The Journal of Arthroplasty, Volume 34, Issue 11, 2594 - 2600

The Role of Malnutrition in Ninety-Day Outcomes After Total Joint Arthroplasty

Black, Collin S. et al.
Hip Knee


Research has linked malnutrition to more complications in total joint arthroplasty (TJA) patients. The role of preoperative albumin in predicting length of stay (LOS) and 90-day outcomes remains understudied. Often, an albumin cut-off ≤3.5 g/dL is used as proxy for malnutrition, although this value remains understudied. This preoperative level may be missing some patients at risk for adverse events post TJA.


TJA patients at a single institution from 2013 to 2018 were reviewed for preoperative albumin level. In total, 4047 cases (total knee arthroplasty: 2058; total hip arthroplasty: 1989) had available data, including 90-day readmissions, 90-day emergency department (ED) visits, and postoperative LOS.


About 5.6% experienced a readmission and 9.6% had at least one ED visit within 90 days. Overall prevalence of malnutrition was 3.6%, and this cohort experienced a longer average LOS (3.5 vs 2.2 days, P < .0001) and was more likely to experience a readmission (16% vs 5%, P < .0001) or ED visit (18% vs 9%, P = .0005). Additionally, albumin ≤3.5 g/dL was correlated with more frequent discharge to skilled nursing facility/rehab (30.8% vs 14.7%, P < .0001), increased risk for 90-day readmission with univariable (odds ratio [OR] 1.79, P < .0001) and multivariable logistic regression (OR 1.55, P < .0001), and increased risk for 90-day ED visits with univariable (OR 1.62, P < .0001) and multivariable regression (OR 1.35, P < .0001). The optimal albumin cut-off was 3.94 g/dL in a univariable model for 90-day readmission.


Screening for malnutrition may serve a role in preoperative evaluation. An albumin cutoff value of 3.5 g/dL may miss some at-risk patients.

Download article