The reasons for readmission were available for 84 of the 93 readmitted patients. The most common reasons for readmission included pneumonia (14 cases, 16.7%), dislocation (7 cases, 8.3%), pulmonary embolism (7 cases, 8.3%), and surgical site infection (6 cases, 7.1%) (Table 4).
Table 4. Reasons for Readmission
Number | Percent | |
Pneumonia | 14 | 16.7% |
Dislocation | 7 | 8.3% |
Pulmonary embolism | 7 | 8.3% |
Surgical site infection | 6 | 7.1% |
Atrial fibrillation | 4 | 4.8% |
Hematoma | 4 | 4.8% |
Altered mental status | 3 | 3.6% |
Chest pain | 3 | 3.6% |
Renal insufficiency/kidney failure | 3 | 3.6% |
Urinary tract infection | 3 | 3.6% |
Acute gastric or duodenal ulcer | 2 | 2.4% |
Dermatitis/other allergic reaction | 2 | 2.4% |
Orthostatic hypotension/syncope | 2 | 2.4% |
Pain | 2 | 2.4% |
Respiratory distress | 2 | 2.4% |
Sepsis | 2 | 2.4% |
Urinary retention | 2 | 2.4% |
Acute cholecystitis | 1 | 1.2% |
Cerebrovascular accident | 1 | 1.2% |
Constipation | 1 | 1.2% |
Contusion of shoulder | 1 | 1.2% |
Deep venous thrombosis requiring therapy | 1 | 1.2% |
Gastrointestinal hemorrhage | 1 | 1.2% |
Gout | 1 | 1.2% |
Hepatic encephalopathy | 1 | 1.2% |
Intestinal infection | 1 | 1.2% |
Narcotic overdose | 1 | 1.2% |
Nausea/vomiting | 1 | 1.2% |
Proximal humerus fracture | 1 | 1.2% |
Rotator cuff tear | 1 | 1.2% |
Seroma | 1 | 1.2% |
Unspecified disease of pericardium | 1 | 1.2% |
Weakness | 1 | 1.2% |
DISCUSSION
Our analysis of 3042 TSAs from the NSQIP database suggests that unplanned readmission to the hospital occurs following about 1 in 40 cases of TSA. The study also suggests that the most common reasons for readmission encompass pneumonia, dislocation, pulmonary embolism, and surgical site infection. Old age, male sex, anemia, and dependent functional status serve as risk factors for readmission, and patients with such factors should be counseled and monitored accordingly.
In recent years, an increasing emphasis has centered on reducing rates of hospital readmission, with programs such as the Hospital Readmissions Reduction Program of the Affordable Care Act cutting reimbursements for hospitals with high 30-day readmission rates.17,18 To date, only a few studies have evaluated the reasons for readmission and readmission rates for TSA.19-23 Initial reports consisted of single-institution TSA registry reviews. For example, Mahoney and colleagues20 retrospectively evaluated shoulder arthroplasty procedures at their institution to document the readmission rates, finding a 5.9% readmission rate at 30 days. Readmission occurred more frequently in the first 30 days following discharge than in the 30- to 90-day period, with the most common reasons for readmission including medical complications, infection, and dislocation. Streubel and colleagues22 evaluated reoperation rates from their institution’s TSA registry, finding a 0.6% reoperation rate for primary TSA at 30 days and 1.5% for revision TSA. Instability and infection were the most common indications for reoperation. Our findings confirm these single-institution results and demonstrate their application to a nationwide sample of TSA, not just to high-volume academic centers. We similarly observed that dislocation, surgical site infection, and medical complications (mostly pneumonia and pulmonary embolism) were common causes of readmission, and that the 30-day readmission rate was about 1 in 40.
Several authors have since used statewide databases to analyze and determine risk factors for readmission following TSA. Lyman and colleagues19 used the New York State Database to show that higher hospital TSA surgical volume was associated with a lower rate of readmission when age and comorbidities were controlled for in a multivariate model. Old age was also associated with an increased readmission rate in their multivariate analysis, but comorbidities (as measured by the Charlson comorbidity index) presented a nonsignificant associative trend. These authors opted not to determine specific causes of readmission. Schairer and colleagues21 used State Inpatient Databases from 7 states, finding a 90-day readmission rate of 7.3%, 82% of which were due to medical complications and 18% of which were due to surgical complications (mostly infection and dislocation). Their multivariate regression revealed that male sex, reverse TSA, Medicaid insurance, patients discharged to inpatient rehabilitation or nursing facilities, medical comorbidities, and low-volume TSA hospitals were associated with readmission. Zhang and colleagues23 used the same source to show that the 90-day readmission rate reached 14% for surgically treated proximal humerus fractures and higher for patients who underwent open reduction internal fixation, were female, were African American, were discharged to a nursing facility, possessed Medicaid insurance, or experienced medical comorbidities. Most recently, Basques and colleagues31 analyzed 1505 TSA cases from 2011 and 2012 in the NSQIP database, finding a 3.3% rate of readmission, with heart disease and hypertension as risk factors for readmission. Although the limitations of the NSQIP database prevented us from analyzing surgeon and hospital TSA volume or reverse vs anatomic TSA, our results confirm that the findings from statewide database studies apply to the United States nationwide NSQIP database. Old patient age, male sex, and medical comorbidities (anemia and dependent functional status) are independent risk factors for TSA readmission. We identified pneumonia, dislocation, pulmonary embolism, and surgical site infection as the most common reasons for readmission.
This study features several limitations that should be considered when interpreting the results. Anatomic and reverse TSA share a CPT code and were not separated using NSQIP data. A number of studies have reported that reverse TSA may place patients at higher risk for readmission;20,21 however, confounding by other patient factors could play a role in this finding. The 30-day timeframe for readmission is another potential limitation; however, this timeframe is frequently used in other studies and is the relevant timeframe for the reduced reimbursement penalties from the Hospital Readmissions Reduction Program of the Affordable Care Act.18 Furthermore, the NSQIP database contains no information on surgeon or hospital TSA volume, which is a result of safeguards for patient and provider privacy. Additionally, readmission data were only available for 2011 to 2013, with causes of readmission only present in 2013. Although provided with such current information, we cannot analyze readmission trends over time, such as in response to the Affordable Care Act of 2010. Finally, although NSQIP surgical clinical reviewers strive to identify readmissions to other hospitals during their reviews of outpatient medical records, proportions of these readmissions are possibly missed. Therefore, our 30-day readmission rate may slightly underestimate the true rate.
Despite these limitations, the NSQIP database offers a unique opportunity to examine risk factors and reasons for readmission following TSA. The prior literature on readmission following TSA stemmed either from limited samples or administrative data, which feature known limitations.32 By utilizing a large, prospective, non-administrative, nationwide sample, our findings are probably both more reliable and generalizable to the country as a whole.
CONCLUSION
Unplanned readmission occurs following about 1 in 40 cases of TSA. The most common causes of readmission include pneumonia, dislocation, pulmonary embolism, and surgical site infection. Patients with old age, male sex, anemia, and dependent functional status are at a higher risk for readmission and should be counseled and monitored accordingly.
This paper will be judged for the Resident Writer’s Award.