Surgical Treatment of Septic Shoulders: A Comparison Between Arthrotomy and Arthroscopy.

PURPOSE: To compare the efficacy, as measured through the rate of reoperation, and rates of other 30-day perioperative complications between arthrotomy and arthroscopy for the treatment of septic native shoulders in a national patient population.

METHODS

Patients who were diagnosed with septic arthritis in a native shoulder and underwent irrigation and debridement through arthrotomy or arthroscopy were identified in the 2005-2016 National Surgical Quality Improvement Program database. Patient preoperative characteristics were characterized. Rate of reoperation, a proxy used to measure treatment efficacy, and other perioperative complications were compared between the 2 procedures.

RESULTS

In total, 100 patients undergoing shoulder arthrotomy and 155 patients undergoing shoulder arthroscopy for septic shoulder were identified. On univariate analysis, there were no statistically significant differences in patient preoperative characteristics, operative time (60 vs. 48 minutes, P = .290), length of stay (7.5 vs. 6.6 days, P = .267), or time to reoperation (8.9 vs. 7.2 days, P = .594) between the 2 surgical groups. On multivariate analysis controlling for patient characteristics, there were no statistically significant differences in risk of reoperation (relative risk [RR] = 1.914, 99% confidence interval [CI] = 0.730-5.016, P = .083), any adverse events (RR = 1.254, 99% CI = 0.860-1.831, P = .122), minor adverse events (RR = 1.304, 99% CI = 0.558-3.047, P = .421), serious adverse events (RR = 1.306, 99% CI = 0.842-2.025, P = .118), or readmission (RR = 0.999, 99% CI = 0.441-2.261, P = .998) comparing arthrotomy with arthroscopy.

CONCLUSIONS

By demonstrating similar rates of reoperation, other postoperative complications, and 30-day readmissions, the current study suggests that arthrotomy and arthroscopic surgery have similar efficacy in treating septic shoulders. However, owing to the small sample size, there is still the possibility of a type II error.

LEVEL OF EVIDENCE: Level III, therapeutic retrospective comparative study.