Spinal Anesthesia Is Associated With Decreased Complications After Total Knee and Hip Arthroplasty

Jared Warren, DO, ATC, CSCS; Kavin Sundaram, MD, MSc; Hiba Anis, MD; Atul F. Kamath, MD; Michael A. Mont, MD; Carlos A. Higuera, MD; Nicolas S. Piuzzi, MD


J Am Acad Orthop Surg. 2020;28(5):e213-e221. 

In This Article

Abstract and Introduction


Background: We compared the following 30-day outcomes for total knee arthroplasty (TKA) and total hip arthroplasty in spinal anesthesia (SA) versus general anesthesia (GA) (1) mortality, (2) major and minor complication rates, and (3) discharge disposition.

Methods: From 2011 to 2016, the American College of Surgeons National Surgical Quality Improvement Program database contained 45,871 SA total hip arthroplasties and 65,092 receiving GA. There were 80,077 SA TKAs and 103,003 GA TKAs. Adjusted multivariate logistic regression evaluated associations between anesthesia type and 30-day outcomes.

Results: Anesthesia modality was not associated with 30-day mortality (P > 0.05). The GA cohorts were at a greater risk for any complication, major complications, and minor complications (P < 0.05). Patients who received GA were at an increased risk for nonhome discharge.

Conclusion: Patients who undergo total joint arthroplasty with SA experience fewer 30-day complications and are less likely to have a nonhome discharge than those with GA.


Total joint arthroplasty (TJA) is a highly successful treatment for the management of advanced osteoarthritis (OA) in both hips and knees.[1–5] Management of both the psychologic and physiologic aspects of pain associated with this routine surgical operation is a cooperative endeavor between the orthopaedic surgeon, patient, and anesthesia provider.[6] The shared decision-making regarding the type of anesthesia to be used comprises multiple factors that balance patient's comorbidities, safety, and preferences. It has long been established that the choice of anesthesia has implications for patient satisfaction as well as the rate of complications.[7–9]

Previous studies have suggested that spinal anesthesia (SA) may be more beneficial than general anesthesia (GA) for preventing complications in elective TJA. A systematic review of 18 randomized control trials involving total knee arthroplasty (TKA) compared regional anesthesia with GA and found that there was markedly less blood loss with regional anesthesia.[10] In addition, SA has been associated with a markedly lower incidence of surgical site infection (SSI) when compared with GA in TKA and total hip arthroplasty (THA).[11] Conversely, a multicenter prospective observational trial of 2,162 patients, of which 425 received SA, found that SA was an independent risk factor for venous thromboembolism,[12] although regional anesthesia was associated with reductions in postoperative pain, morphine consumption, nausea, and vomiting. A Taiwanese database study found that patients receiving SA had an increased risk of coronary artery disease after TKA.[13]

Outcomes associated with the type of anesthesia have been equivocal, despite overall benefits favoring SA. A few studies have used the National Improvement Quality Improvement Program (NSQIP) database to analyze SA versus GA in TJA; however, these studies were insufficiently powered to draw conclusions on low-frequency complications, such as pulmonary embolism (PE) and cardiac arrest.[14,15] In addition, meta-analyses and systematic reviews have been undertaken, but none have looked a number of specific complications such as, revision surgery, cardiac arrest, sepsis/septic shock, and renal complications.[8,16] To our knowledge, the literature lacks data on short-term and perioperative outcomes in a large United States sample. As such, the aims of this study were to investigate whether GA is an independent risk factor when compared with SA for (1) 30-day mortality, (2) major and minor complication rates, (3) increased hospital length of stay (LOS), and (4) discharge disposition.