ABSTRACT
Background
Colorectal cancer (CRC) is a significant contributor to preventable deaths worldwide. The assessment of potentially avoidable mortality following elective CRC surgery has yet to be reported. The Australian and New Zealand Audit of Surgical Mortality (ANZASM) seeks to peer‐review all deaths associated with surgical care. This study aims to determine the incidence of clinical management issues (CMIs) in potentially preventable deaths following elective CRC surgery.
Methods
ANZASM data from 2013 to 2023 were reviewed, and all in‐hospital deaths following elective CRC surgery were selected for analysis. A predefined list of patient‐care factors filled in by the treating surgeon and independent assessors was examined to identify CMIs. Deaths were categorized into two groups: potentially preventable and non‐preventable.
Results
A total of 243 patients died after elective CRC resection over the 10‐year study period. Of these, 134 (55.1%) deaths were considered potentially preventable. Compared to the non‐preventable group, patients with potentially preventable deaths had higher rates of overall intraoperative (15.7% vs. 3.7%) and postoperative complications (88.8% vs. 63.3%). Potentially preventable deaths also had a greater number of CMIs per patient [median (range): 2 (1–6) vs. 1 (0–3)], with higher CMI rates during preoperative (25.0% vs. 12.8%), intraoperative (32.6% vs. 3.7%) and postoperative (58.8% vs. 28.6%) phases of surgical care.
Conclusion
Over half of deaths following elective CRC surgery were potentially preventable. This study highlights deficiencies at all stages of surgical care and provides essential feedback to colorectal surgeons and healthcare professionals to further improve the safety and quality of care.
Keywords: clinical management issues, colorectal cancer, colorectal surgery, perioperative complications, potentially preventable deaths, surgical mortality audit
1. Introduction
Colorectal cancer (CRC) is a significant contributor to preventable cancer mortality across all countries [1]. In Australia, CRC was the leading cause of potentially avoidable cancer deaths, with 2335 such deaths recorded in 2023 [2]. Surgical resection remains the cornerstone of curative treatment for CRC [3]. Although outcomes have improved over time, postoperative mortality still occurs, particularly in high‐risk patients [4, 5]. Preventing avoidable harm in surgical care remains a central focus of quality improvement in colorectal surgery. In 1998, Macarthur et al. used data from the Scottish Audit of Surgical Mortality to quantify preventable deaths after large bowel resection [6]. They reported that approximately half of the patients who died had identifiable deficiencies in clinical management. Although 60% of resections were performed for malignancy, no subgroup analysis specific to the CRC population was undertaken. While indicators of preventable mortality are not intended to be a definitive measure of health system performance, they provide a starting point to assess the effectiveness of healthcare policies and systems in avoiding premature mortality from preventable and treatable causes of death.
The Royal Australasian College of Surgeons (RACS) requires all surgeons in operative‐based practice who have had a surgical death to participate in the Australian and New Zealand Audit of Surgical Mortality (ANZASM) as part of their continuing professional development. It aims to improve care and reduce surgical deaths through regular, extensive education and feedback to surgeons and hospitals. The data collected by ANZASM presents a unique opportunity to examine the gaps and pitfalls leading up to a potentially preventable death [7, 8]. To date, no studies have examined factors associated with potentially avoidable death following CRC surgery. Consequently, this study aims to determine the incidence of potentially preventable mortality following elective CRC surgery and identify clinical management issues (CMIs) associated with these cases.
2. Methods
2.1. Study Design
This retrospective cohort study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement [9]. Data extracted from the ANZASM database included all patients who died following elective CRC surgery in Australia (except New South Wales) between May 1, 2013, and May 1, 2023. CRC cases were defined using procedure and diagnostic Read codes recorded within ANZASM, a standardised hierarchical clinical coding system used to classify diagnoses, procedures and comorbidities. Patients were excluded if they were under 18 years of age, underwent polypectomies, required emergency procedures, or underwent CRC surgery not involving resection (e.g., colostomy formation, colonic bypass or exploratory operations without resection). Emergency procedures were excluded to reduce clinical heterogeneity when assessing potentially preventable mortality, as deaths in this setting are more strongly influenced by acute physiological derangement and limited opportunity for preoperative optimisation. Non‐resectional procedures were excluded to focus on patients undergoing definitive surgical management, allowing more meaningful comparison of CMIs within a relatively homogeneous cohort.
2.2. ANZASM Data Collection
The functioning, governance and objectives of ANZASM have been previously described [10]. In brief, ANZASM is notified of a patient's death (whether a surgical procedure was performed or not) by hospital staff, departments of health, coroner's office, or by surgeons themselves. Clinical details are collected via a standardised Surgical Case Form (SCF) completed by the treating surgeon [11]. The SCF is then de‐identified and sent for first‐line assessment (FLA) to a consultant surgeon from the same specialty working at a different hospital. During FLA, assessors determine if the case requires a more forensic second‐line assessment (SLA). SLA is conducted by another surgeon from the same specialty who is provided full access to the patient's medical records for that admission. As part of the audit process, ANZASM assessors determine whether aspects of a patient's management could have been improved or contributed to the patient's mortality, deemed CMIs. ANZASM uses three grades for CMIs in ascending levels of severity: ‘area of consideration’, ‘area of concern’ or ‘adverse event’. Each CMI is also assessed for clinical impact classified as: ‘made no difference to the outcome’, ‘may have contributed to death’ or ‘an event that caused the death of a patient otherwise expected to survive’. Additionally, a judgment regarding the preventability of the CMI is made, classified as ‘definitely preventable’, ‘probably preventable’, ‘probably not preventable’ or ‘definitely not preventable’.
2.3. Extraction
Patient demographics, operative details, postoperative outcomes, and CMIs were retrieved from the ANZASM database. For each case that had completed the audit process, all CMIs were analysed irrespective of severity. CMIs from FLA and SLA were combined. Where both assessors described the same issue, it was counted only once. In cases of disagreement between assessors' grading, the highest severity grade was selected [7]. Information regarding intraoperative and postoperative complications was manually extracted from the treating surgeon's course‐to‐death narrative summary. During completion of the SCF, treating surgeons identify potential CMIs by selecting from a predefined list of patient‐care factors, including transfer delay, inappropriate transfer, inappropriate transfer level of care, delay in recognising complications and communication issues. Likewise, ANZASM assessors record CMIs using a predefined list covering domains such as preoperative management, decision to operate, choice of operation, timing of operation (too late, too soon or wrong time of day), intraoperative management, operating surgeon's experience level, postoperative management, inappropriate use of deep vein thrombosis (DVT) prophylaxis and fluid balance issues.
2.4. Definitions
Deaths were grouped into potentially preventable or non‐preventable according to the definition proposed by Liu et al. [7] Potentially preventable deaths were defined as any case with CMIs that were deemed ‘definitely’ or ‘probably’ preventable and ‘caused’ or ‘may have caused’ the death of the patient. Non‐preventable deaths were defined as deaths without any CMIs or where these issues were deemed ‘probably not’ or ‘definitely not’ preventable or made ‘no difference’ to the likelihood of death. Colonic tumours were defined as right‐sided cancers (caecum, ascending colon, hepatic flexure, or transverse colon) and left‐sided cancers (splenic flexure, descending colon, or sigmoid colon). Hospitals were divided based on the Modified Monash Model (MMM) into metropolitan hospitals (MMM1) and regional or remote hospitals (MMM2‐7) [12]. Intraoperative anastomotic leak was defined as a defect in the anastomosis (diagnosed via methods such as air leak testing, endoscopic visualisation, doppler ultrasound, near‐infrared fluorescence imaging and assessment of completeness of anastomotic doughnuts) requiring redo of the anastomosis if technically feasible, performing a repair or proximal diversion if repeat anastomosis is not possible. Postoperative anastomotic leak was graded applying the definition and severity grading system developed by the International Study Group of Rectal Cancer (ISREC) [13].
2.5. Statistical Analysis
Statistical analyses were undertaken using Jamovi for Windows (The Jamovi Project, Version 2.6.44). Continuous variables were assessed for normal distribution using the Kolmogorov–Smirnov test, and results were expressed as mean (standard deviation) for normally distributed data and median (range) for non‐normally distributed data. Continuous variables were compared using the Student t‐test or Mann–Whitney U‐Test, which were used for continuous variables depending on the type of distribution. Categorical variables were presented as absolute numbers and percentages. Categorical variables were compared using the Pearson's χ 2 test or Fisher's exact test where applicable. All statistical tests were two‐sided, with significance set at p < 0.05.
2.6. Ethics Approval
This study was conducted as a quality assurance activity as defined by the National Health and Medical Research Council and did not require formal ethics approval [14]. Approval for the study was obtained from the RACS ANZASM Data Management Committee prior to commencement. As ANZASM collects de‐identified data on deceased patients, individual informed consent was not required, and the study was conducted in accordance with ANZASM governance processes.
3. Results
Over the 10‐year period, 423 deaths were audited by ANZASM in which the patient had undergone surgical resection for CRC. After excluding 132 emergency cases, 25 non‐resection procedures, 13 with missing data and 10 polypectomies, 243 cases were included in the analysis (Figure 1). Of these, 209 (86.0%) patients were identified to have at least one CMI. Cases were resolved after FLA in 119 (56.9%) cases, with the remaining 90 (43.1%) cases undergoing an SLA.
FIGURE 1.
Flowchart of colorectal cancer cases audited by the Australian and New Zealand Audit of Surgical Mortality and the number of potentially preventable deaths identified.
CMIs were graded in terms of severity, clinical impact, and preventability. In total, 221 (58.3%) CMIs were flagged by assessors as an area of consideration, 90 (23.7%) as an area of concern and 68 (17.9%) as an adverse event. The median (range) number of CMIs per patient was 1 (0–6). Of the 379 CMIs in 209 patients, 260 (68.6%) may have contributed to the death of the patient, and 60 (15.8%) were deemed to have directly caused death in patients expected to survive. Additionally, 254 (67.0%) CMIs were deemed preventable or potentially preventable. En masse, 134 (55.1%) deaths in this cohort were classified as potentially preventable based on our definition.
Patient demographics and operative details are presented in Table 1. The median age was 80 years (range: 37–96 years), and two‐thirds of the patients were male (66.3%). The most frequent tumour location was the right colon (48%), followed by the left colon and rectum (27% and 24.7%, respectively). There were no significant differences in patient demographics and operative details between potentially preventable and non‐preventable deaths.
TABLE 1.
Patient demographics and operative details.
| Variables | Total N = 243 | Potentially preventable deaths N = 134 | Non‐preventable deaths N = 109 | p |
|---|---|---|---|---|
| Age (years), median (range) | 80 (37–96) | 79 (51–93) | 80 (37–96) | 0.161 |
| Sex | 0.380 | |||
| Male | 161 (66.3) | 92 (68.7) | 69 (63.3) | |
| Female | 82 (33.7) | 42 (31.3) | 40 (36.7) | |
| Length of hospital stay, days, median (range) | 11 (0–102) | 12 (0–83) | 11 (0–102) | 0.303 |
| Tumour location | 0.126 | |||
| Right colon | 117 (48.0) | 64 (47.8) | 53 (48.6) | |
| Left colon | 66 (27.0) | 31 (23.1) | 35 (32.1) | |
| Rectum | 60 (24.7) | 39 (29.1) | 21 (19.3) | |
| Presentation | 0.704 | |||
| Primary | 236 (97.1) | 131 (97.8) | 105 (96.3) | |
| Recurrence | 7 (2.9) | 3 (2.2) | 4 (3.7) | |
| No. of lesions | 0.824 | |||
| Solitary | 221 (90.9) | 123 (91.8) | 98 (89.9) | |
| Synchronous | 20 (8.2) | 10 (7.5) | 10 (9.2) | |
| Metachronous | 2 (0.8) | 1 (0.7) | 1 (0.9) | |
| AJCC stage | 0.515 | |||
| I–III | 222 (91.4) | 121 (90.3) | 101 (92.7) | |
| IV | 21 (8.6) | 13 (9.7) | 8 (7.3) | |
| Neoadjuvant therapy for rectal cancer | 0.374 | |||
| No | 38 (63.3) | 25 (64.1) | 13 (61.9) | |
| SCRT | 10 (16.7) | 7 (17.9) | 3 (14.3) | |
| CRT | 10 (16.7) | 7 (17.9) | 3 (14.3) | |
| TNT | 2 (3.3) | 0 (0.0) | 2 (9.5) | |
| Hospital type | 0.215 | |||
| Private | 88 (36.2) | 43 (32.1) | 45 (41.3) | |
| Public | 150 (61.7) | 89 (66.4) | 61 (56.0) | |
| Co‐location | 5 (2.1) | 2 (1.5) | 3 (2.8) | |
| Patient insurance status | 0.066 | |||
| Private | 97 (39.9) | 48 (35.8) | 49 (45.0) | |
| Public | 144 (59.3) | 86 (64.2) | 58 (53.2) | |
| Veteran | 2 (0.8) | 0 (0.0) | 2 (1.8) | |
| MMM classification | 0.530 | |||
| Metropolitan hospitals | 178 (73.3) | 96 (71.6) | 82 (75.2) | |
| Regional or remote hospitals | 65 (26.7) | 38 (28.4) | 27 (24.8) | |
| ASA score | 0.239 | |||
| I | 1 (0.4) | 1 (0.7) | 0 (0.0) | |
| II | 37 (15.2) | 24 (17.9) | 13 (11.9) | |
| III | 147 (60.5) | 79 (59.0) | 68 (62.4) | |
| IV | 49 (20.2) | 24 (17.9) | 25 (22.9) | |
| V | 3 (1.2) | 3 (2.2) | 0 (0.0) | |
| Unknown | 6 (2.5) | 3 (2.2) | 3 (2.8) | |
| Number of comorbidities, median (range) | 2 (0–7) | 2 (0–7) | 2 (0–6) | 0.957 |
| Type of co‐morbidities | ||||
| Cardiovascular | 154 (63.4) | 79 (59.0) | 75 (68.8) | 0.113 |
| Respiratory | 83 (34.2) | 47 (35.1) | 36 (33.0) | 0.738 |
| Renal | 55 (22.6) | 31 (23.1) | 24 (22.0) | 0.836 |
| Hepatic | 20 (8.2) | 10 (7.5) | 10 (9.2) | 0.629 |
| Neurological | 33 (13.6) | 18 (13.4) | 15 (13.8) | 0.941 |
| Advanced malignancy | 85 (35.0) | 44 (32.8) | 41 (37.6) | 0.437 |
| Diabetes | 53 (21.8) | 35 (26.1) | 18 (16.5) | 0.071 |
| Obesity | 42 (17.3) | 26 (19.4) | 16 (14.7) | 0.333 |
| Other | 69 (28.4) | 39 (29.1) | 30 (27.5) | 0.786 |
| Surgeon's assessment of overall risk of death before surgery | 0.490 | |||
| Minimal | 8 (3.3) | 6 (4.5) | 2 (1.8) | |
| Small | 72 (29.6) | 43 (32.1) | 29 (26.6) | |
| Moderate | 108 (44.4) | 57 (42.5) | 51 (46.8) | |
| Considerable | 55 (22.6) | 28 (20.9) | 27 (24.8) | |
| Procedure | 0.633 | |||
| Right colectomy a | 107 (44.0) | 54 (40.3) | 53 (48.6) | |
| Anterior resection b | 83 (34.2) | 47 (35.1) | 36 (33.0) | |
| Total colectomy c | 27 (11.1) | 18 (13.4) | 9 (8.3) | |
| Hartmann's | 7 (2.9) | 5 (3.7) | 2 (1.8) | |
| APR | 14 (5.8) | 7 (5.2) | 7 (6.4) | |
| Proctocolectomy | 1 (0.4) | 0 (0.0) | 1 (0.9) | |
| Proctectomy | 1 (0.4) | 1 (0.7) | 0 (0.0) | |
| Pelvic exenteration | 3 (1.2) | 2 (1.5) | 1 (0.9) | |
| Surgical approach | 0.946 | |||
| Robotic | 2 (0.8) | 1 (0.7) | 1 (0.9) | |
| Laparoscopic | 109 (44.9) | 61 (45.5) | 48 (44.0) | |
| Laparoscopic assisted | 22 (9.1) | 13 (9.7) | 9 (8.3) | |
| Open | 110 (45.3) | 59 (44.0) | 51 (46.8) | |
| Conversion to open surgery d | 27 (24.3) | 15 (24.2) | 12 (24.5) | 0.971 |
| Primary surgeon | 0.459 | |||
| Consultant | 201 (82.7) | 110 (82.1) | 91 (83.5) | |
| Fellow | 32 (13.2) | 20 (14.9) | 12 (11.0) | |
| Surgical trainee | 10 (4.1) | 4 (3.0) | 6 (5.5) | |
Note: Values are n (%) unless otherwise indicated.
Abbreviations: AJCC, American Joint Committee on Cancer; APR, abdominoperineal resection; ASA, American Society of Anasethesiologists; CRT, chemoradiotherapy; MMM, Modified Monash Model; SCRT, short course radiotherapy; TNT, total neoadjuvant therapy.
Includes right hemicolectomy, extended right hemicolectomy, and ileocolic resection.
Includes left hemicolectomy and sigmoidectomy.
Includes subtotal colectomy.
Conversion of laparoscopic or robotic to open resection.
Intraoperative and postoperative complications are detailed in Table 2. The potentially preventable deaths group had a significantly higher rate of intraoperative complications compared with the non‐preventable deaths group (15.7% vs. 3.7%, p = 0.002). Similarly, postoperative complications were significantly more frequent in the potentially preventable deaths group (88.8% vs. 63.3%, p < 0.001). These included anastomotic leak (36.1% vs. 20.2%, p = 0.011), wound dehiscence (8.2% vs. 1.8%, p = 0.042), ileus (30.6% vs. 17.4%, p = 0.018) and respiratory complications (41.0% vs. 20.2%, p < 0.001). ICU requirement (90.3% vs. 79.8%, p = 0.021), unplanned ICU admission (59.0% vs. 42.2%, p = 0.009) and unplanned return to theatre (53.0% vs. 32.1%, p = 0.001) occurred significantly more often in the potentially preventable deaths group compared to the non‐preventable deaths group.
TABLE 2.
Intraoperative and postoperative complications.
| Variables | Total N = 243 | Potentially preventable deaths N = 134 | Non‐preventable deaths N = 109 | p |
|---|---|---|---|---|
| Operative time, minutes | 0.311 | |||
| Median (range) | 180 (48–900) | 180 (48–900) | 180 (60–480) | |
| Mean (SD) | 193 (106) | 200 (120) | 183 (85) | |
| Intraoperative complications | 25 (10.3) | 21 (15.7) | 4 (3.7) | 0.002* |
| Anastomotic leak | 4 (1.6) | 2 (1.5) | 2 (1.8) | 1.00 |
| Bleeding | 3 (1.2) | 2 (1.5) | 1 (0.9) | 1.00 |
| Vascular injury | 5 (2.1) | 5 (3.7) | 0 (0.0) | 0.067 |
| Small bowel injury | 11 (4.5) | 9 (6.7) | 2 (1.8) | 0.118 |
| Ureteric injury | 1 (0.4) | 1 (0.7) | 0 (0.0) | 1.00 |
| Splenic injury | 2 (0.8) | 0 (0.0) | 2 (1.5) | 0.503 |
| Postoperative complications | 188 (77.4) | 119 (88.8) | 69 (63.3) | < 0.001* |
| Surgical complications | ||||
| Anastomotic leak a | 62 (29.1) | 43 (36.1) | 19 (20.2) | 0.011* |
| Grade A | 2 (3.2) | 2 (4.7) | 0 (0.0) | |
| Grade B | 13 (21.0) | 8 (18.6) | 5 (26.3) | |
| Grade C | 47 (75.8) | 33 (76.7) | 14 (73.7) | |
| Wound dehiscence | 13 (5.3) | 11 (8.2) | 2 (1.8) | 0.042* |
| Wound infection | 8 (3.3) | 6 (4.5) | 2 (1.8) | 0.302 |
| Postoperative Ileus | 60 (24.7) | 41 (30.6) | 19 (17.4) | 0.018* |
| Bleeding | 30 (12.3) | 19 (14.2) | 11 (10.1) | 0.335 |
| Bowel obstruction | 11 (4.5) | 6 (4.5) | 5 (4.6) | 0.967 |
| Bowel perforation | 8 (3.3) | 6 (4.5) | 2 (1.8) | 0.302 |
| Bowel ischemia | 24 (9.9) | 17 (12.7) | 7 (6.4) | 0.104 |
| Stoma related complications | 8 (3.3) | 7 (5.2) | 1 (0.9) | 0.077 |
| Medical complications | ||||
| Sepsis | 74 (30.5) | 47 (35.1) | 27 (24.8) | 0.083 |
| Multiple organ dysfunction syndrome | 20 (8.2) | 15 (11.2) | 5 (4.6) | 0.062 |
| Thrombotic | 20 (8.2) | 11 (8.2) | 9 (8.3) | 0.989 |
| Urological | 41 (16.9) | 26 (19.4) | 15 (13.8) | 0.243 |
| Hepatic | 5 (2.1) | 3 (2.2) | 2 (1.8) | 0.654 |
| Respiratory | 77 (31.7) | 55 (41.0) | 22 (20.2) | < 0.001* |
| Cardiac | 62 (25.5) | 39 (29.1) | 23 (21.1) | 0.155 |
| Neurological | 26 (10.7) | 16 (11.9) | 10 (9.2) | 0.488 |
| ICU requirement | 208 (85.6) | 121 (90.3) | 87 (79.8) | 0.021* |
| Unplanned ICU admission | 125 (51.4) | 79 (59.0) | 46 (42.2) | 0.009* |
| Unplanned return to theatre | 106 (43.6) | 71 (53.0) | 35 (32.1) | 0.001* |
| Unplanned readmission | 7 (2.9) | 5 (3.7) | 2 (1.8) | 0.464 |
Note: Values are n (%) unless otherwise indicated. *p value < 0.05.
Abbreviations: DVT, deep vein thrombosis; ICU, intensive care unit; SD, standard deviation.
Anastomotic leaks were calculated including only patients with an anastomosis in the denominator.
The nature of CMIs is summarised in Table 3. The potentially preventable deaths group had a significantly higher number of CMIs per patient than the non‐preventable deaths group [median (range): 2 (1–6) vs. 1 (1–3), p < 0.001]. This higher number of CMIs adversely impacted preoperative (25.0% vs. 12.8%, p = 0.017), intraoperative (32.6% vs. 3.7%, p < 0.001) and postoperative (58.8% vs. 28.6%, p < 0.001) phases of surgical care, as well as the choice of operation (20.9% vs. 8.3%, p = 0.006), fluid balance (13.4% vs. 3.7%, p = 0.012) and delayed recognition of complications (22.4% vs. 7.3%, p = 0.001).
TABLE 3.
Nature of clinical management issues.
| Variables | Total N = 243 | Potentially preventable deaths N = 134 | Non‐preventable deaths N = 109 | p |
|---|---|---|---|---|
| Number of clinical management issues per patient, median (range) | 1 (0–6) | 2 (1–6) | 1 (0–3) | < 0.001* |
| Clinical management issues a | ||||
| Preoperative | 48 (19.6) | 34 (25.0) | 14 (12.8) | 0.017* |
| Decision to operate at all | 23 (9.5) | 13 (9.7) | 10 (9.2) | 0.898 |
| Choice of operation | 37 (15.2) | 28 (20.9) | 9 (8.3) | 0.006* |
| Timing of operation | 17 (7.0) | 12 (9.0) | 5 (4.6) | 0.184 |
| Intraoperative | 49 (19.8) | 45 (32.6) | 4 (3.7) | < 0.001* |
| Operating surgeon's experience level | 6 (2.5) | 4 (3.0) | 2 (1.8) | 0.566 |
| Postoperative | 119 (45.8) | 87 (58.8) | 32 (28.6) | < 0.001* |
| Inappropriate DVT prophylaxis | 7 (2.9) | 5 (3.7) | 2 (1.8) | 0.379 |
| Fluid balance issues | 22 (9.1) | 18 (13.4) | 4 (3.7) | 0.012* |
| Delay in recognising complications | 38 (15.6) | 30 (22.4) | 8 (7.3) | 0.001* |
| Communication | 13 (5.3) | 9 (6.7) | 4 (3.7) | 0.394 |
Note: Values are n (%) unless otherwise indicated. *p value < 0.05.
A patient can have more than one CMI.
The causes of death are shown in Table 4, with sepsis being the most common (32%), followed by respiratory events (31.5%). Cardiac arrest was significantly more frequent in the non‐preventable deaths group than in the potentially preventable deaths group (7.4% vs. 1.5%, p = 0.046). The distribution of other causes of death was similar between the two groups.
TABLE 4.
Causes of death.
| Variables | Total N = 243 | Potentially preventable deaths N = 134 | Non‐preventable deaths N = 109 | p |
|---|---|---|---|---|
| Unknown | 2 (0.8) | 1 (0.7) | 1 (0.9) | |
| Cardiac | ||||
| Total | 40 (16.6) | 17 (12.8) | 23 (21.3) | 0.077 |
| Myocardial infarction | 23 (9.5) | 11 (8.3) | 12 (11.1) | 0.455 |
| Cardiac arrest | 10 (4.1) | 2 (1.5) | 8 (7.4) | 0.046* |
| Ischemic heart disease | 2 (0.8) | 0 (0.0) | 2 (1.9) | 0.200 |
| Decompensated heart failure | 3 (1.2) | 3 (2.3) | 0 (0.0) | 0.225 |
| Cardiogenic shock | 1 (0.4) | 1 (0.8) | 0 (0.0) | 1.00 |
| Endocarditis | 1 (0.4) | 0 (0.0) | 1 (0.9) | 0.448 |
| Respiratory | ||||
| Total | 76 (31.5) | 48 (36.1) | 28 (25.9) | 0.091 |
| Hospital acquired pneumonia | 10 (4.1) | 6 (4.5) | 4 (3.7) | 1.00 |
| Aspiration pneumonia | 26 (10.8) | 17 (12.8) | 9 (8.3) | 0.268 |
| Aspiration of gastric content | 14 (5.8) | 9 (6.8) | 5 (4.6) | 0.481 |
| Acute pulmonary oedema | 3 (1.2) | 0 (0.0) | 3 (2.8) | 0.089 |
| Respiratory failure | 23 (9.5) | 16 (12.0) | 7 (6.5) | 0.145 |
| Vascular | ||||
| Total | 25 (10.4) | 13 (9.8) | 12 (11.1) | 0.735 |
| Pulmonary embolism | 13 (5.4) | 4 (3.0) | 9 (8.3) | 0.087 |
| Cerebrovascular accident | 6 (2.5) | 4 (3.0) | 2 (1.9) | 0.694 |
| Bleeding | 6 (2.5) | 5 (3.8) | 1 (0.9) | 0.228 |
| Renal and hepatic | ||||
| Total | 14 (5.8) | 7 (5.3) | 7 (6.5) | 0.688 |
| Acute tubular necrosis | 1 (0.4) | 1 (0.8) | 0 (0.0) | 1.00 |
| Acute kidney injury | 9 (3.7) | 5 (3.8) | 4 (3.7) | 1.00 |
| Acute hepatic failure | 3 (1.2) | 1 (0.8) | 2 (1.9) | 0.589 |
| Hepatorenal syndrome | 1 (0.4) | 0 (0.0) | 1 (0.9) | 0.448 |
| Malignancy | ||||
| Total | 8 (3.3) | 2 (1.5) | 6 (5.6) | 0.081 |
| Disseminated malignancy | 4 (1.7) | 1 (0.8) | 3 (2.8) | 0.328 |
| Recurrence | 2 (0.8) | 0 (0.0) | 2 (1.9) | 0.200 |
| Secondary malignant neoplasm | 2 (0.8) | 1 (0.8) | 1 (0.9) | 1.00 |
| Orthopaedic | ||||
| Hip fracture from hospital bed fall | 1 (0.4) | 1 (0.8) | 0 (0.0) | 1.00 |
| Sepsis | ||||
| Total | 77 (32.0) | 45 (33.8) | 32 (29.6) | 0.486 |
| Anastomotic leak | 31 (12.9) | 19 (14.3) | 12 (11.1) | 0.464 |
| Bowel obstruction | 4 (1.7) | 2 (1.5) | 2 (1.9) | 1.00 |
| Gastrointestinal perforation | 3 (1.2) | 3 (2.3) | 0 (0.0) | 0.255 |
| Abdominal sepsis (no leak) | 15 (6.2) | 8 (6.0) | 7 (6.5) | 0.882 |
| Ischemic gut | 15 (6.2) | 9 (6.8) | 6 (5.6) | 0.699 |
| Generalised sepsis | 9 (3.7) | 4 (3.0) | 5 (4.6) | 0.519 |
Note: Values are n (%) unless otherwise indicated. *p value < 0.05.
Of the 243 patients, 285 CMIs were identified in the potentially preventable deaths group. In total, 57 (42.5%) patients had one CMI identified, 30 (22.4%) had two, 30 (22.4%) had three, 9 (6.7%) had four, 6 (4.5%) had five and two (1.5%) patients had six CMIs identified (Figure 2). These were categorised as having impacted preoperative management/preparation (34 CMIs, 11.9%), decision to operate (13 CMIs, 4.6%), choice of operation (28 CMIs, 9.8%), timing of operation (12 CMIs, 4.2%), intraoperative management/technical management of surgery (45, 15.8%), operating surgeon's experience level (4, 1.4%), postoperative management (87, 30.5%), inappropriate DVT prophylaxis (5, 1.8%), fluid balance (18, 6.3%), delay in recognising complications (30, 10.5%) and communication (9, 3.2%) (Figure 3).
FIGURE 2.
Percentage of colorectal cancer cases with at least one clinical management issue (CMI).
FIGURE 3.
Clinical management issues across 14 areas of patient care in patients with potentially preventable deaths.
4. Discussion
This 10‐year analysis of national surgical mortality audit data identified underlying causes, complications and CMIs that contributed to a patient's death following elective CRC surgery. Over 50% of deaths were potentially preventable. Potentially preventable deaths were associated with significantly higher rates of intraoperative and postoperative complications than non‐preventable deaths. Potentially preventable deaths were also characterised by a high proportion of CMIs during preoperative, intraoperative, and postoperative phases of care. These findings may inform clinical modifications at the hospital, local health network and broader system level to possibly reduce potentially preventable deaths following elective CRC surgery.
The Bowel Cancer Outcomes Registry has reported the inpatient mortality rate after elective CRC surgery to be less than 1%; the 243 deaths presented in this study were derived from approximately 24 300 resections performed across multiple institutions in Australia [15]. Although each death is probably thoroughly examined within their respective colorectal units, a comprehensive assessment of all these cases has not been available until now. In the present study, preoperative CMIs were significantly more common in the potentially preventable group compared with the non‐preventable group, and this pattern was consistent across all phases of the patient journey (pre‐, intra‐ and post‐operative). This suggests that adverse outcomes are not attributed to isolated phase‐specific events, but rather to deficiencies in care processes and compounding complications across the perioperative pathway. Patients who experience preoperative CMIs may be more susceptible to subsequent intra‐ and postoperative CMIs, reflecting a cascade of risk that culminates in poor outcomes. While multidisciplinary team (MDT) decision‐making has improved the coordination and selection of treatment strategies, it does not appear to fully mitigate this cumulative risk [16]. Consistent with this, clinical judgment alone has been found to underpredict mortality in patients undergoing major surgery [17]. Taken together, these findings highlight the importance of accurate preoperative patient prognostication to identify high‐risk individuals and reduce the likelihood of downstream complications across the patient journey. Implementation of prognostic scores and attendance at a high‐risk preoperative assessment clinic have been proposed as possible ways to lower mortality following CRC surgery [18, 19]. However, despite their promise, uptake in clinical practice remains limited, partly due to the low level of supporting evidence [20].
Importantly, these findings should not be interpreted as advocating restriction of surgery in high‐risk patients to achieve zero mortality. In appropriately selected and consented patients, high‐risk CRC resection remains justified, particularly where there is potential for cure or meaningful survival benefit [21]. In the present study, there were no significant differences in baseline demographics, comorbidity burden, or surgeon‐assessed preoperative risk between potentially preventable and non‐preventable deaths, suggesting that case selection alone does not explain the observed variation in outcomes. The higher burden of CMIs across all phases of care indicates that modifiable aspects of perioperative management, rather than patient selection, are more likely to be the primary drivers of potentially preventable mortality.
In patients with potentially preventable deaths, CMIs were identified in the decision to operate, choice of operation and timing of operation. These findings are consistent with a recent analysis of ANZASM data by Ey et al., who demonstrated that non‐technical errors, particularly in operative decision‐making, are key contributors to surgical mortality in Australia [22]. Operative management CMIs are mostly related to the choice of operation. Assessors raised issues concerning the failure to consider alternative surgical procedures and the decision‐making behind creating a primary anastomosis versus an end colostomy. The timing of the operation was another concern, with a recent meta‐analysis reporting that elective CRC surgery should not be postponed longer than 4 weeks, as delays were associated with poorer outcomes [23].
Similar to previous studies, postoperative CMIs contributed the most to potentially preventable mortality [6, 7]. This pattern differs somewhat from findings reported in other surgical specialties captured by ANZASM. In endocrine surgery, urology and neurosurgical mortality cases, the most common potentially preventable CMI was inadequate assessment in the preoperative phase [24, 25, 26]. In cardiac surgery, the most common preventable CMI was a technical factor in the intraoperative phase, whereas in thoracic surgery the most common preventable CMI was infection in the postoperative phase [27]. In upper gastrointestinal cancer surgery, the most common potentially preventable CMI was failure to recognise and respond to deterioration in the postoperative phase of care [7]. These findings suggest that the predominant contributors to potentially preventable mortality vary across surgical specialties and phases of care. In the present study, the CMIs in postoperative care included inappropriate DVT prophylaxis, fluid balance issues and delay in recognising postoperative complications. To address these CMIs, many centres have adopted enhanced recovery after surgery (ERAS) protocols, which have been proven to reduce both recovery time and postoperative complication rates following CRC surgery [28]. However, successfully implementing an ERAS protocol can be challenging, especially at smaller institutions, as it requires institutional and MDT support [29, 30]. To overcome these barriers, studies have emphasised that ERAS protocols supported by audit and feedback approaches lead to improved outcomes, even at smaller hospitals [29, 31].
Our data showed that potentially preventable mortality was characterised by a higher rate of postoperative complications. Equally noteworthy are the significantly higher rates of anastomotic leak, wound dehiscence, and postoperative ileus in patients with potentially preventable mortality. Anastomotic leak is one of the most severe complications of CRC surgery owing to its negative impact on both short‐ and long‐term outcomes [32]. The increased incidence of intraoperative complications, a known risk factor for anastomotic leak, likely contributed to the high rate of anastomotic leaks in the potentially preventable death group [33]. We would recommend using a predefined risk reduction strategy to decrease the risk of anastomotic leak following CRC surgery. This involves risk assessment using a validated anastomotic leak calculator, medication optimization by stopping immunosuppressants, anti‐inflammatories, and anticoagulants prior to surgery, and the use of a previously described standardized anastomotic technique [34]. Prophylactic negative pressure wound therapy has been shown to effectively prevent surgical wound complications after CRC surgery [35]. Many efforts have been undertaken in recent years to shift the paradigm from viewing postoperative ileus as an inevitable part of abdominal surgery to considering it a preventable event [36]. Patients who developed postoperative ileus in the presence of other complications had significantly higher mortality rates [37]. While the most important risk factors for postoperative ileus are not recorded in the ANZASM database, avoidance and treatment of risk factors can be best realised by the application of ERAS pathways incorporating minimally invasive surgery, a stringent fluid regimen, modern opioid‐sparing pain strategies and avoiding routine nasogastric tube insertion [20]. Among medical complications, respiratory complications were more frequent in patients with potentially preventable deaths, possibly reflecting a higher incidence of fluid balance issues, consistent with evidence linking fluid overload to respiratory complications [20].
While the ANZASM dataset does not permit formal calculation of failure to rescue, delayed recognition of complications was significantly more common in potentially preventable deaths (22.4% vs. 7.3%) [38]. Given that the predominant causes of death, particularly sepsis, including anastomotic leak, and respiratory complications, are typically amenable to early intervention, this suggests that deficiencies in rescue processes may have contributed to mortality [39]. Therefore, improving early recognition, escalation, and management of postoperative complications is essential to prevent the conversion of morbidity into mortality [40].
Poor communication was a contributing factor in CMIs in potentially preventable deaths. Three recurrent themes were identified. The most common occurred between the ICU staff and surgeons (four cases) and involved a lack of urgent and collaborative communication, resulting in refusal of ICU admission, refusal to treat potentially reversible complications, failure to contact the consultant surgeon prior to withdrawal of treatment, and transfer of patients back to the ward without notifying the surgical team. A second theme involved failure by junior doctors to recognise and escalate patient deterioration to consultant surgeons (three cases), including failure to notify the consultant surgeon of deterioration and proceeding to surgical intervention without prior consultant input. The third theme involved communication issues between surgical and medical teams (two cases), characterised by a lack of coordinated management plans for investigations and medication administration. Similar patterns of communication failure have been reported in a prior ANZASM analysis of cardiothoracic surgery patients [41]. To reduce communication failures, studies have suggested implementing ward round checklists, which have been shown to reduce ICU length of stay and improve documentation without affecting ward round duration [42, 43].
This study has several limitations. ANZASM, although primarily intended for auditing surgical mortality, was not originally designed for clinical research purposes. Furthermore, ANZASM is a mortality audit, and therefore, its design excludes patients who survived CRC resection. Additionally, ANZASM is not a national registry for all patients who undergo CRC surgery. Hence, we could not provide a population estimate for some of the comparisons. Furthermore, the data gathered relies on self‐reports from the treating surgeons, potentially introducing bias. Additionally, the assessors themselves may inadvertently introduce their own biases, albeit mitigated to some extent by the independent and blinded selection of assessors. Another limitation was the inaccessibility of mortality data from the state of New South Wales. Given that this represents approximately 30% of the Australian population, there is the potential for an under‐reporting bias. The exclusion of emergency and non‐resectional CRC procedures limits the generalisability of our findings to the broader CRC surgical population. However, this approach was intended to reduce heterogeneity and isolate potentially preventable CMIs within elective CRC surgery. Future studies examining emergency CRC surgery as a distinct cohort are warranted. The present study includes cases over a 10‐year period, during which clinical practices may have evolved in these centres. Finally, ANZASM does not capture all details from case notes; thus, manually collected variables from case notes may be subject to bias and may well be under‐reported.
In the ANZASM database, over half of deaths after elective CRC surgery were potentially preventable. A large proportion of potentially preventable mortality was preceded by intraoperative and postoperative complications. In addition, approximately 30% of CMIs that contributed to potentially preventable deaths occurred during postoperative care. These findings provide important feedback to colorectal surgeons and may help guide more focused interventions to reduce potentially preventable deaths in patients who underwent CRC surgery.
Author Contributions
Sergei Bedrikovetski: conceptualization, data curation, investigation, methodology, formal analysis, writing – original draft. Luke Traeger: conceptualization, investigation, formal analysis, visualization, writing – review and editing. Nathan Procter: conceptualization, investigation, methodology, data curation, writing – review and editing. Helena S. Kopunic: conceptualization, project administration, methodology, resources, writing – review and editing. Guy J. Maddern: conceptualization, project administration, resources, writing – review and editing.
Funding
The authors have nothing to report.
Conflicts of Interest
G.J.M. is the Chair of the ANZASM Steering Committee and a member of the Editorial Board of the ANZ Journal of Surgery. He had no role in the editorial handling of this manuscript. The other authors declare no conflicts of interest.
Data Availability Statement
The data that support the findings of this study are available from Australian and New Zealand Audit of Surgical Mortality. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the author(s) with the permission of Australian and New Zealand Audit of Surgical Mortality.
References
- 1. Frick C., Rumgay H., Vignat J., et al., “Quantitative Estimates of Preventable and Treatable Deaths From 36 Cancers Worldwide: A Population‐Based Study,” Lancet Global Health 11 (2023): e1700–e1712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. AIHW , Deaths in Australia (AIHW, Australian Government, 2025). [Google Scholar]
- 3. Mao J. X., Gao R., Wang Y., Yan X. B., and Wang H. H., “Surgical Treatment of Colorectal Cancer: A Multidimensional Review,” World Journal of Gastrointestinal Surgery 17 (2025): 107785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Ketelaers S. H. J., Orsini R. G., Burger J. W. A., Nieuwenhuijzen G. A. P., and Rutten H. J. T., “Significant Improvement in Postoperative and 1‐Year Mortality After Colorectal Cancer Surgery in Recent Years,” European Journal of Surgical Oncology 45 (2019): 2052–2058. [DOI] [PubMed] [Google Scholar]
- 5. Theile D. E., Philpot S., Blake M., Harrington J., and Youl P. H., “Outcomes Following Colorectal Cancer Surgery: Results From a Population‐Based Study in Queensland, Australia, Using Quality Indicators,” Journal of Evaluation in Clinical Practice 25 (2019): 834–842. [DOI] [PubMed] [Google Scholar]
- 6. Macarthur D. C., Nixon S. J., and Aitken R. J., “Avoidable Deaths Still Occur After Large Bowel Surgery. Scottish Audit of Surgical Mortality, Royal College of Surgeons of Edinburgh,” British Journal of Surgery 85 (1998): 80–83. [DOI] [PubMed] [Google Scholar]
- 7. Liu D. S., Fayed A., Evans P., et al., “Understanding Potentially Preventable Mortality Following Oesophago‐Gastric Cancer Surgery: Analysis of a National Audit of Surgical Mortality,” Annals of Surgical Oncology 30 (2023): 4950–4961. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Bedrikovetski S., Kopunic H. S., Procter N., et al., “Use of the Australian and New Zealand Audit of Surgical Mortality (ANZASM) to Determine Potentially Preventable Deaths After Emergency General Surgery,” Surgery 184 (2025): 109444. [DOI] [PubMed] [Google Scholar]
- 9. von Elm E., Altman D. G., Egger M., et al., “The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies,” Annals of Internal Medicine 147 (2007): 573–577. [DOI] [PubMed] [Google Scholar]
- 10. Raju R. S., Guy G. S., Majid A. J., Babidge W., and Maddern G. J., “The Australian and New Zealand Audit of Surgical Mortality‐Birth, Deaths, and Carriage,” Annals of Surgery 261 (2015): 304–308. [DOI] [PubMed] [Google Scholar]
- 11. Surgeons RACo , ANZASM National Report 2017–2018 (Royal Australasian College of Surgeons Melbourne, 2022). [Google Scholar]
- 12. Health AGDo , Modified Monash Model (Australian Government Department of Health Canberra, 2019). [Google Scholar]
- 13. Kulu Y., Ulrich A., Bruckner T., et al., “Validation of the International Study Group of Rectal Cancer Definition and Severity Grading of Anastomotic Leakage,” Surgery 153 (2013): 753–761. [DOI] [PubMed] [Google Scholar]
- 14. Health N, Medical Research Council N , Ethical Considerations in Quality Assurance and Evaluation Activities (NHMRC Canberra, 2014). [Google Scholar]
- 15. Dagher H., Murphy E., Hoh S. M., et al., “The 2022 Data Bowel Cancer Outcomes Registry Report,” 2023.
- 16. Fehervari M., Hamrang‐Yousefi S., Fadel M. G., et al., “A Systematic Review of Colorectal Multidisciplinary Team Meetings: An International Comparison,” BJS Open 5 (2021): zrab044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wong D. J. N., Harris S., Sahni A., et al., “Developing and Validating Subjective and Objective Risk‐Assessment Measures for Predicting Mortality After Major Surgery: An International Prospective Cohort Study,” PLoS Medicine 17 (2020): e1003253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. de Nes L. C. F., Hannink G., 't Lam‐Boer J., et al., “Postoperative Mortality Risk Assessment in Colorectal Cancer: Development and Validation of a Clinical Prediction Model Using Data From the Dutch ColoRectal Audit,” BJS Open 6 (2022): zrac014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Carlisle J., Swart M., Dawe E. J., and Chadwick M., “Factors Associated With Survival After Resection of Colorectal Adenocarcinoma in 314 Patients,” British Journal of Anaesthesia 108 (2012): 430–435. [DOI] [PubMed] [Google Scholar]
- 20. Gustafsson U. O., Rockall T. A., Wexner S., et al., “Guidelines for Perioperative Care in Elective Colorectal Surgery: Enhanced Recovery After Surgery (ERAS) Society Recommendations 2025,” Surgery 184 (2025): 109397. [DOI] [PubMed] [Google Scholar]
- 21. Rizzuto A., Palaia I., Vescio G., Serra R., Malanga D., and Sacco R., “Multivisceral Resection for Occlusive Colorectal Cancer: Is It Justified?,” International Journal of Surgery 33, no. Suppl 1 (2016): S142–S147. [DOI] [PubMed] [Google Scholar]
- 22. Ey J. D., Kollias V., Lee O., et al., “Non‐Technical Error Leading to Patient Fatalities in the Australian Surgical Population,” British Journal of Surgery 112 (2025): znaf083. [DOI] [PubMed] [Google Scholar]
- 23. Whittaker T. M., Abdelrazek M. E. G., Fitzpatrick A. J., et al., “Delay to Elective Colorectal Cancer Surgery and Implications for Survival: A Systematic Review and Meta‐Analysis,” Colorectal Disease 23 (2021): 1699–1711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Gupta A. K., Stewart S. K., Cottell K., et al., “Potentially Avoidable Issues in Urology Mortality Cases in Australia: Identification and Improvements,” ANZ Journal of Surgery 90 (2020): 719–724. [DOI] [PubMed] [Google Scholar]
- 25. Chui J. N., Papachristos A. J., Mechera R., et al., “Unexpected Deaths After Endocrine Surgery: Learning From Rare Events Using a National Audit of Surgical Mortality,” British Journal of Surgery 109 (2022): 1164–1171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Gupta A. K., Stewart S. K., Cottell K., McCulloch G. A., Babidge W., and Maddern G. J., “Potentially Avoidable Issues in Neurosurgical Mortality Cases in Australia: Identification and Improvements,” ANZ Journal of Surgery 87 (2017): 86–91. [DOI] [PubMed] [Google Scholar]
- 27. Chan J. C. Y., Gupta A. K., Stewart S. K., et al., “Mortality in Australian Cardiothoracic Surgery: Findings From a National Audit,” Annals of Thoracic Surgery 109 (2020): 1880–1888. [DOI] [PubMed] [Google Scholar]
- 28. Kannan V., Ullah N., Geddada S., et al., “Impact of “Enhanced Recovery After Surgery” (ERAS) Protocols vs. Traditional Perioperative Care on Patient Outcomes After Colorectal Surgery: A Systematic Review,” Patient Safety in Surgery 19 (2025): 4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Justiniano C. F., Loria A., Hellenthal N. J., et al., “The Accumulation of ERAS (Enhanced Recovery After Surgery) Components Reduces Post‐Colectomy Length of Stay at Small and Low Volume Hospitals,” American Journal of Surgery 223 (2022): 744–752. [DOI] [PubMed] [Google Scholar]
- 30. Ayinde B. O., Chokshi P., Adhikari S., et al., “Challenges and Elements Hindering the Adoption of Enhanced Recovery After Surgery (ERAS) Protocols in Colorectal Surgery and Their Resolutions: A Systematic Review,” Cureus 16 (2024): e63222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Pagano E., Pellegrino L., Robella M., et al., “Implementation of an Enhanced Recovery After Surgery Protocol for Colorectal Cancer in a Regional Hospital Network Supported by Audit and Feedback: A Stepped Wedge, Cluster Randomised Trial,” BMJ Quality and Safety 33 (2024): 363–374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Tonini V. and Zanni M., “Impact of Anastomotic Leakage on Long‐Term Prognosis After Colorectal Cancer Surgery,” World Journal of Gastrointestinal Surgery 15 (2023): 745–756. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. van Rooijen S. J., Huisman D., Stuijvenberg M., et al., “Intraoperative Modifiable Risk Factors of Colorectal Anastomotic Leakage: Why Surgeons and Anesthesiologists Should Act Together,” International Journal of Surgery 36 (2016): 183–200. [DOI] [PubMed] [Google Scholar]
- 34. Bedrikovetski S., Dudi‐Venkata N. N., Kroon H. M., Vather R., and Sammour T., “Towards a Zero Percent Anastomotic Leak Rate Using a Defined Risk Reduction Strategy,” European Surgery 53 (2021): 317–322. [Google Scholar]
- 35. Kacmaz H. Y., Baser M., and Sozuer E. M., “Effect of Prophylactic Negative‐Pressure Wound Therapy for High‐Risk Wounds in Colorectal Cancer Surgery: A Randomized Controlled Trial,” Advances in Skin & Wound Care 35 (2022): 597–603. [DOI] [PubMed] [Google Scholar]
- 36. Holte K. and Kehlet H., “Postoperative Ileus: A Preventable Event,” British Journal of Surgery 87 (2000): 1480–1493. [DOI] [PubMed] [Google Scholar]
- 37. Tevis S. E., Carchman E. H., Foley E. F., Harms B. A., Heise C. P., and Kennedy G. D., “Postoperative Ileus—More Than Just Prolonged Length of Stay?,” Journal of Gastrointestinal Surgery 19 (2015): 1684–1690. [DOI] [PubMed] [Google Scholar]
- 38. Wells C. I., Bhat S., Xu W., et al., “Variation in the Definition of ‘Failure to Rescue’ From Postoperative Complications: A Systematic Review and Recommendations for Outcome Reporting,” Surgery 175 (2024): 1103–1110. [DOI] [PubMed] [Google Scholar]
- 39. Wells C. I., Baraza W., O'Grady G., and Bissett I. P., “‘Failure to Rescue’ From Anastomotic Leak Following Colorectal Cancer Resection: An Observational Study From a Binational Registry,” Colorectal Disease 27 (2025): e70262. [DOI] [PubMed] [Google Scholar]
- 40. Wells C. I., Varghese C., Boyle L. J., et al., ““Failure to Rescue” Following Colorectal Cancer Resection: Variation and Improvements in a National Study of Postoperative Mortality,” Annals of Surgery 278 (2023): 87–95. [DOI] [PubMed] [Google Scholar]
- 41. Chan J. C. Y., Gupta A. K., Stewart S., et al., ““Nobody Told Me”: Communication Issues Affecting Australian Cardiothoracic Surgery Patients,” Annals of Thoracic Surgery 108 (2019): 1801–1806. [DOI] [PubMed] [Google Scholar]
- 42. Ey J. D., Herath M. B., Reid J. L., et al., “Improving Communication and Patient Information Recall via a Question Prompt List: Randomized Clinical Trial,” British Journal of Surgery 110 (2023): 1793–1799. [DOI] [PubMed] [Google Scholar]
- 43. Treloar E. C., Ey J. D., Herath M., et al., “Optimizing Ward Rounds: Systematic Review and Meta‐Analysis of Interventions to Enhance Patient Safety,” British Journal of Surgery 112 (2025): znaf041. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from Australian and New Zealand Audit of Surgical Mortality. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the author(s) with the permission of Australian and New Zealand Audit of Surgical Mortality.