Reverse Shoulder Arthroplasty in Nonagenarians - NIS-based study of perioperative and postoperative complications

Abstract

PurposeOur aim is to investigate the length of stay (LOS), cost of care (COC) and postoperative complications associated with reverse shoulder arthroplasty (RSA) in nonagenarians (people aged 90–99 years old). MethodsWe used the National Inpatient Sample (NIS) database to identify 59,925 patients who underwent RSA between 2016 and 2019, including 555 nonagenarians. We investigated the incidences of various medical and orthopedic postoperative complications in nonagenarians compared to their younger counterparts, as well as compared nonagenarians undergoing elective and non-elective surgery. ResultsNonagenarians were less likely admitted for elective surgery (69.9% in nonagenarians vs 92.8% in controls, p < 0.001) and were subject to longer LOS (3.5 days in nonagenarians vs 1.89 days in controls, p < 0.001) and greater COC ($91,794.69 US in nonagenarians vs $79,574.12 US in controls, p < 0.001). Nonagenarians had increased incidences of hospital mortality (0.72% in nonagenarians vs 0.06% in controls, p < 0.001), pneumonia (1.44% in nonagenarians vs 0.37% in controls, p < 0.001), blood loss anemia (22.34% in nonagenarians vs 10.12% in controls, p < 0.001), and ARF (6.85% in nonagenarians vs 2.18% in controls, p < 0.001). Nonagenarians undergoing elective RSA had fewer complications than those requiring non-elective RSA. ConclusionNonagenarians undergoing RSA are subject to increased LOS, COC, and postoperative complications. Despite this, we feel that the associated complications can be deemed acceptable and that with adequate preparation, the benefits of a successful RSA may outweigh the associated complications in elderly patients. This is important to aid clinicians and patients in making informed decisions for patient care and resource allocation, as well as highlights room for improvement in costs and hospital stay, as well as sheds light on persistent health disparities in orthopedic surgery. Level of evidenceIV