Author + information
- Paul L. Hess, MD,
- Melissa A. Greiner, MS,
- Sana M. Al-Khatib, MD, MHS,
- Frederick A. Masoudi, MD, MSPH,
- Paul D. Varosy, MD,
- Richard I. Fogel, MD,
- Lesley H. Curtis, PhD and
- Adrian F. Hernandez, MD, MHS∗ ()
- ↵∗Duke Clinical Research Institute, P.O. Box 17969, Durham, North Carolina 27715
More than 100,000 patients annually receive an implantable cardioverter-defibrillator (ICD) in the United States. Placement techniques for ICDs have evolved and become safer, and hospital length of stay has declined accordingly. Patients who receive an ICD electively now are routinely discharged the next day.
Benefits of shorter length of stay after ICD placement include patient convenience, increased bed availability, and the potential for lower costs. However, early discharge may accentuate the impact of known complications, including mechanical complications requiring system revision, hemothorax or pneumothorax, cardiac perforation, infection, bleeding, and death. Evaluation of same-day discharge after ICD placement has been limited to small, single-center studies in select patients (1,2). Using the National Cardiovascular Data Registry’s ICD Registry linked with Medicare claims, we assessed the prevalence of same-day discharge among older patients after elective ICD placement in the United States; variation in discharge strategy over time and across hospitals; and associations with death, all-cause readmission, and device-specific readmission.
We included patients ≥65 years of age who had a registry record for an ICD implantation between April 1, 2006, and December 31, 2009; those who were enrolled in fee-for-service Medicare at the time of the procedure; and those who could be linked to Medicare claims data (n = 206,611). We excluded those patients who had a prior ICD implantation (n = 60,667); underwent implantation for secondary prevention, including sustained ventricular tachycardia (n = 30,589); had New York Heart Association functional class IV symptoms (n = 4,835); were admitted for a reason other than ICD implantation (n = 32,880); died in the hospital during the index admission (n = 91); were discharged to a facility other than home (n = 4,651); or had a hospital length of stay >1 day (n = 14,703). We defined same-day discharge on the basis of an admission date equal to the discharge date on the Medicare inpatient or outpatient claim. Overnight observation was based on a discharge date 1 day after the admission date on the claim. We determined all-cause death from death dates in the Medicare denominator files. Readmission outcomes were based on subsequent inpatient claims. We used Kaplan-Meier methods to estimate death and log-rank tests to assess differences in death between groups. We used the cumulative incidence function for other outcomes and Gray tests to assess differences in outcomes between the groups. We used Cox proportional hazards models with robust standard errors for hospital clustering to examine univariate and multivariable associations between same-day discharge and outcomes.
The final study sample consisted of 58,195 patients from 1,314 sites; 3,083 (5.3%) were discharged on the same day as the ICD placement. Compared with patients who were observed overnight, those who were discharged the day after the procedure were less frequently under the care of physicians trained in electrophysiology or thoracic surgery and more commonly underwent device implantation in private-practice hospitals rather than government or academic centers. Hospitals performing same-day discharges were less often located in rural areas and more commonly suburban areas. Rates of same-day discharge varied by hospital, from 0% in 846 hospitals (64%) to ≥50% in 27 hospitals (2%; interquartile range 0.0% to 3.0%; range 0.0% to 100%). The overall rate increased from 4.1% in 2006 to 6.6% in 2009. Table 1 shows the observed cumulative incidence of death, all-cause readmission, and device-related readmission among patients with same-day discharge and those observed overnight, as well as associations with study outcomes according to the number of days after discharge. There were no significant differences in observed rates between the groups. After multivariable adjustment, same-day discharge was not significantly associated with death, all-cause readmission, or device-related readmission 90 days after discharge.
Practitioners routinely perform a physical examination, obtain a chest x-ray, and perform device interrogation the day after an elective procedure and prior to patient discharge. For appropriate patients, a new practice model may incorporate novel care processes but not require overnight observation, particularly if the patient undergoes ICD placement early in the day. These care processes include monitoring for several hours after the procedure while the patient recovers from sedation, follow-up chest x-ray and device interrogation, effective communication regarding wound care, and arrangement of close outpatient follow-up.
A number of ICD recipients presently monitored overnight may undergo same-day discharge with appropriate follow-up without increased risk of adverse events. In fact, patients who undergo ICD placement late in the day followed by discharge the next morning may receive a comparable level of care. Centers considering implementation of same-day discharge may benefit from developing protocols to identify eligible patients, monitor them adequately after the procedure, and arrange close outpatient follow-up. Same-day discharge is not suitable for all patients, because some patients are at sufficiently high risk to warrant observation for longer periods. Patients who undergo cardiac resynchronization therapy often have advanced heart failure and significant comorbidity and may require close monitoring. Patients with known procedural complications also may benefit from a longer observation period. Standardized criteria for same-day discharge developed by professional societies or hospitals would be valuable. In the absence of such criteria, discharge strategy should be carefully selected on a patient-by-patient basis by the treating physician.
Similar to same-day discharge after percutaneous coronary intervention, same-day discharge after ICD placement may lead to greater patient satisfaction, increased bed availability for patients with greater clinical need, and cost savings without adversely affecting readmission rates (3). Rates of readmission are particularly relevant, as they are already considered a quality metric for disease states such as heart failure and may be bundled to reimbursement for the index hospitalization in the future. Discharge policy changes may be implemented in select patients without adversely affecting readmission rates and in turn public perception regarding quality of care or reimbursement. New healthcare models should consider the value and outcomes of same-day discharge for appropriate patients.
Please note: This work was supported internally by the Duke Clinical Research Institute. Dr. Hess was supported by grant T32HL069749-09 from the National Heart, Lung, and Blood Institute; and has served as a consultant for Sanofi. Dr. Masoudi has received grant and contract funding through his institution from the American College of Cardiology and the Oklahoma Foundation for Medical Quality; and has served as the senior medical officer for the National Cardiovascular Data Registries as well as a clinical consultant for the Oklahoma Foundation for Medical Quality. Dr. Fogel has served as a speaker for Medtronic; and has served as an expert witness for St. Jude and Biotronik. Dr. Curtis has received research grants from GlaxoSmithKline, GE Healthcare, Johnson & Johnson, Boston Scientific, and Novartis. Dr. Hernandez has served as a consultant for AstraZeneca, Bristol-Myers Squibb, Corthera, Cytokinetics, and Johnson & Johnson; and received grants from Amylin, Boston Scientific, Bristol-Myers Squibb, Johnson & Johnson, Medtronic, and Portola. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the National, Heart, Lung, and Blood Institute or the National Institutes of Health. Damon M. Seils, MA, Duke University, provided editorial assistance and prepared the manuscript. Seils did not receive compensation for his assistance apart from his employment at the institution where the study was conducted.
- American College of Cardiology Foundation