Impact of mood disorders on postoperative outcomes following ankle open reduction and internal fixation

Introduction

Mood disorders (MD), including depression and anxiety, represent a substantial and growing global health burden, with recent estimates indicating over 359 million cases of anxiety and 332 million cases of depression worldwide, accounting for more than 60% of all mental health disorders. ¹ In the United States, a nationally representative analysis from the National Epidemiologic Survey on Alcohol and Related Conditions-III survey found that the 12-months prevalence of major depressive disorder is approximately 10.4%, with higher rates observed in younger adults and females. ² Large-scale epidemiologic studies confirm that MD are the most common psychiatric diagnoses, affecting over 22% of adults, and frequently co-occur with anxiety and other psychiatric conditions. ³ Psychiatric diagnoses are an especially important consideration in orthopedic surgical populations. Notably, orthopedic patients, particularly those with foot and ankle pathology, exhibit even higher rates of depression and anxiety, with some reports of prevalence up to three times greater than the general population.^4,5

Orthopedic trauma, including ankle fractures, is associated with a marked increase in psychiatric symptoms, with up to 56% of patients reporting depression, anxiety, or posttraumatic stress disorder during their postoperative course. ⁶ The American Academy of Orthopedic Surgeons (AAOS) guidelines emphasize that anxiety, depression, and premorbid psychiatric conditions are independently associated with worse pain, decreased functional outcomes, and impaired return to work following orthopedic trauma. ⁷ These psychosocial factors are now recognized as critical determinants of recovery, often exerting a greater influence on long-term disability and patient-reported outcomes than objective physical measures.^8,9

Open reduction and internal fixation (ORIF) is the standard surgical intervention with extensive evidence supporting its efficacy in restoring alignment and function.^10,11 However, postoperative recovery can be complex and multifactorial, with MD emerging as significant predictors of patient satisfaction, pain, and functional outcomes. ¹² Patients with depression or anxiety report lower preoperative and postoperative outcome scores, less improvement, and higher dissatisfaction rates after foot and ankle surgery. ⁹ Longitudinal studies reveal that more than half of patients undergoing ankle ORIF experience mild to moderate depressive symptoms postoperatively, particularly those with a prior history of mental health disorders or limited ability to work during recovery. ¹³ Furthermore, persistent psychological distress is associated with a twofold increased risk of unsatisfactory surgical outcomes, independent of fracture severity or demographic factors. ¹⁴

Despite mounting evidence of the interplay between MD and orthopedic outcomes, there remains a paucity of data specifically examining the effect of MD on postoperative outcomes following ankle ORIF. Addressing this gap is essential for optimizing patient-centered care, improving functional outcomes, and reducing postoperative complications in individuals undergoing ankle ORIF. The purpose of this study is to evaluate the impact of MD on clinical outcomes and postoperative complications following ankle ORIF.

Methods

Statistical analysis

Cohorts were matched in a 1:1 ratio using a caliper of 0.1 pooled standard deviations. Matching variables included age at the time of the procedure, sex, ethnicity, race, chronic kidney disease, diabetes mellitus, nicotine dependence, and body mass index (Table 1).

OPEN IN VIEWER

Table 1.

Baseline demographics between mood disorder (MD) and non-MD cohorts before propensity-score matching.

Variables	MD cohort (n = 35,475)	Non-MD cohort (n = 86,233)	P-value	SMD
Age at index, years (SD)	52.5 +/− 17.6	44.9 +/− 19.7	<0.001	0.404
Male, n (%)	9530 (27.0%)	42,010 (49.5%)	<0.001	0.476
Caucasian, n (%)	28,759 (81.4%)	56,879 (67.0%)	<0.001	0.335
African American, n (%)	4108 (11.6%)	14,057 (16.6%)	<0.001	0.142
Hispanic or Latino, n (%)	2147 (6.1%)	10,650 (12.5%)	<0.001	0.224
Chronic kidney disease, n (%)	4104 (11.6%)	2953 (3.5%)	<0.001	<0.001
Diabetes mellitus, n (%)	7974 (22.6%)	7590 (8.9%)	<0.001	0.381
Nicotine dependence	10,087 (28.6%)	11,123 (13.1%)	<0.001	0.388
Body Mass index	31.6 +/− 7.6	29.7 +/− 6.8	<0.001	0.273

Demographics of both cohorts before propensity-score matching. Abbreviations: SMD, standardized mean difference; SD, standard deviation; MD, mood disorder.

For patient privacy, TriNetX does not display exact counts for outcomes with fewer than 10 patients and instead rounds values up to 10; however, matching is performed using the exact patient counts. All statistical analyses were conducted within the TriNetX analytics environment. Measures of association (risk ratios, odds ratios, hazard ratios with 95% confidence intervals) were calculated to compare complication rates between groups. P-values were derived using z-tests for categorical variables, with p < 0.05 considered statistically significant.

Results

Patient characteristics

After propensity score matching, the final analytic sample included 31,040 patients in each cohort. Baseline demographics and comorbidities were well balanced between patients with MD and those without MD. Balancing was assessed primarily using the standardized mean difference (SMD). (Table 2).

OPEN IN VIEWER

Table 2.

Demographics between mood disorder (MD) and non-MD cohorts after propensity-score matching.

Variable	MD cohort (n = 31,040)	Non-MD cohort (n = 31,040)	P-value	SMD
Age at index, years (SD)	51.4 +/− 17.7	51.7 +/− 18.8	0.062	0.015
Male, n (%)	9233 (29.7%)	9297 (30.0%)	0.575	0.005
Caucasian, n (%)	24,815 (79.9%)	25,234 (81.3%)	<0.001	0.034
African American, n (%)	3861 (12.4%)	3751 (12.1%)	0.178	0.011
Hispanic or Latino, n (%)	2077 (6.7%)	1995 (6.4%)	0.184	0.011
Chronic kidney disease, n (%)	2534 (8.2%)	2400 (7.7%)	0.047	0.016
Diabetes mellitus, n (%)	5479 (17.7%)	5296 (17.1%)	0.052	0.016
Nicotine dependence	7350 (23.7%)	7024 (22.6%)	0.002	0.025
Body Mass index	31.5 +/− 7.5	30.1 +/− 6.9	<0.001	0.184

Demographics of both cohorts after propensity-score matching. Abbreviations: SMD, standardized mean difference; SD, standard deviation; MD, mood disorder.

Discussion

This study demonstrates that patients with pre-existing MD who undergo ankle ORIF experience significantly worse outcomes across multiple clinical outcomes compared to those without MD. Patients with MD exhibited elevated opioid utilization at 2 weeks, 6 weeks, 3 months, 6 months, and 1 year postoperatively. Additionally, at both 3-months and 6-months follow-up, the MD cohort demonstrated higher rates of hospital readmissions, ED visits, and physical therapy utilization. Furthermore, significantly elevated rates of SSI and DVT/PE at both the 3-months and 6-months time points were observed. The observed findings are consistent with the established literature on postoperative consequences of MD in other foot and ankle surgeries. Depression in particular has been associated with both significantly worse outcomes and limited postoperative improvement in the setting of total ankle arthroplasty (TAA).^15,16 However, uniquely to TAA, a 2018 study Cunningham et al. found that depression was not independently associated with 90-days readmissions, emergency department utilization, or wound complications, suggesting that the effect of mood disorders on postoperative outcomes may be procedure-specific. ¹⁷

The persistent elevation in opioid use among MD patients at all postoperative timepoints represents a clinically significant finding with important implications for pain management protocols following ankle ORIF. This aligns with evidence from prior studies indicating that mental health disorders are independent risk factors for new persistent opioid use after ankle fracture treatment, regardless of injury severity.^18,19 Previous literature has also established strong associations between depression, anxiety, and chronic opioid use, with psychiatric disorders serving as independent risk factors for prolonged opioid consumption and opioid use disorder development.^20,21 Patients with MD have been reported to experience altered pain perception, lower pain thresholds, and diminished analgesic response to standard pharmacologic interventions, potentially driving higher opioid requirements.^22,23 Furthermore, the psychological distress associated with MD may amplify the subjective pain experience through central sensitization mechanisms and maladaptive coping strategies. ²⁴ The sustained increase opioid use among MD patients at 1 year is also notable, as it potentially suggests progression toward chronic opioid dependence rather than appropriate tapering following acute postoperative recovery. It is notable that opioid use was elevated in both cohorts across all timepoints, which likely reflects broad prescribing patterns following ankle ORIF, a procedure known to carry a high baseline burden of postoperative pain requiring pharmacologic management. The clinically meaningful finding is the consistently and significantly greater opioid utilization in the MD cohort relative to matched controls at every timepoint assessed.

The significantly elevated rates of ED visits, hospital readmissions, and physical therapy utilization at both 3 and 6 months among MD patients reflect the complex interplay between psychiatric illness and postoperative recovery trajectories. Patients with depression and anxiety disorders have demonstrated lower treatment adherence, decreased engagement with rehabilitation protocols, and reduced self-efficacy regarding recovery, all of which can compromise healing and functional restoration.^25,26 The increased ED utilization may stem from inadequate pain control, heightened anxiety, or complications that arise more frequently in this population due to suboptimal self-care and delayed help-seeking behavior for emerging problems.^27–29 Similarly, higher readmission rates suggest either increased complication incidence or difficulty managing recovery in the outpatient setting. The increase in physical therapy utilization at 3 months and 6 months, despite generally poorer adherence patterns in MD populations, may reflect either more intensive rehabilitation requirements due to delayed recovery or referral patterns aimed at addressing functional deficits that persist longer in these patients. The literature suggests that even after symptomatic improvement, individuals with MD may continue to experience functional impairment, necessitating prolonged or intensified rehabilitation efforts.^30,31

The significantly higher rates of SSI and DVT/PE at both 3 and 6 months in the MD cohort represent serious complications with substantial morbidity implications. The association between depression and impaired immune function is well-established, with dysregulation of inflammatory cytokines, altered neutrophil function, and compromised wound healing documented in multiple studies.^32,33 Additionally, MD are associated with health behaviors that increase infection risk, including smoking, poor nutrition, decreased physical activity, and suboptimal medication adherence, particularly with prophylactic antibiotics. ³⁴ The elevated DVT/PE rates may similarly reflect multiple contributory factors, including reduced mobility secondary to pain catastrophizing or motivational deficits, inadequate adherence to thromboprophylaxis regimens, and potential prothrombotic inflammatory states associated with depression. ³⁵ A prior retrospective analysis further supports these findings, concluding that patients with psychiatric diagnoses had higher odds of SSI and DVT/PE within 1 month and up to 1 year after fixation, mirroring the elevated risks observed in the current cohort. ³⁶

Pain catastrophizing, a psychological construct characterized by magnification of pain, rumination, and feelings of helplessness, has been independently associated with worse postoperative outcomes across multiple orthopedic domains.^37,38 Validated instruments such as the Pain Catastrophizing Scale (PCS) are increasingly used in surgical risk stratification. ³⁹ However, because this study utilized a large, federated administrative database, patient-reported outcome measures and psychometric assessments including catastrophizing scales were not available for capture or analysis. The TriNetX platform does not contain standardized psychological survey data, precluding the incorporation of catastrophizing scores into the study design.

The findings of this study should also be interpreted in the context of the well-established pain catastrophizing literature. Pain catastrophizing, defined as a negative cognitive-affective orientation toward pain involving magnification, rumination, and perceived helplessness, has been consistently shown to predict worse pain intensity, functional outcomes, opioid consumption, and patient satisfaction following orthopedic surgery, including foot and ankle procedures.^37,40,41 Patients with mood disorders, particularly depression and anxiety, demonstrate significantly higher rates of catastrophizing behavior, and this psychological trait may serve as a key mediating mechanism linking pre-existing MD to the adverse outcomes observed in the current study.^42,43 Prior work has demonstrated that catastrophizing is a stronger predictor of postoperative pain and disability than radiographic or surgical findings alone,^40,44 and that even moderate reductions in catastrophizing through cognitive behavioral therapy or perioperative psychological interventions are associated with meaningful improvements in surgical outcomes. ⁴⁵ Future prospective investigations should incorporate validated catastrophizing instruments such as the PCS to better delineate the relative contributions of mood disorder diagnosis versus catastrophizing behavior to postoperative outcomes following ankle ORIF.

These findings have several important implications for clinicians caring for patients undergoing ankle ORIF. First, routine preoperative screening for MD should be standard practice to identify high-risk patients and guide perioperative optimization strategies. Second, patients with MD should be counseled regarding their elevated risk for postoperative complications, including the potential for prolonged opioid use and hospital readmission, emphasizing the importance of psychiatric optimization and multimodal pain management. Third, a multidisciplinary approach incorporating psychiatric consultation, enhanced wound monitoring, aggressive thromboprophylaxis, early mobilization, and closer outpatient follow-up may help mitigate these risks. Finally, these results highlight the need for further research to determine how preoperative psychiatric intervention can improve surgical outcomes and to identify which specific interventions are most effective in reducing complications in this high-risk population.

Several limitations must be considered when interpreting these results. First, the retrospective design and reliance on electronic health record data from the TriNetX database may introduce misclassification bias, particularly with respect to the accurate identification and documentation of MD, as many cases remain undiagnosed or inadequately captured in administrative datasets. Additionally, the database does not allow differentiation of opioid prescriptions written specifically for ankle-related pain from those prescribed for other concurrent pain conditions, which may contribute to the overall elevation in opioid use observed across both cohorts. The same limitation applies to physical therapy utilization, which may reflect rehabilitation for comorbid conditions rather than exclusively ankle ORIF recovery. Second, residual confounding by unmeasured or incompletely captured variables such as MD severity, psychiatric medication adherence, adequacy of preoperative psychiatric optimization, socioeconomic status, nutritional status, or substance use cannot be excluded despite propensity score matching and adjustment for known covariates. Furthermore, the analysis did not account for other important coexisting variables that could function as confounders, including medical comorbidities such as cardiac disease and rheumatoid disorders, which may both worsen outcomes and be directly or indirectly associated with mood disorders. Third, despite propensity score matching on nine clinical and demographic variables, BMI remained modestly imbalanced between cohorts following matching (SMD 0.184), representing a potential limitation of this study. Notably, both groups fell within the Class I obesity range (31.5 vs 30.1 kg/m²), minimizing the likelihood that this difference meaningfully influenced clinical outcomes. Additionally the database does not capture patient-reported outcome measures, functional scores, or pain assessments, limiting our ability to assess the full impact of MD on patient experience and quality of life beyond objective clinical endpoints. Finally, the generalizability of these findings may be limited to populations represented within the TriNetX network, which may differ in demographic composition, healthcare access, or clinical practice patterns compared to other institutions or international settings.

Conclusion

In conclusion, this retrospective database cohort study found that pre-existing MDs were associated with significantly higher rates of postoperative opioid utilization, healthcare resource utilization including ED visits and hospital readmissions, and medical complications including SSI and DVT/PE following ankle ORIF. These findings suggest that MD exert substantial and sustained effects on multiple dimensions of postoperative recovery, extending well beyond the acute perioperative period. Clinically, this underscores the importance of routine preoperative screening for MD, comprehensive patient counseling regarding elevated complication risks, and implementation of multidisciplinary perioperative optimization strategies incorporating psychiatric consultation, multimodal analgesia, and enhanced postoperative monitoring.

Supplemental material