Objectives: Preoperative depression has been associated with worse outcomes following orthopaedic surgery including lower patient-reported outcomes, increased pain and impairment, and a higher rate of complications. Patient Health Questionnaire-2 (PHQ-2) is a commonly administered screening tool for measuring depressive symptoms, however the relationship between PHQ-2 and preoperative characteristics of patients presenting for arthroscopic knee and shoulder surgery have not been established. The purpose of this study was to investigate the association between depression and preoperative demographics and patient recorded outcomes of patients presenting for arthroscopic knee and shoulder surgery. Methods: This retrospective chart review evaluated data from all patients who underwent elective primary arthroscopic shoulder and knee surgery at 1 health system. Current Procedural Terminology codes were used to identify patients undergoing anterior cruciate ligament reconstruction (ACLR) (29888), meniscectomy (29880/29881), meniscus repair (29882/29883), knee chondroplasty (29877), labral repair (29807), rotator cuff repair (29827), and capsulorrhaphy (29806). Patients were included if they had completed Patient-Reported Outcomes Measurement Information System - Pain Interference (PROMIS-PI) and a PHQ-2 score at least 3 months prior to their surgery. The PHQ-2 is a validated tool used to screen for depression with scores ranging from 0-6. Patients were categorized as either depressed (PHQ-2 ≥ 3) or non-depressed (PHQ-2 < 3). Demographic characteristics and PROMIS scores were compared and a univariate and multivariate regression analysis was performed. Results: A total of 979 patients were included with 128 (13%) depressed patients and 851 nondepressed patients. There was no significant differences between the depressed and nondepressed group for age (49.18 ± 16.13 vs. 49.01 ± 15.71; P=0.926), sex (53.1% vs. 47.8% female; P = 0.263), median household income ($65,230 ± $24,684 vs. $66,810 ± $22,414; P=0.463), area deprivation index national percentile (62.48 ± 24.31 vs. 64.12 ± 24.58; P = 0.494), and depression diagnosis (4.7% vs. 3.1%; P = 0.333). Preoperative PROMIS-PI was significantly worse in the depressed cohort versus non-depressed (66.1 ± 5.9 vs. 63.6 ± 6.3; P < 0.001). Preoperative PROMIS-Depression score (R2 = 0.34) was the highest correlated variable to preoperative PHQ2 score in the univariate analysis. In the multivariate regression analysis, higher preoperative PROMIS-PI scores were found to be a risk factor for a PHQ2 score >3 (P < 0.001). Patients undergoing knee chondroplasty were less likely to be in the depressed cohort (odds ratio: 0.092 (0.009-0.914); P = 0.042). No other surgeries were more or less likely to be depressed. There was no difference between groups for age, sex, smoking status, insurance type, Area Deprivation Index, Metabolic Health Index, comorbid depression, and anxiety diagnosis. Conclusions: Overall, a larger number of patients screened positive for depression amongst patients undergoing knee and shoulder arthroscopy than were formally diagnosed. Demographic and socioeconomic variables were similar indicating these patients may be hard to identify without a screening tool like PHQ-2. Preoperative pain scores were higher amongst the depressed patients, indicating higher pain scores may be associated with increased risk of depression. This study highlights the need for further research on preoperative mood screening before arthroscopic surgery.
Read full abstract