Evaluating the utility of perioperative laboratory tests in primary shoulder arthroplasty

Abstract

BackgroundThe rising incidence of primary total shoulder arthroplasty (TSA) predicts a $4 billion increase in cost associated with this procedure by 2030. Studies have evaluated sources of cost containment, including inpatient services, namely routine perioperative laboratory tests. This study evaluates the utility of routine perioperative complete blood count (CBC) and basic metabolic panel (BMP) in TSA and identifies predictors of abnormal postoperative laboratory results and the need for medical intervention (NMI). MethodsAll patients undergoing primary TSA from 2013 to 2020 at a single institution were included. Demographics, comorbidities, procedural details, and preoperative and postoperative CBC and BMP values were recorded. Abnormal values were identified for each laboratory parameter. NMI was defined as postoperative blood transfusion, electrolyte repletion, fluid resuscitation, and inpatient medical consultation. Univariate and multivariate analyses were performed to determine independent predictors for abnormal postoperative CBC and BMP (AbBMP) results and NMI. A P value < .05 was considered statistically significant. ResultsA total of 328 patients were included in the study, with 202 (61.6%) female patients and a median age of 70.5 years. Medical intervention was required in 115 (35.1%). There were no differences in rates of NMI based on patient demographics, comorbidities, and preoperative CBC and BMP results, although there was a significant difference in rates of NMI when comparing postoperative serum potassium (4.1 mmol/L in no NMI group; 3.9 mmol/L in NMI group; P = .039). Ultimately, there were no demographic, surgical, or laboratory factors associated with postoperative NMI on predictive modeling. Increasing body mass index (odds ratio [OR] 1.08, 95% confidence interval [CI] 1.01-1.15, P < .03), preoperative serum glucose (OR 1.05, 95% CI 1.02-1.08, P = .003), and preoperative serum blood urea nitrogen (OR 1.12, 95% CI 1.05-1.12, P = .01) are associated with an 8.0%, 5.0%, and 12% increase in odds of an AbBMP, respectively. Preoperative potassium is inversely correlated with odds of AbBMP (OR 0.19, 95% CI 0.05-0.67, P = .01). ConclusionsAlthough postoperative laboratory tests were routinely obtained following primary shoulder arthroplasty, abnormal postoperative CBC and BMP were not associated with NMI. Postoperative BMP tests should be considered in patients with elevated body mass index, elevated preoperative serum glucose, elevated preoperative serum blood urea nitrogen, and those with preoperative serum hypokalemia. Postoperative CBC tests are likely not necessary after primary shoulder arthroplasty.