Institutional review board approval was obtained prior to the initiation of this retrospective observational cohort study. Inclusion criteria for this study consisted of all patients who underwent unilateral HRA between January 2012 and March 2018 at a single, urban, tertiary academic care center. A query was performed of the electronic data warehouse of the authors' institution, Caboodle (version 15; Epic, Verona, Wisconsin). Demographics characteristics (patient age, sex, body mass index [BMI], race, marital status, smoking status, insurance type, and American Society of Anesthesiologists score) and perioperative variables (anesthesia type, laterality, LOS, surgical duration, and discharge disposition) were obtained directly from the authors' electronic data warehouse. Complications and quality outcomes were obtained by a combination of manual chart review and through their institution's database, which included 1-year HRA-related readmission, peri-prosthetic fractures, 90-day infection, and 60-day emergency department visits.
Patients were subsequently divided into SDD HRA enrolled candidates vs non-SDD HRA candidates. Patients who enrolled into the SDD protocol but were discharged beyond postoperative day 0 were categorized as program failures. Patients in the non-SDD HRA cohort had either undergone surgery prior to the implementation of the SDD pathway or opted out of the program.
SDD Perioperative Management
All patients in this study underwent HRA performed by 1 of 5 orthopedic surgeons at the authors' institution. Each patient receiving SDD HRA was risk stratified and medically optimized prior to surgery. To qualify for SDD HRA, patients had to meet all the following criteria: no history of coronary artery disease or arrhythmias, no history of deep venous thrombosis or pulmonary embolism, not currently receiving chronic anticoagulation or antithrombotic agents, no history of moderate or severe sleep apnea, hemoglobin of 12 g/dL or greater, and a BMI of 40 kg/m2 or less. Patients were only included in the SDD group if they were discharged before midnight on the day of surgery.
Physicians discussed the risks, benefits, and alternatives to HRA with all patients. In addition, each patient had an individual 2-hour session with a clinical care coordinator and a physical and occupational therapist prior to his or her surgery. During this encounter, the patient was educated on the expected recovery course, pain management modalities, physical therapy exercises, postoperative expectations, and discharge planning. Inclusion in the SDD group required the presence of a designated family member or acquaintance who could transport the patient home and assist with activities of daily living in the immediate postoperative period. A maximum travel time of 1.5 hours from the institution to the patient's residence was also required for inclusion. Patients were free to withdraw from the program at any time prior to surgery.
All HRAs were performed using the Birmingham Hip Resurfacing System (Smith & Nephew, Memphis, Tennessee). Spinal anesthesia was attempted in all surgical candidates, with general anesthesia reserved only for patients with technically challenging spinal access. Additional anesthetic medications administered during the surgery included intravenous fentanyl, propofol, midazolam, and dexamethasone. All patients received standardized antibiotics for 24 hours initiated within 1 hour prior to the surgical start time. The SDD patients received 1 dose of intravenous cefazolin perioperatively. If they were present in-house at the time for administration of the second dose, they would receive it then; if not, they were discharged with 2 oral doses of cephalexin to take every 8 hours.
A uniform perioperative multimodal pain regimen was established, which reduced the need for narcotics. Patients were administered several nonnarcotic medications, including oral acetaminophen and celecoxib, pre- and postoperatively. An intra-articular injection of 30 mL of 0.25% bupivacaine and liposomal bupivacaine was administered prior to closure. Oral and intravenous opioid administration was strongly discouraged unless all nonnarcotic alternatives were deemed ineffective. Patient- controlled analgesia was reserved for severe refractory cases.
Postoperatively, all patients received thromboprophylaxis agents consisting of 81 or 325 mg of aspirin orally twice daily for 4 weeks and mechanical compression devices for 2 weeks. Active smokers and patients in the non-SDD HRA group with a history of a thrombotic event or a hyper-coagulable state were prescribed 40 mg of enoxaparin sodium subcutaneously daily for 4 weeks instead of aspirin and mechanical compression devices, as per the authors' institutional protocol. Of note, no patients with a history of thrombotic event or hypercoagulable state were included in the SDD group.
After the patients had been deemed safe for discharge, all SDD patients were discharged home with either home health care services or self-care. On postoperative day 1, patients received a call from the clinical care coordinator to ensure that they were recovering appropriately. In addition, physical therapists and visiting nurses followed-up intermittently with each HRA recipient at the patient's residence to assist with wound care and rehabilitation for 2 weeks. All other details of postoperative management were surgeon-specific and identical to their standard of care practices for inpatient HRA.
All data were collected and transformed using Excel for Mac 2011 version 14.6.2 (Microsoft Corp, Redmond, Washington). Statistical analysis was performed using Prism 7 software (GraphPad, La Jolla, California). Baseline and demographic characteristics were summarized by standard descriptive summaries (eg, means and SD for continuous variables). An analysis was performed comparing baseline patient characteristics and outcomes between SDD and non-SDD HRA recipients. Categorical baseline characteristics and outcomes were analyzed using a chi-square test with Fisher's P reported between the 2 groups. A Student's t test was used to compare group means between the 2 groups. In all instances, P<.05 was deemed significant.