gabapentanoids plus opioids = higher death rate

The same research group has independently published articles finding that opioid-related deaths were higher with co-prescription of either pregabalin or gabapentin (gabapentinoids).

1. pregabalin (see opioid death with pregabalin AIM2018 in dropbox, or doi:10.7326/M18-1136 in dropbox, or doi:10.7326/M18-1136)

Details:

--population-based nested case-control study of patients aged 15-105 in Ontario with public drug coverage who received prescription opioids from 1997-2017
--1417 case patients (those who died from opioids) were matched to 5097 controls by age, history of CKD and Charlson comorbidity index score

--median age 28, 44% female, 12% residing in long-term care facility, 44% on SSRIs/50% other antidepressants/76% on benzos/35% other psychotropic meds/6% opioid-agonists, 24% alcohol use disorder, 4% CKD, 17% diabetes, 24% COPD

--opioid dose: 13% 1-19 MME/d, 23% 20-49 MME/d, 20% 50-99 MME/d, 17% 100-199 MME/d, 27% >=200 MME/d (MME=morphine milligram equivalents)

--case patients were more likely to have been prescribed other CNS depressants, receive more total meds annually (11 vs 9), and have more comorbidities than the controls

Results:
--after multivariate adjustment: concommitant exposure to pregabalin (script within the prior 120 days) with opioids led to a 68% increased opioid-related death​, adjusted odds ratio (aOR) 1.68 (1.19-2.36), n=69 cases

--other analyses:

    --taking pregabalin vs other CNS depressants (eg benzodiazepines or tricyclics): aOR 2.00 (1.39-2.88), with n=59

    ​--high dose pregabalin (>300mg/d): aOR 2.51 (1.24-5.06), with n=17 (statistically significant, but wide confidence intervals from the small number of cases)

    --low or moderate dose pregabalin (<=300mg/d): aOR 1.52 (1.04-2.22), with n=52

    --those with recent NSAID exposure: aOR 1.04 (0.90-1.19), with n=531 (ie, no increased death rate in those taking both opioids and NSAIDs together)

2. gabapentin (see https://journals.plos.org/plosmedicine/article/file?id=10.1371/journal.pmed.1002396&type=printable )

Details:

--population-based nested case-control study of patients aged 15-105 in Ontario with public drug coverage who received prescription opioids from 1997-2014
--1256 case patients (those who died from opioids) were matched to 4619 controls by age, history of CKD and Charlson comorbidity index score

--median age 48, 43% female, 46% of SSRIs/20% other antidepressants/77% on benzos/36% other psychotropic meds, 26% alcohol use disorder, 5% CKD, 16% diabetes, 81% anxiety or sleep disorders/20% affective disorders/14% psychoses/72% other psych disorders, COPD 23%

--opioid dose: 11% 1-19 MME/d, 18% 20-49 MME/d, 16% 50-99 MME/d, 15% 100-199 MME/d, 40% >=200 MME/d​ (the controls in general were on lower MMEs than the cases)

--46% of patients receiving gabapentin in 2013 had at least 1 concomitant prescription for an opioid

Results:
--after multivariate adjustment: concommitant exposure to gabapentin (script within the prior 120 days) with opioids led to a 49% increased opioid-related death​, adjusted odds ratio (aOR) 1.49 (1.18-1.88), p<0.001; n=155cases

--other analyses:

    ​--high dose gabapentin (>=1800 mg/d): aOR 1.58 (1.09-2.27), p=0.015, with n=57

    --moderate dose gabapentin (900-1799mg/d): aOR 1.56 (1.06-2.28), p<0.001, with n=57

    --low dose gabapentin (<900 mg/d): aOR 1.32 (0.89-1.97), with n=480 (non-significant)

    --those with recent NSAID exposure: aOR 1.14 (0.98-1.32), p=0.113, with n=531 (ie, no increased death rate in those taking both opioids and NSAIDs together)​

Commentary:
--gabapentinoids have been increasingly prescribed as an adjunct for the management of neuropathic and chronic pain syndromes, though recent analyses suggest that they are not terribly effective (see http://gmodestmedblogs.blogspot.com/2018/07/gabapentinoids-still-not-help-low-back.html , http://gmodestmedblogs.blogspot.com/2017/08/gabapentinoids-not-indicated-for.html , and http://gmodestmedblogs.blogspot.com/2017/04/diabetic-peripheral-neuropathy-and-more.html ​ ) . for example, more than 1/2 of Ontario residents beginning pregabalin therapy are concurrently on opioids

--the concern here, compounded by the finding that these drugs do not seem to have much benefit, is that the co-prescription of CNS depressants might well lead to a higher mortality rate, as has also been found with benzodiazepines (see https://blogs.bmj.com/bmjebmspotlight/2017/03/28/primary-care-corner-with-geoffrey-modest-md-opiates-and-benzos-assoc-with-inc-mortality/ ), though another study found less of an association (see https://blogs.bmj.com/bmjebmspotlight/2017/07/17/primary-care-corner-with-geoffrey-modest-md-benzos-may-not-increase-mortality-risk/​ )

    --there were a notably high number of patients concomitantly on opiates and benzos, and quite high opioid doses were prescribed (would also be good to know if there were a dose-response curve here: was their higher mortality with increasing MMEs?)

    --and, surprisingly in the above study, those on pregabalin seemed to have twice the mortality as those on other CNS depressants (benzos and tricyclics, though it is quite likely that the risk of benzos and tricyclics is the same; and that data not provided)

--there may also be another potentiator: a study found a 44% increased in systemic gabapentin exposure after the administration of morphine (?increased gabapentin absorption from opiate-induced decreased intestinal motility). The finding of increased mortality with increasing doses of gabapentinoids is consistent with this finding

--the lack of association of opioids plus NSAIDs and increased mortality suggests that there is a fundamental difference between using NSAIDs vs gabapentinoids as adjunctive therapy to opioids in those with pain not adequately treated with opiates alone. (The NSAID comparison was a surrogate for pain not controlled sufficiently by opioids alone).

--of course, these are observational studies: perhaps those prescribed gabapentinoids had more uncontrolled pain and were more likely to intentionally overdose?? And more pain than NSAIDs could control?? (we have no data on the level of pain control in the different groups), or there were other unaccounted-for biases (they did not control for many potential confounders in the studies above, details about the pain syndromes, length of pain symptoms, sites of pain, use of other adjuncts to pain control, more detailed psych/support issues…..)

so, these studies do raise the pretty likely specter that gabapentinoids increase mortality in those on opiates. This, combined with the above-mentioned studies showing no clear benefit of gabapentinoids in chronic pain or neuropathic pain syndromes, should make us reconsider whether these meds should be given as adjuncts to opiates in patients with otherwise insufficiently responding pain syndromes (ie, harm without benefit is not what the doctor ordered...or, perhaps, should not be)