Postoperative laboratory testing in the era of outpatient total joint arthroplasty: Targeted patient selection and associated cost savings.

ABSTRACT

Background: With the advent of outpatient total joint arthroplasty (TJA), the days of routinely drawing postoperative labs (complete blood counts [CBCs] and metabolic panels [CMPs/BMPs]) to monitor for complications are behind us. However, there does exist a subset of at-risk patients that may benefit from diligent postoperative monitoring, though the circumstances under which labs should be ordered remains unclear and subject to surgeon discretion. A systematic review of the literature was therefore conducted to evaluate the utility of postoperative laboratory testing, approaches to targeted patient selection and associated cost-savings. Methods: The PubMed, MEDLINE, EBSCOhost, and Google Scholar electronic databases were searched on August 17, 2023, to identify all studies published since January 1, 2000, that evaluated the role of postoperative lab testing in TJA. (PROSPERO study protocol registration CRD42023437334). Articles were included if a full-text English manuscript was available and the study assessed the utility of routine postoperative labs in TJA. 19 studies were included comprising 34,166 procedures. The mean Methodological index for Nonrandomized Studies score was 18.2 ± 1.5. Results: Abnormal postoperative lab results were common and infrequently required clinical intervention. Among several identified risk factors for patients that may benefit from postoperative laboratory monitoring, preoperative lab values proved excellent discriminators of transfusion requirement and metabolite-associated intervention. Selective testing demonstrated the ability to generate substantial cost-savings. Conclusion: Routine postoperative laboratory testing offers little clinical utility and produces unnecessary expenditures. Preoperative lab values offer the greatest predictive utility for postoperative transfusion requirement and metabolite-associated clinical intervention, with a preoperative hemoglobin threshold of 111.5 g/L offering an area under the curve (AUC) of 0.93 for predicting postoperative transfusion. Further investigations are needed for metabolic panel predictive models and should incorporate preoperative lab values. The refinement of such models can enable targeted patient selection to avoid unnecessary labs and generate substantial cost savings without compromising patient safety.