The role of a multidisciplinary approach to opioid reduction for patients with operatively treated upper extremity fractures

Abstract

Background

The aim of this study was to determine whether a multimodal program for opioid reduction in upper extremity fracture patients decreased (1) inpatient opioid consumption during post-operative days (PODs) 1-4 and (2) outpatient opioid prescription volume.

Methods

We conducted a retrospective comparative study of patients with isolated upper extremity fractures admitted to the inpatient orthopedic unit at an academic level 1 trauma center in 2018 and 2020 (pre- and post-program implementation). We calculated total inpatient opioid consumption (measured in morphine milligram equivalents [MMEs]) during POD 1-4 and from opioid prescriptions within 90 days postdischarge. Patients with multiple concurrent fractures, preinjury opioid use, or less than 90-day follow-up were excluded. Descriptive statistics in the form of frequencies (%) and means (standard deviation) were calculated. Linear mixed modeling with repeated measures revealing P values of .05 or lower was considered statistically significant. Multiple concurrent fractures were defined as bilateral upper extremity injuries, ipsilateral fractures involving separate regions (eg, hand and shoulder), or combined radius and ulna fractures. This modeling was used to evaluate independent predictors of opioid use, including age, American Society of Anesthesiologists class, fracture type, and gabapentin use.

Results

A total of 128 patients with an operatively treated upper extremity fracture were identified (76 in 2018, 52 in 2020). No differences were seen in patient demographics. Pre-operative peripheral nerve block use was similar between groups (P = .570). Gabapentin was used in a significantly higher proportion of patients in 2020 than it was in 2018 (34% vs. 18.3%, P = .050). Inpatient opioid consumption significantly decreased from 2018 to 2020 (P = .003), most notably on POD 1 and 2, with age and American Society of Anesthesiologists class being significant predictors of opioid use (P = .0012 and P = .0315, respectively). Mean MMEs on POD 1 were 68.21 in 2018 vs. 50.32 in 2020, whereas mean MMEs on POD 2 were 46.76 in 2018 vs. 12.15 in 2020. Forty-six point zero five percentage of patients in 2018 were taking opioids on POD 2 vs. 26.92% in 2020 (P = .029). Pain scores on POD 1-4 showed no differences (P ≥ .4875). 90-day postdischarge opioid prescriptions decreased in 2020, though not significantly. Fracture location and concomitant gabapentin use were significant predictors of outpatient opioid use (P = .001 and P = .013, respectively.)

Conclusion

A multidisciplinary approach is effective for reducing inpatient opioid consumption in operatively treated upper extremity fracture patients, with the greatest reduction seen on POD 1 and 2. Increases in gabapentin prescriptions following protocol implementation may indicate the beneficial role of gabapentin administration in decreasing inpatient opioid consumption.

While opiates were previously used primarily as analgesics in the treatment of cancer patients, the use (and abuse) of opioid medications has steadily increased since the beginning of the 1990's. Pharmaceutical companies downplayed the potential for long-term negative consequences of these medications, which led to dramatic increases in their prescription for more general treatment of pain. Long before the opioid crisis of the 1990s, heroin had already played a central role in America's troubled relationship with narcotics; first synthesized in the late 19th century and marketed as a nonaddictive substitute for morphine, it quickly became associated with addiction and social stigma by the mid-20th century.¹² This earlier wave of opioid use set the stage for later crises. The first major surge in prescription opioid use occurred in the 1990s, followed by a second wave around 2010, driven by the resurgence of heroin—now cheaper and more accessible—and the rise of synthetic opioids like fentanyl.^8,13 The opioid epidemic has represented a major public health crisis in the United States, accounting for hundreds of thousands of deaths since 1999,^6,7 though recent trends suggest a potential decline in opioid-related mortality.²

The opioid crisis in the United States is especially pertinent to the field of orthopedic surgery, as orthopedic surgeons were reported in 2017 to be the third highest in the prescription of opioids among all physicians.¹⁶ Many orthopedic procedures are inherently painful because they require deep dissection through the soft tissue envelope and instrumentation of the bone; therefore, patients frequently experience post-operative pain following orthopedic surgery.⁹ With the amount of continued pain perceived by patients as one of the most important factor to their satisfaction after surgery, surgeons often feel pressure to alleviate this pain.⁴ Guidelines published by the American Academy of Orthopaedic Surgeons suggest practice-based approaches to pain management, leaving room for significant discrepancies in the proper post-operative treatment of pain.¹ These discrepancies have led to a growing body of research on the importance of multidisciplinary opioid reduction programs.

The American Pain Society recently published guidelines on the multidisciplinary management of post-operative pain, which emphasize the importance of pain control in the pre-operative and post-operative periods.¹ With proper pre-operative education and individualized post-operative care, a multidisciplinary approach is an effective tool to reduce opioid use among orthopedic fracture patients and provide exemplary pain management.

The aim of this study was to investigate the effectiveness of a multidisciplinary approach to opioid reduction in patients admitted for operative treatment of an isolated upper extremity fracture. Specifically, we sought to measure inpatient opioid consumption during post-operative days (PODs) 1-4 and outpatient opioid prescriptions provided after discharge. We hypothesized that, when comparing the year following implementation of a multidisciplinary opioid reduction program to the year before implementation, there would be a decrease in post-operative inpatient opioid prescriptions ordered and volume of outpatient opioid prescriptions ordered and filled within 90 days of discharge.

Methods

In 2018, a multidisciplinary program was developed to address the current state of opioid use in the post-operative management of orthopedic surgery patients at our institution, an academic level 1 trauma center. This program was implemented in 2019 and included 4 main components: (1) changes to standardized orthopedic order sets; (2) standardization of language used to talk with patients about pain; (3) adoption of a functional pain assessment scale and comfort menu; and (4) administration of educational computer-based training modules to all staff members involved in patient care on our unit.

Key changes to standardized order sets included 3 scheduled acetaminophen doses during the day as well as an “as-needed” dose for overnight hours. In addition, oxycodone dose ranges were changed from 5-15 mg to 5-10 mg for nongeriatric patients, and for patients 65 years and older, the dose range was set at 2.5-5 mg. The frequency of administration was lengthened from every 3 hours as needed to every 4 hours as needed. Finally, intravenous narcotic orders were automatically removed from the order sets 72 hours after the orders were placed.

The key principle underlying the recommended language for speaking with patients about their pain was to focus on the functional consequences of pain (ie, what their pain is preventing them from accomplishing) instead of the severity of the pain. The functional pain scale complements this underlying principle, providing a function-centered description for each number on the 0-10 visual analog scale commonly used in pain assessment. The comfort menu summarizes the wide range of nonmedication alternatives for managing pain, including deep breathing exercises, hallway walking, and quiet activities such as books/magazines, movies, playing cards, and coloring/art activities.

We conducted a retrospective comparative study on patients with isolated upper extremity fractures admitted to the inpatient orthopedic unit at an academic level 1 trauma center in 2018 and 2020 (pre- and post-program implementation). We calculated total inpatient opioid consumption during POD 1-4 and within 90 days postdischarge (measured in morphine milligram equivalents [MMEs]). These data were extrapolated from patient records based on the number of prescriptions that were ordered and subsequently filled, both inpatient and outpatient. Prescription data were derived from patient records based on prescriptions ordered and filled in inpatient and outpatient settings.

Patients were excluded if they had multiple concurrent fractures, including bilateral fractures or multiple ipsilateral fractures, or less than 90-day follow-up. All patients with preinjury opioid use were excluded, including patients with pre-existing neuropathic pain syndrome diagnoses requiring opioids, such as diabetic neuropathy, ulnar neuropathy, and complex regional pain syndrome. Descriptive statistics in the form of frequencies (%) and means (standard deviation) were calculated. Linear mixed-effects modeling with repeated measures was performed using SAS (v 9.4), with statistical significance defined as P < .05. The model incorporated a random intercept for each subject to account for between-subject variability and to allow for correlation among repeated MME measurements across multiple time points within individual subjects. Multiple concurrent fractures were defined as bilateral upper extremity injuries, ipsilateral fractures involving separate regions (eg, hand and shoulder), or combined radius and ulna fractures. This modeling was used to evaluate independent predictors of opioid use, including age, American Society of Anesthesiologists (ASA) class, fracture type, pre-operative nerve block, and gabapentin use. Age was defined as the mean age at surgery. ASA class was classified as healthy, mild systemic disease, severe systemic disease, or threat to life systemic disease, fracture type was identified using Current Procedural Terminology codes, and gabapentin use was a binary result.

Results

A total of 128 patients (76 in 2018, 52 in 2020) with an operatively treated upper extremity fracture were identified. No differences were identified in patient demographics between groups, as shown in Table I.

Table I.

Comparison of patient demographics and clinical characteristics between 2018 and 2020 cohorts.

Demographics pre–post mean ± SD or count (col %)
Variable	2018	2020	P value
Gender			.62
Male	37 (48.7)	23 (44.2)
Female	39 (51.3)	29 (55.8)
Age at surgery	50.7 ± 17.6	50.4 ± 20.8	.977
BMI	27.8 ± 6.7	29.5 ± 9.4	.245
Inpatient LOS	3.5 ± 5.5 (median = 2)	3.4 ± 3.7 (median = 2)	.66
Anxiety Dx			.78
Yes	9 (11.8)	7 (13.5)
No	67 (88.2)	45 (86.5)
Depression Dx			.61
Yes	12 (15.8)	10 (19.2)
No	64 (84.2)	42 (80.8)
Gabapentin used			.05
Yes	13 (18.3)	17 (34.0)
No	58 (81.7)	33 (66.0)
Nerve blocker used			.57
Yes	40 (52.6)	30 (57.7)
No	36 (47.4)	22 (42.3)
ASA class			.53
Healthy	21 (28.0)	10 (19.2)
Mild systemic disease	42 (56.0)	32 (61.5)
Severe systemic disease	11 (14.7)	10 (19.2)
Threat to life systemic disease	1 (1.3)	0 (0.0)
Primary diagnosis
Open AC joint dislocation repair	12	10
Open elbow dislocation repair	16	6
Open humeral shaft fracture fixation	15	6
Open both bone forearm fracture fixation	8	8
Open supracondylar humerus fracture fixation	8	2
Open supracondylar humerus fracture (no fixation)	3	5
Open humeral condyle fracture fixation	7	0
Open radial head/neck fracture fixation or replacement	5	0
Open coronoid fracture fixation	1	4
Upper arm/elbow tendon repair	0	4
Open ulnar shaft fracture fixation	0	1
Open humeral shaft IM nailing	1	0
Open radial head/neck fracture (no fixation)	0	1
Open radial shaft fracture fixation	0	1
Open ulnar shaft fracture fixation	0	1
Open distal radius fracture fixation	0	1

BMI, body mass index; ASA, American Society of Anesthesiologists; LOS, length of stay; Dx, diagnosis; AC, acromioclavicular; IM, intramedullary.

This table presents a comparison of demographic and clinical variables for patients treated in 2018 (n = 76) and 2020 (n = 52). Variables include gender distribution, age, BMI, length of stay, psychiatric diagnoses (anxiety and depression), use of gabapentin, nerve block usage, and ASA classification. There were no statistically significant differences between groups for most variables. However, gabapentin usage was significantly higher in 2020 (34.0%) compared to 2018 (18.3%), with a P value of .05, indicating a trend toward increased use.

Bold value indicates statistically significant difference.

Primary outcomes

Inpatient opioid consumption significantly decreased from 2018 to 2020 (P = .003). As shown in Fig. 1, this difference was most notable on POD 1 and 2. Mean MMEs on POD 1 in 2018 were 68.21 vs. 50.32 in 2020, whereas mean MMEs on POD 2 in 2018 were 46.76 vs. 12.15 in 2020 (P < .01; P = .01). Forty-six point zero five percentage of patients in 2018 were taking opioids on POD 2 vs. 26.92% in 2020 (P = .029). There were no significant differences in pain scores on POD 1-4 (P ≥ .4875). Pre-operative peripheral nerve block use was similar between groups, with 52.6% receiving a nerve block in 2018 vs. 57.7% in 2020 (P = .570). Age and ASA class were found to be independent risk factors for increased opioid use (P = .0012 and P = .0315, respectively).

Figure 1.

Inpatient opioid consumption (in MMEs) decreased from the year prior to the protocol (2018, N = 76) to the first year of implementation (2020, N = 52). MME, morphine milligram equivalent.

Secondary outcomes

Ninety-day postdischarge opioid prescriptions decreased in 2020 compared to 2018, but this decline was not significant (P = .29). Mean opioid use in MMEs at 30 days, 60 days, and 90 days were 9.5, 1.3, and 0.8, respectively, in the 2018 cohort. These averages decreased to 7.3, 0.8, and 0.5 in 2020. There were no significant differences with regard to age at time of surgery, gender, ASA class, or use of a nerve block. (Fig. 2). Gabapentin was used significantly more frequently in 2020 (34%) than in 2018 (18.3%, P = .050). Primary fracture diagnosis and concomitant gabapentin use were significant predictors of outpatient opioid use (P = .001 and P = .013, respectively.) (Fig. 3). Fracture distribution (clavicle, humerus, elbow, forearm, wrist/hand) is presented in Table I, and distal radius/ulna fractures (Current Procedural Terminology 25620) were significant predictors of higher opioid use both for 90-day prescriptions (P < .05).

Figure 2.

Outpatient opioid consumption (in MMEs) decreased from the year prior to the protocol (2018, N = 76) to the first year of implementation (2020, N = 52). MME, morphine milligram equivalent.

Figure 3.

Outpatient gabapentin prescriptions increased from the year prior to the protocol (2018, N = 71) to the first year of implementation (2020, N = 50).

Discussion

The present study demonstrated that a multidisciplinary pain management program is an effective strategy in reducing inpatient opioid consumption in operatively treated upper extremity fracture patients. Implementation of this protocol in our institution decreased inpatient opioid consumption post-operatively, driven most strongly by a significant decrease in opioid use on POD 1.

Importantly, our study found that, while opioid consumption was significantly reduced post-operatively, there was no change in patient-reported pain scores. Previous studies¹¹ have shown concern that protocols prioritizing decreased opioid utilization may lead to decreased patient satisfaction scores, which are imperative to hospital outcome metrics in addition to individual patient satisfaction. In addition, Rane (2019) found that patient-reported-pain, as measured by the Patient-Reported Outcomes Measurement Information System, corresponded with a significant decrease in overall patient satisfaction. In contrast, our study demonstrated that adequate post-operative pain control can still be achieved while significantly reducing opioid consumption, as measured in MMEs, thus maintaining patient satisfaction. Our multimodal pain management program was aimed not only at reducing inpatient opioid consumption but also decreasing outpatient opioid prescriptions in operative upper extremity fracture patients.

Previous studies have demonstrated the effectiveness of multimodal pain management following orthopedic procedures, including the use of nerve blocks, local anesthetic, and muscle relaxants.^10,14 Gabapentin has emerged as a particularly effective medication in managing post-operative pain, particularly neuropathic pain. Blaber et al (2023) demonstrated that in anterior cruciate ligament reconstruction patients, perioperative use of gabapentin was shown to decrease opioid consumption without negatively impacting pain scores.³ In our study, gabapentin use was significantly higher after implementation of the multimodal pain management program. Further exploration of the use of gabapentin is required to understand the optimal dosing to prescribe pre-operatively and/or post-operatively for adequate analgesia.

There are several limitations of this study. First, the current study is retrospective in nature. However, pain scores were prospectively recorded in real-time by patients' primary bedside nurses. Secondly, specific fracture patterns and mechanisms of injury were not classified, and therefore could not be compared between the 2018 and 2020 groups. In theory, more complex fracture patterns or higher-energy mechanisms of injury may have led to increased pain levels, thereby requiring higher opioid quantities. However, there were no significant differences between groups in the distribution of procedures performed for various diagnoses between groups (see Table I). Finally, the relatively small sample size from one geographic region limits the generalizability of this study's findings. Regional and health system–level shifts in post-operative pain management may have occurred alongside the implementation of this program. Future multicenter studies exploring multidisciplinary approaches to decreasing post-operative opioid use could further investigate the efficacy of this method among diverse populations. In addition, fracture classification details were not fully available in the initial analysis; however, fracture distribution data are now provided in Table I to improve interpretability.

The main components of our multidisciplinary program included changes to order sets, implementation of a functional pain scale, and staff education. We did not incorporate the addition of certain other medical professionals in the program. In recent literature, the incorporation of behavioral health providers and pharmacists has been shown to be particularly effective at decreasing opioid use. When pharmacists prepared discharge medication lists, there was a significant reduction in outpatient opioid consumption, notably without change in post-operative pain.¹⁵ Similarly, integrating behavioral health providers into the primary care management of opioid use was shown to improve access to care, reduce barriers for primary care clinicians, and increase patients' feelings of support.⁵

Conclusion

A multidisciplinary approach to opioid reduction is an effective strategy in reducing inpatient opioid consumption in operatively treated upper extremity fracture patients, especially on POD 1 and 2. Gabapentin was prescribed more frequently following protocol implementation, indicating a potential synergistic role.

Disclaimers:

Funding: Research support was provided by an award from the University of Wisconsin School of Medicine and Public Health and the Herman and Gwendolyn Shapiro Foundation.

Conflicts of interest: The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.