Recent studies have described safe outcomes for short-stays in the hospital after total shoulder arthroplasty. The purpose of this study is to identify pre-operative and operative risk factors for hospital admissions exceeding 24 hours.

The American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database was queried from 2006 to 2016 for the current procedural terminology (CPT) billing code related to total shoulder arthroplasty. Patients were then grouped as either having a length of stay (LOS) equal to or less than 24 hours or greater than 24 hours. Patients admitted to the hospital prior to the day of surgery were excluded. Patient demographics, co-morbidities, and operative time were then analyzed as risk factors for a hospital stay exceeding 24 hours. Pre-operative co-morbidities included body mass index (BMI), diabetes, smoking, chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), hypertension, dialysis, chronic steroid or immunosuppressant use, bleeding disorders, and American Society of Anesthesiologists (ASA) Classification. Univariate and multivariate analyses were then performed to identify risk factors associated with 30-day readmission.

14,339 patients met inclusion criteria and 6,507 (45.3%) had a hospital LOS less than or equal to 24 hours. The mean length of hospitalization was 1.95 ± 1.88 days, the average age was 69 ± 9.7 years old, and 56.9% of the patients were female. Following a risk adjusted multivariate analysis, increasing age (odds ratio [OR], 1.03, 95% confidence interval [CI], 1.02–1.03), ASA classification (OR, 1.50, 95% CI, 1.41–1.60), diabetes (OR, 1.69, 95% CI, 1.43–1.99), COPD (OR, 1.35, 95% CI, 1.16–1.57), CHF (OR, 2.67, 95% CI, 1.34–5.33), dialysis (OR, 2.47, 95% CI, 1.28, 4.77), history of a bleeding disorder (OR, 1.50, 95% CI, 1.20–1.88), or increasing operative time (OR, 1.01, 95% CI, 1.01–1.01) were identified as independent risk factors for hospital lengths of stay exceeding 24 hours. Male gender was identified as a protective factor for prolonged hospitalization (OR, 0.50, 95% CI, 0.46–0.53).

This study identifies patient demographics, co-morbidities, and operative-relative risk factors that are associated with increased risk for a prolonged hospitalization following total shoulder arthroplasty. Female gender, increasing age, ASA classification, operative time, or a history of diabetes, COPD, CHF, or history of a bleeding disorder are risk factors hospitalizations exceeding 24 hours.

Malnutrition is an important consideration during the perioperative period and albumin is the most common laboratory surrogate for nutritional status. The purpose of this study is to identify if preoperative serum albumin measurements are predictive of infection following arthroscopic procedures.

Patients undergoing knee, shoulder or hip arthroscopy between 2006–2016 were identified in the American College of Surgeons National Surgical Quality Improvement Program database. Patients with an arthroscopic current procedural terminology code and a preoperative serum albumin measurement were included. Patients with a history of prior infection, including a non-clean wound class, pre-existing wound infection or systemic sepsis were excluded. Independent t-tests where used to compare albumin values in patients with and without the occurrence of a postoperative infection. Pre-operative albumin levels were subsequently evaluated as predictors of infection with logistic regression models.

There were 31,906 patients who met the inclusion criteria. The average age was 55.7 years (standard deviation (SD) 14.62) and average BMI was 31.7 (SD 7.21). The most prevalent comorbidities were hypertension (49.2%), diabetes (18.4%) and smoking history (16.9%). The average preoperative albumin was 4.18 (SD 0.42). There were 45 cases of superficial infection (0.14%), 10 cases of wound dehiscence (0.03%), 17 cases of deep infection (0.05%), 27 cases of septic arthritis or other organ space infection (0.08%) and 95 cases of any infection (0.30%). The preoperative albumin levels for patients who developed septic arthritis (mean difference (MD) 0.20, 95% CI, 0.038, 0.35, P = 0.015) or any infection (MD 0.14, 95% CI 0.05, 0.22, P = 0.002) were significantly lower than the normal population. Additionally, disseminated cancer, Hispanic race, inpatient status and smoking history were significant independent risk factors for infection, while female sex and increasing albumin were protective towards developing any infection. Rates of all infections were found to increase exponentially with decreasing albumin. The relative risk of infection with an albumin of 2 was 3.46 (95% CI, 2.74–4.38) when compared to a normal albumin of 4. For each albumin increase of 0.69, the odds of developing any infection decreases by a factor of 0.52.

This study suggests that preoperative serum albumin is an independent predictor of septic arthritis and all infection following elective arthroscopic procedures. Although the effect of albumin on infection is modest, malnutrition may represent a modifiable risk factor with regard to preventing infection following arthroscopy.