2025 ISAKOS Congress in Munich, Germany

2025 ISAKOS Biennial Congress Paper


Current Comorbidity Indices Inadequately Identify Patients Who Experience Poor Outcomes Following Total Shoulder Arthroplasty

Stephen C. Weber, MD, San Diego, CA UNITED STATES
Alexander R. Zhu, BA, Baltimore, Maryland UNITED STATES
Eve R. Glenn, ScB, Baltimore, Maryland UNITED STATES
Laurence Okeke, BA, Washington, D.C. UNITED STATES
Edward G McFarland, MD, FAAOS, Baltimore, Maryland UNITED STATES

The Johns Hopkins School of Medicine, Baltimore, Maryland, UNITED STATES

FDA Status Cleared

Summary

Current Comorbidity Indices do not Identify Patients Who Experience Poor Outcomes Following Total Shoulder Arthroplasty

Abstract

Background

There is increasing emphasis on patient-related outcomes for total shoulder arthroplasty (TSA), with a focus on identifying patients at increased risk of adverse postoperative outcomes. While some complications are inherent to surgery, many can be mitigated by identifying at-risk patients. Several comorbidity indices offer prognostic utility for postoperative complications, but few studies have assessed their performance for TSA. This study aimed to compare the prognostic value of four comorbidity indices in predicting 30-day outcomes post-TSA.

Methods

We retrospectively queried the National Surgical Quality Improvement Program (NSQIP) database, including patients who underwent primary TSA (CPT 23472) between 2011 and 2022 (N = 39,855). Logistic regression models were used to assess six outcomes: discharge to a non-home setting, length of stay >1 day, any 30-day major or minor complication, any readmissions within 30 days, and 30-day mortality. The concordance statistic (C-statistic) was calculated to measure the ability of these models to rank patients from low to high risk.

Results

Only the modified Charlson Comorbidity Index (mCCI) was a good predictor of discharge to a non-home setting (C-statistic = 0.711, 95% confidence interval 0.702–0.720). The American Society of Anesthesiologists Physical Status Classification system, the Elixhauser Comorbidity Measure, and the 5-Factor Modified Frailty Index were poor predictors of all other outcomes (C-statistic <0.7).

Conclusions

Of the indices evaluated, only the mCCI effectively predicted discharge destination after TSA. The poor performance of all four indices in predicting other outcomes underscores the need for a comorbidity scoring system specific to TSA.