Key Points
Question
What are the long-term outcomes associated with an opioid-sparing protocol after kidney transplant?
Findings
In this quality improvement project that included 743 kidney graft recipients, patients in the opioid-sparing protocol group had 99% lower odds of filling more than 100 morphine milligram equivalents during the 1-year follow-up.
Meaning
These findings suggest that a significant reduction in opioid use after kidney transplant was associated with the implementation of a multimodal opioid-sparing pain protocol.
This quality improvement study evaluated the long-term outcomes associated with an opioid-sparing protocol in adult kidney graft recipients.
Abstract
Importance
Opioid use following kidney transplant is associated with an increased risk of graft loss and mortality. Opioid minimization strategies and protocols have shown reductions in short-term opioid use after kidney transplant.
Objective
To evaluate the long-term outcomes associated with an opioid minimization protocol following kidney transplant.
Design, Setting, and Participants
This single-center quality improvement study evaluated postoperative and long-term opioid use before and after the implementation of a multidisciplinary, multimodal pain regimen and education process in adult kidney graft recipients from August 1, 2017, through June 30, 2020. Patient data were collected from a retrospective chart review.
Exposures
Preprotocol and postprotocol implementation use of opioids.
Main Outcomes and Measures
Between November 7 and 23, 2022, opioid use before and after protocol implementation was evaluated up to 1 year after transplant using multivariable linear and logistic regression.
Results
A total of 743 patients were included, with 245 patients in the preprotocol group (39.2% female and 60.8% male; mean [SD] age, 52.8 [13.1 years]) vs 498 in the postprotocol group (45.4% female and 54.6% male; mean [SD] age, 52.4 [12.9 years]). The total morphine milligram equivalents (MME) in the 1-year follow-up in the preprotocol group was 1203.7 vs 581.9 in the postprotocol group. In the postprotocol group, 313 patients (62.9%) had 0 MME in the 1-year follow-up vs 7 (2.9%) in the preprotocol group (odds ratio [OR], 57.52; 95% CI, 26.55-124.65). Patients in the postprotocol group had 99% lower odds of filling more than 100 MME in the 1-year follow-up (adjusted OR, 0.01; 95% CI, 0.01-0.02; P < .001). Opioid-naive patients postprotocol were one-half as likely to become long-term opioid users vs preprotocol (OR, 0.44; 95% CI, 0.20-0.98; P = .04).
Conclusions and Relevance
The study’s findings show a significant reduction in opioid use in kidney graft recipients associated with the implementation of a multimodal opioid-sparing pain protocol.
Introduction
Nearly 500 000 people in the US died of opioid overdoses between 1999 and 2019. Almost 247 000 of these deaths involved a prescription opioid, with an average of 38 deaths per day.1,2 In October 2017, the US government officially declared the opioid epidemic a public health emergency.3 This declaration unveiled a 5-point strategy to combat the opioid crisis, which may have contributed to the decline in prescription opioid overdose deaths over the past few years.
It is well established that misuse and overdose of prescription opioids can lead to substantial morbidity and mortality; however, opioids still remain the standard of care for management of acute postoperative pain.4 Data have suggested that persistent and long-term opioid use after surgery is rising in patients who have previously taken or never taken opioids5 and that more than 80% of patients receive opioids after low-risk surgery.6 Excluding misuse and overdose, accumulating evidence has suggested that any level of opioid use is associated with comorbidities and mortality. Three large registry analyses showed that perioperative opioid use is associated with an increased risk of adverse outcomes, graft failure, and death in graft recipients7,8,9; however, there appears to be a dose- and exposure-associated increase in risk. Results from a large cohort study of kidney graft recipients showed that long-term prescription opioid use may be associated with an increased risk of death and graft loss compared with no or short-term opioid prescriptions.10
In response to this growing evidence, as well as based on our own experience, our institution implemented a quality improvement project in 2018 that included a multidiscipline, multimodal pain regimen and education process (MMPRE) to minimize opioid use after kidney transplant. Initial results of this initiative revealed a reduction in opioid exposure in the early postoperative period while maintaining appropriate pain control.11 As a follow-up to the initial analysis, we evaluated the long-term durability and effectiveness of our MMPRE initiative.
Methods
Design and Patient Population
This quality improvement study was a single-center, long-term, retrospective cohort analysis of a quality assurance and process improvement (QAPI) endeavor in kidney graft recipients to measure postoperative opioid use following the implementation of MMPREP compared with historic control patients who did not receive this protocol. The analysis consisted of a before-and-after design such that patient characteristics and outcomes were compared between those who received a kidney graft before the protocol was initiated vs those who received a graft after protocol initiation.
We expanded our original population to include a larger number of patients both before and after the initial protocol implementation. Adult kidney graft recipients (aged ≥18 years) who underwent transplant between August 1, 2017, and June 30, 2020, were included in this analysis. Patients were excluded if they underwent a multiorgan transplant or died within 30 days following transplant. The MMPREP was continued from the original protocol, which has been described in depth.11 Patients received a preoperative education handout on the reasons for opioid minimization (eFigure 1 in Supplement 1). Preoperative transversus abdominis plane or quadratus lumborum block with bupivacaine, 0.25%, was administered by anesthesia. The block was chosen based on anesthesia preference. Type of intraoperative anesthesia was not changed throughout the preprotocol and postprotocol period. The surgical technique and mobilization process were well established and did not change throughout the 3-year period. Postoperatively, patients received acetaminophen 1 g orally every 6 hours and gabapentin initiated at 100 mg every 24 hours and titrated to kidney function and pain control (eFigure 2 in Supplement 1). Oral opioids were used if the patient reported pain scores of 6 or greater out of 10 (10 being the worst pain ever experienced), and intravenous opioids were administered if the patient’s pain scores were 8 or greater out of 10. The postoperative clinical pathway did not change during either phase of this analysis. The pathway is well structured, including ambulation and education beginning postoperative day 1, with anticipation of discharge on postoperative day 3. On discharge, patients were prescribed acetaminophen 1 g orally every 8 hours as needed and gabapentin titrated to effect and adjusted for renal function continued for 5 days (eFigure 3 in Supplement 1). All posttransplant medications, including immunosuppressants and opioids, were managed and prescribed by the transplant surgeons and surgical advanced practice clinicians for the first 45 days after transplant. Following postoperative day 45, a transplant nephrologist began managing immunosuppression and pain. Although we did not explicitly capture the individual who prescribed opioids, we assumed it to be highly likely that if opioids were prescribed in the first 45 days after surgery, it was done by the surgeon or transplant advanced practice clinician. After this period, patients could be prescribed opioids by a nephrologist or outside clinician; however, preoperative patient education regarding the risks associated with opioids may have deterred them from seeking these medications from other clinicians. The Medical University of South Carolina institutional review board determined that this analysis was not a research study but an extension of the original QAPI endeavor and, therefore, exempt from institutional review board approval and informed consent. The study followed the Standards for Quality Improvement Reporting Excellence (SQUIRE) reporting guideline.
Definitions and Data Collection
Outpatient prescription opioid and benzodiazepine data from 1 year before to 1 year after protocol initiation were collected for all patients through the electronic medical record (EMR) linked to the state prescription drug monitoring program South Carolina Reporting and Identification Prescription Tracking System, which includes all retail and outpatient hospital pharmacy dispensing of schedules II to IV controlled substances. Pretransplant naive opioid use was defined as 0 opioid prescription fills in the year before transplant. Long-term pretransplant opioid use was defined as having 1 or more opioid prescription fills up to 1 year before transplant. Long-term opioid use was defined as 2 or more opioid prescription fills after 90 days posttransplant, and non–long-term opioid use was defined as no opioid prescription fills after 90 days posttransplant. Data from patients who had 1 opioid prescription fill after transplant were collected, but these individuals were not allocated to a group.
Baseline demographic characteristics, including age, sex, race, and ethnicity, for recipients were collected using the center-specific Standard Transplant Analysis and Research file. Demographic characteristics for donors included age and sex, but specific information on donor race and ethnicity was not available. Data on type of block, total inpatient opioid use, prior medication history, and discharge medications were manually collected from the EMR. Outpatient prescription opioid data before and after transplant were collected using an online reporting service as part of the prescription drug monitoring program. All opioid prescriptions were converted to a standardized scale of morphine milligram equivalents (MME) using conversions previously reported in the literature (eTable in Supplement 1). Opioid use while patients were in the postanesthesia care unit was not included in the analysis.
Statistical Analysis
Standard descriptive statistics were used to display the data, including percentages for categorical data and means with SDs or medians with IQRs for continuous data based on distribution. Categorical data were compared by χ2 or Fisher exact test. Continuous data were compared by t test or Mann-Whitney U test. Multivariable linear and logistic regressions were conducted for the outcomes associated with long-term opioid use after transplant and included variables that were statistically associated with outcomes from univariable analyses or those deemed clinically relevant as potential confounders by clinicians. Covariates in these models included long-term gabapentin or pregabalin use before transplant, working at the time of transplant, living outside the state of South Carolina, body mass index (BMI), and delayed graft function (DGF). Due to the nonnormal distribution of MMEs, the median MMEs were modeled using an interrupted time series with autoregression. In this model, month of transplant (time) is modeled for the outcome of median MME, with a dummy variable coded for preprotocol vs postprotocol. Correlation by time is accounted for in the autoregression procedure by incorporating up to 12 lag terms (t-1, t-2, t-3, etc) into the model using backward elimination. This procedure was also conducted for the outcome of the proportion of patients with no outpatient opioid prescriptions. Data were analyzed between November 7 and 23, 2022, using SPSS, version 25.0 (IBM) and SAS, version 9.4 (SAS Institute) statistical software. A 2-sided P < .05 was considered statistically significant.
Results
Patient Characteristics
A total of 745 patients underwent a transplant between August 1, 2017, and June 30, 2020. The preprotocol group included 96 females (39.2%) and 149 males (60.8%) (mean [SD] age, 52.8 [13.1] years; African American, 145 [59.2%]; Asian, 4 [1.6%]; Hispanic, 6 [2.5%]; White, 89 [36.3%]; other race and ethnicity, 1 [0.4%]), and the postprotocol group included 226 females (45.4%) and 272 males (54.6%) (mean [SD] age, 52.4 [12.9] years; African American, 300 [60.2%]; Asian, 10 [2.0%]; Hispanic, 8 [1.6%]; White, 177 [35.5%]; other race and ethnicity, 3 [0.6%]). Two patients died within 30 days after transplant and were not included in the analysis. As a result, 743 kidney graft recipients were included in the final analysis, with 245 in the preprotocol group and 498 in the postprotocol group. Baseline characteristics were similar across cohorts (Table 1). There were no differences in pretransplant MME opioid use per month between the preprotocol and postprotocol groups (mean [SD], 0.15 [0.42] vs 0.11 [0.25], respectively; P = .19) or number of patients overall with benzodiazepine use (39 [15.9%] vs 76 [15.3%], respectively; P = .82). Body mass index (weight in kilograms divided by height in meters squared) was the only statistically significant factor that differed between cohorts, with patients in the postprotocol group having a higher BMI (mean [SD], 30.1 [5.9] vs 29.3 [4.7] preprotocol; P = .02). The mean date of transplant in the preprotocol group was February 7, 2018 (range, August 3, 2017, to September 2, 2018), and the mean date of transplant for the postprotocol group was August 11, 2019 (range, September 4, 2018, to June 28, 2020). Donor characteristics were also similar between groups (Table 1).
Table 1. Baseline Recipient and Donor Characteristics Before and After Multidisciplinary, Multimodal Pain Regimen and Education Process Implementation.
| Characteristic | No. (%) | P value | |
|---|---|---|---|
| Preprotocol (n = 245) | Postprotocol (n = 498) | ||
| Recipient | |||
| Age, mean (SD), y | 52.8 (13.1) | 52.4 (12.9) | .80 |
| Sex | |||
| Female | 96 (39.2) | 226 (45.4) | .11 |
| Male | 149 (60.8) | 272 (54.6) | |
| BMI, mean (SD) | 29.3 (4.7) | 30.1 (5.9) | .02 |
| Race and ethnicity | .78 | ||
| African American | 145 (59.2) | 300 (60.2) | |
| Asian | 4 (1.6) | 10 (2.0) | |
| Hispanic | 6 (2.5) | 8 (1.6) | |
| White | 89 (36.3) | 177 (35.5) | |
| Othera | 1 (0.4) | 3 (0.6) | |
| DGF | 50 (20.4) | 130 (26.1) | .09 |
| cPRA, %, mean (SD) | 47.3 (40.1) | 42.4 (38.4) | .13 |
| EPTS, mean (SD) | 0.47 (0.30) | 0.46 (0.30) | .60 |
| Pretransplant | |||
| Diabetes | 99 (40.4) | 214 (43.1) | .49 |
| Gabapentin or pregabalin use | 39 (15.9) | 79 (15.9) | .99 |
| Benzodiazepine use | 39 (15.9) | 76 (15.3) | .82 |
| Opioid use per mo, MME, mean (SD) | 0.15 (0.42) | 0.11 (0.25) | .19 |
| Donor b | |||
| Deceased donor | 210 (85.7) | 418 (83.9) | .53 |
| DCD | 40 (16.3) | 101 (20.3) | .20 |
| Age, mean (SD) | 38.8 (12.6) | 38.2 (13.8) | .86 |
| KDPI, %, mean (SD) | 34.5 (0.26) | 35.4 (0.27) | .47 |
| Sex | |||
| Female | 108 (44.1) | 194 (39.0) | .18 |
| Male | 137 (55.9) | 304 (61.0) | |
Abbreviations: BMI, body mass index measured as weight in kilograms divided by height in meters squared; cPRA, calculated panel reactive antibodies; DCD, donation after cardiac death; DGF, delayed graft function; EPTS, estimated posttransplant survival; KDPI, kidney donor profile index; MME, morphine milligram equivalents.
Not further specified.
Specific data on donor race and ethnicity were not available.
Perioperative Outcomes
The total MME in the 1-year follow-up in the preprotocol group was 1203.7 vs 581.9 in the postprotocol group. After implementation of the MMPREP, 34 patients (6.8%) were discharged with an opioid prescription following kidney transplant compared with 236 (96.3%) preprotocol (P < .001). In-hospital median MME per day in the preprotocol group was 30.6 mg (IQR, 17.5-45.2 mg) vs 0.0 mg (IQR, 0.0-3.8 mg) in the postprotocol group (P < .001) (Table 2). Before MMPREP implementation, 6 patients (2.4%) were opioid free while hospitalized. After protocol implementation, 199 patients (40%) were opioid free during hospitalization (P < .001).
Table 2. Perioperative and Long-term Opioid Use Outcomes Before and After Multidisciplinary, Multimodal Pain Regimen and Education Process Implementation.
| Opioid use outcome | No. (%) | P value | Adjusted OR (95% CI)a | |
|---|---|---|---|---|
| Preprotocol (n = 245) | Postprotocol (n = 498) | |||
| Perioperative | ||||
| MME in hospital per day, median (IQR) | 30.6 (17.5-45.2) | 0.0 (0.0-3.8) | <.001 | NA |
| Opioid free in hospital | 6 (2.4) | 199 (40) | <.001 | NA |
| Discharged with opioid prescription | 236 (96.3) | 34 (6.8) | <.001 | NA |
| Long-term | ||||
| Discharged with opioid prescription | 236 (96.3) | 34 (6.8) | <.001 | 0.00 (0.00-0.01) |
| >2 Opioid prescription fills (months 10-12) | 19 (7.8) | 21 (4.2) | .04 | 0.52 (0.28-0.99) |
| No MME during 1-y follow-up | 7 (2.9) | 313 (62.9) | <.001 | 57.52 (26.55-124.65) |
| Total MME during follow-up, median (IQR) | 450 (225-505) | 0.0 (0.0-112.5) | <.001 | NA |
| >100 MME during follow-up | 238 (97.1) | 125 (25.1) | <.001 | 0.01 (0.01-0.02) |
| Long-term opioid users to non–long-term users | 50 (20.4) | 193 (38.8) | <.001 | 2.47 (1.72-3.53) |
| Naive opioid users to long-term users | 13 (5.3) | 12 (2.4) | .04 | 0.44 (0.20-0.98) |
Abbreviations: MME, morphine milligram equivalents; NA, not applicable; OR, odds ratio.
Variables adjusted for include long-term gabapentin or pregabalin use before transplant, working at the time of transplant, living outside the state of South Carolina, body mass index, and delayed graft function.
Durable adherence to the MMPREP was observed, with nearly all patients in the postprotocol group receiving a preoperative nerve block and being discharged with a prescription for gabapentin or pregabalin and acetaminophen (Table 3). Rates of adherence to protocol components remained consistent over time. The most common type of nerve block was quadratus lumborum (405 patients [81.3%]).
Table 3. Opioid Minimization Protocol Adherence Before and After Multidisciplinary, Multimodal Pain Regimen and Education Process Implementation.
| Protocol component | No. (%) | P value | |
|---|---|---|---|
| Preprotocol (n = 245) | Postprotocol (n = 498) | ||
| Nerve block | 7 (2.9) | 424 (85.1) | <.001 |
| At discharge | |||
| Gabapentin or pregabalin | 27 (11.0) | 450 (90.4) | <.001 |
| Acetaminophen | 30 (12.2) | 473 (95.0) | <.001 |
Long-term Outcomes
The median MME during the 1-year follow-up in the preprotocol group was 450 mg (IQR, 225-505 mg) vs 0 mg (IQR, 0-112.5 mg) in the postprotocol group, which includes the opioid prescription at the time of discharge. In the postprotocol group, 313 patients (62.9%) had no opioid use (0 MME) during the 1-year follow-up vs 7 patients (2.9%) in the preprotocol group (adjusted odds ratio [OR], 57.52; 95% CI, 26.55-124.65; P < .001) (Table 2). Patients in the postprotocol group had 99% lower odds of filling opioid prescriptions that totaled more than 100 MME during the 1-year follow-up (adjusted OR, 0.01; 95% CI, 0.01-0.02; P < .001). Postprotocol patients were one-half as likely to fill 2 or more opioid prescriptions in the final 3 months of the follow-up year (OR, 0.52; 95% CI, 0.28-0.99; P = .04), which is how we defined long-term opioid use.
Postprotocol patients who were opioid naive at the time of transplant were one-half as likely to become long-term opioid users compared with the preprotocol group (OR, 0.44; 95% CI, 0.20-0.98; P = .04) (Table 2 and Figure 1). There was no significant difference between groups for patients who were opioid naive before transplant and remained naive during the 1-year follow-up. Kidney graft recipients who were classified as long-term opioid users before transplant were significantly more likely to become opioid naive after implementation of the MMPREP (OR, 2.47; 95% CI, 1.72-3.53; P < .001) (Table 2). There was no significant difference between groups for patients classified as pretransplant long-term opioid users who remained long-term users after transplant. An interrupted time series analysis for median total MME during the 1-year follow-up and the percentage of patients who were opioid free is shown in Figure 2. The results demonstrate that the MMPREP produced a 246-MME reduction (95% CI, 90.6-401.4; P = .01) after accounting for natural trends in opioid prescribing during this 3-year period (reduction of 17.4 MME per quarter; P = .16). The time series analysis also demonstrated a 54.4% increase in the proportion of patients who were opioid free in the outpatient setting (P < .001), after accounting for natural trends in opioid prescribing (0.64% per quarter; P = .63).
Figure 1. Interrupted Time Series Analysis for Outpatient Opioid Use in Kidney Graft Recipients.
This figure displays an interrupted time series analysis for outpatient opioid utilization in kidney transplant recipients over 3 years, including 1 year before and 2 years after protocol implementation. The blue line represents the proportion of patients with no outpatient postoperative opioid use, increasing from less than 5% before protocol implementation to more than 60% after implementation. The bars represent the median outpatient morphine milligram equivalents (MME) for 1 year after transplant, decreasing from 450 MME before protocol implementation to 0 to 15 MME after implementation. Q indicates quarter.
Figure 2. Trends of Opioid Use During 1-Year Follow-up.
This figure shows the percentage of patients who were long-term opioid users before protocol implementation who became non–long-term users after implementation vs patients who were naive opioid users (no use) who became long-term users. Long-term use was measured as 2 or more opioid prescription fills after 90 days postsurgery; non–long-term use as 0 opioid prescription fills after 90 days postsurgery; and naive use as 0 opioid prescription fills in the year before surgery.
Discussion
In this quality improvement study, we report follow-up results of our original QAPI endeavor using MMPREP in a large population of kidney graft recipients. The study took place over 3 years to evaluate the durability and long-term effectiveness of the protocol in reducing opioid use. We found that MME was substantially minimized both perioperatively and up to 1 year posttransplant. Most importantly, long-term opioid use was significantly reduced and nonnaive opioid users going into transplant were more than twice as likely to become non–long-term opioid users after implementation of the MMPREP.
We were able to confirm and strengthen the initial results of the protocol11 by demonstrating a substantial and durable decrease in the number of opioids used during the initial postoperative period and decrease in the number of opioid prescriptions at hospital discharge. With the risks associated with opioid use following kidney transplant, several centers have adopted opioid-sparing pain protocols. By using a multimodal approach consisting of an intraoperative nerve block and scheduled acetaminophen, Dualeh et al12 were able to decrease the quantity of opioids prescribed at hospital discharge. Similar results were published by Lichvar et al,13 who found that opioid prescriptions at discharge in kidney graft recipients was significantly lower after an opioid-sparing protocol and educational session were implemented. Other transplant centers across the US have shown comparable results after implementation of similar opioid-sparing protocols in the perioperative transplant period.12,14 However, these studies, along with others, have solely focused on the short-term outcomes following opioid minimization protocols with limited sample sizes.
In this analysis, we were able to show a significant decrease in total MME and overall number of opioid prescriptions in the 1 year after implementation of the MMPREP. One of the more clinically important findings is the number of patients who had no prescription fills during the 1-year follow-up. We found that the majority of patients who underwent transplant after protocol implementation had no opioid prescriptions in the year following transplant, compared with only 2.9% of patients prior to this protocol being implemented. These findings are important, considering that epidemiologic research has shown that opioid use in the first year after transplant is associated with a 2-fold increased risk of death and a 68% increased risk of graft failure.8 Other complications associated with opioid use in the post–kidney transplant setting may include immunosuppression nonadherence. A single-center, retrospective study of US veterans found that kidney graft recipients who used opioids during the first year after transplant had a lower probability of adhering to timely refills of tacrolimus.15 Concurrent use of benzodiazepines and opioids after kidney transplant has also been associated with an increased risk of death and graft failure.15 Thus, the findings from this long-term analysis of an opioid minimization protocol suggest that graft and patient outcomes may be improved with comprehensive adoption of such pathways.
These data suggest that long-term opioid use can be reduced through deliberately applied nonopioid pain management strategies and patient involvement implemented at the time of surgery. In the general surgical population, Aalberg et al16 found that patients with persistent postoperative opioid use were at increased risk of both opioid use disorder and overdose compared with postoperative patients with no long-term opioid use. When evaluating long-term opioid use following kidney transplant, Kulshrestha et al17 found that patients with long-term use more often had hospital readmissions (P < .001) during the first year after transplant but similar rates of acute rejection (P = .31). Across other organs, a single-center cohort study of lung graft recipients revealed that long-term opioid use after transplant was strongly associated with decreased lung function and increased risk for mortality.18 Pretransplant opioid use was independently associated with time to graft loss or mortality in liver graft recipients in a single-center retrospective study by Fleming et al.19 Similarly, Lentine et al20 found that higher levels of opioids before and after heart transplant were a marker for increased risk of adverse outcomes. In both transplant as well as other surgical patient populations, postoperative long-term opioid use has been associated with poor long-term outcomes, emphasizing the importance of opioid reduction strategies following surgery as seen in our multimodal opioid-sparing protocol.
Limitations
This study has several limitations. First, this study was not a parallel-arm controlled clinical trial; thus, the findings should not be construed as causal. Natural trends in opioid prescribing were somewhat accounted for in the interrupted time series analysis, but residual confounding by time could still be an issue. Second, the data were retrospective, collected as an extension of a previous quality improvement project, which makes the study prone to selection bias. However, to minimize the potential for bias and missingness, data were collected manually from detailed and comprehensive chart abstraction and opioid prescription fill information gathered from a statewide database that is the most comprehensive assessment available. Third, this analysis was not randomized; however, baseline characteristics were gathered from validated Standard Transplant Analysis and Research files, which contain detailed sociodemographic characteristics, and these characteristics were similar between groups. Fourth, because of boundaries within the online opioid reporting system, we were only able to ascertain opioid prescription fill history backdated to 4 years from the search date. Therefore, some patients in the preprotocol cohort only had 1 month of a pretransplant opioid prescription fill history available. For these patients, it is not fully known whether they were long-term opioid users before transplant. Other limitations from the online opioid reporting system include a lag time up to 48 hours in reporting (not an issue with this retrospective collection and long-term follow-up), excluding prescriptions for less than a 48-hour supply, and missing prescription fills outside the state of South Carolina. We attempted to control for this factor by including living outside the state of South Carolina in our multivariable models. Opioid use during hospitalization was collected by manual chart abstraction, and postanesthesia care unit opioid use was not captured due to reporting limitations in the EMR. Fifth, we were unable to evaluate patient pain scores because this analysis was retrospective and these data are not documented in a robust manner. However, because all opioid use within the hospital is tied to specific pain scores, in-hospital opioid use should have reflected pain scores. Documentation of patient pain scores throughout hospitalization and on discharge would be valuable in future studies. Sixth, due to increased public awareness of the opioid epidemic during this quality improvement project, clinicians may have been less likely to prescribe opioids during the postprotocol period. Seventh, illicit drug use, including opioids, was not captured as part of this analysis. Eighth, we did not explicitly document which practitioner (surgeon, nephrologist, or outside clinician) prescribed opioids and could not link this to diagnoses or reason for prescription. Ninth, defining long-term opioid use is difficult, and there are clear limitations to most definitions, as a gold standard is lacking and it is not feasible to specifically link long-term opioid use to the surgery in a causal manner. We attempted to address this limitation by using several different measures of long-term opioid use, including both dichotomous (yes/no) and quantifiable (MME) measures while also using previously validated measures, such as opioid prescription fills after 90 days postsurgery.
Conclusions
Results of this long-term, large-scale QAPI endeavor demonstrate a substantial reduction in opioid use in kidney graft recipients associated with the implementation of a multimodal opioid-sparing pain protocol. The protocol was associated with a significant reduction in 1-year postoperative opioid use, including in patients who used opioids at the time of transplant. In opioid-naive patients, the protocol was associated with a reduced risk of becoming long-term opioid users. A parallel-arm, randomized clinical trial is needed to establish causation and confirm these findings.
eTable. Morphine Milligram Equivalents (MME) Conversions
eFigure 1. Opioid Information Handout
eFigure 2. Inpatient Opioid Minimization Protocol
eFigure 3. Discharge Opioid Minimization Protocol
Data Sharing Statement
References
- 1.CDC WONDER. Centers for Disease Control and Prevention, 2020. Accessed May 12, 2022. http://wonder.cdc.gov
- 2.Mattson CL, Tanz LJ, Quinn K, Kariisa M, Patel P, Davis NL. Trends and geographic patterns in drug and synthetic opioid overdose deaths—United States, 2013-2019. MMWR Morb Mortal Wkly Rep. 2021;70(6):202-207. doi: 10.15585/mmwr.mm7006a4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Jones MR, Viswanath O, Peck J, Kaye AD, Gill JS, Simopoulos TT. A brief history of the opioid epidemic and strategies for pain medicine. Pain Ther. 2018;7(1):13-21. doi: 10.1007/s40122-018-0097-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lawal OD, Gold J, Murthy A, et al. Rate and risk factors associated with prolonged opioid use after surgery: a systematic review and meta-analysis. JAMA Netw Open. 2020;3(6):e207367. doi: 10.1001/jamanetworkopen.2020.7367 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hah JM, Bateman BT, Ratliff J, Curtin C, Sun E. Chronic opioid use after surgery: implications for perioperative management in the face of the opioid epidemic. Anesth Analg. 2017;125(5):1733-1740. doi: 10.1213/ANE.0000000000002458 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wunsch H, Wijeysundera DN, Passarella MA, Neuman MD. Opioids prescribed after low-risk surgical procedures in the United States, 2004-2012. JAMA. 2016;315(15):1654-1657. doi: 10.1001/jama.2016.0130 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lentine KL, Yuan H, Tuttle-Newhall JE, et al. Quantifying prognostic impact of prescription opioid use before kidney transplantation through linked registry and pharmaceutical claims data. Transplantation. 2015;99(1):187-196. doi: 10.1097/TP.0000000000000248 [DOI] [PubMed] [Google Scholar]
- 8.Lentine KL, Lam NN, Naik AS, et al. Prescription opioid use before and after kidney transplant: implications for posttransplant outcomes. Am J Transplant. 2018;18(12):2987-2999. doi: 10.1111/ajt.14714 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Wise B, Wilson LZ, Taber DJ, Pilch NA, Rohan V, Fleming JN. The impact of pretransplant opioid exposure on healthcare utilization and costs in kidney transplant. Pharmacotherapy. 2021;41(1):6-13. doi: 10.1002/phar.2479 [DOI] [PubMed] [Google Scholar]
- 10.Abbott KC, Fwu CW, Eggers PW, Eggers AW, Kline PP, Kimmel PL. Opioid prescription, morbidity, and mortality in US transplant recipients. Transplantation. 2018;102(6):994-1004. doi: 10.1097/TP.0000000000002057 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Rohan VS, Taber DJ, Patel N, et al. Impact of a multidisciplinary multimodal opioid minimization initiative in kidney transplant recipients. Clin Transplant. 2020;34(10):e14006. doi: 10.1111/ctr.14006 [DOI] [PubMed] [Google Scholar]
- 12.Dualeh SHA, McMurry K, Herman AE, et al. Evaluation of an opioid restrictive pain management initiative in adult kidney transplant recipients. Clin Transplant. 2021;35(7):e14313. doi: 10.1111/ctr.14313 [DOI] [PubMed] [Google Scholar]
- 13.Lichvar A, Campara M, Benedetti E, Benken J. The implementation and impact of an opioid minimization protocol in an urban renal transplant population. Am J Transplant. 2019;19(suppl 3). Accessed May 20, 2022. https://atcmeetingabstracts.com/abstract/the-implementation-and-impact-of-an-opioid-minimization-protocol-in-an-urban-renal-transplant-population [Google Scholar]
- 14.Schwab ME, Braun HJ, Ascher NL, Hirose R. Implementing an opioid reduction protocol in renal transplant recipients. Am J Surg. 2020;220(5):1284-1289. doi: 10.1016/j.amjsurg.2020.06.055 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lam NN, Schnitzler MA, Axelrod DA, et al. Outcome implications of benzodiazepine and opioid co-prescription in kidney transplant recipients. Clin Transplant. 2020;34(9):e14005. doi: 10.1111/ctr.14005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Aalberg JJ, Kimball MD, McIntire TR, McCullen GM. Long-term outcomes of persistent post-operative opioid use: a retrospective cohort study. Ann Surg. Published online January 25, 2022. doi: 10.1097/SLA.0000000000005372 [DOI] [PubMed] [Google Scholar]
- 17.Kulshrestha S, Barrantes F, Samaniego M, Luan FL. Chronic opioid analgesic usage post-kidney transplantation and clinical outcomes. Clin Transplant. 2014;28(9):1041-1046. doi: 10.1111/ctr.12414 [DOI] [PubMed] [Google Scholar]
- 18.Drees D, Tumin D, Miller R, et al. Chronic opioid use and clinical outcomes in lung transplant recipients: a single-center cohort study. Clin Respir J. 2018;12(9):2446-2453. doi: 10.1111/crj.12948 [DOI] [PubMed] [Google Scholar]
- 19.Fleming JN, Taber DJ, Pilch NA, et al. Association of pretransplantation opioid use with graft loss or death in liver transplantation patients with model of end-stage liver disease exceptions. J Am Coll Surg. 2018;226(4):651-659. doi: 10.1016/j.jamcollsurg.2017.12.025 [DOI] [PubMed] [Google Scholar]
- 20.Lentine KL, Shah KS, Kobashigawa JA, et al. Prescription opioid use before and after heart transplant: associations with posttransplant outcomes. Am J Transplant. 2019;19(12):3405-3414. doi: 10.1111/ajt.15565 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable. Morphine Milligram Equivalents (MME) Conversions
eFigure 1. Opioid Information Handout
eFigure 2. Inpatient Opioid Minimization Protocol
eFigure 3. Discharge Opioid Minimization Protocol
Data Sharing Statement
