gabapentinoids increase asthma and copd

  

 

given my general feelings that gabapentinoids (gabapentin and pregabalin) are dramatically over-prescribed and often for inappropriate indications, this blog will address issues of major complications in terms of asthma and COPD as well as heart failure associated with their use (see gabapentinoids inc asthma Thorax2025 in dropbox, or https:// doi. org/ 10. 1136/ thorax- 2025- 223240). 

Details:

-- researchers in Japan accessed their National Database of Health Insurance Claims and Specific Health Check-ups (NDB) from 2015 to 2023

    -- this database has data from all public insurers in Japan, resulting in nearly complete coverage of hospitals nationwide, with demographic data on each patient as well as diagnoses, examinations, and treatments from both inpatient and outpatient settings

-- the patients included were at least 40 years old (that is when routine checkups are done in Japan, and a lot of baseline data are elicited) and who had asthma comorbid with neuropathic or chronic pain

-- in Japan gabapentinoids are indicated in those with epilepsy, restless leg syndrome, neuropathic pain and pain associated with fibromyalgia

    -- but, the major prescriptions for gabapentinoids are for managing neuropathic and chronic pain

-- the gabapentinoids users were compared to two active comparator cohorts, those on tricyclic antidepressants (TCAs) or on serotonin-norepinephrine reuptake inhibitors (SNRIs). drug exposure was the time from the day prescription was filled to 28 days following the exhaustion of dispensed days' supply

-- 171,393 patients were found in the database who initiated gabapentinoids, 5916 initiated TCAs, and 19,800 initiated SNRIs

-- in the group comparing those on gabapentinoids to TCAs: mean age 55.4, 73.4% female, 13.5% current smokers, BMI <18 in 8.8%/BMI 18.5-24.9 in 58.2%/BMI 25-29.9 in 24.5%/BMI at least 30 in 8.5%

-- duration of pain: less than one year in 21.3%/1 to 2 years in 6.6%/at least two years in 17.1% (the rest unspecified)

-- duration of asthma: less than one year in 14.9%/1 to 2 years an in 7.4%/at least two years in 16.7% (the rest unspecified)

    -- overall, prior to propensity score-matching, gabapentinoids users were likely to be older and have neuropathic pain but less likely to have depression or use psych medications

-- in the group comparing those on gabapentinoids to SNRIs: mean age 60, 65% female, 14.4% current smokers, BMI <18 in 6%/BMI 18.5-24.9 in 55.5%/BMI 25-29.9 in 27.8%/BMI at least 30 in 10.7%

-- duration of pain: less than one year in 19.1%/1 to 2 years in 6.3%/at least two years in 15.9% (the rest unspecified)

-- duration of asthma: less than one year in 14.4% /1 to 2 years an in 7.6%/at least two years in 14.8% (the rest unspecified)

-- confounders assessed included patient demographics, comorbidities, treatments, healthcare resources, health-related behavior, blood test results, and asthma-related variables such as smoking status, duration of asthma, type II inflammation diseases (such as conditions associated with overactive immune response specifically of T-helper 2 lymphocytes, as in asthma, atopic dermatitis, allergic conjunctivitis, chronic urticaria), asthma-related hospitalizations in the prior year, and exercise capacity

-- Propensity score-weighting was used to control for the identified confounders (ie, to mathematically make the disparate cohorts similar)

-- primary outcome: the initial occurrence of an asthma exacerbation requiring systemic steroids, defined as a prescription of oral corticosteroids equivalent to at least 15 mg/day of prednisolone, or intravenous steroids for three or more days

-- secondary outcomes: 

    -- the number of asthma exacerbations requiring systemic corticosteroids during the follow-up

    -- the initial occurrence of an asthma exacerbation requiring hospitalization

    -- the number of asthma exacerbations requiring hospitalization during the follow-up

-- subgroup analyses included stratification by age <65 versus older, sex, BMI (<18.5, 18.5 to 24.9, 25 to 29.9, and at least 30), comorbid COPD, previous use of weak opioids, previous use of strong opioids, previous use of NSAIDs, which of the gabapentinoids were used (in Japan they also use mirogabalin, which is touted as having potentially greater therapeutic effects for neuropathic pain and fewer adverse effects than the other gabapentinoids; this med is not available in the US), and the presence of neuropathic pain

-- in addition a negative outcome analysis was done to assess for residual confounding (e.g. health seeking behavior) by selecting herpes zoster infection, felt to be unrelated to asthma or exacerbations

Results, after propensity score-weighting:

-- comparing those on gabapentinoids versus TCAs:

    -- first exacerbation requiring systemic corticosteroids: incidence rates 59.4 versus 33.7, 46% increase with the gabapentinoids, HR 1.46 (1.34-1.60)

    -- number of exacerbations requiring systemic corticosteroids: incidence rates 109.5 versus 72.3, 52% increase with gabapentinoids, HR 1.52 (1.40-1.64)

    -- first exacerbation requiring hospitalization: incidence rates of 0.91 versus 0.42, twice as much in those on gabapentinoids, HR 2.02 (1.11-3.68)

    -- number of exacerbations requiring hospitalization: incidence rates of 1.12 versus 0.53, more than twice as much in those on gabapentinoids, HR 2.11 (1.20-3.71)

        -- of note, the curves of cumulative incidences over time tended to increase until one year and then remain with unchanged significant difference

        -- there were no significant differences by subgroups stratified by age, sex, BMI, COPD, opioid use, NSAIDs, or neuropathic pain

        -- gabapentin was not associated with a statistically higher rate of outcomes, though pregabalin and mirogabalin were (gabapentin is not prescribed much in Japan)

        -- exploratory analyses demonstrated dose-related hazard ratios that were higher in the medium- and high-dose groups compared with the low-dose group; and no association between gabapentinoids and a higher incidence of negative outcomes (with herpes zoster infection), which helps validate their results

-- comparing those on gabapentinoids versus SNRIs:

    -- first exacerbation requiring systemic corticosteroids: incidence rates 63.5 versus 42.8, 24% increase with the gabapentinoids, HR 1.24 (1.34-1.60)

    -- number of exacerbations requiring systemic corticosteroids: incidence rates 119.7 versus 100.4, 19% increase with gabapentinoids, HR 1.19 (1.15-1.24)

    -- first exacerbation requiring hospitalization: incidence rates of 0.93 versus 0.68, 24% increase in those on gabapentinoids, HR 1.24 (0.94-1.63), a strong trend but not statistically significant

    -- number of exacerbations requiring hospitalization: incidence rates of 1.27 versus 1.21, 5% increase in those on gabapentinoids, HR 1.05 (0.81-1.37), not statistically significant

    -- no subgroup differences were found for age, sex, BMI, COPD, opioid use, NSAIDs, or neuropathic pain. gabapentin was not associated with a higher incidence of the outcome, though pregabalin and mirogabalin were  (gabapentin is not prescribed much in Japan)
    -- exploratory analysis did also find that hazard ratios were dose-related and higher in those on medium and high dose of gabapentinoids than if on low dose; and no association between gabapentinoids and a higher incidence of negative outcomes (with herpes zoster infection), which helps validate their results

Commentary:

-- gabapentinoid prescriptions have increased dramatically, with gabapentin the fifth most prescribed US medication in 2024, with an increase from 2010 to 2024 from 79.5 to 177.6 per thousand people and mostly given to older adults and women:  in the UK, there has been more than a 10-fold increase in the past two decades, and concerns of misuse has led to their reclassification of gabapentin as controlled substances with potentials for harm and misuse (https://www.medpagetoday.com/neurology/generalneurology/117705)

-- in the US there was an FDA warning in 2019 about serious breathing difficulties with gabapentinoids noting that "reports submitted to FDA and data from the medical literature show that serious breathing difficulties can occur when gabapentinoids are taken by patients with preexisting respiratory risk factors": https://www.fda.gov/media/133681/download?attachment

-- the current study did find that there was a significant increased risk of corticosteroid-requiring asthma exacerbations and hospitalizations for asthma exacerbations with gabapentinoids as compared with TCAs, and an increased risk but less so of corticosteroid-requiring asthma exacerbations compared with SNRI’s

-- subgroup analysis suggested that pregabalin and mirogabalin were associated with these increased risks while gabapentin was not

    -- before giving gabapentin a "bye" based on this statement, gabapentin is prescribed infrequently in Japan for neuropathic or chronic pain, but it is about 83% of the gabapentinoids prescribed in the US (https://pmc.ncbi.nlm.nih.gov/articles/PMC5838608/).

    -- it is also possible that the lack of association with gabapentin in this trial might be the lower and much more variable blood levels achieved than with pregabalin (which also suggests that if we do want to prescribe a gabapentinoid, pregabalin has a more predictable and reproducible blood level than gabapentin)

-- the authors propose several mechanisms to explain this, including sedation and hypoventilation, and the potential for gabapentinoids to lead to acute heart failure (see below), potentially confounding asthma with respiratory distress from heart failure

-- another study by the same group assessed gabapentinoid association with COPD, in a pre-publication article (https://doi.org/10.1513/AnnalsATS.202411-1230OC), for which i could only get access to the summary:

    -- they utilized the nationwide administrative claims database (as in asthma study) for this retrospective cohort study, identifying patients with COPD and neuropathic or chronic pain who initiated gabapentinoid treatment between 2015 and 2022

    -- they employed the active comparator new-user cohorts, one with tricyclic antidepressants (TCAs) and the other with serotonin-noradrenaline reuptake inhibitors (SNRIs), as in the above study

    -- patient characteristics were balanced using overlapping propensity score-weighting

    -- the primary outcome was the initial occurrence of COPD exacerbations requiring systemic corticosteroids

    -- as compared to TCAs, gabapentinoids were associated with a 21% higher incidence of the primary outcome (67.8 vs. 46.7 per 100 person-years; HR: 1.21, 1.03-1.42)

    -- as compared to SNRIs, gabapentinoids had an 18% higher incidence of the primary outcome (68.8 vs. 51.4 per 100 person-years; HR: 1.18, 1.10-1.28)

    -- their conclusion: "gabapentinoids may increase the risk of COPD exacerbations compared to other central nervous system-active medications at the same treatment stage for neuropathic or chronic pain, suggesting that their use should be limited to clearly beneficial cases"

-- the Annals of Internal Medicine (see gabapentinoids inc risk COPD exacerb AnnIntMed2024 in dropbox, or doi:10.7326/M23-0849) found similar results in a Canadian study of patients with COPD from 1994-2015 in a propensity score-matched new-user cohort study; this study assessed severe COPD exacerbations leading to hospitalization, comparing people starting  gabapentinoids for epilepsy, neuropathic pain, or other chronic pain vs nonusers, by gabapentinoid indication (they did not include patients on opioids or benzos, since these meds are associated with respiratory depression)

    -- compared with nonuse, gabapentinoid use was associated with an increased risk for severe COPD exacerbation across all of these indications:

        -- for epilepsy: 58% increase, HR 1.58 (1.08 to 2.30)

        -- for neuropathic pain, 35% increase, HR 1.35 (1.24 to 1.48)

        -- for other chronic pain, 49% increase, HR 1.49 (1.27 to 1.73)

        -- summation for all indications, 39% increase, HR 1.39 (1.29 to 1.50)

        -- in general, the increased risk of severe COPD exacerbation began within 4-6 months of starting the gabapentinoids and continued to increase over 4-5 years

        -- it should be noted that a placebo-controlled trial as is this one might have more biases that one with an active comparator

-- a French study found an association between gabapentinoid-induced peripheral edema (a well-documented side effect, found in 7 to 10% of users), quantifying the risk for acute heart failure (see gabapentinoids heart failure BoimedPharmacoth2022 in dropbox, or doi.org/10.1016/j.biopha.2022.112807)

    -- 58 reports were analyzed, gabapentin was used in 5 and pregabalin in 53 patients. Median age 77, four times as many women [a pretty small study]

    -- patients with non-cardiogenic edema had edema onset a median time of 23 days after starting the gabapentinoid

    -- patients with acute heart failure had onset a median time of 17 days after starting the gabapentinoid. heart failure only occurred in those on pregabalin (and of the 9 people, 3 had hypertension, 3 had prior heart failure and 2 had no known cardiovascular risk factors) and no person on gabapentin (but only 5 people were on gabapentin)

    -- both of these types of edema also increased after a gabapentinoid dose escalation and rapidly responded to a discontinuation of the gabapentinoid in a median of seven days

-- this current study adds some negatives to the use of gabapentinoids. there are several studies demonstrating the lack of positives, especially for chronic and neuropathic pains:

    -- eg, a blog on a recent article finding that gabapentinoids vs placebo did not improve pain control in patients with peripheral nerve injury,  https://gmodestmedblogs.blogspot.com/2025/04/pain-and-gabapentinoids-no-clear.html . this blog has links to several other blogs highlighting drug company shenanigans/malfeasance in studies leading the gabapentin approval, a meta-analysis of 9 studies finding that gabapentinoids did not help with low back or radicular pain, an RCT that gabapentin did not help with chronic pelvic pain, a systematic review that gabapentinoids did not help diabetic peripheral neuropathy, an increase in opioid death rate when opioids are used in combination with gabapentinoids, a study finding that gabapentinoids are associated with a 2- to 3-fold increase for substance misuse and overdoses, and the high likelihood (60%) of patients having adverse CNS adverse effects including dizziness and increased falls and fractures.

Limitations:

-- this study took place in Japan and may not be generalizable to other areas. for example, they did not include people younger than age 40. And gabapentin is used rarely for neuropathic or chronic pain there

-- the study was based on diagnostic codes that have not been validated, though the numbers of asthma-related deaths did agree with the national death certificate data

-- they did not include emergency department visits/outcomes, since this was not in their database (and a lot of people with asthma may well be evaluateed and treated there)

-- this was an observational study and with potentially unassessed confounders, limiting the accuracy of the results as  well as causality.

    -- for example, though they did take into account smoking, it was classified as current vs past/never. there are important differences that are obscured here (how much did people smoke? if they were past smokers who quit a few weeks ago, they would be in a different bucket than those who just started a few weeks ago. and no information about environmental or occupational exposures. or exercise. or stressors that might provoke asthma.

-- the median followup in the study was 60 days, with most follow-ups ending because of treatment discontinuations (though this might reflect the real-world situations, given the relatively high level of adverse effects)

-- it is a bit surprising that the benefit of TCAs was more than SNRIs. one factor is that there are almost no direct comparisons between these 2 meds. TCAs have been used for decades for neuropathic pain (and work quite well in my experience, especially for patients with diabetic neuropathies but also for other neuropathies) as well as an adjunct for chronic pain. TCAs are used now less often. And there is not much incentive by drug companies to do more studies given that both are generic meds. there was one trial i have seen published in 2025 finding similar response to these meds in women with chronic pelvic though amitriptyline was less well tolerated: https://onlinelibrary.wiley.com/doi/10.1111/luts.70023 . i have seen no studies of the more tolerable TCAs such as nortriptyline and desipramine

    -- and there is no comment on the doses of the meds taken, which could affect the effectiveness and the adverse effects of the meds, or which TCAs or SNRIs were taken (even in my scouring the supplementary material)

so,

-- it is pretty clear that gabapentinoids are inferior to other meds in terms of chronic pain and neuropathy, with reviews noted above suggesting that duloxetine and venlafaxine seem to be the best

-- the indications for gabapentinoids are mixed. some studies suggesting more benefit (and the Cochrane review was more positive about benefits in patients with postherpetic neuralgia and peripheral diabetic neuropathy: https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD007938.pub4/full ), though other studies suggesting zero benefit (as referenced above)

-- and there is the potential for increasing respiratory diseases (asthma and COPD, as above) as well as heart failure, all very common underlying conditions, especially in older people who are most likely to be prescribed gabapentinoids

-- and it is pretty clear that some meds (eg duloxetine) work very well for fibromyalgia, chronic pain and neuropathic pain

-- ?? are these additional studies putting another nail into the gabapentinoid coffin??? perhaps crenation??

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them