Use of total shoulder arthroplasty (TSA) and reverse TSA for shoulder conditions has increased dramatically in recent years.1 Approximately 27,000 standard TSAs were performed in the United States in 2008, and this number is expected to double by 2015.2 TSA provides excellent pain relief, restoration of function, and patient satisfaction.3 The evolution of implant design over the past 25 years has contributed to excellent long-term implant survival, with rates comparable to those of total knee and hip arthroplasty.4 Similarly, compared with previous designs, contemporary designs and techniques have resulted in fewer complications.5
Several studies have investigated the long-term complications of TSA. These complications include prosthetic loosening, instability, periprosthetic fracture, rotator cuff tears, nerve injury, and deltoid dysfunction.6-11 In addition, Waterman and colleagues11 very recently assessed the influence of risk factors on short-term postoperative complications of TSA. However, none of these studies has assessed the influence of multiple risk factors on postoperative length of stay (LOS) after TSA. Only 1 study, using data from 2005 and earlier, has analyzed the potential effect of multiple patient characteristics on readmission after TSA12; other studies have been only descriptive.13-16
We conducted a retrospective cohort study to characterize the risk factors for extended LOS and readmission after TSA in a large sample of patients drawn from a national database. We hypothesized that patient factors, including age, sex, and obesity, would be significantly associated with postoperative LOS and readmission after TSA. National databases have been increasingly used in orthopedic research, as they offer particular advantages. Large sample sizes allow for powerful analyses of associations—analyses previously not possible in single-surgeon and single-institution studies. In addition, use of a large, national patient sample allows us to draw generalizable conclusions to better define patients’ and physicians’ postoperative expectations.
Methods
We conducted a retrospective cohort study using the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database. ACS-NSQIP collects 150 patient variables from 374 participating US hospitals.17 Patients are prospectively identified, and information is collected from operative reports, medical records, and patient interviews by trained clinical reviewers.17,18 Routine auditing by the program ensures high-quality data, with reported interrater disagreement below 2% for all variables. Data are collected through the 30th postoperative day, including after discharge.
This study was granted an exemption from our institutional review board, as we used a deidentified and publicly available database. Patients who were 60 years or older and underwent TSA between 2011 and 2012 were identified in the ACS-NSQIP database. TSA patients were identified using Current Procedural Terminology (CPT) code 23472, which includes TSA and reverse TSA procedures.
Patients were divided into groups based on surgical indications, which were available as International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes. Patients with postoperative ICD-9 codes 714.0 (rheumatoid arthritis), 715.0-9 (osteoarthritis), 716.61/716.81/716.91 (unspecified arthropathy), 718.01 (articular cartilage disorder), 718.31 (recurrent dislocation of shoulder), 718.81 (other joint derangement of shoulder), 719.41/719.91 (unspecified shoulder pain/disorder), 726.0-2 (disorder of shoulder tendons and bursa), 727.61 (rotator cuff rupture), and 840.3-9 (rotator cuff sprain) were classified as having a nonfracture indication. Patients with postoperative ICD-9 codes 716.11 (traumatic arthropathy), 833.80-89 (malunion/nonunion of fracture), and 812.00-20 (fracture of proximal humerus) were classified as having a fracture-associated indication. Patients with incomplete perioperative data were excluded from the study, leaving 1505 patients for the study (out of an initial 1726).
Patient characteristics, including sex, age, height, weight, and history of smoking, were collected from the ACS-NSQIP database. Body mass index (BMI) was calculated from each patient’s height and weight. Information about medical comorbidities was also collected from the ACS-NSQIP database. History of pulmonary disease was defined as a history of dyspnea, severe chronic obstructive pulmonary disease, ventilator-assisted respiration within 48 hours before surgery, or current pneumonia. History of heart disease was defined as a history of congestive heart failure or angina within 1 month before admission, myocardial infarction within 6 months before admission, cardiac surgery, or percutaneous coronary intervention. American Society of Anesthesiologists (ASA) class 3 or higher indicates severe systemic disease. Steroid use was defined as regular administration of corticosteroid medications within 30 days before surgery. Functional status was defined as the ability to perform activities of daily living (ADLs) within 30 days before surgery, with the patient’s best functional status during this period recorded. Similar to how other variables were collected from the database, this information was obtained through medical record abstraction and patient interviews by trained personnel. ADLs are defined in the ACS-NSQIP as “activities usually performed in the course of a normal day in a person’s life” and include bathing, feeding, dressing, toileting, and mobility. An independent patient does not require assistance for any ADLs, a partially dependent patient requires assistance for some ADLs, and a totally dependent patient requires assistance in completing all ADLs. Partially and totally dependent patients were grouped for analysis. Information about a patient’s discharge destination (to home or a facility) was also available in the database.17