General surgical emergency admissions are the largest group of all surgical admissions to UK hospitals1 and account for a large percentage of all surgical deaths.2 Within this group, emergency major bowel surgery has one of the highest mortalities,1–7 which can reach 50% in the over 80s;6 many of the patients undergoing this type of surgery are elderly and have multiple comorbidities.1–7 A review of 94 UK hospitals' surgical outcomes (excluding specialist surgical procedures such as neurosurgery) using centrally validated data reported that 12.5% of surgical procedures accounted for more than 80% of surgical deaths with the three procedures with the highest mortality all being intraabdominal surgeries.2 The postoperative needs of this large group of patients were highlighted by the Association of Surgeons consensus statement of June 2007.1
A recent paper and accompanying editorial highlighted the problem of critical care bed access for patients following high-risk non-cardiac surgery.8,9 Although inadequate provision of beds, funding and resources were cited as causes, these may be compounded by an underestimation, by both surgeons and anaesthetists, of the mortality associated with such surgery.
The aims of this observational study were to estimate mortality after emergency laparotomy and to explore disparities between survivors and non-survivors. Secondary endpoint was to describe postoperative care and its potential impact on mortality.
The study was approved by the local audit committee of the Royal United Hospitals (Bath); therefore, the requirement for patient consent was waived.
The study was performed in a 650-bed general hospital serving a population of over 500 000. The hospital has a fully staffed 24-h emergency (non-orthopaedic) theatre. Patients were included if they underwent a midline exploration of the abdomen classified as ‘urgent’ defined as needing surgery within hours of diagnosis; or ‘emergency’, requiring surgery within minutes of diagnosis according to the classification used by the UK National Confidential Enquiry into Patient Outcome and Death (NCEPOD).3 Emergency vascular surgery, including aortic aneurysm surgery was excluded. Patients undergoing a ‘re-look laparotomy’ were excluded, as were patients undergoing a simple appendectomy.
The decision as to the postoperative care and destination of the patients was made by the operating surgeon, anaesthetist and intensivist as is standard UK practice. There was no lack of critical care beds during the study. Patients sent to the ward had standard postanaesthesia care unit (PACU) stays of approximately 2 h with no enhanced recovery. There is no facility for extended PACU stay (e.g. overnight) in this hospital. Patients sent to the critical care unit received high dependency unit (HDU) care (level 2 care, i.e. requiring only single organ support, not including invasive ventilation) or intensive care (ICU; level 3 care, i.e. requiring either invasive ventilation or support of two or more organ systems). The critical care unit admits over 500 patients a year of whom at least 65% are mechanically ventilated, the median Acute Physiology and Chronic Health Evaluation II (APACHE II) score is 12 and the standardised mortality ratio (SMR) is 0.86.
Data were collected prospectively on consecutive patients undergoing emergency laparotomy over a 5-month period. Procedural data were collected retrospectively from hospital notes the day after surgery and included the following: patient demographics, surgical procedure performed, time of day anaesthesia started, grade of surgeon and anaesthetist, type of anaesthesia and monitoring used, blood loss and location of immediate postoperative care. Intraoperative lactate measurements were recorded.
Patients were followed up and length of in-patient stay, survival at 30 days [as P-POSSUM (Portsmouth predictor - Physiological and Operative Severity Score for the enUmeration of Mortality and Morbidity) predicts 30-day mortality10,11] and in-hospital survival were recorded. Data were collected by investigators, who were not directly involved in the care of the patients.
A P-POSSUM score and predicted death rate (PDR) were calculated retrospectively using the above data and the last blood results available prior to surgery. P-POSSUM is a risk score developed in 1991 that consists of 18 variables (12 physiological and six operative). These values are weighted and give a score and predicted 30-day mortality. It has been validated in many different surgery types.10,11 Individual patient P-POSSUM score PDRs ranged from 0.6 to 99.9% [median 13.0, interquartile range (IQR) 4.0–40.9]. A cut-off of more than 10% for PDR was arbitrarily chosen to distinguish between high-risk and low-risk patients. The observed number of deaths in the first 30 days was divided by the predicted death rate as defined by the P-POSSUM score to give a SMR.
Results were expressed as arithmetic mean ± SD, median with IQR or frequencies with percentages. Parametric values were analysed using the paired t-test; non-parametric samples were analysed using either Mann–Whitney U-tests or χ2 test in contingency tables. Data analysis was performed with SPSS for Windows version 15.0 software (SPSS Inc., Chicago, Illinois, USA).
A P value of less than 0.05 was considered statistically significant.
One hundred and twenty-four consecutive patients were included. One patient had two laparotomies; the second procedure was deemed to be unrelated to the original laparotomy and the patient survived to leave hospital.
Observed and predicted overall mortality
One hundred patients survived and 24 patients (19.4%) died during their hospital stay. Ten of 26 patients over 80 years died (38%). Fifteen of all deaths (62%) occurred in the first week after surgery. The 30-day mortality was 16.9% (21 of 124 patients); 19 patients (15%) remained in hospital longer than 30 days postoperatively, of whom 16 survived to discharge. The PDR for all 124 patients based on the mean P-POSSUM score was 27.4%. All patients with a PDR of more than 90% died. The overall SMR was 0.71.
The demography of survivors and non-survivors was different (Table 1). The median duration of hospital stay for patients who survived was 11 days (range 2–188, IQR 6–20), and for non-survivors 4.5 days (range 1–40, IQR 2–12).
Postoperative pathway and mortality
The distribution of the patients after surgery to the different postoperative care pathways was unequal (Table 2). Eighty-seven patients (70.2%) went to the PACU and then ward after surgery, six to PACU followed by HDU and then ward (4.8%), and 30 to ICU followed by HDU and then ward (24.2%; Table 2).
Patients sent to ICU+HDU+ward pathway or PACU+HDU+ward pathway had a higher PDR (median 50.4%; range 11.3–99.7%, IQR 34.8–79.0%) than those discharged to a general surgical ward after PACU (median 6.7%; range 0.6–97.0%, IQR 2.2–15.7).
Thirty-six patients with a PDR more than 10% were admitted to the PACU+ward after surgery (Table 2); 15 of these 36 patients were 80 years or older, with a mean PDR of 43.9% (median 34.6%, IQR 18.9–63.7%). In only one patient's notes was it documented that critical care was not offered for reasons of futility; this patient had a PDR of 82.7% and died 15 days after surgery.
Similar numbers of non-survivors followed either PACU+ward (11) or ICU+HDU+ward (12) pathways (Tables 1 and 2). Three patients were subsequently admitted to critical care having started on the PACU+ward pathway, two of whom died. The SMR for patients managed in ICU+HDU+ward pathway was 0.69; for patients managed in PACU+ward, 0.82 (Table 2).
There were two main results of this study. First, in a large District General Hospital, patients undergoing high-risk emergency abdominal surgery had high mortality. Second, most patients were cared postoperatively via the pathway PACU and surgical ward; however, this pathway had higher SMR than the pathway ICU+HDU+ward.
The overall mortality reported here is high but unlikely to be an outlying result in comparison with other similar hospitals.1–5,7 For similar emergency surgery, mortality rates range between 7% for lysis of adhesions12 and 36% for Hartmann's procedure.13 In our cohort of patients, the in-hospital mortality rate was approximately one in five rising to almost two in five in the patients aged over 80. This is consistent with results published previously.2–7,14 In addition, patients over 80 years of age are more likely to present for emergency surgery than for elective surgery.7,14
Our mortality rates were in all cases lower than that predicted by P-POSSUM scoring. Two explicatory models may be considered: first, P-POSSUM may overestimate mortality; and second, the investigated hospital had high care standards. Three arguments are in favour of the second hypothesis: this hospital can provide a high proportion of highly skilled consultants during these emergency interventions. This hospital has a high performance in their ICU with a consistently low SMR, confirmed by Intensive Care National Audit and Research Centre, which collects data from English and Welsh ICUs (September 2006),15 and the investigated hospital is a medium-sized and manageable hospital, and which may facilitate communication channels between key players.
In this investigation, the distribution to the different postoperative care levels could be suboptimal, as observed by the patients discharged to PACU and then to ward having a higher SMR. This result is in concordance with a recent study observing increased survival in ICU and intermediate care or high-density units, including prolonged stay in PACU as compared with a general surgical ward.16 It is highly probable that postoperative complications are detected earlier and treated in well managed ICU and intermediate care units, and therefore, these units increase survival.17 The decision-making for the postoperative pathway after emergency laparotomy was purely clinical and often based on incomplete patient history; therefore, decisions may be heterogeneous. To reduce subjectivity in this important decision-making process, an Apgar score for surgery18 and other end of intervention evaluations for high-risk surgery19,20 have been proposed. However, none of these instruments have been tested in emergency conditions.
This study has a number of limitations. It is a single-centre study; therefore, data cannot be generalised. It is an observational study in a small number of patients with a variety of different surgical disorders, grouped under the term ‘emergency laparotomy’. The small number does not allow performing a multivariate regression analysis to identify independent risk factors for death in this setting; the results of this study have to be considered more as hypothesis generating than as evidence.
Based on the generated hypotheses, further research is required in the form of large cohort studies testing whether the screened risk factors are predictive for postoperative mortality in this particular setting. Furthermore, the use of P-POSSUM scoring and the PDR for risk stratification of patients undergoing emergency laparotomy should be investigated for optimal postoperative resource allocation and corresponding outcome. The P-POSSUM score and PDR are freely accessible21 and there is no restriction in its use. Furthermore, research is needed to investigate whether more intermediate or more critical care beds are needed for this type of patient or whether only a prolonged PACU stay may increase survival.
In conclusion, high mortality after emergency laparotomy was observed in this small observational study. The postoperative pathway may be a modifiable risk factor. The potential impact of postoperative pathways should be investigated in large cohort studies.
We thank Mr Jeremy Tate, Consultant Surgeon; Dr Gareth Parry PhD, Statistician; and Dr Andy Padkin and Dr Jules Cranshaw, Consultant Anaesthetists who commented on the article.
1 Association of Surgeons of Great Britain and Ireland. Emergency general surgery
: the future. A consensus statement; June 2007. www.asgbi.co.uk
2 Pearse RM, Harrison DA, James P, et al
. Identification and characterization of the high-risk surgical population in the United Kingdom. Crit Care 2006; 10:R81.
3 National confidential enquiries into patient outcome and death
4 Semmens JB, Aitken RJ, Sanfilippo FM, et al
. The western Australian audit of surgical mortality: advancing surgical accountability. Med J Aust 2005; 183:504–508.
5 Cook TM, Day CJE. Hospital mortality after urgent and emergency laparotomy
in patients aged 65 yr and over. Risk and prediction of risk using multiple regression analysis. Br J Anaesth 1998; 80:776–781.
6 Ford PNR, Thomas I, Cook TM, et al
. Outcome in critically ill octogenarians after surgery
: an observational study. Br J Anaesth 2007; 99:824–829.
7 Scottish Audit of Surgical Mortality Report 2008 (2007 data). www.sasm.org.uk
8 Jhanji S, Thomas B, Ely A, et al
. Mortality and utilization of critical care resources amongst high-risk surgical patients in a large NHS trust. Anaesthesia
9 Goldhill DR, Down JF. Are we operating as well as we can? Critical care to minimize postoperative mortality and morbidity. Anaesthesia
10 Copeland GP, Jones D, Walters M. POSSUM: a scoring system for surgical audit. Br J Surg 1991; 78:355–360.
11 Prytherch DR, Whiteley MS, Higgins B, et al
. POSSUM and Portsmouth POSSUM for predicting mortality. Physiological and Operative Severity Score for the enumeration of mortality and morbidity. Br J Surg 1998; 85:1217–1220.
12 Grafen FC, Neuhaus V, Schob O, et al
. Management of acute small bowel obstruction from intestinal adhesions: indications for laparoscopic surgery
in a community teaching hospital. Langenbecks Arch Surg 2010; 395:57–63.
13 Gastinger I, Marush F, Koch A, et al
. Hartmann's procedure indication in colorectal carcinoma. Chirurg 2004; 75:1191–1198.
14 Turrentine FE, Wang H, Simpson VB, Jones RS. Surgical risk factors, morbidity and mortality in elderly patients. J Am Coll Surg 2006; 6:865–877.
15 Intensive Care National Audit and Research Centre. www.icnarc.org
16 Simchen E, Sprung CL, Galai N, et al
. Survival of critically ill patients hospitalized in and out of intensive care. Crit Care Med 2007; 35:449–457.
17 Grocott M. Improving outcomes after surgery
. BMJ 2009; 339:b5173.
18 Gawande AA, Kwaan MR, Rogenbogen SE, et al
. An Apgar score for surgery
. J Am Coll Surg 2007; 204:201–208.
19 Ziai WC, Varelas PN, Zeger SL, et al
. Neurologic intensive care resource use after brain tumor surgery
: an analysis of indications and alternative strategies. Crit Care Med 2003; 31:2782–2787.
20 Blanshard H, Chung F, Manninen P, et al
. Awake craniotomy for removal of intracranial tumor: consideration for early discharge. Anesth Analg 2001; 92:89–94.
21 P-POSSUM score calculator. Societe Francaise d'Anesthesia et de Reanimation. http://www.sfar.org/scores2