The Relationship Between Body Mass Index and Acute Postoperative Opioid Consumption Following Primary Total Joint Arthroplasty: A Retrospective Cohort Study

1. Introduction

Obesity is characterized as a pro-inflammatory condition, frequently associated with chronic pain.¹ Across myriad investigations, obesity has been positively correlated with pain; this relationship is multifactorial, with mechanical and physiological factors contributing.^2,3 Excess body weight increases joint stress and alters joint loading vectors, while metabolic dysfunction in obesity reduces muscular control and further amplifies joint loads.⁴ Additionally, the endocrine function of adipose tissue leads to elevated levels of inflammatory cytokines, which may induce a hyperalgesia state.^4–6

Acute pain control following total joint arthroplasty (TJA) is a critical aspect of patient care with significant medical and economic implications.^7–9 Etcheson et al. demonstrated that postoperative pain is integral to the patient experience, which is having increasing influence on care reimbursement models.^8,9 Pain control also entered the national spotlight during the opioid crisis, which prompted increased regulation and attention on opioid prescribing practices. Recent evidence has demonstrated efficacy for dexamethasone and multimodal analgesic protocols in reducing postoperative pain and opioid consumption.^10–14

Prior investigations have associated BMI and obesity with acute pain following primary TJA.^15–18 However, the strength of this association differs across studies, likely due to methodological variances and inherent sample differences, which reinforce the inadequacies of BMI as a proxy for overall obesity and its metabolic consequences. Despite the recent attention on pain control in TJA, the relationship between BMI and postoperative opioid consumption remains infrequently and inconsistently reported. This issue is of increasing importance, as the proportion of obese patients undergoing TJA is expected to grow substantially in the coming years.^19,20 Additionally, improved understanding of obesity in TJA is critical to determining the efficacy of BMI based restrictions on access to TJA.

The primary objective of this study was to describe the relationship between obesity and acute postoperative opioid consumption following primary TJA. A secondary objective was to determine whether this relationship differed between total knee arthroplasty (TKA) and total hip arthroplasty (THA).

2. Methods

2.1. Study Design and Ethical Approval

Following appropriate institutional approvals (Protocol R-1023CS/YF), a retrospective chart review was performed in accordance with established ethical parameters and the study inclusion criteria. The FINER criteria provided a template for study parameter justification: the study is Feasible using routinely collected perioperative data; Interesting given the increasing prevalence of obesity; Novel in its focus on a minority predominant sample; Ethical following appropriate approvals; and Relevant given projections for increasing proportions of obese patients in TJA. All cases were performed by a single fellowship trained joint reconstruction surgeon at a single institution.

2.2. Case Identification

Cases were identified using CPT codes, and surgical notes were reviewed to confirm alignment with the inclusion criteria. The PICO framework was applied as follows: Population - adults over 18 years of age without preoperative opioid use; Intervention -primary TKA or THA; Comparison - stratification by BMI; Outcome- postoperative opioid consumption (MME) at 24 hours, 48 hours, and at discharge.

Cases of revision arthroplasty were excluded. Patients with current or recent opioid use were excluded to remove the potential confounder of nociceptive sensitization, often termed opioid-induced hyperalgesia.

2.3. Data Collection

The following case characteristics were collected: patient age, sex, ethnicity, BMI, and surgical intervention. Outcome data included postoperative opioid consumption at 24 hours, 48 hours, and the total cumulative amount at discharge, with all opioid doses converted to morphine milligram equivalents (MME). Patient ethnicity was collected because obesity and its common comorbidities are more prevalent in minority populations,^21–23 accordingly, the relationship between BMI and opioid consumption may differ based on sample composition.

2.4. Statistical Analysis

Pearson’s correlation coefficient (r) was used to determine the relationship between BMI and MME, as it is commonly regarded as the standard measure of linear predictive accuracy for continuous variables. A two-sample Student’s t-test was used to evaluate statistical significance. Statistical significance was set at p<0.05.

3. Results

3.1. Cohort Demographics

Following exclusions for inadequate documentation and preoperative opioid consumption, 544 cases were included in the final analysis. The mean patient age was 68.2 years, the mean BMI was 31.2 kg/m², and 65% of the series were female. Regarding self-identified ethnicity, 86% identified as Hispanic American, 9% as non-Hispanic White, and 5% as African American.

3.2. Overall TJA Cohort: BMI and Opioid Consumption

At 24 hours, Pearson’s r=0.025 (p=0.56, n=544), indicating a weak correlation between BMI and MME. A t 48 hours, Pearson’s r=0.068 (p=0.11, n=430), again indicating a weak correlation. At discharge, Pearson’s r=0.042 (p=0.22, n=544), consistent with the prior time points. Table 1 summarizes the correlation statistics at each time point.

3.3. Total Hip Arthroplasty Subgroup

In the THA series, there were 196 cases with a mean BMI of 30.4 kg/m², and 51% were classified as obese (BMI ≥30 kg/m²). At 24 hours, Pearson’s r=0.111 (p=0.71, n=196); at 48 hours, r=0.137 (p=0.13, n=148); and at discharge, r=0.061 (p=0.22, n=196). These results indicate a weak but marginally stronger correlation in the THA group compared with the overall cohort.

3.4. Total Knee Arthroplasty Subgroup

In the TKA series, there were 348 cases with a mean BMI of 31.5 kg/m², and 59% were classified as obese. At 24 hours, Pearson’s r=0.02 (p=0.42, n=348); at 48 hours, r=0.05 (p=0.28, n=266); and at discharge, r=0.005 (p=0.31, n=348). Table 2 summarizes correlation statistics stratified by procedure type.

4. Discussion

The relationship between BMI and acute postoperative opioid consumption in primary TJA patients is an emerging concept without a consensus position in the literature. The current results indicate that patient BMI is a very weak predictor of opioid consumption following primary TJA within a series primarily composed of Hispanic American patients. This finding is notable, as obesity is often characterized as a chronic pain state^1,2,6 However, the present results suggest that obese patients may not require greater levels of opioids in the immediate postoperative period following primary TJA compared with non-obese patients.

The literature displays conflicting findings regarding the impact of obesity on pain levels following TJA. Lendrum et al. reported an increase in opioid consumption with increasing BMI category following TJA.¹⁵ Aggarwal et al. reported increased narcotic use in obese patients compared with non-obese patients following THA.¹⁶ However, Pinto et al. reported that BMI was not strongly correlated with acute pain at 48 hours following primary TJA, which is consistent with the current results.¹⁷ Lo et al. concluded that increasing BMI was an independent predictor for major pain (visual analog scale ≥5) within two days following TKA.¹⁸

There is a lack of clarity discriminating the postoperative pain course between THA and TKA. Lendrum et al. reported a stronger association between BMI and opioid consumption within the THA subgroup, which is consistent with the current results showing a marginally stronger correlation in the THA series.¹⁵ However, in both reports, the overall relationship inclusive of TKA and THA remained very weak. These findings are in contrast to those of Dwyer et al. who reported that TKA patients had higher total opioid dosing and longer courses of postoperative opioids than a matched cohort of THA patients.²⁴ Dattilo et al. similarly reported that total opioid consumption and length of use were greater in TKA patients compared with THA patients.²⁵ Wylde et al. described severe acute pain as being more frequent following TKA than THA, and noted that pain in TKA patients did not subside at a comparable rate.²⁶ Kugelman et al. reported significantly higher opioid consumption at 24 and 48 hours in TKA patients compared with THA patients.²⁷

Obesity is often characterized as a chronic pain state, with physiological, mechanical, and psychosocial factors contributing.^1,6 The pathogenesis of obesity encompasses a spectrum from genetic predisposition to environmental and behavioral influences.²⁸ Adipose tissue is the primary driver of the aberrant physiology and metabolic detriment in obese patients. Furthermore, the magnitude of adipose tissue may alter medication distribution and clearance, thereby impacting pharmacokinetics and pharmacodynamics.^29,30 However, BMI is an indirect measure of adipose tissue with variable predictive capacity for TJA outcomes.^31,32

Factors to be considered when administering opioids to obese patients include preoperative opioid consumption, existing pulmonary function, and concomitant obstructive sleep apnea (OSA). Preoperative opioid consumption may be higher in obese patients who commonly suffer from chronic pain, suggesting a stronger potential for pain sensitization and increased risk of acute pain following TJA.^33,34 In a series of TJA patients evaluated preoperatively, Thomazeau et al. reported significantly higher daily consumption of opioids and NSAIDs in obese patients in order to attain pain relief comparable to that of non-obese patients.³⁵

Controlling postoperative pain in the obese population is especially important, as early mobilization reduces the risk of medical complications including pressure ulcerations, deep vein thrombosis, and respiratory infection. The commonality of OSA in obesity must be carefully considered when determining postoperative analgesia. Opioid administration increases the risk of hypoxic episodes, which has contributed to the drive for opioid sparing multimodal pain control in obese patients.^36,37 Close monitoring of obese patients in the perioperative period is warranted regardless of whether a formal OSA diagnosis has been established.

5. Limitation

Retrospective study designs are inherently weakened by limitations including the potential for error during data input, aggregation, and analysis. Lendrum et al. performed a similar analysis across a larger sample; however, the current study’s 24-hour and 48-hour data sets provide a more granular understanding of the postoperative pain course.¹⁵ Furthermore, 41% of the Lendrum et al. series had BMIs classified as obese compared with 55% of the current series. This discrepancy is likely representative of the underlying population from which each sample was derived, as the current series was taken from a community with a high proportion of minority patients, among whom obesity and its comorbidities are increasingly prevalent.^22,23 Approximately 5% of both series were classified as morbidly obese.

The 48-hour data may be subject to bias, as patients who remained hospitalized at 48 hours postoperatively were likely to have experienced complicating clinical factors that could confound the analysis. However, the consistent results across the 24-hour and 48-hour data sets suggest that this additional risk may not have materially impacted the findings. There is also an increasing prevalence of outpatient TJA, to which the current results may not be fully generalizable.

6. Conclusion

Acute pain following primary TJA is a complex, multifactorial problem. Our results indicate that patient’s BMI is a very weak predictor of opioid consumption following primary TJA within a series primarily composed of minority patients. These findings contribute to the growing body of evidence against BMI based restrictions to TJA and underscore the need for individualized, multimodal analgesic strategies rather than BMI driven opioid prescribing.

Conflict of Interest

The authors declare no conflicts of interest.

Funding

This research received no external funding.

Ethical Approval

Institutional Review Board approval was obtained (Protocol R-1023CS/YF).

Data Availability

Data are available upon reasonable request to the corresponding author.