Higher rates of mortality and perioperative complications in patients undergoing primary shoulder arthroplasty and a history of previous stroke.

Abstract

Cerebrovascular accidents (CVAs), or strokes, are the second most common cause of mortality and third most common cause of disability worldwide. Although advances in the treatment of strokes have improved survivorship following these events, there remains a limited understanding of the effect of a prior stroke and sequelae on patients undergoing shoulder arthroplasty (SA). This study aimed to determine the outcomes of patients with a history of stroke with sequela undergoing primary SA. Over a 30-year time period (1990-2020), 205 primary SAs (32 hemiarthroplasties [HAs], 56 anatomic total shoulder arthroplasties [aTSAs], and 117 reverse shoulder arthroplasties [RSAs]) were performed in patients who sustained a previous stroke with sequela and were followed for a minimum of 2 years. This cohort was matched (1:2) according to age, sex, body mass index, implant, and year of surgery with patients who had undergone HA or aTSA for osteoarthritis or RSA for cuff tear arthropathy. Mortality after primary SA was individually calculated through a cumulative incidence analysis. Implant survivorship was analyzed with a competing risk model selecting death as the competing risk. The stroke cohort sustained 38 (18.5%) surgical and 42 (20.5%) medical perioperative complications. Compared with the control group, the stroke cohort demonstrated higher rates of any surgical complication (18.5% vs 10.7%; P=.007), instability (6.3 % vs 1.7%; P=.002), venous thromboembolism (3.4% vs 0.5%; P=.004), pulmonary embolus (2.0% vs 0%; P=.005), postoperative stroke (2.4% vs 0%; P=.004), respiratory failure (1.0% vs 0%; P=.045), any medical complication (20.5% vs 7.3%; P<.001), and 90-day readmission (16.6% vs 4.9%; P<.001). Additionally, RSA in the stroke cohort was associated with higher reoperation (8.5% vs 2.6%; P=.011) and revision rates (6.8% vs 1.7%; P=.013) compared with the matched cohort. Subsequent cumulative incidences of death at 1, 2, 5, 10, 15, and 20 years were 4.4% vs 3.4%, 10.7% vs 5.1%, 25.6% vs 14.7%, 51.6% vs 39.3%, 74.3% vs 58.6%, and 92.6% vs 58.6% between the stroke and matched cohorts, respectively (P<.001). A preoperative diagnosis of a stroke in patients undergoing primary SA is associated with higher rates of perioperative complications and mortality when compared to a matched cohort. This information should be considered to counsel patients and surgeons to optimize care and help mitigate risks associated with the perioperative period.