Safety of Same-Day Discharge Compared to Overnight Observation Following Laparoscopic Appendectomy in the Pediatric-Aged Patient: A National Surgical Quality Improvement Program Project

Introduction

Acute appendicitis is one of the most common surgical conditions in pediatric-aged patients and often requires surgical intervention. In the United States, more than 80,000 appendectomies are performed annually. ¹ Several centers have made efforts to reduce hospital stay after appendectomy with the performance of outpatient appendectomy and same-day discharge (SDD).^2–4 The process is feasible as postoperative complications are generally minor and uncommon with an infection rate of 3.1% and a 30-day readmission rate ranging from 1.1% to 3.9%.^5–7

Recent quality improvement (QI) projects, primarily single-center studies, have focused on increasing SDD after appendectomy, as overnight observation is still common after these procedures. In response to the move toward SDD, Litz et al. evaluated the safety of pediatric outpatient appendectomy by comparing their single-center outcomes from a cohort of 154 patients with the National Surgical Quality Improvement Program-Pediatric (NSQIP-P) registry, a national database that includes patient information from a sample of pediatric procedures performed at participating institutions.^2,8 The institutional rate of outpatient management was higher compared to the NSQIP-P data (84% versus 48%, P < .0001), and surgical length of stay (LOS) was shorter (0.3 ± 0.7 versus 1.1 ± 1.9 days, P < .0001). There was no significant difference in the incidence of superficial, deep, or organ/space surgical site infections. The incidences of other complications (1.3% versus 0.6%, P = .26) and 30-day readmission rate (3.2% versus 2.6%, P = .61) were similar.

A second issue that has been raised by previous studies using the National Surgical Quality Improvement Program (NSQIP) registry is that hospitals may use outpatient status based on their own internal definition and it may still include one or more nights spent in the hospital.^9–12 Bovonratwet et al. reported that the definition of “inpatient” and “outpatient” status used by hospital in the United States often was not synonymous with LOS. ¹³ Varying institutional definitions of outpatient and inpatient may have influenced the results of these previous studies using the NSQIP database with the terms inpatient and outpatient.

Since the latest QI projects in pediatric appendectomy have specifically emphasized SDD, we used the NSQIP-P registry to compare pediatric laparoscopic appendectomy outcomes according to LOS. The primary purpose of this study was to evaluate the safety of appendectomy performed with SDD in pediatric patients when compared with appendectomy followed by an overnight stay. We hypothesized that 30-day unplanned readmissions and postsurgical complications are equivalent between patients discharged on the day of operation and patients discharged after one night of hospital stay.

Materials and Methods

This study was deemed exempt from review by the Institutional Review Board at Nationwide Children's Hospital. The analysis included de-identified data from the years 2012 through 2016 of the NSQIP-P registry. The NSQIP database is maintained by the American College of Surgeons and includes more than 120 patient perioperative variables, including preoperative risk factors, intraoperative variables, and 30-day postoperative mortality and morbidity outcomes for patients undergoing major surgical procedures in both the inpatient and outpatient setting from 100 participating sites. ⁸ The data were collected by certified Surgical Clinical Reviewers, and interrater reliability and validity of the data have been previously reported. ¹⁴

We included patients younger than 18 years of age, who were categorized as “outpatient” in the registry, and underwent laparoscopic appendectomy from 2012 to 2016. Patients were selected for inclusion if they had a Current Procedural Terminology code of 44970, corresponding to laparoscopic appendectomy. Patients were classified as not undergoing overnight observation if the time from operation to discharge was 0 days (SDD). Patients with SDD were compared with patients discharged after one night of hospital stay. We excluded patients admitted before the day of surgery, patients discharged after more than one night of hospital stay, and patients with missing data on outcomes or characteristics.

The primary outcome was 30-day unplanned readmission, and the secondary outcome was the occurrence of any 30-day complication, including readmission, wound complications (including superficial, deep, and organ/space surgical site infections), cardiac arrest, tracheal reintubation, urinary tract infection, pneumonia, renal failure, neurological complication (including seizure, coma, cerebrovascular events, or intraventricular hemorrhage), thromboembolic complication, or 30-day mortality.

Patient characteristics included gender, age, race (non-Hispanic white, non-Hispanic black, other), body-mass index, American Society of Anesthesiologists physical classification, cardiac risk factors (none, minor, major, and severe), and year of service. Continuous variables were summarized as mean with standard deviation and compared using unpaired t-tests, and categorical data were summarized as count with percentages compared using chi-square tests or Fisher's exact tests, as appropriate.

We used propensity score matching to match each patient discharged on the day of operation to a similar patient discharged after one night of hospital stay. The propensity model was a logistic regression of time to discharge (same-day versus next-day discharge), including all patient characteristics as independent variables, and the propensity score was the linear prediction of the model. ¹⁵ To obtain the most similar unique next-day discharge patient for each SDD patient, we used one-to-one nearest neighbor matching without replacement. A caliper width of 0.2 standard deviations of the propensity score was used to ensure similarity between patients in the matched sample who were discharged the same day and who were discharged the next day. The balance of characteristics between these groups was checked using standardized difference, where a standardized difference <0.1 indicated acceptable balance.^16,17 The outcomes, including 30-day unplanned readmission and any 30-day complication, were evaluated in the matched sample using conditional logistic regression (stratified on matched same-day and next-day discharge pairs). Data analysis was performed in STATA/IC 14.2 (StataCorp LP, College Station, TX), and two-tailed P < .05 was considered statistically significant.

Results

We identified 37,326 patients who underwent laparoscopic appendectomy from 2012 to 2016, of whom 15,785 were excluded due to a LOS >1 day, 2451 were excluded due to admission before date of surgery, 9131 were excluded for undergoing an inpatient procedure, and 2267 were excluded due to missing data on outcomes or characteristics. Of the remaining 7692 patients, 2647 were discharged on the day of surgery, and 5045 were discharged after one night of hospital stay. After propensity score matching on patient characteristics, 2443 patients discharged on the same day were matched to controls discharged following one night of hospital stay.

Patient characteristics of the total population are summarized in Table 1. The overnight patients were more likely to be non-Hispanic black and other races. In the SDD group, there was a significant increase in the number of cases between 2014 and 2015, showing a movement toward SDD following laparoscopic appendectomy. At least one complication occurred in 246 (3%) patients, including 141 (2%) cases of unplanned readmission, 147 (2%) cases of wound complications, 2 (0.03%) cases of pneumonia, 6 (0.1%) cases of urinary tract infection, and 1 (0.01%) case of renal failure. There were no instances of tracheal reintubation, neurological complications, thromboembolic complications, cardiac arrest, or 30-day mortality.

Characteristics of the propensity-matched cohort are summarized in Table 2. After propensity score matching on patient characteristics, 2443 patients discharged on the same day were matched to controls discharged following 1 night of hospital stay. Covariates were well balanced between SDD and overnight stay. On stratified conditional logistic regression, SDD was not associated with an increased risk of 30-day unplanned readmission (odds ratio [OR] = 0.9, 95% confidence interval [CI] = 0.6–1.4; P = .667) or the occurrence of any 30-day complication (OR = 0.8, 95% CI: 0.6–1.1; P = .258).

Discussion

Given the recent trends in fast-tracking pediatric patients following laparoscopic appendectomy with the goal of SDD, we sought to evaluate the safety of this technique using a large national database. This study investigated two outcomes (30-day readmission rate following surgery and the occurrence of complications) following laparoscopic appendectomy performed with SDD, compared with that followed by an overnight stay. Using propensity matching within the NSQIP database to adjust for demographic and clinical characteristics, we noted that SDD was not associated with an increased postsurgery 30-day readmission rate or an increased likelihood of complications, compared with patients discharged after one night of hospital stay.

Even after using a strict definition for SDD in a very large cohort of patients, this outcome is consistent with previous retrospective and prospective studies evaluating SDD for uncomplicated appendicitis in pediatric patients.^2–4,12,18 Previous studies have found that SDD for uncomplicated appendectomy provides safe outcomes that are comparable with overnight stay after the procedure, thereby demonstrating the safety of the technique with a reduction in medical cost with improved patient and family satisfaction.^2,4,12,18,19

Aguayo et al. performed a single center, retrospective study of 588 laparoscopic appendectomies for nonperforated appendicitis in pediatric patients and reported that readmission rate and complication rate for SDD patients were not different from patients who had an overnight stay. ³ A prospective observational study of 158 children reported no difference in readmission and complication rate in SDD patients after laparoscopic appendectomy. ⁴ Similar results were reported by Alkhoury et al. and Halter et al. in cohorts of 158 and 236 pediatric patients following SDD after laparoscopic appendectomy.^4,12 However, all these studies were conducted as single-center studies with small cohorts.

Before propensity matching, we noted that non-Hispanic white patients were more likely to have SDD compared to nonwhite patients, which is similar to what has been previously reported regarding patient demographics and variability in SDD following appendectomy.^11,18 Oyetunji et al. reviewed a total of 56,077 children with nonperforated appendicitis and found that African American children (OR 1.57, 95% CI 1.42–1.73) and Hispanic patients (OR 1.44, 95% CI 1.36–1.56) were less likely to be SDD compared with Caucasian patients. ¹¹

Although SDD is generally appropriate for patients with uncomplicated appendectomy, the patient's status and comorbid conditions rather than the surgical procedure itself may lead to the decision for an overnight stay. ⁴ While the surgical procedure itself may also impact this decision, a recent retrospective cohort study using the NSQIP registry demonstrated the safety of SDD after either open or laparoscopic appendectomy for uncomplicated appendicitis. ¹⁹

The year by year numbers from the database demonstrate the recent trend favoring SDD following appendectomy. Many hospitals have introduced perioperative pathways to facilitate SDD following simple appendectomy. Although these factors could lead to a selection bias in various study designs, the propensity score matching method enabled us to remove these confounding variables and still demonstrate no difference in the 30-day readmission rate and a decreased complication rate following SDD compared with overnight admission. The decreased use of hospital resources may result in a decrease in cost of approximately $4000 per patient. ²⁰

Since we used the propensity score matching method, there are certain patients who were not matched in the paired analysis. As a result, external validity may be limited. Furthermore, propensity score methods can reduce the risk of bias due to balancing the covariates between the groups; however, there are still unobserved variables which could cause confounding. The procedure length, a patient's socioeconomic factors (health insurance or parents' educational background), the distance from hospital to the patient's home, hospital region, or hospital size could be potential variables that impact the admission decision process. We excluded patients who were admitted before the day of surgery because these patients may have had underlying factors unrelated to the surgical procedure that nevertheless impacted the perioperative course. We also excluded patients who were discharged after more than one night of hospital stay, as multiple nights of hospital stay are likely a result of postoperative complications rather than family preference or hospital culture.

We used the large multicenter registry database, but it does not necessarily represent the national estimate since the data were only from participating centers. However, the data were clinically validated and more than 100 sites are participating from all over the United States. Therefore, the result of our study—SDD after laparoscopic appendectomy does not increase the 30-day readmission rate and complications—is applicable for the pediatric patients who have simple appendicitis.

In summary, we performed a propensity-score matched analysis using a large nationwide cohort of patients who underwent laparoscopic appendectomy for uncomplicated appendicitis. We compared the outcome for SDD and more than one-day overnight stay after the surgical procedure, and found that SDD was not associated with an increased risk of 30-day readmission or any postoperative complication. Protocols to expedite perioperative care, including standardization of intraoperative care, may facilitate same-day hospital discharge in appropriately selected patients, resulting in a decrease in health care costs and improved family satisfaction.