To the Editor:
In a large retrospective study, Memtsoudis et al.1 demonstrated that patients receiving neuraxial anesthesia for primary hip and knee arthroplasties had lower costs of hospitalization. Presumably, neuraxial anesthesia may provide a clinical benefit (e.g., decreased incidence of deep vein thrombosis, reduced cardiopulmonary complications, reduced opiate consumption, and lower incidence of postoperative delirium) when compared to general anesthesia for total joint replacements. Curiously, none of the clinical outcomes assessed in this study reached a level of significance, leaving three possible explanations for their results: a type II error, an accounting error, or some other downstream clinical benefit not measured in this trial. We believe that the first explanation is unlikely given the size of the study population and the third will require further study. However, the second explanation deserves a closer examination.
Previously, Adam et al.2 established that variability in costing methods applied raises questions about the validity of study results. Memtsoudis et al. extracted all the study data from the Premier Healthcare (Charlotte, North Carolina) database where the data elements associated with costs were dependent on each hospital’s accounting methodology. Further complicating the picture is the fact that a smaller number of hospitals submitted charges, which were subsequently converted using Medicare cost-to-charge ratios, for the purposes of the study. Incorporating charge data into costing analysis can be misleading because of the lack of fixed relationship between costs and charges. Although it is possible that the clinical benefits of neuraxial anesthesia extend beyond the intraoperative period, it is unlikely that a reduction in the length of stay or lower anesthesia costs led to the results from the study. Basques et al.3 showed no difference in length of stay between general and spinal anesthesia for total hip arthroplasty in more than 20,000 patients included from the American College of Surgeons National Surgical Quality Improvement Program (Chicago, Illinois) database. Even when it comes to anesthesia supply costs for neuraxial versus general anesthesia, Wanderer et al.4 recently argued that intraoperative “savings resulting from interventions focused on the clinical practice of attending anesthesiologists may be negligible.”
Long ago, Macario et al.5 suggested that “anesthesia practice patterns may influence downstream events in the hospitalization, some of which may have substantial economic impact.” Indeed, Memtsoudis et al. build upon the work examining the operative, safety, and patient-centered outcomes for a perioperative surgical home focused on total joint replacements.6 By including a cost analysis, however inaccurate, the results show that the preferred anesthetic of choice for primary arthroplasties may be some form of neuraxial anesthesia. Whether or not the administration of spinal and/or epidural anesthesia will result in indelible cost savings will remain a mystery until more rigorous cost accounting is undertaken.
The authors declare no competing interests.