxia.

Gabapentin is indicated for seizures and nerve pain associated with shingles. Pregabalin is indicated for fibromyalgia and pain associated with diabetic peripheral neuropathy or spinal cord injury.

Both drugs are widely prescribed off-label as a non-opioid alternative for various other pain syndromes. “Gabapentin is prescribed more than 40 million times each year,” Chaitoff noted.

Newer evidence has suggested, however, that gabapentin may not work for all pain conditions and is associated with falls. As a result, advocates have called for efforts to reduce gabapentin use, which requires an accurate assessment of its risks, Chaitoff and colleagues pointed out.

The largest studies that have estimated gabapentin’s risks have compared its use against patients who didn’t use medications, the researchers noted. This could mean gabapentin’s risks could be overestimated, which could lead to pain being undertreated, they observed.

Including an active comparator like duloxetine in a study can reduce the chance that outcomes are driven by confounding by indication, “given that we know patients with neuropathy are more likely to fall even without medications,” Chaitoff said.

“The active comparator is also important because in routine clinical practice, patients with neuropathic pain are not offered gabapentin or nothing,” he added.

Chaitoff and co-authors used a target trial emulation framework to assess falls in MarketScan claims data after patients started gabapentin or duloxetine. Duloxetine is approved for diabetic peripheral neuropathic pain, fibromyalgia, and chronic musculoskeletal pain, in addition to major depressive disorder and generalized anxiety disorder.

The study included 57,086 adults ages 65 or older with a diagnosis of diabetic neuropathy, postherpetic neuralgia, or fibromyalgia in the year before cohort entry. Of these, 52,152 were newly prescribed gabapentin and 4,934 were newly prescribed duloxetine from January 2014 to December 2021. Differences in covariates between the groups were small after weights were applied.

Patients with depression, anxiety, seizures, or cancer were excluded from the study. The median initial dose of gabapentin was 200 mg. The overall median follow-up duration was 30 days.

New user, active comparator, and target trial emulation studies can overcome selection bias and reduce the risk for unmeasured confounding, Chaitoff and co-authors observed. “Most notably, whereas prior literature reports higher rates of falls in gabapentin users versus nonusers, we found that new gabapentin versus duloxetine users had a lower hazard of any fall-related healthcare visit and no increase in severe fall-related healthcare events,” they wrote.

One explanation for the difference in outcomes may be dosing: gabapentin is highly titratable and many new users started on low doses. “Alternatively, duloxetine is usually titrated only once or twice,” the researchers noted. It’s possible that during peri-initiation and titration periods, gabapentin may be less likely to cause falls than duloxetine, they suggested.

The analysis has other limitations, they acknowledged. Despite its design, the study still was subject to confounding because pain is often undercoded in administrative records. The researchers also could not stratify results by drug dosage.

“It is important to note that this study does not address all clinical use situations,” Chaitoff said. “For example, it did not specifically assess whether gabapentin is associated with fewer falls among those at higher risk for acute kidney injury.” It also did not measure differences in other outcomes, like somnolence.

Please enable JavaScript to view the

comments powered by Disqus.