Comparison of appendicectomy outcomes: acute surgical versus traditional pathway

Introduction

The constantly increasing acute surgical workload has challenged traditional care delivery models. A range of new models have been developed including the introduction of acute surgical units (ASUs) that aim to focus resources in a timely way on patients admitted acutely.

In the traditional model, the on-call consultant would accept acutely admitted patients under their care and be on-call for emergency operating while engaged in regular elective commitments, including being offsite for clinics or operating lists.

The concept of the ASU was first introduced by the Prince of Wales Hospital, Sydney, in 2005,1 and has since been implemented at other hospitals across Australasia and worldwide.2-4 The ASU has been established at Auckland City Hospital (ACH) since June 2008.

The ASU is consultant-led model in the care of acute surgical patients that aims to optimize resources and improve key performance indicators (KPIs) by keeping the acute surgical workload independent of the elective workload.

The ASU model at ACH

ACH is a metropolitan, tertiary, teaching Hospital in Auckland, New Zealand. The Department of General Surgery at ACH has five sub-speciality teams: Head, Neck, Breast and Endocrine (five consultants), Upper Gastrointestinal (four consultants), Colorectal (four consultants), Trauma (three consultants) and the Acute Surgical Unit (three consultants).

The ASU model at our institution has previously been described by Hsee et al.5 In this model at ACH (Fig. 1), there is a dedicated acute surgeon who remains on site with no conflicting elective commitments (0730–1630), a dedicated team of junior staff (two registrars, one specialist nurse practitioner, one house officer) and a dedicated theatre and theatre staff for acute operating. There are two handovers (0730 and 1630) and two consultant-led ward rounds per day (0800 and 1400), where all ASU admissions and ward referrals are discussed and reviewed by the ASU consultant surgeon. After 1630 on weekdays and on weekends, the on-call consultant takes over responsibility for acute admissions and acute operating.

Methods

Patient selection

A retrospective study was conducted, comparing patients who had appendicectomies acutely in the 29 months prior to and 31 months following the establishment of the ASU at ACH.

A dataset of 1982 patients who had an appendicectomy between 1 January 2006 and 31 December 2010 was acquired from the decision support team, clinical records department at ACH (Fig. 2). Prior to data acquisition, 125 entries were excluded as these entries were either duplicates or other elective procedures. Of the 1857 appendicectomies, 875 appendectomies were in the 29 months prior to the establishment of ASU (from January 2006 to May 2008) and 982 appendectomies in the 31 months following the inception of ASU (June 2008 to December 2010).

Sixty-five data points were retrospectively acquired by two independent data collectors according to clear pre-determined category definitions for each data point from patient discharge summaries, blood tests, scan results, operation notes, clinic letters and pathology reports.

Results

One thousand and fifty-seven patients who underwent appendicectomy were admitted during the study period, with a 12% increase in the post-ASU era.

The two groups were comparable in their demographics (Table 2), with no significant differences with respect to age, gender ratio, ASA and the rate of perforated appendicitis.

The rate of laparoscopic appendicectomy (79.0% (n = 691) versus 87.9% (n = 863); P < 0.0001) and preoperative imaging (24.2% (n = 212) versus 37.1% (n = 364); P < 0.0001) has significantly increased. However, 31.4% (188/598) of scans did not identify the appendix and those who had scans preoperatively had a longer mean preoperative LOS (0.79 ± 0.53 days versus 1.40 ± 0.82 days; p = 0.000). The clinical operative diagnostic accuracy was 78.0% (n = 1448).

Table 2 compares the KPIs for the traditional model with those for the ASU model. There was a significant reduction in the LOS and the proportion of daytime operations. Although there were trends towards improvement in most of the other KPIs, these differences were not statistically significant. There was one death in a co-morbid elderly patient who had severe sepsis at the time of presentation.

Discussion

There has been a 12% increase in appendicectomies since ASU was formed and a 15.7% increase in the catchment population of ACH (404 619–468 000).10, 11 Studies from other centres have also shown an increase in appendicectomy workload.1, 2 The KPIs in our study have remained the same or improved despite there being no increase in the radiological, theatre or bed resources for this 12% increase.

Until 1981, when Semm performed the first laparoscopic appendicectomy,12 open appendicectomy remained the gold standard. Since then, several randomized control trials and meta-analyses have shown advantages to laparoscopic appendicectomy with less post-operative pain and wound infections and shorter recovery times,13, 14 resulting in a worldwide trend towards an increase in the popularity of laparoscopic appendicectomy,15, 16 similar to that seen in our study and the study by Gandy et al.1

The role of preoperative imaging in appendicitis is still debated. There are studies that show a decrease in the rate of perforation and negative appendicectomy17, 18 and others studies that show no significant differences in outcome and increase in the TTT19-22 as a result of preoperative scanning. In addition to confirming the increase in the preoperative LOS for those patients who have preoperative imaging, our study shows that the rate of detection of the appendix on preoperative scans (68.6%) was lower than the clinical operative diagnostic accuracy (78.0%, n = 1448). The findings of our study do not support the routine use of scans for the diagnosis of appendicitis. There may be a subgroup of patients with atypical presentations for suspected appendicitis in whom preoperative scanning may be useful.23 This study was not designed to investigate the role of radiological investigations in appendicitis and this topic is the focus of further study.

Acute surgical units are considered the future of emergency surgery.24, 25 There are a number of postulated benefits to an ASU model over the traditional model, not all of which can be measured. KPIs are a means to quantify the outcomes of ASU when compared with the traditional model.3 Our study has shown significant improvements in the LOS and proportion of daytime operations. The implication of these improvements in the efficiency of provision of care for appendicectomy patients are as follows: treatment when there is more expertise available during the daylight hours and shorter hospital stays. The decrease in the LOS also provides a direct cost saving to the establishment.

This study is a retrospective, single institution study and has the limitations associated with such a study design. We were unable to study the admission to consultation time as the records for this parameter were not available prior to June 2008. The benefits to training could not be assessed as data regarding the degree of supervision could not be ascertained from the database. It is inherent to the consultant-led ASU model that there is an increase in supervised training and other studies have demonstrated an improvement in training opportunities.3

As the concept of acute care surgery is relatively novel, there are few studies comparing the traditional model with the acute care model.1-3, 26, 27 KPIs have been identified as valuable in the evaluation of performance of ASU.28 However, there is no established standardized approach to the comparison of outcomes following the introduction of an acute surgical model.1 Table 3 is a comparison of the findings of our study with the two similar benchmark studies by Gandy et al.1 and Earley et al.26

All three studies similarly show improvements in LOS, proportion of daytime operating and with trends towards improvements in other KPIs. The differences between groups are smaller in our study compared with the two previous studies. This may be due to small study effects with the smaller population studies. For some KPIs, there was less scope for improvement given the already favourable baseline parameters in the traditional model at our institution (our study, Gandy et al.,1 Earley et al.26): surgical complications (7.4%, 8.5% and 17.4%), TTT (8.2, 10.5 and 7.6 h) and trainee first operator (95.9%, 68.2% and 30%).

Our study is one of the larger studies comparing ASU and traditional care pathways for appendicectomy outcomes across a wider range of KPIs.

Conclusion

The introduction of ASU to ACH has resulted in a significant improvement in the LOS and proportion of daytime operations for appendicectomy patients. Improvements in other KPIs may be possible with a more detailed analysis of the contributory factors to each KPI.