In-Hospital Complications following Arthrotomy versus Arthroscopy for Septic Knee Arthritis: A Cohort-Matched Comparison.

Abstract

There is a paucity of literature comparing the relative merits of open arthrotomy versus arthroscopy for the surgical treatment of septic knee arthritis. The primary goal of this study is to compare the risk of perioperative complications between these two surgical techniques. To this end, 560 patients treated for septic arthritis of the native knee with arthroscopy were statistically matched 1:1 with 560 patients treated with open arthrotomy. The outcome measures included major complications, minor complications, mortality, inpatient hospital charges, and length of stay (LOS). Major complications were defined as myocardial infarction, cardiac arrest, stroke, deep vein thrombosis, pulmonary embolism, pneumonia, postoperative shock, unplanned ventilation, deep surgical site infection, wound dehiscence, infected postoperative seroma, hospital acquired urinary tract infection, and retained surgical item. Minor complications included phlebitis and thrombophlebitis, postprocedural emphysema, minor surgical site infection, peripheral nerve complication, and intraoperative hemorrhage. Mortality data were extracted from the database using the Uniform Bill patient disposition. Complications were analyzed using univariate and multivariate logistic regression models, whereas mean costs and LOS were compared using the Kruskal-Wallis H-test. Major complications occurred in 3.8% of the patients in the arthroscopy cohort and 5.4% of the patients in the arthrotomy cohort (p = 0.20). Too few patients in our sample died to report based on National (Nationwide) Impatient Sample (NIS) minimum reporting standards. Rates of minor complications were similar for the arthroscopy and arthrotomy cohorts (12.5 vs. 13.9%; p = 0.48). Multivariate analysis did not reveal any greater risk of minor or major complication between the two procedures. Inpatient hospital cost was similar for arthroscopy ( = $15,917; standard deviation [SD] = 14,424) and arthrotomy ( = $16,020; SD = 18,665; p = 0.42). LOS was also similar for both arthrotomy (6.78 days, SD = 6.75) and arthroscopy (6.24 days, SD = 5.95; p = 0.23). Patients undergoing arthroscopic treatment of septic arthritis of the knee showed no difference in relative risk of perioperative complications, LOS, or hospital cost compared with patients who underwent open arthrotomy.