Perioperative predictive factors of failure to rescue following highly advanced hepatobiliary-pancreatic surgery: a single-institution retrospective study

Failure to rescue (FTR) is defined as a severe postoperative complication leading to death [1‐5]. When surgery is a key component of treatment, there may be surgery-related complications that lead to FTR in some cases. In particular, highly advanced hepatobiliary-pancreatic surgery (HBPS) is more likely than general gastroenterological surgery to induce severe complications leading to FTR. The FTR rate can be used as a quality indicator of the management of postoperative complications rather than simply being indicative of complication severity. Thus, FTR is an important outcome to consider when seeking to improve treatment quality.

World Health Organization (WHO) guidelines [6] state that postoperative mortality and complication rates decreased in departments that used the WHO surgical check list [7]. In the USA, use of the National Surgical Quality Improvement Program (NSQIP) has been shown to improve surgical outcomes [8]. The Japanese Society of Hepato-Biliary-Pancreatic Surgery (JSHBPS) established systems for board certification in relation to both instructors and training institutions in Japan in 2008, which were reported to have improved highly advanced HBPS mortality rates [9].

Silber et al. suggested that both patient- and hospital-specific factors affect potential prevention of FTR [10]. However, subsequent studies have focused only on hospital-specific risk factors [11‐13], whereas patient-specific predictors of FTR in highly advanced HBPS have not been adequately investigated. Recognition of factors significantly associated with FTR may improve protocols that attempt to rescue patients with severe postoperative complications. Therefore, this study aimed to identify perioperative (pre-, intra-, and postoperative) risk factors to help predict FTR following highly advanced HBPS.

This single-center retrospective study was conducted in accordance with the World Medical Association Declaration of Helsinki and was approved by the Ethics Committee of Gifu University (approval number: 2023–018).

This retrospective study involved 686 patients who had undergone highly advanced HBPS at the Department of Gastroenterological Surgery, Gifu University Hospital, between January 2010 and October 2022. Gifu University Hospital is a JSHBPS-certified training institution. All highly advanced HBPS surgical procedures were conducted by experienced board-certified JSHBPS-qualified surgeons.

At least one postoperative complication occurred in 348 (50.7%) patients. According to the CD grading system, 42 (6.1%) patients with grade I complications recovered without any treatment, 129 (18.8%) patients with grade II complications required antibiotic therapy, 156 (22.7%) patients with grade III complications needed radiologic intervention or re-operation, and 12 (1.7%) patients with grade IV complications and 9 (1.3%) patients with grade V complications died in the hospital. We excluded 509 patients (no postoperative complications, n = 338; no severe complications, n = 171). In total, 177 (25.8%) patients who experienced at least one postoperative severe complication, defined as grades ≥ III, were included in our study (Fig. 1).

FTR is defined as a postoperative complication leading to death [1‐5]. While a higher complication rate might appear likely to lead to an increased postoperative mortality rate, Ghaferi et al. showed that differences in mortality rates were not associated with large differences in postoperative complication rates [12, 14], but with the ability of hospitals to effectively rescue patients from complications. Therefore, FTR can be considered a quality indicator of the management of postoperative complications rather than of the extent of postoperative complications alone. FTR rates have been reported to vary widely across hospitals for all procedures and are highly correlated with postoperative mortality. Hospital bed size, intensive care unit (ICU) availability, rapid response system (RRS) availability, hospital technology, nurse-to-patient ratios, average daily census, and teaching status have been found to be associated with differences in FTR rates between hospitals with very low and very high mortality rates [11‐15]. These findings suggest that FTR rates might be influenced by the extent to which hospitals have well-organized multi-disciplinary teams enabling early intervention through involving endoscopists, radiologists, infection control doctors, and intensivists. Our institution, with > 600 beds and classified as a high-volume center, is a university-affiliated hospital with highly advanced technology as well as being a JSHBPS-certified training institution. Furthermore, endoscopists, radiologists, infection control doctors, and intensivists are on staff. Both ICU and RRS are available, and nursing care is provided at a ratio of 7:1. These specific characteristics of our institution are likely to have contributed to the lower FTR rates than those reported in previous studies [16‐21]. Therefore, analysis of data obtained in this highly technical and well-equipped medical environment may help identify issues that need to be addressed to further reduce FTR rates following highly advanced HBPS.

Some studies have reported non-hospital-related risk factors for FTR following highly advanced HBPS. Elfrink et al. showed that factors independently associated with FTR following liver resection were age (65–80 years), an ASA physical status classification of 3, liver cirrhosis, biliary cancer, major liver resection, postoperative liver failure, cardiac complications, and thromboembolic complications [16]. Lei et al. reported that the factors predicting postoperative mortality following liver resection for hepatocellular carcinoma were the Child–Pugh score, intraoperative blood loss, and postoperative liver failure [22]. Gleeson et al. identified the following independent risk factors in FTR following PD: age, ≥ 65 years; albumin level, < 3.5 g/dl; and the development of shock, postoperative renal failure, or postoperative respiratory failure [19]. Endo et al. found that major liver resection and blood transfusion were independently associated with FTR following hepatopancreatectomy [23]. The results of this study are consistent with those of previous reports concerning perioperative predictive factors for FTR following highly advanced HBPS.

First, intraoperative blood loss of > 1600 ml and blood transfusion have been found to be predictive factors for FTR. Intraoperative blood loss is an essential consideration in surgery and has been reported to have both short- and long-term outcomes [22‐25]. Nonami et al. [26] reported that blood loss was independently associated with postoperative liver failure and mortality. A strong correlation has been observed between intraoperative blood loss and blood transfusions. Yamamoto et al. [27] reported that all patients with blood loss > 1500 ml received blood transfusions in their study, including 251 liver resection cases. Therefore, considering the cutoff value for blood loss calculated in this study, it is possible to conclude that both massive blood loss and transfusion are risk factors for FTR. Homologous blood transfusion is known to increase the rate of postoperative infectious complications, owing ostensibly to immunosuppression [28, 29], with homologous blood transfusion being a significant risk factor for bacterial infection and a possible risk factor for FTR in surgically treated patients. Therefore, surgeons should maximize their efforts to decrease intraoperative blood loss and avoid unnecessary blood transfusions through the application of sophisticated surgical skills and communication with anesthesiologists.

Second, postoperative organ failure is one of the most serious postoperative complications that can lead to poor outcomes in all surgeries [30‐32]. Organ failure involves organ dysfunction to such a degree that homeostasis cannot be maintained without external clinical intervention. MOF is defined as the involvement of two or more organ systems. Postoperative organ dysfunction can occur in any organ; however, the pulmonary, hepatic, cardiac, renal, and cerebral vessels are more commonly involved. In this study, we selected liver and respiratory failure as risk factors for FTR owing to their high rate of postoperative organ failure. Both types of organ failure significantly correlated with FTR in the univariate analysis. Postoperative liver failure has previously been reported to be an independent risk factor for FTR following liver resection [16‐18, 22]. In addition, massive intraoperative bleeding and liver-related disease comorbidities are known causes of postoperative liver failure [16, 26], and the confounding relationship between these factors may have influenced our results.

Postoperative respiratory failure is a significant risk factor for FTR [32‐34]. Postoperative respiratory failure is defined as an unplanned postoperative reintubation or prolonged postoperative intubation. Several previous studies have reported postoperative respiratory failure incidence rates ranging from 2.7 to 3.4%. Older age, ASA, pulmonary-related disease comorbidity, longer surgery, pneumonia, abdominal surgery, and diaphragmatic dysfunction have also been reported to be risk factors [32‐35]. Diaphragmatic dysfunction may develop following prolonged mechanical ventilation, damage to the muscles and nerves of the diaphragm, and irritation from subdiaphragmatic abscesses or thoracoabdominal effusions. More caution may be needed to avoid respiratory failure for patients with complications that may lead to diaphragmatic dysfunction.

This study had some limitations. First, this single-center retrospective study involved a small number of FTR events, which may have resulted in selection bias and multiplicity issues in the statistical analysis. A multicenter study with a larger number of patients is required to obtain more accurate results. However, multicenter studies may show a large effect on inter-institutional disparities in terms of FTR rates. Therefore, our study concerning FTR at a single institution with a well-developed medical environment and uniform surgical indications may be of particular value. Second, this study included all patients with highly advanced HBPS and hepatobiliary and pancreatic surgeries. Each type of surgery may be associated with different risk factors for FTR. This limitation should be considered when evaluating our study results.

FTR was shown to be associated with perioperative factors. Well-coordinated surgical procedures to avoid intra- and postoperative bleeding and unnecessary blood transfusions, as well as postoperative team management with attention to the occurrence of organ failure, may decrease FTR rates.