Impact of hospital resection volume and service capability on post‐operative mortality following gastrectomy

Achieving a textbook outcome in patients undergoing gastric resections in a low incidence, high-volume Australian Upper GI unit

BACKGROUND

Textbook outcome (TBO) has been proposed as a composite measure of quality in esophagogastric surgery, and achieving a TBO has been associated with improved overall survival (OS). The Dutch Upper Gastrointestinal Cancer Audit group determined their TBO rate for gastrectomy to be 32.1%, using 10 parameters. Our study aimed to assess the TBO rate in patients who had a gastrectomy for cancer in an Australian Upper GI unit, allowing for comparisons with international specialist centers.

METHODS

Retrospective analysis of a prospectively maintained database of patients who had a gastrectomy for cancer performed by the surgeons in a single Australian center between 2013 and 2018. Postoperative complications were analyzed using Clavien-Dindo (CD) ≥2 and CD ≥3 definitions. Baseline factors and their association with TBO were analyzed using multivariable logistical regression. The association between TBO and survival rates was determined by Cox proportional hazards regression analysis.

RESULTS

In 136 patients, 84 (62%) achieved a TBO when complications were graded as CD ≥2. Greatest negative impact on TBO was the complication rate, lymph node yield, and length of stay. Patients more likely to achieve a TBO were younger, with an increased body mass index and absence of underlying respiratory disease. A nonsignificant trend toward improved OS was seen when TBO was achieved.

CONCLUSION

Our TBO rate compares favorably with published data from high-volume centers. Assessment of a unit's TBO may provide a stronger evaluation of quality when assessing where complex surgery should be performed within Australia.

Failure to rescue following oesophagectomy in Australia: a multi-site retrospective study using American College of Surgeons National Surgical Quality Improvement Program

BACKGROUND

Failure to rescue (FTR), defined as death after a major complication, is increasingly being used as a surrogate for assessing quality of care following major cancer resection. The aim of this paper is to determine the failure to rescue (FTR) rate after oesophagectomy and explore factors that may contribute to FTR within Australia.

METHODS

A retrospective review of the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database from 2015 to 2023 at five Australian hospitals was conducted to identify patients who underwent an oesophagectomy. The primary outcome was FTR rate. Perioperative parameters were examined to evaluate predictive factors for FTR. Secondary outcomes include major complications, overall morbidity, mortality, length of stay and 30-day readmissions.

RESULTS

A total of 155 patients were included with a median age of 65.2 years, 74.8% being male. The FTR rate was 6.3%. In total, 50.3% of patients (n = 78) developed at least one postoperative complication with the most common complication being pneumonia (20.6%) followed by prolonged intubation (12.9%) and organ space SSI/anastomotic leak (11.0%). Multivariate logistic regression analysis was performed to determine any factors that were predictive for FTR however none reached statistical significance.

CONCLUSION

This study is the first to evaluate the FTR rates following oesophagectomy within Australia, with FTR rates and complication profile comparable to international benchmarks. Integration of multi-institutional national databases such as ACS NSQIP into units is essential to monitor and compare patient outcomes following major cancer surgery, especially in low to moderate volume centres.

Gastrectomy mortality in Australia

BACKGROUND

Despite advances in medical management and endoscopic therapy, gastrectomy remains an important yet high-risk procedure for a range of benign and malignant upper gastrointestinal pathologies. No study has previously analysed Australian gastrectomy perioperative mortality rate (POMR). This retrospective, population-based cohort study was conducted to determine the Australian national gastrectomy POMR, allowing state-based and regional trends and outcomes to be assessed.

METHODS

Logistic regression models were compared using de-identified procedural data between 1 July 2005 and 30 June 2017 from the Australian Institute of Health and Welfare. Codes relating to total and subtotal gastrectomy contained in the Australian Classification of Health Interventions were used to extract patient data. Mortality rates were risk adjusted for age and gender. Temporal trends and differences between states/territories and regions were investigated.

RESULTS

The national average POMR throughout the study period was 2.1%. For subtotal gastrectomy, the national mean POMR was 1.1%, decreasing from 2.7% (2005) to 1.3% (2017). For total gastrectomy, the national mean POMR was 2.8%, decreasing from 3.3% (2005) to 1.7% (2017). POMR significantly reduced over time without variation between states or regions. Procedure volume steadily reduced in rural centres with a concomitant increase in metropolitan centres over time.

CONCLUSION

Pleasingly, the Australian gastrectomy POMR is favourable when compared to international cohorts. Improved outcomes were consistent between states and territories, and metropolitan and regional centres. Progressive metropolitan centralization of gastrectomy was demonstrated without evidence of improved outcomes.

Procedure-specific outcomes following gastrectomy for cancer compared by hospital volume and service capability

BACKGROUND

International literature recommends centralising gastric cancer surgery, however, with volumes that define 'high-volume resection' being higher than those in most major centres in Australia and New Zealand. These reports rarely focus on the difference between total (TG) and partial gastrectomy (PG). We assessed the impact of resection volume and service capability on operative mortality, morbidity and surgical quality in patients who had a PG and TG.

METHODS

Patients who had gastrectomy for adenocarcinoma, between 2001 and 2015, were collected from the Queensland Oncology Repository. Hospitals were characterised by cases-per-annum (high-volume [HV] ≥ 5 and low-volume [LV] < 5) and hospital service capability as (high-service [HS] and low-service [LS]), giving three hospital groups: HVHS, LVHS and LVLS. Chi-squared tests were used to compare post-operative mortality, morbidity, failure to rescue (FTR) from complications and surgical quality between these three groups.

RESULTS

There were 426 patients who had a TG and 827 having PG. HVHS centres performed 59% of PG with high surgical quality rates of: HVHS = 53%, LVHS = 34% and LVLS = 46% (p < 0.01). Surgical complications were highest in LVLS (LVLS = 19%, LVHS = 11%, HVHS = 11%; p = 0.02). There was no difference in 30-day mortality nor in FTR. For TG, HVHS performed 67% of these procedures, with lower 30-day mortality (2%) and FTR rates (5%) compared with LVHS (7%, 22%) and LVLS (12%, 28%; p < 0.01). There was no difference in operative morbidity and surgical quality between hospital groups.

CONCLUSION

Despite the 'high-volume' threshold for gastrectomy being the lowest described in the literature, we have shown that centralisation to HVHS centres was associated with lower operative mortality for TG and improved quality of surgery for PG.

Hospital characteristics associated with better 'quality of surgery' and survival following oesophagogastric cancer surgery in Queensland: a population-level study

BACKGROUND

The impact of hospital characteristics on the quality of surgery and survival following oesophagogastric cancer surgery has not been well established in Australia. We assessed the interaction between hospital volume, service capability and surgical outcomes, with the hypothesis that both the quality of surgery and survival are better following treatment in high-volume, high service capability hospitals.

METHODS

All patients undergoing oesophagectomy and gastrectomy for cancer in Queensland, between 2001 and 2015, were included. Demographic, pathology and outcome data were collected. Hospitals were categorized into high (HV) (≥5 gastrectomies; ≥6 oesophagectomies) and low volume (LV). Hospital service capability was defined as high (HS) and low (LS), and then linked to hospital volume: HVHS, LVHS and LVLS. Higher quality surgery was defined using six perioperative parameters. Univariable comparisons of quality of surgery between hospital groups used chi-squared tests. The 5-year overall survival was compared using log-rank tests and Cox proportional hazard models.

RESULTS

For both gastrectomy and oesophagectomy, higher quality surgery occurred more frequently in HVHS hospitals (gastrectomy: HVHS = 44.2%, LVHS = 23.1%, LVLS = 29.1% (P < 0.01); oesophagectomy: HVHS = 34.5%, LVHS = 24.4%, LVLS = 21.7% (P = 0.01)). Following oesophagectomy, the 3- and 5-year overall survival was better following treatment in HVHS (P < 0.01). There was no difference between the groups following gastrectomy.

CONCLUSION

In Queensland, the quality of surgery was higher in HVHS hospitals performing gastrectomy and oesophagectomy; however, the impact on cancer survival was only seen following oesophagectomy.