Abstract
Background:
Minimal pain and opioid use after operative treatment for pediatric supracondylar humeral fractures have been previously described; however, opioid-prescribing practices in the United States remain variable. We hypothesized that children without an opioid prescription would report similar postoperative pain compared with children prescribed opioids following closed reduction and percutaneous pinning (CRPP) of supracondylar humeral fractures.
Methods:
Children who were 3 to 12 years of age and were undergoing CRPP for a closed supracondylar humeral fracture were prospectively enrolled in a multicenter, comparative study. Following a standardized dosing protocol, oxycodone, ibuprofen, and acetaminophen were prescribed at 2 hospitals (opioid cohort), and 2 other hospitals prescribed ibuprofen and acetaminophen alone (non-opioid cohort). The children’s medication use and the daily pain that they experienced (scored on the Wong-Baker FACES Scale) were recorded at postoperative days 1 to 7, 10, 14, and 21, using validated text-message protocols. Based on an a priori power analysis, at least 64 evaluable subjects were recruited per cohort.
Results:
A total of 157 patients were evaluated (81 [52%] in the opioid cohort and 76 [48%] in the non-opioid cohort). The median age at the time of the surgical procedure was 6.2 years, and 50% of the subjects were male. The mean postoperative pain scores were low overall (<4 of 10), and there were no significant differences in pain ratings between cohorts at any time point. No patient demographic or injury characteristics were correlated with increased pain or medication use. Notably, of the 81 patients in the opioid cohort, 28 (35%) took no oxycodone and 40 (49%) took 1 to 3 total doses across the postoperative period. Patients rarely took opioids after postoperative day 2. A single patient in the non-opioid cohort (1 [1%] of 76) received a rescue prescription of opioids after presenting to the emergency department with postoperative cast discomfort.
Conclusions:
Non-opioid analgesia following CRPP for pediatric supracondylar humeral fractures was equally effective as opioid analgesia. When oxycodone was prescribed, 84% of children took 0 to 3 total doses, and opioid use fell precipitously after postoperative day 2. To improve opioid stewardship, providers and institutions can consider discontinuing the routine prescription of opioids following this procedure.
Level of Evidence:
Therapeutic Level II. See Instructions for Authors for a complete description of levels of evidence.
Supracondylar humeral fractures are a common pediatric injury with excellent clinical outcomes1,2. Postoperative pain and opioid use following closed reduction and percutaneous pinning (CRPP) for pediatric supracondylar humeral fractures have been reported in single-center and retrospective cohorts, with multiple reports questioning the need for opioids following this procedure3-6. Standardizing postoperative medication orders in this population has been shown to reduce variation in orthopaedic resident prescribing and to decrease the amount of unnecessary opioids7. Despite a growing body of retrospective evidence, high-quality prospective studies have been limited, and prescription variation still exists in the United States at the national, institutional, and provider levels.
Promoting opioid stewardship without compromising postoperative analgesia remains a focus for quality, safety, and value initiatives in pediatric orthopaedics. Multiple investigations have shown that children are overprescribed opioids following orthopaedic procedures, increasing the potential for opioid diversion into communities, misuse, and accidental poisonings8-11. A 2016 report in JAMA Pediatrics found that pediatric hospitalizations due to opioid poisonings rose nearly twofold from 1999 to 2012 and identified young children as an age group at increased risk12. Finally, although national opioid-prescribing rates have decreased in recent years, inconsistencies still exist, highlighting the need for additional policy and practice initiatives13.
The purpose of this investigation was to compare the efficacy of opioid analgesia with that of non-opioid analgesia following CRPP for pediatric supracondylar humeral fractures. The primary aim was to develop evidence-based guidelines for postoperative analgesia in this population. We hypothesized that children taking non-opioid analgesia would report similar postoperative pain control compared with children taking opioid analgesics following this procedure.
Materials and Methods
A prospective, comparative cohort study was conducted at 4 tertiary referral children’s hospitals in the United States from 2021 to 2022. Institutional review board approval was obtained at each hospital.
Subject Recruitment
Children who were 3 to 12 years of age and presented with a closed, modified-Gartland Type-II or III supracondylar humeral fracture were eligible for participation. Only children who actually underwent CRPP were included; those who underwent conversion to open reduction and internal fixation (ORIF) intraoperatively were excluded. Mini-incisions used to guide medial pin placement were not considered ORIF and such cases remained eligible. Children with open fractures, vascular injuries requiring exploration or repair, polytrauma, and refractures were excluded. Only English-speaking families were enrolled, as the questionnaires were only validated in English. Informed consent from parents or legal guardians was obtained.
Medication Dosing Protocol
Following the local standards of care, all children at 2 hospitals were prescribed oxycodone, acetaminophen, and ibuprofen at discharge; these children comprised the opioid cohort. Standard instructions were provided to take oxycodone only as needed for breakthrough pain. It was the local standard of care at the remaining 2 hospitals to prescribe acetaminophen and ibuprofen alone; enrolled children at these institutions comprised the non-opioid cohort. A standardized dosing regimen was developed according to the existing prescription trends at all 4 hospitals (Table I). Children prescribed pain medications other than ibuprofen, acetaminophen, and oxycodone were excluded from the final analysis.
TABLE I.
Outpatient Medication Dosing Protocol
| Medication | Dosage | Liquid |
|---|---|---|
| Ibuprofen | 7.5 to 10 mg per kg every 6 hours as needed | Dispense approximately 237-mL bottle |
| Acetaminophen | 10 to 15 mg per kg every 4 hours as needed | Dispense approximately 237-mL bottle |
| Oxycodone | 0.05 to 0.1 mg per kg every 4 hours as needed for breakthrough pain | 7 doses in the initial prescription |
Data Collection
Caregivers received text-message questionnaires on postoperative days 1 to 7, 10, 14, and 21. The distribution of questionnaires was automated through REDCap (Research Electronic Data Capture; Vanderbilt University), using validated text message-based protocols3,14,15. At 7:30 a.m. local time, caregivers reported the pain that their child was experiencing (from 0 to 10) on the Wong-Baker FACES Pain Rating Scale. At 6 p.m. local time, caregivers again reported the pain that their child was experiencing as well as the doses of oxycodone (if applicable), ibuprofen, and acetaminophen taken in the previous 24 hours. Children’s morning and evening pain scores were averaged to a single daily mean.
Perioperative data were abstracted from the medical record and included age, sex, race, ethnicity, height, weight, body mass index (BMI), nerve injury, fracture type, inpatient pain rating, number of pins, pin configuration, inpatient medications administered, procedure start and end times, postoperative immobilization, length of hospitalization, and complications.
Statistical Analysis
An a priori power analysis was performed. To detect differences in mean pain with a medium effect size, α = 0.05, and 80% power, a minimum of 64 evaluable subjects were required per cohort. Pearson chi-square tests and Fisher exact tests were used to compare categorical variables. Spearman rho and point biserial correlations were used to examine relationships between demographic characteristics, pain scores, and medication use. Tests of normality were used to evaluate the distributions of continuous variables. For non-normally distributed variables, the median and interquartile range (IQR) were reported. Nonparametric tests were used to compare pain and medication data across the cohorts; however, for ease of visualization and clinical application for the reader, the figures were created with mean values. To account for repeated measures, a mixed-effects model was performed. Analyses were performed using SPSS Statistics (version 28; IBM) and STATA version 16 (StataCorp), with the significance threshold set at p < 0.05.
Source of Funding
This investigation was funded by the Pediatric Orthopaedic Society of North America (POSNA) 2020 Angela S.M. Kuo Memorial Award.
Results
Patient and Injury Characteristics
From May 2021 to August 2022, 190 children were recruited (44.7% in the non-opioid cohort and 55.3% in the opioid cohort). Twelve children were later excluded because of intraoperative or postoperative factors that violated eligibility criteria. An additional 21 families were nonresponsive to questionnaires and were excluded. Ultimately, 157 children were evaluated (48.4% in the non-opioid cohort and 51.6% in the opioid cohort) (Fig. 1).
Fig. 1.
Study recruitment flowchart.
The median age at the time of the surgical procedure was 6.2 years (IQR, 5.1 to 7.7 years), and 50.3% of the children were male. There were no significant differences in age, sex, race, or BMI between the cohorts (Table II). The non-opioid cohort included more Hispanic and Latino patients (33%) than the opioid cohort (11%) (p < 0.001); however, across the study population, ethnicity was not correlated with children’s mean postoperative pain scores or total medication use. Four children (2.5%) had a nerve injury at the initial presentation, and 1 child (0.6%) had postoperative median nerve neurapraxia (Table II); all recovered uneventfully. No patient demographic variables (age, sex, race, ethnicity, BMI) or injury characteristics (modified-Gartland classification, presence of nerve injury) had significant correlations with postoperative pain or medication use.
TABLE II.
Patient Demographic and Operative Characteristics
| All Patients (N = 157) | Non-Opioid Cohort (N = 76) | Opioid Cohort (N = 81) | |
|---|---|---|---|
| Demographic characteristics | |||
| Age at time of surgery* (yr) | 6.2 (5.1 to 7.7) | 6.0 (5.0 to 7.6) | 6.5 (5.4 to 7.8) |
| Sex† | |||
| Male | 79 (50.3%) | 39 (51.3%) | 40 (49.4%) |
| Female | 78 (49.7%) | 37 (48.7%) | 41 (50.6%) |
| Race†‡ | |||
| White | 107 (77.0%) | 48 (78.7%) | 59 (75.6%) |
| Black | 5 (3.6%) | 0 (0%) | 5 (6.4%) |
| Asian | 14 (10.1%) | 9 (14.8%) | 5 (6.4%) |
| Multiracial | 4 (2.9%) | 2 (3.3%) | 2 (2.6%) |
| Other | 9 (6.5%) | 2 (3.3%) | 7 (9.0%) |
| Ethnicity†‡ | |||
| Hispanic or Latino | 34 (22.7%) | 25 (35.7%) | 9 (11.3%) |
| Not Hispanic or Latino | 116 (77.3%) | 45 (64.3%) | 71 (88.8%) |
| BMI categories†‡ | |||
| Underweight, <5th percentile | 4 (3.8%) | 2 (4.2%) | 2 (3.4%) |
| Healthy weight, 5th to 84th percentile | 73 (68.9%) | 33 (68.8%) | 40 (69.0%) |
| Overweight, 85th to 95th percentile | 18 (17.0%) | 7 (14.6%) | 11 (19.0%) |
| Obese, >95th percentile | 11 (10.4%) | 6 (12.5%) | 5 (8.6%) |
| BMI percentile* | 61 (28 to 86) | 70 (25 to 85) | 61 (28 to 86) |
| Perioperative characteristics | |||
| Preoperative | |||
| Preoperative length of stay* (hr) | 10.7 (3.9 to 14.2) | 9.5 (2.6 to 14.2) | 11.2 (6.2 to 13.3) |
| Preoperative nerve injury† | |||
| No | 153 (97.5%) | 76 (100%) | 77 (95.1%) |
| Yes | 4 (2.5%) | 0 (0%) | 4 (4.9%) |
| Operative | |||
| Modified-Gartland classification† | |||
| Type II | 67 (42.7%) | 37 (48.7%) | 30 (37.0%) |
| Type III | 90 (57.3%) | 39 (51.3%) | 51 (63.0%) |
| Procedure time* (min) | 30 (23 to 39) | 27 (21 to 36) | 34 (25 to 41) |
| No. of pins† | |||
| 2 | 68 (43.3%) | 37 (48.7%) | 31 (38.3%) |
| 3 | 85 (54.1%) | 38 (50.0%) | 47 (58.0%) |
| ≥4 | 4 (2.5%) | 1 (1.3%) | 3 (3.7%) |
| Pin configuration† | |||
| Lateral entry only | 135 (86.0%) | 61 (80.3%) | 74 (91.4%) |
| Cross-pinning (medial and lateral) | 22 (14.0%) | 15 (19.7%) | 7 (8.6%) |
| Postoperative | |||
| Postoperative length of stay* (hr) | 4.9 (2.8 to 8.1) | 5.7 (2.3 to 11.8) | 4.9 (3.3 to 7.1) |
| Postoperative nerve injury† | |||
| No | 156 (99.4%) | 76 (100%) | 80 (98.8%) |
| Yes | 1 (0.6%) | 0 (0%) | 1 (1.2%) |
| Immobilization† | |||
| Long arm cast | 137 (87.3%) | 65 (85.5%) | 72 (88.9%) |
| Long arm posterior splint | 20 (12.7%) | 11 (14.5%) | 9 (11.1%) |
| Prophylactic cast bivalving† (if applicable) | |||
| No | 65 (47.4%) | 13 (20.0%) | 52 (72.2%) |
| Yes | 72 (52.6%) | 52 (80.0%) | 20 (27.8%) |
The values are given as the median, with the IQR in parentheses.
The values are given as the number of patients, with the percentage in parentheses.
Subjects for whom demographic data were missing from the electronic medical record were not included in these categories in this table.
Operative Characteristics
Pain scores documented at the initial hospital presentation were not significantly different between cohorts (median, 4.0 [IQR, 1.5 to 6.5] in both; p = 0.903). The opioid cohort had longer documented median procedure times (34 minutes) than the non-opioid cohort (27 minutes) (p = 0.006) (Table II). Nearly all fractures required 2 or 3 pins, and the non-opioid cohort had a higher rate of cross-pinning (Table II). Most children (87.3%) had cast immobilization, and there were no differences in the rate of casting compared with that of splinting between the cohorts. The rate of prophylactic cast bivalving in the operating room was higher in the non-opioid cohort (80.0%) compared with the opioid cohort (27.8%) (p < 0.001) (Table II). Most children (86.6%) received ketorolac for perioperative analgesia; however, multimodal analgesia strategies did vary between cohorts (Table III).
TABLE III.
Perioperative Analgesia*
| Medication Name | Non-Opioid Cohort† (N = 76) | Opioid Cohort† (N = 81) | P Value‡ |
|---|---|---|---|
| Ketorolac | 67 (88.2%) | 69 (85.2%) | 0.584 |
| Morphine or fentanyl | 67 (88.2%) | 59 (72.8%) | 0.016 |
| Intravenous acetaminophen | 35 (46.1%) | 19 (23.5%) | 0.003 |
| Other§ | 7 (9.2%) | 45 (55.6%) | <0.001 |
| Dexamethasone | 0 | 10 | |
| Oral acetaminophen | 0 | 9 | |
| Oral oxycodone | 1 | 12 | |
| Alfentanil | 1 | 1 | |
| Bupivacaine | 1 | 6 | |
| Hydromorphone | 5 | 2 | |
| Lidocaine | 0 | 8 |
Perioperative analgesia was defined as medications received during the operation or within 1 hour after leaving the operating room.
The values are given as the number of patients, with or without the percentage in parentheses.
Significant p values are shown in bold.
Some patients had >1 other medication.
There were no significant correlations between postoperative pain and operative time, number of pins, pin configuration, or immobilization type. There were similarly no correlations between these variables and medication use. Similarly, prophylactic cast bivalving had no significant correlation with early (postoperative days 1 to 3) postoperative pain or medication use.
Postoperative Pain and Analgesia
There were no significant differences in pain ratings between the non-opioid cohort and the opioid cohort from postoperative days 1 to 21 (Fig. 2, Table IV). A repeated-measures mixed-model multivariable analysis was performed to compare the effect of access to opioids on the mean postoperative pain that the subjects experienced. Variables with significant differences between the cohorts were included in the model. The interactions of opioid access and time were examined in our model. There was no significant difference in postoperative pain experienced by the opioid cohort compared with the non-opioid cohorts (p = 0.569) (Table V). That is, when controlled for clinical and demographic differences, access to an opioid did not significantly impact the mean postoperative pain experienced by the subjects (Table V).
Fig. 2.
Mean daily pain ratings in the non-opioid cohort and the opioid cohort.
TABLE IV.
Postoperative Pain Ratings for Each Cohort
| Postoperative Day | Non-Opioid Cohort | Opioid Cohort | P Value | ||
|---|---|---|---|---|---|
| Pain Rating* | Response Rate | Pain Rating* | Response Rate | ||
| 1 | 3.3 | 84.2% | 3.6 | 86.4% | 0.361 |
| 2 | 3.2 | 92.1% | 2.4 | 92.6% | 0.060 |
| 3 | 2.4 | 88.2% | 1.8 | 92.6% | 0.088 |
| 4 | 1.7 | 86.8% | 1.6 | 92.6% | 0.802 |
| 5 | 1.3 | 89.5% | 1.0 | 88.9% | 0.627 |
| 6 | 1.1 | 84.2% | 0.9 | 87.7% | 0.864 |
| 7 | 0.7 | 84.2% | 0.9 | 88.9% | 0.114 |
| 10 | 0.7 | 82.9% | 0.5 | 81.5% | 0.585 |
| 14 | 0.1 | 78.9% | 0.2 | 82.7% | 0.310 |
| 21 | 0.2 | 75.0% | 0.2 | 79.0% | 0.969 |
The values are given as the mean.
TABLE V.
Results of Repeated-Measures Mixed-Model Multivariable Analysis
| Predictors for Subjects’ Mean Postoperative Pain* | Coefficient† | P Value |
|---|---|---|
| Prophylactic cast bivalving | ||
| No | Reference | — |
| Yes | 0.38 ± 0.27 (−0.15 to 0.91) | 0.160 |
| Pin configuration | ||
| Lateral entry only | Reference | — |
| Cross-pinning | −0.26 ± 0.37 (−0.99 to 0.47) | 0.484 |
| Operative time, per minute | 0.01 ± 0.01 (−0.01 to 0.02) | 0.482 |
| Patient ethnicity | ||
| Not Hispanic or Latino | Reference | — |
| Hispanic or Latino | 0.31 ± 0.28 (−0.23 to 0.85) | 0.267 |
| Access to opioids | ||
| No (i.e., non-opioid cohort) | Reference | — |
| Yes (i.e., opioid cohort) | 0.12 ± 0.35 (−0.48 to 0.87) | 0.569 |
Patient demographic and operative variables that were significant on bivariate analysis (see the Results section) were included in the multivariable analysis. The model was also tested for interactions of the variable access to opioids with the time variable (postoperative day), and we found no collinearity.
The values are given as the coefficient and the standard error, with the 95% confidence interval in parentheses.
Upper outliers were defined as pain scores above the 95th percentile of responders in our data set. On postoperative days 1 to 2, pain scores of >6.0 out of 10 were found to be outliers. On postoperative days 3 to 7, pain scores of >4.0 were outliers; after postoperative day 7, pain scores of >2.0 were outliers. There were no significant differences in the number of outliers between the non-opioid cohort and the opioid cohort.
Postoperative analgesia followed similar trends as pain, with most medications taken on postoperative days 1 to 2 (Fig. 3). On postoperative day 1, the non-opioid cohort took a mean of 0.7 more doses of ibuprofen (p = 0.030) (Table VI) and 0.4 fewer doses of acetaminophen (p < 0.001) (Table VII) compared with the opioid cohort. Children in the opioid cohort took a mean of 0.9 doses of oxycodone on postoperative day 1 (mean morphine milligram equivalents [MME] per day = 0.1) (Table VIII). After postoperative day 1, there were no significant differences in ibuprofen use between the cohorts (Table VI). There were significant but minimal differences in acetaminophen use between the cohorts on postoperative days 2 and 6 (Table VII).
Fig. 3.
Mean daily use of oxycodone (OXY), ibuprofen (IBU), and acetaminophen (ACET) in the non-opioid cohort and the opioid cohort.
TABLE VI.
Mean Daily Ibuprofen Use for Each Cohort
| Postoperative Day | Non-Opioid Cohort | Opioid Cohort | P Value† | ||
|---|---|---|---|---|---|
| Ibuprofen Doses* | Response Rate | Ibuprofen Doses* | Response Rate | ||
| 1 | 3.1 | 72.4% | 2.4 | 76.5% | 0.030 |
| 2 | 2.3 | 81.6% | 2.3 | 84.0% | 0.979 |
| 3 | 1,9 | 78.9% | 1.7 | 82.7% | 0.535 |
| 4 | 1.1 | 78.9% | 1.2 | 86.4% | 0.458 |
| 5 | 1.0 | 81.6% | 1.0 | 81.5% | 0.940 |
| 6 | 0.7 | 76.3% | 0.7 | 82.7% | 0.723 |
| 7 | 0.6 | 78.9% | 0.6 | 79.0% | 0.584 |
| 10 | 0.3 | 76.3% | 0.3 | 72.8% | 0.479 |
| 14 | 0.1 | 75.0% | 0.1 | 74.1% | 0.910 |
| 21 | 0.1 | 71.1% | 0.0 | 70.4% | 0.073 |
The values are given as the mean.
Values shown in bold are significant.
TABLE VII.
Mean Daily Acetaminophen Use for Each Cohort
| Postoperative Day | Non-Opioid Cohort | Opioid Cohort | P Value† | ||
|---|---|---|---|---|---|
| Acetaminophen Doses* | Response Rate | Acetaminophen Doses* | Response Rate | ||
| 1 | 2.2 | 76.5% | 2.6 | 72.4% | <0.001 |
| 2 | 1.9 | 85.2% | 2.3 | 81.6% | 0.01 |
| 3 | 1.4 | 82.7% | 1.4 | 82.9% | 0.279 |
| 4 | 1.2 | 86.4% | 1.0 | 73.7% | 0.973 |
| 5 | 0.7 | 81.5% | 0.8 | 80.3% | 0.856 |
| 6 | 0.4 | 82.7% | 0.5 | 73.7% | 0.038 |
| 7 | 0.5 | 79.0% | 0.4 | 75.0% | 0.118 |
| 10 | 0.2 | 74.1% | 0.4 | 72.4% | 0.187 |
| 14 | 0.0 | 75.3% | 0.1 | 73.7% | 0.349 |
| 21 | 0.0 | 70.4% | 0.0 | 59.2% | 0.335 |
The values are given as the mean.
Values shown in bold are significant.
TABLE VIII.
Mean Daily Oxycodone Use for the Opioid Cohort
| Postoperative Day | Oxycodone Doses* | Response Rate | Mean MME per Day† |
|---|---|---|---|
| 1 | 0.9 | 77.8% | 0.1 |
| 2 | 0.6 | 85.2% | 0.1 |
| 3 | 0.2 | 84.0% | <0.1 |
| 4 | 0.3 | 86.4% | <0.1 |
| 5 | 0.1 | 81.5% | <0.1 |
| 6 | 0.0 | 82.7% | <0.1 |
| 7 | 0.2 | 80.2% | <0.1 |
| 10 | 0.0 | 74.1% | <0.1 |
| 14 | 0.0 | 75.3% | <0.1 |
| 21 | 0.0 | 70.4% | <0.1 |
The values are given as the mean.
Calculated assuming that the maximum study dosage of 0.1 mg per kg per dose of oxycodone was taken. MME = morphine milligram equivalents.
In the opioid cohort, the mean daily oxycodone use fell to <0.5 doses after postoperative day 2 (Fig. 3). Twenty-eight children (34.6%) took no oxycodone, and 40 children (49.4%) took 1 to 3 total doses (Fig. 4). On average, 75.7% of prescribed opioids were unused. Only 3 (3.7%) of the 81 children reported taking >7 doses of oxycodone, although, on chart review, none of these families requested oxycodone refills. Only 1 child (1.3%) in the non-opioid cohort, a 9-year-old girl with a Type-II supracondylar humeral fracture, required rescue opioids. The child presented to the emergency department on postoperative day 1 reporting stinging pain, worst in the hand and wrist. To address the symptoms, the cast was bivalved and the patient received an oxycodone prescription. Her medication use was not consistently recorded, but, at the first outpatient orthopaedics visit, the family reported that she had discontinued all medications by postoperative day 5.
Fig. 4.
Cumulative postoperative medication use among study subjects. Total oxycodone (OXY), ibuprofen (IBU), and acetaminophen (ACET) taken from postoperative days (PODs) 1 to 21 are reported. The percentages reported above each bar represent the cumulative percentage of patients who took that number of doses, or fewer, across the study period.
Including the cast pain that required rescue opioids, a total of 6 postoperative complications (3.8%) were noted. Cast bivalving due to postoperative swelling was required in 2 additional children (1 in the opioid cohort and 1 in the non-opioid cohort). The non-opioid cohort had 3 superficial pin-site infections that required antibiotics; 1 of these patients returned to the operating room on postoperative day 15 for irrigation and debridement. Clinical healing was within the expected course, occurring within 6 weeks after the date of the surgical procedure, for all patients.
Discussion
Postoperative Pain: Expectations and Exceptions to the Rule
In accordance with previous reports3,5, we found that children had overall mild pain following CRPP for supracondylar humeral fractures, with the greatest pain and analgesia needs reported in the first 48 hours. There were no significant differences in pain between the opioid cohort and the non-opioid cohort from postoperative days 1 to 21. On multivariable analysis controlling for time and clinical and demographic variables, access to an opioid prescription did not impact the mean postoperative pain that subjects experienced. The greatest observed difference in pain between the cohorts was noted on postoperative day 2 and was <1 point on the Wong-Baker FACES scale; this fails to reach the minimal clinically important difference (MCID) for this scale16,17. Taken together, these results support retaining our null hypothesis that children with an opioid prescription have similar postoperative pain compared with children without opioids prescribed following this procedure.
Whether or not children were prescribed an opioid, daily postoperative pain was low (mean, <4 of 10), even on postoperative day 1. Surgeons who wish to continue prescribing opioids should recognize that pain is low after postoperative day 3, and opioid prescriptions should be accordingly limited. As a secondary finding, postoperative pain was not related to the modified-Gartland fracture type, number or configuration of pins, postoperative immobilization type, or any other patient demographic variables. This evidence reaffirms prior retrospective reports5. Ultimately, clinicians need not vary postoperative medication prescriptions based on patient demographic or injury characteristics.
Pain is a common driver of unplanned emergency department visits after pediatric ambulatory surgery18. Establishing evidence-based thresholds for normative pain after procedures helps clinicians and families to set expectations for postoperative pain management. We found that mean pain scores of >6 out of 10 across the first 48 hours were outliers; this aligns with previous reports3,5. Clinicians should be advised that persistent pain of >6 out of 10 after CRPP is abnormal; these children merit a callback to assess for swelling, cast tightness, elevated compartment pressures, or other potentially serious complications. In this study, only 1 child in the non-opioid cohort required rescue opioids; however, that child’s symptoms were most consistent with cast discomfort due to swelling. In instances of immobilization discomfort, family education on the recommended amount and frequency of ibuprofen and acetaminophen doses, strict elevation, and other pain relief strategies may avoid unnecessary visits and opioids.
Postoperative Analgesia: A Call for Opioid Stewardship
At the provider level, reducing the amount of opioids prescribed, via standardized order sets and family education, is critical to reduce opioid misuse. Previous work by this study team demonstrated that children take <25% of prescribed opioids after CRPP (mean [and standard deviation], 4.8 ± 5.6 doses used compared with 19.8 ± 7.1 doses prescribed)3. Expanding on these results, our current investigation shows that, when only 7 doses of oxycodone were provided with instructions to take a dose only for breakthrough pain, 84% of children took ≤3 total doses and 76% of prescribed doses remained unused. Although pediatric volumes of liquid oxycodone are unlikely to be abused, there remains the potential for diversion or accidental ingestion and poisoning in young children, especially given that families tend to store unused opioids in unlocked locations with no immediate plan for disposal19. Counseling on safe opioid disposal methods and institutional pathways to return unused opioids should remain a focus for pediatric orthopaedics.
Regarding the importance of family education, children in this study were instructed to take ibuprofen every 6 hours (4 daily doses) and acetaminophen every 4 hours (6 daily doses). However, medication tracking revealed that, on average, children in both cohorts were taking fewer than the recommended maximum daily doses at every time point. This suggests that families were not maximizing non-opioid analgesia before administering oxycodone. This highlights the importance of standardized order sets and counseling by providers regarding appropriate postoperative analgesia strategies, both of which have been demonstrated to reduce opioid use after orthopaedic surgery7,20-22. When families are informed partners in postoperative pain management, unnecessary opioid exposure in children may be avoided.
At the policy level, practices such as prescription drug monitoring programs, state-mandated preoperative opioid informed consent, and electronic prescribing are increasingly widespread23. Some reports in the literature have indicated positive effects of these programs in decreasing rates of opioid prescriptions overall24,25 and specifically in pediatric orthopaedics26,27, reducing overdoses28, and decreasing prescriptions for unnecessarily high doses29. However, others have reported mixed results and poor implementation30-33. Our investigation showed that non-opioid analgesia was effective following CRPP for pediatric supracondylar humeral fractures; thus, electronic prescribing pathways and standardized institutional protocols could be updated such that oxycodone prescriptions are not automatically prompted for these patients.
Limitations and Future Directions
A limitation of this study was the lack of randomization and the potential for selection bias. Selection bias may have been partially mitigated by selecting 4 geographically diverse hospitals. Nevertheless, the nonrandomized design of the study resulted in cohort-level differences in patient characteristics (ethnicity) and surgical characteristics (procedure time, pin configuration, decision for a postoperative cast bivalving). The overall difference in ethnicity between the cohorts likely reflects underlying demographic differences between the populations served by the 4 hospitals. Decisions for cross-pinning, postoperative immobilization, and prophylactic cast bivalving were outside the scope of our study design and were left to the treating surgeon’s discretion. However, the cohort-level differences in these variables had no correlations with postoperative pain or medication use and were not significant on multivariable analysis; thus, they were unlikely to have impacted the study results.
Additionally, this investigation was not designed to provide guidelines for inpatient analgesia. In this study, perioperative analgesia varied between cohorts, reflecting differences in anesthesia practices. This lack of standardization may have impacted postoperative pain, particularly on postoperative day 1. Additional investigations in these areas could permit expanding our recommendations to include standardization of inpatient medications.
A final consideration is generalizability. It is unlikely that CRPP for other upper-extremity fractures would be more painful than CRPP for supracondylar humeral fractures. Thus, it is plausible that there is no utility for routine opioid prescriptions for these patients. In the orthopaedic literature, adult patients undergoing most upper-extremity ambulatory procedures demonstrated low opioid needs and high overprescription rates34; these procedures in children would benefit from further prospective investigation. Research into more invasive pediatric orthopaedic procedures and adolescent populations may demonstrate additional areas for improving opioid stewardship. Investigation into these areas will ultimately improve the quality, safety, and value of surgical care in pediatric orthopaedics.
Conclusion
This prospective, multicenter, comparative cohort study supports the hypothesis that, compared with children who take opioid analgesia, children who take non-opioid analgesia report similar postoperative pain following CRPP for supracondylar humeral fractures. No substantial benefit to opioid analgesia was found in this population. To improve opioid stewardship for this archetypical pediatric procedure, providers and institutions should consider discontinuing the routine prescription of opioids following CRPP for supracondylar humeral fractures.
Footnotes
Investigation performed at the Division of Orthopaedics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania
A commentary by R. Dale Blasier, MD, is linked to the online version of this article.
Disclosure: The Disclosure of Potential Conflicts of Interest forms are provided with the online version of the article (http://links.lww.com/JBJS/H740).
References
- 1.Holt JB, Glass NA, Shah AS. Understanding the epidemiology of pediatric supracondylar humeral fractures in the United States: identifying opportunities for intervention. J Pediatr Orthop. 2018. May/Jun;38(5):e245-51. [DOI] [PubMed] [Google Scholar]
- 2.Howard A, Mulpuri K, Abel MF, Braun S, Bueche M, Epps H, Hosalkar H, Mehlman CT, Scherl S, Goldberg M, Turkelson CM, Wies JL, Boyer K; American Academy of Orthopaedic Surgeons. The treatment of pediatric supracondylar humerus fractures. J Am Acad Orthop Surg. 2012. May;20(5):320-7. [DOI] [PubMed] [Google Scholar]
- 3.Nelson SE, Adams AJ, Buczek MJ, Anthony CA, Shah AS. Postoperative pain and opioid use in children with supracondylar humeral fractures: balancing analgesia and opioid stewardship. J Bone Joint Surg Am. 2019. Jan 16;101(2):119-26. [DOI] [PubMed] [Google Scholar]
- 4.Adams AJ, Buczek MJ, Flynn JM, Shah AS. Perioperative ketorolac for supracondylar humerus fracture in children decreases postoperative pain, opioid usage, hospitalization cost, and length-of-stay. J Pediatr Orthop. 2019. Jul;39(6):e447-51. [DOI] [PubMed] [Google Scholar]
- 5.Stillwagon MR, Feinstein S, Nichols B, Andrews PN, Vergun AD. Pain control and medication use in children following closed reduction and percutaneous pinning of supracondylar humerus fractures: are we still overprescribing opioids? J Pediatr Orthop. 2020. Nov/Dec;40(10):543-8. [DOI] [PubMed] [Google Scholar]
- 6.Swanson CE, Chang K, Schleyer E, Pizzutillo PD, Herman MJ. Postoperative pain control after supracondylar humerus fracture fixation. J Pediatr Orthop. 2012. Jul-Aug;32(5):452-5. [DOI] [PubMed] [Google Scholar]
- 7.Goodloe JB, Bailey EP, Luce LT, Corrigan CS, Dow MA, Barfield WR, Murphy RF. A standardized order-set improves variability in opioid discharge prescribing patterns after surgical fixation of pediatric supracondylar humerus fractures. J Surg Educ. 2021. Sep-Oct;78(5):1660-5. [DOI] [PubMed] [Google Scholar]
- 8.Sabatino MJ, Kunkel ST, Ramkumar DB, Keeney BJ, Jevsevar DS. Excess opioid medication and variation in prescribing patterns following common orthopaedic procedures. J Bone Joint Surg Am. 2018. Feb 7;100(3):180-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Pruitt LCC, Swords DS, Russell KW, Rollins MD, Skarda DE. Prescription vs. consumption: opioid overprescription to children after common surgical procedures. J Pediatr Surg. 2019. Nov;54(11):2195-9. [DOI] [PubMed] [Google Scholar]
- 10.Metz AK, Tomasevich KM, Froerer DL, Rosenthal RM, Featherall J, Aoki SK. Postoperative pain medication utilization in pediatric patients undergoing sports orthopaedic surgery: characterizing patient usage patterns and opioid retention. J Am Acad Orthop Surg Glob Res Rev. 2022. Oct 24;6(10):e22.00206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Keil LG, Sullivan MH, Dadoo S, Stillwagon MR, Vergun AD. How much opioid do kids actually need? A prospective study of analgesic prescribing and postdischarge opioid use among pediatric patients with operative fractures. J Pediatr Orthop. 2021. Nov-Dec 1;41(10):e871-6. [DOI] [PubMed] [Google Scholar]
- 12.Gaither JR, Leventhal JM, Ryan SA, Camenga DR. National trends in hospitalizations for opioid poisonings among children and adolescents, 1997 to 2012. JAMA Pediatr. 2016. Dec 1;170(12):1195-201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tomaszewski DM, Arbuckle C, Yang S, Linstead E. Trends in opioid use in pediatric patients in US emergency departments from 2006 to 2015. JAMA Netw Open. 2018. Dec 7;1(8):e186161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Anthony CA, Lawler EA, Ward CM, Lin IC, Shah AS. Use of an automated mobile phone messaging robot in postoperative patient monitoring. Telemed J E Health. 2018. Jan;24(1):61-6. [DOI] [PubMed] [Google Scholar]
- 15.Anthony CA, Volkmar AJ, Shah AS, Willey M, Karam M, Marsh JL. Communication with orthopedic trauma patients via an automated mobile phone messaging robot. Telemed J E Health. 2018. Jul;24(7):504-9. [DOI] [PubMed] [Google Scholar]
- 16.Williamson A, Hoggart B. Pain: a review of three commonly used pain rating scales. J Clin Nurs. 2005. Aug;14(7):798-804. [DOI] [PubMed] [Google Scholar]
- 17.Tsze DS, Hirschfeld G, von Baeyer CL, Bulloch B, Dayan PS. Clinically significant differences in acute pain measured on self-report pain scales in children. Acad Emerg Med. 2015. Apr;22(4):415-22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sawhney M, VanDenKerkhof EG, Goldstein DH, Wei X, Pare G, Mayne I, Tranmer J. Emergency department use and hospital admission in children following ambulatory surgery: a retrospective population-based cohort study. BMJ Paediatr Open. 2021. Nov 23;5(1):e001188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Delamerced A, Zonfrillo MR, Monteiro K, Watson-Smith D, Wills HE. Factors affecting opioid management for injured children after hospital discharge. J Pediatr Surg. 2021. Mar;56(3):506-11. [DOI] [PubMed] [Google Scholar]
- 20.Vincent S, Paskey T, Critchlow E, Mann E, Chapman T, Abboudi J, Jones C, Kirkpatrick W, Namdari S, Hammoud S, Ilyas AM. Prospective randomized study examining preoperative opioid counseling on postoperative opioid consumption after upper extremity surgery. Hand (N Y). 2022. Mar;17(2):200-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jones K, Engler L, Fonte E, Farid I, Bigham MT. Opioid reduction through postoperative pain management in pediatric orthopedic surgery. Pediatrics. 2021. Dec 1;148(6):e2020001487. [DOI] [PubMed] [Google Scholar]
- 22.Holman JE, Stoddard GJ, Horwitz DS, Higgins TF. The effect of preoperative counseling on duration of postoperative opiate use in orthopaedic trauma surgery: a surgeon-based comparative cohort study. J Orthop Trauma. 2014. Sep;28(9):502-6. [DOI] [PubMed] [Google Scholar]
- 23.Soelberg CD, Brown RE, Jr, Du Vivier D, Meyer JE, Ramachandran BK. The US opioid crisis: current federal and state legal issues. Anesth Analg. 2017. Nov;125(5):1675-81. [DOI] [PubMed] [Google Scholar]
- 24.Danovich D, Greenstein J, Chacko J, Hahn B, Ardolic B, Ilyaguyev B, Berwald N. Effect of New York State electronic prescribing mandate on opioid prescribing patterns. J Emerg Med. 2019. Aug;57(2):156-61. [DOI] [PubMed] [Google Scholar]
- 25.Horvat CM, Martin B, Wu L, Fabio A, Empey PE, Hagos F, Bigelow S, Kantawala S, Au AK, Kochanek PM, Clark RSB. Opioid e-prescribing trends at discharge in a large pediatric health system. J Opioid Manag. 2019. Mar/Apr;15(2):119-27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Malige A, Bram JT, Maguire KJ, McNeely LW, Ganley TJ, Williams BA. Decreased prescribing of postoperative opioids in pediatric ACL reconstruction: treatment trends at a single center. Orthop J Sports Med. 2021. Feb 1;9(2):2325967120979993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Williams BA, Magee LC, Makarewich CA, Swarup I, McNeely LW, Shah AS. Preoperative opioid informed consent and prescribing practices in children undergoing orthopaedic trauma surgery. J Am Acad Orthop Surg Glob Res Rev. 2022. Jan 24;6(1):e21.00309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Abouk R, Powell D. Can electronic prescribing mandates reduce opioid-related overdoses? Econ Hum Biol. 2021. Aug;42:101000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Gooptu A, Smith-Ray R, Singh T, John C. PNS185 E-prescribing mandate in Maine reduces high opioid dosing. Value in Health. 2019;22:S316. [Google Scholar]
- 30.Rhodes E, Wilson M, Robinson A, Hayden JA, Asbridge M. The effectiveness of prescription drug monitoring programs at reducing opioid-related harms and consequences: a systematic review. BMC Health Serv Res. 2019. Nov 1;19(1):784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Everson J, Cheng AK, Patrick SW, Dusetzina SB. Association of electronic prescribing of controlled substances with opioid prescribing rates. JAMA Netw Open. 2020. Dec 1;3(12):e2027951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Deiana C, Giua L. The intended and unintended effects of opioid policies on prescription opioids and crime. BE J Econ Anal Policy. 2021;21(2):751-92. [Google Scholar]
- 33.Brady JE, Wunsch H, DiMaggio C, Lang BH, Giglio J, Li G. Prescription drug monitoring and dispensing of prescription opioids. Public Health Rep. 2014. Mar-Apr;129(2):139-47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kim N, Matzon JL, Abboudi J, Jones C, Kirkpatrick W, Leinberry CF, Liss FE, Lutsky KF, Wang ML, Maltenfort M, Ilyas AM. A prospective evaluation of opioid utilization after upper-extremity surgical procedures: identifying consumption patterns and determining prescribing guidelines. J Bone Joint Surg Am. 2016. Oct 19;98(20):e89. [DOI] [PubMed] [Google Scholar]