Textbook outcomes for liver resection: can a medium sized centre have acceptable outcomes?

BACKGROUND: Textbook outcome (TO) is an objective, composite measure of clinical outcomes in surgery. TO in liver surgery has been used in previous international studies to define and compare performance across centres. This study aimed to review TO rates following liver resection at a single institution. The secondary aim was to use a CuSum analysis to evaluate monitoring of performance quality over time for colorectal cancer liver metastases (CRCLM). METHODS: All patients undergoing liver resection for benign and malignant causes from Christchurch Hospital hepatobiliary unit between 2005 and 2022 were included. Textbook outcomes measures were the absence of; intraoperative incidents, Clavien-Dindo >3 complication, 90 day re-admission, 90 day mortality, R1 resection, and post-operative bile leak/liver failure. Sequential CuSum analysis was performed to review achievement of TO in liver resections for colorectal cancer liver metastases (CRCLM). RESULTS: Four hundred and seventy-eight patients were included in this study, 54 had resection for benign pathology, 290 for CRCLM and 134 for other malignancies. TO was achieved in 74% of cases overall, with rates for benign, CRCLM and other malignancy being 82%, 73% and 74% respectively (P = 0.405). CuSum analysis documented a deterioration in performance after patient 60, with return to baseline by end of study period. CONCLUSIONS: TO for liver resection in a medium sized centre in New Zealand are comparable to published rates. It is possible to use process control techniques like CuSum with the binary result of TO to monitor performance, providing opportunity for continuous improvement in surgical units.

Using Textbook Outcomes to benchmark practice in pancreatic surgery.

BACKGROUND: Textbook Outcome (TO) is a novel composite measure of clinical outcomes that can be used to measure the quality of surgical outcomes. TOs for pancreatic surgery were published by the Dutch Pancreatic Cancer Group (DPCG) in 2020. The aim of this study was to explore how a medium volume hepatopancreaticobiliary unit could use TO to benchmark local outcomes following pancreatic surgery. METHODS: Retrospective analysis of prospectively collected data from patients who underwent pancreaticoduodenectomy (PD) or distal pancreatectomy (DP) for all indications between March 2005 and February 2020 at Christchurch Hospital (CH). Analysis of TO items as defined by the DPCG was performed and compared to nationwide Dutch outcomes (2014-2017), including cumulative analysis using CuSum. RESULTS: In total, 273 patients were included (median age 63 years; 51% female) of which 182 (67%) underwent PD and 91 (33%) underwent DP (median annual volume 12 PDs/6 DPs). Overall, 58% of patients undergoing PD and 74% of patients undergoing DP achieved TO, compared with 58% and 67%, P = 0.944 and P = 0.231, respectively, for the Netherlands (median annual volume 33 PDs/8 DPs per hospital). CONCLUSIONS: TO offers a useful quality measure to benchmark local outcomes following pancreatic surgery against an external nationwide analysis. The results show that as a medium volume centre performance was comparable to previously published Dutch results, which included high volume centres. Applying CuSum methodology to the TO metric allows a continuous measure of performance. This offers the potential to provide feedback for quality improvement strategies.

Reducing length of stay for patients presenting to general surgery with acute non-surgical abdominal pain.

BACKGROUND: Acute abdominal pain is a common surgical presentation. We previously found that over the last decade, more patients were admitted to hospital with non-surgical diagnoses (e.g. gastroenteritis, constipation and non-specific abdominal pain) and length of stay and use of imaging (mainly computed tomography scan) for these patients increased. This study aimed to reduce length of stay and use of imaging for patients admitted with non-surgical abdominal pain. METHODS: A prospective study was undertaken in a tertiary centre evaluating length of stay and use of additional imaging in patients with a non-surgical diagnosis after a quality improvement intervention was implemented. RESULTS: A total of 454 patients were included; 204 (44.9%) presented with non-surgical abdominal pain. During the study period, a significant reduction in computed tomography scan requests was observed (38.5-25.0%, P = 0.037) and an increasing proportion of these patients were discharged within 12 h (33.3-57.1%, P = 0.018). The number of re-presentations remained unchanged (P = 0.358). CONCLUSIONS: The study intervention increased the proportion of patients with non-surgical diagnoses that were successfully discharged within 12 h and reduced the use of additional imaging in this group. This may lead to improved use of health care resources for patients with more urgent diagnoses.

Can Surgical Registrars Accurately Identify the Urgent from the Non-Urgent Patient Presenting with Acute Abdominal Pain?

BACKGROUND: Acute abdominal pain is a common surgical presentation with a wide range of causes. Differentiating urgent patients from non-urgent patients is important to optimise patient outcomes and the use of hospital resources. The aim of this study was to determine how accurately urgent and non-urgent patients presenting with abdominal pain can be identified. METHODS: A prospective study of consecutive patients admitted with abdominal pain was undertaken. Urgent patients were classified as requiring treatment (theatre, intensive care unit, endoscopy, or radiologic drainage) within 24 h. Differentiation between urgent and non-urgent was made on the basis of the initial assessment prior to the use of advanced imaging. Outcomes were compared to a final classification based on final diagnosis as adjudicated by an expert panel. RESULTS: Of the 301 patients included, 93 (30.9%) were deemed urgent based on initial assessment, compared to 83 (27.6%) on final diagnosis. Overall sensitivity for recognising urgent patients was 74.7% and specificity 89.9%, and overall accuracy was higher for senior registrars compared to junior registrars (p = 0.015). Urgent patients more often looked unwell or had peritonism on examination (39.8 vs. 17.4% and 56.6 vs. 14.7%, respectively, p < 0.001 for both). CONCLUSIONS: Registrars can accurately differentiate urgent from non-urgent patients with acute abdominal pain in the majority of cases. Accuracy was higher amongst senior registrars. The "end-of-the-bed-o-gram" and clinical examination are the most important features used for making this differentiation. This demonstrates that there is no substitute for exposure to acute presentations to improve a trainee's diagnostic skill.

Patients requiring an acute operation: where are the delays in the process?

BACKGROUND: Delays to surgery for patients requiring an acute operation are associated with increased morbidity and mortality. A recent study from our institution observed long waiting times for patients booked for an acute operation. The aim of this study was to evaluate the patient's progress from presentation to arrival in the operating theatre and to identify where delays occurred. METHODS: Patients undergoing acute general surgery between July 2016 and May 2017 were studied. Data were obtained for time of presentation, imaging, theatre and booking. A time interval from presentation to booking for theatre of greater than 6 h was defined as a diagnostic delay. A time interval from booking to theatre greater than the category defined time (four-level priority system) was defined as a logistic delay. RESULTS: A total of 683 patients were included. A diagnostic delay was observed in 55.1%. This occurred more frequently in patients who required imaging prior to their operation (82.5 versus 41.1%, P < 0.001). Logistic delay occurred in 31.0% of the patients, and this was most common for patients booked as a category 3 (requiring surgery within 6 h, 41.8%, P < 0.001). Patients who had a diagnostic delay were significantly more likely to have a post-operative complication compared to patients who did not (17.2 versus 10.0%, P = 0.009). CONCLUSION: There are significant delays associated with patients presenting to the acute general surgery service and their transition to theatre. Addressing both the diagnostic and the logistic delays in our institution should result in a significant improvement in patient care.

Appropriateness of CT scans for patients with non-traumatic acute abdominal pain.

OBJECTIVE: Acute abdominal pain is the most common reason for surgical admission. CT scans are increasingly used to aid early diagnosis. Excessive use of CT scans is associated with increased length of stay, healthcare costs and radiation. The aim of this study was to evaluate the appropriateness of CT scans for patients presenting with acute abdominal pain. METHODS: We examined 100 consecutive patients presenting with new acute abdominal pain who underwent a CT scan. Clinical information available at the time the scan was ordered, was summarised and reviewed independently by five consultant general surgeons and five consultant radiologists. RESULTS: A CT scan was judged to be not indicated in a median of 21% of cases (range 12-53%), more information was required in a median of 16% (0-41%) and in a median of 58% (37-88%) the CT scan was considered indicated. There was a good level of agreement (Cronbach's α 0.704) across the 10 experts. CONCLUSION: These data suggest that a large proportion of CT scans for patients with acute abdominal pain are not clinically indicated or are being performed prior to adequate clinical work-up. Optimising CT scan requests for this patient group will improve use of healthcare resources. Advances in knowledge: Both radiologists and general surgeons agree that there is no indication for an abdominal CT scan for a patient presenting with acute abdominal pain in a median of 21% of the cases.

Systematic review of diagnostic pathways for patients presenting with acute abdominal pain.

PURPOSE: Abdominal pain is the most common reason for surgical referral. Imaging, aids early diagnosis and treatment. However unnecessary requests are associated with increased costs, radiation exposure and increased length of stay. Pathways can improve the quality of the diagnostic process. The aim of this systematic review was to identify the current evidence for diagnostic pathways and their use of imaging and effect on final outcomes. DATA SOURCES: A systematic search of Embase, Medline and Cochrane databases was performed using keywords and MeSH terms for abdominal pain. STUDY SELECTION: All papers describing a pathway and published between January 2000 and January 2017 were included. DATA EXTRACTION: Data was obtained about the use of imaging, complications and length of stay. Quality assessment was performed using MINORS and Level of Evidence. RESULTS: Ten articles were included, each describing a different pathway. Five studies based the pathway on literature reviews alone and five studies on the results of their prospective study. Of the latter five studies, four showed that routine imaging increased diagnostic accuracy, but without showing a reduction in length of stay, complication rate or mortality. None of the studies included evaluated use of hospital resources or costs. CONCLUSION: Pathways incorporating routine imaging will improve early diagnosis, but has not been proven to reduce complication rates or hospital length of stay. On the basis of this systematic review conclusions can therefore not be drawn about the pathways described and their benefit to the diagnostic process for patients presenting with abdominal pain.

Review of appendicectomies over a decade in a tertiary hospital in New Zealand.

BACKGROUND: Acute appendicectomy is the most common emergency operation for patients with abdominal pain. In the last decade, computed tomography (CT) scans have increasingly been used to aid in the diagnosis in order to reduce the negative appendicectomy rate. The aim of this study was to evaluate our institution's negative appendicectomy rate and the use of pre-operative imaging. METHODS: A retrospective review was undertaken for all adult patients (>16 years), who underwent an appendicectomy on emergency basis in the years 2004, 2009 and 2014. Cases were identified from the hospital electronic theatre record system. Data were also obtained from the patients records and laboratory reports. RESULTS: A total of 874 patients were included, 227 in 2004, 308 in 2009 and 339 in 2014. The negative appendicectomy rate was 29.1% in 2004, 20.1% in 2009 and 19.5% in 2014 (P = 0.014). Negative appendicectomies were more common in women (P = <0.001), patients between the ages of 16-30 years (P = <0.001) and were associated with low inflammatory markers (median white cell count of 10.2, C-reactive protein of 8, P = <0.001). The use of CT scan prior to operation increased between 2009 and 2014 (34 (11.0%) versus 64 (18.9%), P = <0.001). CONCLUSION: Though the number of appendicectomies being performed in our institution has increased over the last decade, the negative appendicectomy rate remains fairly static and the increased use of CT scans did not further decrease the proportion of negative appendicectomies between 2009 and 2014.

Acute abdominal pain-changes in the way we assess it over a decade.

AIMS: Acute abdominal pain accounts for 5-10% of all emergency department visits. Rapid and accurate diagnosis is critical to ensure optimal outcomes. In the last decade, increased use of CT scans and the introduction of surgical short stay units has changed the way this group of patients is managed. The aim of this study was to evaluate the effects of these changes on patient management. METHODS: A retrospective clinical study was undertaken including all patients admitted with abdominal pain under general surgery in the years 2004, 2009 and 2014. Two hundred from each of the three years were randomly selected and their care was reviewed. RESULTS: During the study period, more patients were admitted under general surgery, from 1,462 in 2004 to 2,737 in 2014 (P=0.001). There was an increase in the proportion of patients admitted with non-surgical abdominal pain (25% in 2004 vs 34% in 2014, P=0.035). More computed tomography (CT) scans were performed (26.0% in 2004 vs 45.0% in 2014, P=0.001). CONCLUSIONS: More patients were admitted under general surgery with abdominal pain and a greater proportion of these patients were admitted with non-surgical problems. Use of CT scans increased during the study period.

Total number of lymph nodes in oncologic resections, is there more to be found?

Pathologic staging of oncologic specimens includes the identification of the accurate lymph node status. Retrieving more lymph nodes leads to a more reliable N0 status in the TNM classification. The aim of this prospective study was to evaluate whether more lymph nodes can be retrieved from oncologic resection specimens when more time is invested in the search and if this contributes to a more reliable N-status in the individual patient. A total of 67 gastrointestinal oncologic specimens were reexamined for additional lymph nodes. The mean number of lymph nodes collected in the prospective group was compared against two retrospective groups, one before minima for lymph node counts were set (retrospective group 1) and one after (retrospective group 2). More lymph nodes were dissected per specimen in the prospective group (24.1 lymph nodes), compared to the retrospective group (14.3 lymph nodes, P = <0.001). During the study period, more patients were diagnosed as pN+ compared to the two retrospective groups (62.7 vs. 47.8 % respectively, P = 0.082). Significantly more lymph nodes can be found in oncologic specimens when more time is invested in the search. This will result in more accurate staging of the tumor.

Ectopic pregnancy: when is expectant management safe?

This study was conducted to evaluate expectant management in asymptomatic patients with an initial serum beta-hCG titer of <2,500 IU/l and to determine the independent ability of initial serum beta-hCG titers and trend of serum beta-hCG to predict successful expectant management. A cohort of patients (N = 418) with suspected ectopic pregnancy (EP) between January 1991 and July 2008 is described. Three groups were defined: group I (n = 182), immediate surgical intervention (<24 h); group IIa (n = 130), unsuccessful expectant management (surgical intervention during follow-up), and group IIb (n = 99), successful expectant management (spontaneous regression of trophoblast). Hospital protocol was not complied in 35 cases (Table 1). Beta-hCG levels >3,000 IU/l occur in our expectant management group; however, none of these cases were successful. Unnecessary surgery was prevented in 14% (n = 7) of asymptomatic patients with initial beta-hCG of >2,000 IU/l. The success rate of expectant management was 49%, without a rise in complication rate or number of acute cases. In conclusion, the initial serum beta-hCG cutoff level of 2,000 IU/l is not a rigid upper limit for accepting expectant management in suspected EP and best practice is case specific. In asymptomatic patients, the serum beta-hCG cutoff level of at least 2,500 IU/l can be used for expectant management. This cutoff could be higher, but interpretation is limited due to censure in follow-up inherent to the predefined clinical protocol. There is no gain in including patients for expectant management with initial serum beta-hCG level >3,000 IU/l.