Featured Articles: Original Clinical Research Report

Day-of-Surgery Gabapentinoids and Prolonged Opioid Use: A Retrospective Cohort Study of Medicare Patients Using Electronic Health Records

Young, Jessica C. PhD^*,†; Dasgupta, Nabarun PhD^‡; Chidgey, Brooke A. MD^§; Stürmer, Til MD, PhD^*; Pate, Virginia MS^*; Hudgens, Michael PhD^‖; Jonsson Funk, Michele PhD^*

Author Information

From the ^*Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina

^†Cecil G. Sheps Center for Health Services Research, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina

^‡Injury Prevention Research Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina

^§Department of Anesthesiology and Pain Management, University of North Carolina School of Medicine, Chapel Hill, North Carolina

^‖Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina.

Published ahead of print July 14, 2021.

Accepted for publication May 28, 2021.

Funding: J.C. Young received tuition, research, and stipend support from National Institutes of Health (NIH)/National Institute on Drug Abuse (NIDA) R36 DA04588501 (principal investigator [PI]: J.C.Y.). This research was also supported by the National Center for Advancing Translational Sciences (NCATS), NIH, through Grant Award Number UL1TR002489, and by a National Research Service Award Predoctoral/Postdoctoral Traineeship from the Agency for Healthcare Research and Quality sponsored by The Cecil G. Sheps Center for Health Services Research, The University of North Carolina at Chapel Hill, Grant No. T32-HS000032. N.D. is supported by the US Food and Drug Administration (FDA; HHSF223201810183C). N.D. does not accept personal compensation of any kind from any pharmaceutical company. T.S. receives investigator-initiated research funding and support as PI (R01 AG056479) from the National Institute on Aging (NIA), and as Coinvestigator (R01 HL118255 and R01MD011680), NIH. He also receives salary support as Director of Comparative Effectiveness Research (CER), NC TraCS Institute, University of North Carolina (UNC) Clinical and Translational Science Award (UL1TR002489), and the Center for Pharmacoepidemiology (current members: GlaxoSmithKline, UCB BioSciences, Takeda, AbbVie, and Boehringer Ingelheim), from pharmaceutical companies (Novo Nordisk), and from a generous contribution from Dr. Nancy A. Dreyer to the Department of Epidemiology, University of North Carolina at Chapel Hill. M.J.F. receives investigator-initiated research funding support as PI from the US FDA (75F40119C10115) and the National Heart, Lung, and Blood Institute (NHLBI: R01 HL118255), and as Coinvestigator from the FDA (CER-2017C3-9230), Centers for Disease Control and Prevention (CDC: 1U01DP006369-01), NCATS (1 U54 TR002255), NIA (1R01AG056479), and Health Resources & Services Administration (HRSA: R40MC29455-01-00). She additionally receives salary support from the Center for Pharmacoepidemiology in the Department of Epidemiology, University of North Carolina (members: GlaxoSmithKline, UCB BioSciences, Merck [past member], Takeda, AbbVie, Boehringer and Ingelheim), and consulting fees via University of North Carolina from GlaxoSmithKline. The database infrastructure used for this project was funded by the Pharmacoepidemiology Gillings Innovation Lab (PEGIL) for the Population-Based Evaluation of Drug Benefits and Harms in Older US Adults (GIL200811.0010), the Center for Pharmacoepidemiology, Department of Epidemiology, University of North Carolina Gillings School of Global Public Health, the CER Strategic Initiative of University of North Carolina’s Clinical Translational Science Award (UL1TR002489), the Cecil G. Sheps Center for Health Services Research, University of North Carolina, and the University of North Carolina School of Medicine.

Conflicts of Interest: See Disclosures at the end of the article.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website.

Reprints will not be available from the authors.

Address correspondence to Jessica C. Young, PhD, Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, 135 Dauer Dr, Chapel Hill, NC 27599. Address e-mail to [email protected].

Anesthesia & Analgesia 133(5):p 1119-1128, November 2021. | DOI: 10.1213/ANE.0000000000005656

Free
SDC
Corresponding Article

Abstract

KEY POINTS

Question: Is preoperative gabapentinoid administration associated with a reduction in prolonged opioid use following surgery?
Findings: In a cohort study of 13,958 patients, preoperative gabapentinoid administration was not associated with a reduced risk of prolonged opioid use (adjusted risk ratio [adjRR], 1.19 [95% confidence interval [CI], 0.67–2.12]); given the limited sample size, the estimate was imprecise, with a wide CI ranging from a 33% reduction to a 212% increase in risk, suggesting potential for a substantial increase in risk of prolonged opioid use.
Meaning: The off-label use of preoperative gabapentinoids for surgical pain should be carefully evaluated, as this study did not find an association between prolonged opioid use and preoperative gabapentin.

See Article, p 1116

As the use of multimodal analgesic techniques increases in response to the opioid crisis, it is important to investigate the role and appropriateness of gabapentinoids in surgical pain management. Between 2006 and 2007, 3 systematic reviews evaluating perioperative gabapentinoids and postoperative pain were published, contributing to widespread acceptance that gabapentinoids could help reduce pain and opioid consumption in the immediate postoperative period.^{^1–3}

More recently, a meta-analysis by Verret et al^⁴ found that perioperative use of gabapentinoids was not associated with a meaningful reduction in acute, subacute, or chronic pain. In December 2019, the US Food and Drug Administration (FDA) issued a safety communication warning of respiratory depression when gabapentinoids are used concurrently with central nervous system depressants such as opioids. This communication also discussed growing rates of gabapentinoid misuse and abuse.^⁵ Despite the fact that gabapentinoid use for preoperative pain is not approved by the FDA, clinical guidelines increasingly recommend use of preoperative gabapentinoids as a component of multimodal analgesia at surgery.^{^6–8} Likewise, as opioid-sparing techniques gained in popularity,^⁹^,^¹⁰ hospital protocols increasingly added off-label gabapentinoid use to surgical protocols.^{^11–13}

It remains unknown to what extent off-label use of gabapentinoids in perioperative pain management can safely reduce opioid consumption and long-term risks of opioid use in different surgical settings and patient populations.^¹⁴ To address this knowledge gap, this study used a large surgical cohort of Medicare patients from an integrated health system to test the hypothesis that preoperative day-of-surgery gabapentinoid administration is associated with reduced prolonged opioid use following major surgical procedures.

METHODS

This study was approved under the University of North Carolina Institutional Review Board (IRB) 18-1248, and the requirement for written informed consent of these retrospective data was waived by the IRB. This article adheres to the applicable STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) guidelines.

Data Source

Electronic health records (EHRs) dating from April 4, 2014 to December 16, 2019 from a large integrated health care system in the United States were used. The EHR contains detailed clinical and administrative data for patient care provided across 11 hospitals and over 700 clinics. These data provide an in-depth view of medical encounters, including longitudinal data on diagnosis and procedure codes from any encounter. Relevant to the current study, these data include date and timestamp for start of surgery, surgical procedure code, preoperative pain scores, outpatient medication orders, and inpatient medication administrations. The EHRS also include demographic data including height, weight, race, and ethnicity, in addition to self-reported alcohol and tobacco use.

Study Population

Patients undergoing major therapeutic surgical procedures (nonocular) between January 1, 2016 and September 16, 2019 within the 2 main surgical facilities in the integrated health care system were identified.^¹⁵ Both inpatient and outpatient surgeries were included. For patients undergoing multiple surgeries, only the index surgery was examined. Inpatient surgeries were limited to those with a total length of stay of ≤4 nights, with patients discharged home for self-care.

Patients with any outpatient medication orders for gabapentinoids or diagnoses of epilepsy or postherpetic neuralgia before surgery were excluded. Patients with a documented history of opioid abuse, addiction, or dependence, or who had evidence of prolonged opioid use (opioid orders in 3 consecutive months) at any time in the 12 months before surgery were excluded. To assess prolonged opioid use, patients were required to have at least 90 days of follow-up after discharge from surgery. Patients who underwent additional surgical procedures, died, or disenrolled from the database during the 90-day follow-up were excluded. The impact of this exclusion on each cohort is reported.

Exposure

The primary exposure was preoperative gabapentinoids, defined using inpatient medication administration records. Administration records for oral gabapentinoids must have been on the day of surgery with an administration start time stamp before the start of the surgery, with a description of “GIVEN.” We identified whether pregabalin or gabapentin was administered and the dosage in milligrams.

Outcome

We examined the proportion of patients with prolonged opioid use following surgery. Prolonged opioid use was defined as at least 1 outpatient opioid order in each of 3 consecutive 30-day windows immediately following surgical discharge.^¹⁶

Potential Confounding Variables

We reported and adjusted for demographic factors that have been found to be associated with health care delivery and opioid-prescribing practices, including patient gender, age, and patient-reported race (Black, White, and other). Because use of preoperative gabapentinoids and rates of opioid prescribing have changed over calendar time and by institution, we controlled for calendar time using 6-month increments and for medical facility of surgery. To address baseline health imbalances, we adjusted for maximum recorded preoperative pain (0, 1–3, 4–6, and 7+), number of outpatient prescriptions in the previous 6 months (0, 1–6, and 7+), patient-reported smoking history (current smoker, former smoker, never smoker, and other), patient-reported alcohol use (yes versus no), and body mass index (BMI) categorized according to the US Centers for Disease Control and Prevention. We also adjusted for pain-related medications and diagnoses using binary variables indicating the presence of prescriptions for opioids, benzodiazepines, and pain-related diagnoses (arthritis, cancer, depression, chronic back pain, fibromyalgia, neuralgia, headache/migraine, and abdominal pain) at baseline.

Our main analyses controlled for calendar time in 6-month intervals. Because calendar time may be an important confounding variable, we conducted 2 additional analyses using different specifications to account for calendar time. The first modeled surgery date as a continuous variable (number of days from the start of the study period) using a quadratic term, and the second used a cubic spline with 3 knots at the tenth, 50th, and 90th percentiles.^¹⁷ Because the health care system in this study first implemented Enhanced Recovery After Surgery (ERAS) protocols including preoperative gabapentinoid recommendations on March 1, 2018, we also conducted analyses examining whether the association between preoperative gabapentinoids and prolonged opioid use differed in the period before any ERAS protocols (surgery between January 1, 2016 and February 28, 2018) and after the implementation of ERAS protocols (surgery between March 1, 2018 and September 16, 2019). We assessed the interaction between the exposure and period and conducted stratified analyses by period.

Statistical Analyses

Because of the rapidly evolving landscape surrounding opioid prescribing and pain management, we report the percentage of patients receiving preoperative gabapentinoids and having prolonged opioid use following surgery by 6-month intervals based on the date of surgery.

Crude and adjusted risk and adjusted risk ratios (adjRR) with 95% confidence intervals (CIs) of prolonged opioid use in the exposed (received gabapentinoids on the day of surgery) and unexposed (no gabapentinoids on day of surgery) were calculated using log-binomial regression.

Logistic regression adjusting for the potential confounders detailed above was used to calculate propensity scores predicting administration of preoperative gabapentinoids. Adjusted estimates were calculated using stabilized inverse probability of treatment weights (IPTWs) to reduce bias due to measured confounders.^¹⁸ Exposed subjects received a weight of (1/ps) ×p, where ps represents the predicted probability of exposure to gabapentinoids and p represents the proportion of patients observed as treated with gabapentinoids. Unexposed patients received a weight of (1 − p)/(1 − ps).^¹⁹ Asymmetric trimming at the first and 99th percentiles of the propensity score was used to define a study population with greater treatment equipoise resulting in more clinically relevant estimates.^²⁰^,^²¹ IPTWs were recalculated among the trimmed population, and balance between the weighted groups was assessed using absolute standardized mean differences (ASMDs), with ASMD <0.1 indicating balance.^²² Separate propensity score models were fit for the main, secondary, and sensitivity analyses described below, and asymmetric trimming with stabilized IPTW was repeated within each analysis. Due to extreme weights, 1.5% asymmetric trimming was used in the calendar time–stratified analyses.

Main Analysis

The main analysis was conducted in the total population of surgeries meeting the inclusion criteria.

Secondary Analysis

A secondary analysis was conducted on a subset of 4 surgical procedures (colorectal resection, hip arthroplasty, knee arthroplasty, and hysterectomy), for which at least 30% of all patients received a preoperative gabapentinoid, in an effort to focus on a clinical population with higher equipoise for whom preoperative gabapentinoids appeared to be a more common part of care. Results in the population undergoing surgical procedures in which <30% of patients received preoperative gabapentinoids are also presented in the Supplemental Digital Content, Supplemental Materials, https://links.lww.com/AA/D593. Due to small sample size, the secondary analysis controlled for calendar time in 1-year increments instead of 6-month increments.

Sensitivity Analysis

Because we observed only the health care and medications received within the health care system from which the EHRs were extracted, we conducted a sensitivity analysis restricting the population to patients with at least 1 outpatient visit and 1 outpatient medication order in the health care system in the 182 days before surgery. This subset of patients represents a group that has a more regular history of interaction with this health care system, for whom we have higher confidence that baseline and follow-up care will be captured in the data. Results in the population who did not meet these criteria are also presented in the Supplemental Digital Content, Supplemental Materials, https://links.lww.com/AA/D593. Due to small sample size, the sensitivity analysis controlled for calendar time in 1-year increments instead of 6-month increments.

Quantitative Bias Analyses

Quantitative bias analyses estimate the impact systematic error (such as outcome misclassification) may have on effect estimates. To examine the potential bias due to imperfect capture of opioid prescriptions during follow-up, we linked a subset of the patients to insurance claims data and conducted 2 sets of quantitative bias analyses. The first analysis addressed the potential of underestimating prolonged opioid use and assumed that any patients with prolonged opioid use in either EHR or Medicare claims data were correctly classified as having prolonged opioid use (“gold standard” was the combined EHR and claims data). The second analysis addressed the potential of overestimating prolonged opioid use in the EHR and treated the Medicare claims data as the “gold standard”.^²³^,^²⁴

For both bias analyses, we estimated the positive predictive value (PPV) and negative predictive value (NPV) of prolonged opioid use measured by the EHR data. Using these estimates and corresponding estimated standard errors (SEs), we conducted a probabilistic bias analysis that reclassifies the data to present bias-adjusted risk ratios (adjRRs) incorporating uncertainty in the measurement of the outcome as well as random error. We resampled the population with replacement to create 1000 pseudopopulations. For each iteration, the NPV and PPV were randomly drawn from normal distributions with mean and standard deviation equal to the estimated NPV and PPV and the corresponding SEs. The mean risk ratio over the 1000 iterations is reported as the bias-adjRR, and the 2.5th and 97.5th percentiles of the 1000 iterations were reported as the 95% CI.^²³

All analyses were conducted using SAS 9.4.

RESULTS

We identified 13,958 Medicare patients undergoing major therapeutic surgical procedures between January 1, 2016 and September 16, 2019 who met study inclusion criteria and had 90 days of follow-up (Figure 1; Supplemental Digital Content, Table 1, https://links.lww.com/AA/D593). Overall, 4.2% (exposed) and 6.0% (unexposed) of patients were excluded for having <90 days of follow-up. The majority of patients (94.6% [exposed] and 93.0% [unexposed]), were excluded because they had surgery during the 90-day follow-up. In both cohorts, <0.4% of patients died during the 90-day follow-up (Supplemental Digital Content, Table 2, https://links.lww.com/AA/D593).

Figure 1.:
Flow diagram describing cohort inclusion criteria.

Table 1. - Baseline Characteristics for Patients Undergoing surgery, Stratified by Receipt of Preoperative Gabapentinoids

Characteristic	Observed patients before IPTW			After IPTW with 1% asymmetric trimming
	No preoperative gabapentinoid	Preoperative gabapentinoid	ASMD	No preoperative gabapentinoid	Preoperative gabapentinoid	ASMD
	n = 11,027	n = 2931	ASMD	n = 8812	n = 2491	ASMD
Female	6216 (56.4%)	1844 (62.9%)	0.13	5278 (59.9%)	1384 (55.5%)	0.09
Age at admission, mean (SD)	72.9 (6.16%)	71.7 (5.40%)	0.22	72.4 (5.85%)	72.6 (5.82%)	0.03
Patient race
White	9021 (81.8%)	2455 (83.8%)	0.06	7280 (82.6%)	2074 (83.3%)	0.06
Black	1419 (12.9%)	324 (11.1%)		1044 (11.9%)	282 (11.3%)
Other	587 (5.3%)	152 (5.2%)		488 (5.5%)	136 (5.4%)
Date of surgery
January 2016 to June 2016	1695 (15.4%)	136 (4.6%)	0.51	638 (7.2%)	157 (6.3%)	0.08
July 2016 to December 2016	1530 (13.9%)	201 (6.9%)		901 (10.2%)	267 (10.7%)
January 2017 to June 2017	1545 (14.0%)	321 (11.0%)		1209 (13.7%)	348 (14.0%)
July 2017 to December 2017	1395 (12.7%)	374 (12.8%)		1261 (14.3%)	322 (12.9%)
January 2018 to June 2018	1407 (12.8%)	416 (14.2%)		1329 (15.1%)	369 (14.8%)
July 2018 to December 2018	1367 (12.4%)	617 (21.1%)		1365 (15.5%)	372 (14.9%)
January 2019 to June 2019	1517 (13.8%)	600 (20.5%)		1545 (17.5%)	498 (20.0%)
July 2019 to September 2019	571 (5.2%)	266 (9.1%)		563 (6.4%)	159 (6.4%)
Location
Facility 1	7818 (70.9%)	1632 (55.7%)	0.32	5476 (62.1%)	1562 (62.7%)	0.04
Facility 2	3209 (29.1%)	1299 (44.3%)	0.32	3336 (37.9%)	929 (37.3%)	0.04
Patient BMI
Missing	69 (0.6%)	21 (0.7%)	0.10	60 (0.7%)	17 (0.7%)	0.05
Low	128 (1.2%)	24 (0.8%)		100 (1.1%)	33 (1.3%)
Optimal	2872 (26.0%)	649 (22.1%)		2170 (24.6%)	579 (23.2%)
Overweight	3975 (36.0%)	1078 (36.8%)		3197 (36.3%)	891 (35.8%)
Obese	3983 (36.1%)	1159 (39.5%)		3284 (37.3%)	972 (39.0%)
Maximum presurgical pain recorded
Missing	34 (0.3%)	4 (0.1%)	0.18	20 (0.2%)	7 (0.2%)	0.04
0	8373 (75.9%)	1712 (58.4%)		6287 (71.3%)	1773 (71.1%)
1–3	996 (9.0%)	388 (13.2%)		930 (10.6%)	281 (11.3%)
4–6	1075 (9.7%)	628 (21.4%)		1088 (12.3%)	303 (12.2%)
7+	549 (5.0%)	199 (6.8%)		487 (5.5%)	128 (5.1%)
Alcohol use	4053 (36.8%)	1195 (40.8%)	0.08	3247 (36.8%)	942 (37.8%)	0.02
Smoking status
Current	614 (5.6%)	102 (3.5%)	0.21	429 (4.9%)	144 (5.8%)	0.09
Former	4256 (38.6%)	1036 (35.3%)		3240 (36.8%)	1001 (40.2%)
Other	1329 (12.1%)	541 (18.5%)		1403 (15.9%)	404 (16.2%)
Never	4828 (43.8%)	1252 (42.7%)		3740 (42.4%)	943 (37.8%)
Baseline outpatient medication orders
No. of unique orders
0	3797 (34.4%)	1072 (36.6%)	0.06	3002 (34.1%)	790 (31.7%)	0.05
1–6	4511 (40.9%)	1142 (39.0%)		3618 (41.1%)	1022 (41.0%)
7+	2719 (24.7%)	717 (24.5%)		2192 (24.9%)	680 (27.3%)
Opioids	1370 (12.4%)	358 (12.2%)	0.01	1171 (13.3%)	413 (16.6%)	0.09
Benzodiazepines	1268 (11.5%)	301 (10.3%)	0.04	1028 (11.7%)	271 (10.9%)	0.03
Baseline health conditions
Recent outpatient visit and outpatient medication order	3956 (35.9%)	1128 (38.5%)	0.05	3399 (38.6%)	970 (39.0%)	0.01
Arthritis	2003 (18.2%)	731 (24.9%)	0.17	1781 (20.2%)	574 (23.0%)	0.07
Cancer	2166 (19.6%)	659 (22.5%)	0.07	1883 (21.4%)	541 (21.7%)	0.01
Depression	456 (4.1%)	164 (5.6%)	0.07	413 (4.7%)	116 (4.7%)	0.00
Chronic back pain	1980 (18.0%)	484 (16.5%)	0.04	1612 (18.3%)	491 (19.7%)	0.04
Fibromyalgia	176 (1.6%)	47 (1.6%)	0.00	144 (1.6%)	42 (1.7%)	0.00
Neuralgia	403 (3.7%)	94 (3.2%)	0.03	325 (3.7%)	88 (3.5%)	0.01
Headache/migraine	454 (4.1%)	141 (4.8%)	0.03	387 (4.4%)	113 (4.5%)	0.01
Abdominal pain	1195 (10.8%)	299 (10.2%)	0.02	914 (10.4%)	284 (11.4%)	0.03
Surgical procedure
Knee arthroplasty	505 (4.6%)	875 (29.9%)	0.71	925 (10.5%)	268 (10.8%)	0.01
Lumpectomy	954 (8.7%)	113 (3.9%)	0.20	672 (7.6%)	92 (3.7%)	0.17
Inguinal/femoral hernia repair	769 (7.0%)	39 (1.3%)	0.29	347 (3.9%)	71 (2.9%)	0.06
Laminectomy	679 (6.2%)	49 (1.7%)	0.23	480 (5.4%)	158 (6.3%)	0.04
Hip arthroplasty	194 (1.8%)	393 (13.4%)	0.45	351 (4.0%)	101 (4.1%)	0.01
Laparoscopic cholecystectomy	504 (4.6%)	19 (0.6%)	0.25	195 (2.2%)	53 (2.1%)	0.01
Prostate surgery	395 (3.6%)	8 (0.3%)	0.24	74 (0.8%)	18 (0.7%)	0.01
Colorectal resection	172 (1.6%)	227 (7.7%)	0.30	244 (2.8%)	71 (2.8%)	0.00
Hysterectomy	146 (1.3%)	75 (2.6%)	0.09	179 (2.0%)	49 (2.0%)	0.00
Other procedure^a	6709 (60.8%)	1133 (38.7%)	0.46	5346 (60.7%)	1610 (64.6%)	0.08

Abbreviations: ASMD, absolute standardized mean difference; BMI, body mass index; IPTW, inverse probability of treatment weights; SD, standard deviation.

^aOther procedures include other hernia repair, thyroidectomy, endocrine procedures, shoulder arthroplasty, muscle tendon procedures, heart valve procedures, endarterectomy, spinal fusion, rotator cuff repair, mastectomy, and other procedures occurring in <1.5% of patients.

The mean age of eligible patients was 72.7 years (SD, 6.0), 57.7% were women, 82.2% reported White race, and 12.4% had opioid orders in the previous 182 days (Table 1). Overall, 21.0% had received preoperative gabapentinoid administration on the day of surgery (84.5% of which was pregabalin). The median dose for pregabalin was 100 mg (interquartile range [IQR], 50–100 mg), while the median dose for gabapentin was 300 mg (IQR, 300–600 mg).

Calendar Trends

The proportion of patients who received preoperative gabapentinoid administration on the day of surgery increased from 7.4% in the first half of 2016 to 28.3% in the first half of 2019 (Table 2).

Table 2. - Percentage of Patients Exposed to Preoperative Day-of-Surgery Gabapentinoids and Observed to Have Prolonged Opioid Use by Time Period

Time period	Total no. of patients	Percentage exposed (95% CI)	Percentage with prolonged opioid use (95% CI)
Time period	Total no. of patients	Percentage exposed (95% CI)	Overall	Unexposed	Exposed
January 2016 to June 2016	1878	7.4 (6.2–8.6)	1.4 (0.9–2.0)	1.2 (0.7–1.8)	3.7 (0.5–6.8)
July 2016 to December 2016	2310	11.6 (10.1–13.1)	0.9 (0.4–1.3)	0.5 (0.2–0.9)	3.5 (0.9–6.0)
January 2017 to June 2017	2390	17.2 (15.5–18.9)	0.9 (0.5–1.3)	0.9 (0.4–1.4)	0.9 (0.0–2.0)
July 2017 to December 2017	2248	21.1 (19.2–23.0)	1.0 (0.5–1.4)	0.7 (0.3–1.2)	1.9 (0.5–3.2)
January 2018 to June 2018	2300	22.8 (20.9–24.7)	1.0 (0.6–1.5)	1.1 (0.5–1.6)	1.0 (0.0–1.9)
July 2018 to December 2018	2484	31.1 (29.1–33.1)	0.9 (0.5–1.3)	0.8 (0.3–1.3)	1.1 (0.3–2.0)
January 2019 to June 2019	2735	28.3 (26.4–30.3)	0.6 (0.3–0.9)	0.7 (0.3–1.2)	0.3 (0.0–0.8)
July 2019 to September 2019	1090	31.8 (28.6–34.9)	0.2 (0.0–0.6)	0.4 (0.0–0.8)	0.0 (0.0–1.1)
Cochran-Armitage test for trend		p < .0001	p = .0072	p = .2167	p < .0001

Abbreviation: CI, confidence interval.

The observed risk of prolonged opioid use decreased throughout the study period from 1.4 (95% CI, 0.9–2.0) in the first half of 2016 to 0.6% (95% CI, 0.3–0.9) in the first half of 2019. This decreasing trend in prolonged opioid use was larger in patients who received preoperative gabapentinoids, decreasing from 3.7% (95% CI, 0.5–6.8) to 0.3% (95% CI, 0.0–0.8).

Main Analysis

The observed 90-day risk of prolonged opioid use across the study period was 0.91% (95% CI, 0.77–1.08) (Table 3).

Table 3. - Crude and Adjusted Risk and Risk Ratios of Prolonged Opioid Use Comparing Patients Who Received Preoperative Gabapentinoids to Those Who Did Not (Referent)

Population	Exposure status	No. of patients	Observed risk (95% CI)	No. of patients after trimming	Adjusted^a risk (95% CI)	Adjusted^a risk ratio (95% CI)
Full population	Overall	13,958	0.91% (0.77–1.08)	11,303	0.87% (0.70–1.09)	1.19% (0.67–2.12)
	Gabapentinoid = 1	2931	1.19% (0.86–1.66)	2491	1.00% (0.59–1.68)
	Gabapentinoid = 0	11,027	0.83% (0.68–1.02)	8812	0.84% (0.66–1.06)
Surgeries with >30% exposed	Overall	2626	0.62% (0.38–1.01)	2410	0.70% (0.41–1.21)	1.01% (0.30–3.33)
	Gabapentinoid = 1	1587	0.76% (0.44–1.34)	1481	0.70% (0.39–1.28)
	Gabapentinoid = 0	1039	0.39% (0.15–1.05)	929	0.70% (0.25–1.98)
Outpatient visit and prescription in previous 182 d	Overall	5084	1.63% (1.32–2.02)	4192	1.85% (1.42–2.40)	1.06% (0.57–1.99)
	Gabapentinoid = 1	1128	1.86% (1.22–2.84)	927	1.94% (1.11–3.36)
	Gabapentinoid = 0	3956	1.57% (1.22–2.01)	3264	1.82% (1.35–2.45)

Abbreviations: CI, confidence interval; IPTW, inverse probability of treatment weights.

^aStabilized IPTWs were calculated within each population after conducting 1% asymmetric trimming to adjust for baseline confounding.

Following IPTW adjustment (Supplemental Digital Content, Figure 1, https://links.lww.com/AA/D593), the estimated risk of prolonged opioid use after receiving preoperative gabapentinoids was 1.00% (95% CI, 0.59–1.68). In contrast, in the absence of receipt of preoperative gabapentinoids, the estimated risk of prolonged opioid use was 0.84% (95% CI, 0.66–1.06). Patients receiving preoperative gabapentinoids were not found to be at higher risk of prolonged opioid use compared to those who did not receive gabapentinoids, with an estimated risk ratio of 1.19 (95% CI, 0.67–2.12; Figure 2; Table 3).

Figure 2.:
Crude and adjusted risk ratios of prolonged opioid use comparing patients who received preoperative gabapentinoids to those who did not receive preoperative gabapentinoids. CI indicates confidence interval; IPTW, inverse probability of treatment weight.

Models adjusting for calendar time as a continuous variable using a quadratic term and cubic splines had similar results and are presented in Supplemental Digital Content, Table 3, https://links.lww.com/AA/D593. Analysis of interaction between exposure and period (pre-ERAS versus post-ERAS) was not significant (P = .20), and results from stratified analyses are presented in Supplemental Digital Content, Table 4, https://links.lww.com/AA/D593.

Secondary Analysis

There were 4 distinct procedures in which over 30% of patients received preoperative gabapentinoids: colorectal resection, hip arthroplasty, knee arthroplasty, and hysterectomy. The analysis subset of patients undergoing one of these procedures included 2587 patients (Supplemental Digital Content, Table 5, https://links.lww.com/AA/D593). After adjustment via IPTW (Supplemental Digital Content, Figure 2, https://links.lww.com/AA/D593), patients who received preoperative gabapentinoids were at 1.01 (95% CI, 0.30–3.33) times the risk of prolonged use compared to those who did not (Figure 2; Table 3). Among those undergoing surgical procedures for which <30% of patients received perioperative gabapentinoids, exposed patients were at higher risk of prolonged opioid use, with an estimated risk ratio of 2.34 (95% CI, 1.02–5.41) (Supplemental Digital Content, Table 4, https://links.lww.com/AA/D593).

Sensitivity Analyses

Overall, 5084 patients had at least 1 outpatient visit and 1 outpatient medication order in the 182 days before surgery, and 22.2% received preoperative gabapentinoids (Supplemental Digital Content, Table 6, https://links.lww.com/AA/D593). After adjustment (Supplemental Digital Content, Figure 3, https://links.lww.com/AA/D593), patients who received preoperative gabapentinoids were at 1.06 (95% CI, 0.57–1.99) times the risk of prolonged use compared to those who did not receive preoperative gabapentinoids (Figure 2; Table 3). Among those without at least 1 outpatient visit and 1 outpatient medication order in the 182 days before surgery, patients who received preoperative gabapentinoids were at 1.73 (0.57–5.30) times the risk of prolonged opioid use compared to those who did not (Supplemental Digital Content, Table 4, https://links.lww.com/AA/D593).

Bias Analyses

A subset of the population (n = 3446) was linked to Medicare insurance claims. The first quantitative bias analysis adjusted for potential underestimation of prolonged opioid use in the EHR (combined EHR and Medicare data used as the gold standard). The bias-adjRR for the risk of prolonged use comparing patients who received preoperative gabapentinoids to those who did not was 1.45 (95% CI, 0.78–2.17). A second quantitative bias analysis using Medicare insurance claims as the gold standard resulted in a bias-adjRR of 1.48 (95% CI, 0.78–2.27; Supplemental Digital Content, Table 7, https://links.lww.com/AA/D593).

DISCUSSION

Our analysis of preoperative gabapentinoid exposure in a cohort of Medicare patients undergoing major therapeutic surgical procedures at a large integrated health care delivery system found that 21.0% of patients were administered preoperative gabapentinoids. The observed risk of prolonged opioid use during the 90 days after surgery was relatively low, at 0.91%. The surgeries with the highest proportion of patients receiving preoperative gabapentinoids were colorectal resection, hip arthroplasty, knee arthroplasty, and hysterectomy.

Use of preoperative gabapentinoids increased throughout the study period, while the observed risk of prolonged opioid use decreased, suggesting that prescribing behaviors for perioperative pain management changed throughout the study period. The decreasing trend of prolonged opioid use was more dramatic in the gabapentinoid-exposed group compared to those who did not receive preoperative gabapentinoids. However, following IPTW adjustment, we did not find that gabapentinoids were associated with a reduced the risk of prolonged opioid use. Given the wide CI (adjRR, 1.19 [95% CI, 0.67–2.12]), neither a protective nor a harmful effect can be ruled out.

While many past studies have found a reduction in opioid consumption in the first 24–48 hours following surgery among patients who receive preoperative gabapentinoids, a recent systematic review found no clinically significant analgesic effect for perioperative gabapentinoid use.^⁴ Our study also did not find that preoperative gabapentinoids were associated with a reduction in the risk of postsurgical prolonged opioid use. A randomized controlled trial (RCT) conducted among patients undergoing a similar mix of surgeries found that perioperative gabapentin promoted opioid cessation (HR, 1.24 [95% CI, 1.00–1.54]). While we focused on gabapentinoids administered on the day of surgery, the RCT continued gabapentin administration for 72 hours following surgery. The mean age of patients in the RCT was also 16 years younger than patients in the current study (56.7 vs 72.7), and opioid cessation was based on self-report instead of prescription data. Further research into the potential impact of postsurgical gabapentinoids on the safety and efficacy of gabapentinoid use on opioid consumption following surgery is warranted.

Making causal inference in nonrandomized settings requires the assumption of no uncontrolled founding, an assumption that is impossible to verify and difficult to obtain in practice. While we controlled for potential confounding variables using propensity score methods, there likely remains unmeasured confounding that was not accounted for in these analyses. We were unable to measure the dosage of opioids administered perioperatively due to limitations in data availability for intravenously administered medications during surgery. Preoperative opioid use could be a proxy for preoperative pain and may also play a role in the amount of opioids prescribed postoperatively. We were unable to assess and account for potential differences in preoperative opioid administration. Unmeasured confounding may also be present due to changes in practice and increased caution with opioid prescribing during this study period. We controlled for time trends using 6-month increments as well as additional dates when ERAS protocols were put into place; however, it is possible that there remains residual confounding by elements associated with calendar time. While we conducted a stratified analysis splitting the surgeries into 2 time periods, a larger sample size allowing for more granular stratifications of calendar time and other factors for which the association may differ, such as surgical procedure, would be informative in future work.

This study used EHR data from a large integrated health care system. These data provide clinical details including inpatient medication orders; preoperative pain scores; patient status on admission; and patient details including BMI, smoking, and alcohol history, which are often unavailable in large population-based epidemiologic studies. However, the current data include only information for care provided and medication orders from a single health care system (which could contribute to underestimation of opioid use) and do not include pharmacy fulfillment information (which could contribute to overestimation of opioid use). To address this, we linked a subset of the cohort to Medicare claims data, conducted probabilistic bias analyses addressing potential misclassification of prolonged opioid use in the EHR data, and found that estimates remained above the null.

We also required that patients had 90 days of follow-up after surgery. Overall, 0.3% of patients died within 90 days after surgery, and examination of medical records found no evidence that any of the deaths were opioid related. This study was limited to patients age ≥65 years undergoing surgery in a single health system in the southeastern United States. Results may not be generalizable to younger populations or other systems and regions with differing surgical and prescribing practices. However, older patients represent an understudied and vulnerable population of interest, and these findings add to the limited evidence of the association of preoperative gabapentinoids on postsurgical opioid use.

Currently, the use of presurgical gabapentinoids has been recommended by diverse professional societies. However, gabapentinoid use for surgical pain is considered off-label use, and in 2019, the FDA issued a warning and labeling updates to address risks of breathing difficulties in patients who use gabapentinoids, particularly in combination with opioids.^⁵ Overall, we did not find that presurgical gabapentinoids were associated with a reduction in risk of prolonged opioid use. Given the limited clinical evidence supporting off-label effectiveness, caution is needed when prescribing these medications.^²⁵^,^²⁶ Attempts to reduce opioid abuse by shifting prescribing toward different drugs for pain management have the potential of unintentionally creating new avenues of abuse.^²⁷^,^²⁸ The off-label use of these medications to manage surgical pain should be carefully balanced against known harm, and more research is needed to understand the efficacy and safety of preoperative gabapentinoid use.

ACKNOWLEDGMENTS

We thank the Editors and Reviewers at Anesthesia and Analgesia for their thoughtful comments during the peer-review process.

DISCLOSURES

Name: Jessica C. Young, PhD.

Contribution: This author helped lead the project planning, analyses, and presentation of results and write the manuscript.

Conflicts of Interest: J. C. Young receives consulting fees from CERobs Consulting, LLC.

Name: Nabarun Dasgupta, PhD.

Contribution: This author helped contribute expertise in pain management research and consult on the study question, design the study, interpret the results, and provide comments and edits to the manuscript.

Conflicts of Interest: N. Dasgupta is a consultant to the RADARS System of Denver Health and Hospital Authority, a political subdivision of the State of Colorado.

Name: Brooke A. Chidgey, MD.

Contribution: This author helped provide dates and specific policy recommendations at the clinical level, contribute knowledge on the mechanisms of action and use of the drugs under study, refine the research question, and provide comments and edits to the manuscript.

Conflicts of Interest: None.

Name: Til Stürmer, MD, PhD.

Contribution: This author helped guide the study design and methodology of this research and provide comments and edits to the manuscript.

Conflicts of Interest: T. Stürmer owns stock in Novartis, Roche, BASF, AstraZeneca, and Novo Nordisk.

Name: Virginia Pate, MS.

Contribution: This author helped with dataset management and infrastructure, statistical coding, and general programming and analytic support.

Conflicts of Interest: None.

Name: Michael Hudgens, PhD.

Contribution: This author helped guide study design and statistical methods and provide guidance on interpretation, substantial comments, and edits to the manuscript.

Conflicts of Interest: None.

Name: Michele Jonsson Funk, PhD.

Contribution: This author helped provide oversight over this research, helped with the study conception, design, analysis, and interpretation; and provided comments and edits to the manuscript.

Conflicts of Interest: M. Jonsson Funk received consulting fees via UNC from GlaxoSmithKline.

This manuscript was handled by: Honorio T. Benzon, MD.

REFERENCES

1. Tiippana EM, Hamunen K, Kontinen VK, Kalso E. Do surgical patients benefit from perioperative gabapentin/pregabalin? A systematic review of efficacy and safety. Anesth Analg. 2007;104:1545–1556.

Cited Here |
View Full Text | PubMed | CrossRef

2. Ho KY, Gan TJ, Habib AS. Gabapentin and postoperative pain–a systematic review of randomized controlled trials. Pain. 2006;126:91–101.

Cited Here |
View Full Text | PubMed | CrossRef

3. Mathiesen O, Møiniche S, Dahl JB. Gabapentin and postoperative pain: a qualitative and quantitative systematic review, with focus on procedure. BMC Anesthesiol. 2007;7:6.

Cited Here |
PubMed | CrossRef

4. Verret M, Lauzier F, Zarychanski R, et al.; Canadian Perioperative Anesthesia Clinical Trials (PACT) Group. Perioperative use of gabapentinoids for the management of postoperative acute pain: a systematic review and meta-analysis. Anesthesiology. 2020;133:265–279.

Cited Here |
View Full Text | PubMed | CrossRef

5. FDA Drug Safety Communication. FDA Warns About Serious Breathing Problems With Seizure and Nerve Pain Medicines Gabapentin (Neurontin, Gralise, Horizant) and Pregabalin (Lyrica, Lyrica CR). 2019. Accessed March 20, 2020. https://www.fda.gov/drugs/drug-safety-and-availability/fda-warns-about-serious-breathing-problems-seizure-and-nerve-pain-medicines-gabapentin-neurontin.

Cited Here

6. Chou R, Gordon DB, de Leon-Casasola OA, et al. Management of postoperative pain: a clinical practice guideline from the American Pain Society, the American Society of Regional Anesthesia and Pain Medicine, and the American Society of Anesthesiologists’ Committee on Regional Anesthesia, Executive Committee, and Administrative Council. J Pain. 2016;17:131–157.

Cited Here |
PubMed | CrossRef

7. Gustafsson UO, Scott MJ, Hubner M, et al. Guidelines for perioperative care in elective colorectal surgery: Enhanced Recovery After Surgery (ERAS®) society recommendations: 2018. World J Surg. 2019;43:659–695.

Cited Here |
PubMed | CrossRef

8. Yorkgitis BK, Paffett C, Brat GA, Crandall M. Effect of surgery-specific opioid-prescribing education in a safety-net hospital. J Surg Res. 2019;243:71–74.

Cited Here |
PubMed | CrossRef

9. Kumar K, Kirksey MA, Duong S, Wu CL. A review of opioid-sparing modalities in perioperative pain management: methods to decrease opioid use postoperatively. Anesth Analg. 2017;125:1749–1760.

Cited Here |
View Full Text | PubMed | CrossRef

10. Gabriel RA, Swisher MW, Sztain JF, Furnish TJ, Ilfeld BM, Said ET. State of the art opioid-sparing strategies for post-operative pain in adult surgical patients. Expert Opin Pharmacother. 2019;20:949–961.

Cited Here |
PubMed | CrossRef

11. Winkelman WD, Kim Y, Erlinger AL, Haviland MJ, Hacker MR, Elkadry EA. Optimizing perioperative pain control after ambulatory urogynecologic surgery. Female Pelvic Med Reconstr Surg. 2020;26:483–487.

Cited Here |
View Full Text | PubMed

12. Militsakh O, Lydiatt W, Lydiatt D, et al. Development of multimodal analgesia pathways in outpatient thyroid and parathyroid surgery and association with postoperative opioid prescription patterns. JAMA Otolaryngol Head Neck Surg. 2018;144:1023–1029.

Cited Here |
View Full Text | PubMed | CrossRef

13. Mujukian A, Truong A, Tran H, Shane R, Fleshner P, Zaghiyan K. A standardized multimodal analgesia protocol reduces perioperative opioid use in minimally invasive colorectal surgery. J Gastrointest Surg. 2020;24:2286–2294.

Cited Here |
PubMed | CrossRef

14. Fabritius ML, Geisler A, Petersen PL, et al. Gabapentin for post-operative pain management - a systematic review with meta-analyses and trial sequential analyses. Acta Anaesthesiol Scand. 2016;60:1188–1208.

Cited Here |
View Full Text | PubMed | CrossRef

15. Surgery Flag Software. Healthcare Cost and Utilization Project (HCUP). 2019. Agency for Healthcare Research and Quality. Accessed April 03, 2020. https://www.hcup-us.ahrq.gov/toolssoftware/surgeryflags_svcproc/surgeryflagssvc_proc.jsp.

Cited Here

16. Young JC, Wu JM, Willis-Gray M, Pate V, Jonsson Funk M. Persistent opioid use after hysterectomy in the United States, 2005-2015. Obstet Gynecol. 2020;135:123–132.

Cited Here |
View Full Text | PubMed | CrossRef

17. Harrell FE. Regression Modeling Strategies. 2006.Springer-Verlag,

Cited Here

18. Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med. 2015;34:3661–3679.

Cited Here |
PubMed | CrossRef

19. Schulte PJ, Mascha EJ. Propensity score methods: theory and practice for anesthesia research. Anesth Analg. 2018;127:1074–1084.

Cited Here |
View Full Text | PubMed | CrossRef

20. Brookhart MA, Wyss R, Layton JB, Stürmer T. Propensity score methods for confounding control in nonexperimental research. Circ Cardiovasc Qual Outcomes. 2013;6:604–611.

Cited Here |
View Full Text | PubMed | CrossRef

21. Glynn RJ, Lunt M, Rothman KJ, Poole C, Schneeweiss S, Stürmer T. Comparison of alternative approaches to trim subjects in the tails of the propensity score distribution. Pharmacoepidemiol Drug Saf. 2019;28:1290–1298.

Cited Here |
View Full Text | PubMed | CrossRef

22. Yang D, Dalton J. A unified approach to measuring the effect size between two groups using SAS. Paper 335-2012. Paper presented at: SAS Global Forum 2012; April 22-25, 2012; Cary, NC. Accessed June 23, 2021. https://support.sas.com/resources/papers/proceedings12/335-2012.pdf.

Cited Here

23. Lash TL, Fox MP, MacLehose RF, Maldonado G, McCandless LC, Greenland S. Good practices for quantitative bias analysis. Int J Epidemiol. 2014;43:1969–1985.

Cited Here |
View Full Text | PubMed | CrossRef

24. Funk MJ, Landi SN. Misclassification in administrative claims data: quantifying the impact on treatment effect estimates. Curr Epidemiol Rep. 2014;1:175–185.

Cited Here

25. Peckham AM, Evoy KE, Ochs L, Covvey JR. Gabapentin for off-label use: evidence-based or cause for concern? Subst Abuse. 2018;12:1178221818801311.

Cited Here

26. Goodman CW, Brett AS. A clinical overview of off-label use of gabapentinoid drugs. JAMA Intern Med. 2019;179:695–701.

Cited Here |
View Full Text | PubMed

27. Faryar KA, Webb AN, Bhandari B, Price TG, Bosse GM. Trending gabapentin exposures in Kentucky after legislation requiring use of the state prescription drug monitoring program for all opioid prescriptions. Clin Toxicol (Phila). 2019;57:398–403.

Cited Here |
PubMed

28. Kim S. The unsuspected threat of three opioid-like substitutes. Arch Psychiatr Nurs. 2019;33:325–328.

Cited Here |
View Full Text | PubMed | CrossRef

Supplemental Digital Content

Supplemental Digital Content; [Word] (76 KB)

Secondary Logo

Journal Logo

Day-of-Surgery Gabapentinoids and Prolonged Opioid Use: A Retrospective Cohort Study of Medicare Patients Using Electronic Health Records

Abstract

BACKGROUND:

METHODS:

RESULTS:

CONCLUSIONS:

METHODS

Data Source

Study Population

Exposure

Outcome

Potential Confounding Variables

Statistical Analyses

Main Analysis

Secondary Analysis

Sensitivity Analysis

Quantitative Bias Analyses

RESULTS

Calendar Trends

Main Analysis

Secondary Analysis

Sensitivity Analyses

Bias Analyses

DISCUSSION

ACKNOWLEDGMENTS

DISCLOSURES

REFERENCES

Supplemental Digital Content

Kudos Resources

Kudos Publication

Kudos Resources

Article Level Metrics

Related Links

Related Videos

Readers Of this Article Also Read

Day-of-Surgery Gabapentinoids and Prolonged Opioid Use: A Retrospective Cohort Study of Medicare Patients Using Electronic Health Records : Anesthesia & Analgesia

Abstract

BACKGROUND:

METHODS:

RESULTS:

CONCLUSIONS:

METHODS

Data Source

Study Population

Exposure

Outcome

Potential Confounding Variables

Statistical Analyses

Main Analysis

Secondary Analysis

Sensitivity Analysis

Quantitative Bias Analyses

RESULTS

Calendar Trends

Main Analysis

Secondary Analysis

Sensitivity Analyses

Bias Analyses

DISCUSSION

ACKNOWLEDGMENTS

DISCLOSURES

REFERENCES

Supplemental Digital Content