Overpredicts Mortality Research Articles

NELA 1 - Actual and predicted 30-day mortality following emergency laparotomy using National Emergency Laparotomy Audit (NELA) and American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) risk prediction tools

Abstract Background Emergency laparotomy (EL) is common, with >30,000 performed annually in the UK. EL has a high 30-day mortality of 9.2%, especially in the elderly. Reliable preoperative risk assessment is essential to aid decision-making and perioperative care arrangements. NELA and ACS-NSQIP risk calculators can both be used to assess risk, and both can under or over-predict mortality. Methods This is a multi-center, retrospective, observational, cohort study investigating the correlation between actual 30-dm (a-30-dm) and predicted 30-dm (p-30-dm) following EL using NELA and ACS-NSQIP in adults over and under 70, over 2-years, in two UK centres. Sample size was calculated at 385 cases. Statistical analysis was carried out using GraphPad Prism 10.1.1. Results 626 cases were analyzed. A-30-dm, NELA, and ACS-NSQIP p-30-dm in the under 70s subgroup were 5.78%, 5.53%, and 3.41% respectively, with NELA and ACS-NSQIP correlating positively with a-30-dm: R²=0.12, p<0.0001 and R²=0.09, p<0.0001. In over 70s, a-30-dm was 13.96%, NELA p-30-dm 14.02% and ACS-NSQIP 9.76%, with a positive correlation between the latter two (R²=0.56, p<0.0001), but a weak one between a-30-dm and NELA (R²=0.04, p<0.0001) or ACS-NSQIP (R²=0.02, p<0.0001). For the entire cohort a-30-dm, NELA and ACS-NSQIP p-30-dm were 9.74%, 9.8%, and 6.59% respectively. NELA and ACS-NSQIP p-30-dm showed a strong positive correlation (R²=0.89; p<0.0001) with only NELA showing a strong positive correlation with a-30-dm (R²=0.31, p<0.0001); and not ACS-NSQIP (R²=0.05, p<0.0001). Conclusion NELA can accurately predict 30-day mortality in patients aged <70 undergoing EL but not >70. ACS-NSQIP and a-30-dm correlated in a very weak manner.

RACHS - ANZ : A Modified Risk Adjustment in Congenital Heart Surgery Model for Outcome Surveillance in Australia and New Zealand.

BackgroundOutcomes for pediatric cardiac surgery are commonly reported from international databases compiled from voluntary data submissions. Surgical outcomes for all children in a country or region are less commonly reported. We aimed to describe the bi‐national population‐based outcome for children undergoing cardiac surgery in Australia and New Zealand and determine whether the Risk Adjustment for Congenital Heart Surgery (RACHS) classification could be used to create a model that accurately predicts in‐hospital mortality in this population.Methods and ResultsThe study was conducted in all children's hospitals performing cardiac surgery in Australia and New Zealand between January 2007 and December 2015. The performance of the original RACHS‐1 model was assessed and compared with an alternative RACHS‐ANZ (Australia and New Zealand) model, developed balancing discrimination with parsimonious variable selection. A total of 14 324 hospital admissions were analyzed. The overall hospital mortality was 2.3%, ranging from 0.5% for RACHS category 1 procedures, to 17.0% for RACHS category 5 or 6 procedures. The original RACHS‐1 model was poorly calibrated with death overpredicted (1161 deaths predicted, 289 deaths observed). The RACHS‐ANZ model had better performance in this population with excellent discrimination (Az‐ROC of 0.830) and acceptable Hosmer and Lemeshow goodness‐of‐fit (P=0.216).ConclusionsThe original RACHS‐1 model overpredicts mortality in children undergoing heart surgery in the current era. The RACHS‐ANZ model requires only 3 risk variables in addition to the RACHS procedure category, can be applied to a wider range of patients than RACHS‐1, and is suitable to use to monitor regional pediatric cardiac surgery outcomes.

Abstract 9972: Do Risk Models Accurately Predict the Coronary Artery Bypass Graft Mortality Risk in Patients of Southeast Asian Origin?

Introduction: Cardiac surgery risk models have been developed based on predominantly Caucasian populations. This study is conducted to validate and evaluate these models for Southeast Asian patients. Methods: We analyzed a prospectively collected dataset of 1769 patients who underwent isolated CABG between 2009 and 2013 in an academic medical center: National University Heart Center, Singapore. Patients were scored by six risk models (Additive EuroSCORE, Logistic EuroSCORE, EuroSCORE II, Society of Thoracic Surgeons (STS), Northern New England Cardiovascular Disease Study Group, and Parsonnet). Discriminatory power was compared using ROC and C-statistics. Goodness of fit test was used to assess model calibration. Results: The overall observed mortality was 2.48% (n=44). C-statistics showed fair discriminatory power for all six risk scores, with STS (C-statistic 0.774, 95% CI 0.697 - 0.851) and EuroSCORE II (C-statistic 0.766, 95% CI 0.690 - 0.843) having the highest values. The optimal thresholds using Youden’s method were: STS threshold 1.33%, sensitivity 79.55%, specificity 68.04%; EuroSCORE II threshold 5.23%, sensitivity 54.55%, specificity 86.14%. Calibration testing revealed that only the STS (p=0.156) and EuroSCORE II (p=0.352) had a good fit with our data. However, EuroSCORE II and STS over-predict and under-predict mortality in high risk patients respectively. Conclusion: The STS and EuroSCORE II have the best predictive value for patients of Southeast Asian origin and can provide our surgeons with a suitable benchmark for outcome tracking. Inaccurate predictions for high-risk patients may be improved with recalibration.

Identification of the high risk emergency surgical patient: Which risk prediction model should be used?

IntroductionNational guidance states that all patients having emergency surgery should have a mortality risk assessment calculated on admission so that the ‘high risk’ patient can receive the appropriate seniority and level of care. We aimed to assess if peri-operative risk scoring tools could accurately calculate mortality and morbidity risk.MethodsMortality risk scores for 86 consecutive emergency laparotomies, were calculated using pre-operative (ASA, Lee index) and post-operative (POSSUM, P-POSSUM and CR-POSSUM) risk calculation tools. Morbidity risk scores were calculated using the POSSUM predicted morbidity and compared against actual morbidity according to the Clavien–Dindo classification.ResultsThe actual mortality was 10.5%. The average predicted risk scores for all laparotomies were: ASA 26.5%, Lee Index 2.5%, POSSUM 29.5%, P-POSSUM 18.5%, CR-POSSUM 10.5%.Complications occurred following 67 laparotomies (78%). The majority (51%) of complications were classified as Clavien–Dindo grade 2–3 (non-life-threatening).Patients having a POSSUM morbidity risk of greater than 50% developed significantly more life-threatening complications (CD 4–5) compared with those who predicted less than or equal to 50% morbidity risk (P = 0.01).DiscussionPre-operative risk stratification remains a challenge because the Lee Index under-predicts and ASA over-predicts mortality risk. Post-operative risk scoring using the CR-POSSUM is more accurate and we suggest can be used to identify patients who require intensive care post-operatively.ConclusionsIn the absence of accurate risk scoring tools that can be used on admission to hospital it is not possible to reliably audit the achievement of national standards of care for the ‘high-risk’ patient.

Validation of the use of Possum0 and P- Possum0 score in perforation peritonitis in Indian population

Background: The Physiological and Operative Severity Score for the enUmeration of Mortality and morbidity (POSSUM) model, its Portsmouth (P-POSSUM) modification are two surgical risk scoring systems used extensively to predict post-operative morbidity and mortality in general surgery. The aim was to undertake the study of validity of these models in Indian patients undergoing exploratory laparotomy for perforation peritonitis. Patients and Methods: A prospective study was performed, in which a total of 103 patients undergoing exploratory laparotomy for perforation peritonitis were included during the period of November 2011 to October 2013. The morbidity and mortality risks were calculated using the POSSUM and P-POSSUM. Results: Around 44/103 patients developed complications (total morbidity rate of 42.72%) in post-operative period; this was compared with POSSUM predicted morbidity. There was no statistical difference between observed and predicted morbidity (χ 2 = 45.607; df = 1; observed/expected ratio (O:E) = 0.82; P value = 0.000). Ten patients died (total mortality rate of 9.7%). The P-POSSUM expected mortality rate was compared. There was no statistical difference between the observed and P-POSSUM predicted mortality rates (χ 2 = 17.444, df = 1; P value = 0.000). However, P-POSSUM overpredicts mortality in our study (O:E = 0.25; P value = 0.000). Conclusion: POSSUM and P-POSSUM appear to be good and valid indices for use in risk prediction of morbidity and mortality, respectively (surgical outcome in perforation peritonitis) in the Indian population. We found that POSSUM accurately predicts morbidity but P-POSSUM is not able to predict mortality accurately.

Development and implementation of prospective outcome monitoring in a regional burns service using cumulative sum (CUSUM) techniques

IntroductionReal-time monitoring of mortality in burns units has the potential to immediately mark when mortality rates are significantly higher or lower than predicted. Rapid feedback from targeted internal audit allows early intervention, to reinforce positive practices, and improve systems where outcomes are unsatisfactory.This is the first study to describe prospective use of cumulative sum (CUSUM) methodology in mortality monitoring outside of cardiac surgery. MethodsAn eight-year retrospective study of mortality was performed on all admissions to a regional burns intensive care unit in the UK. Risk-adjusted CUSUM charts, variable life adjusted displays (VLADs) and zeroed VLADs were produced to track mortality against that predicted by the Belgium burns score. The same techniques were implemented prospectively for one year (76 admissions) using the Osler modification of the Baux score for risk adjustment. ResultsInternal audit would have been triggered on nine occasions using zeroed VLAD monitoring in the retrospective study. The Belgium score overpredicts mortality in the elderly.Internal audit was triggered for better than predicted outcomes on two occasions in the prospective study. DiscussionThis study describes a successful design for an early-warning system to monitor outcomes in a burns intensive care setting.

Calibration of the EuroSCORE II risk stratification model: is the Hosmer-Lemeshow test acceptable any more?

We read with great interest the article by Nashef et al .[ 1] regarding the construction and validation of the EuroSCORE II risk model. This manuscript is an outstanding contribution in the creation of a risk stratification model that is robust enough for use in cardiac surgery worldwide. However, we would like to know which statistical procedure should be used to determine calibration of the EuroSCORE II? The Hosmer–Lemeshow goodness-of-fit test has been the most popular test to validate calibration [2, 3], measuring the differences between expected and observed outcomes (mortality) over deciles (test results are acceptable with cohort divided in at least terciles) of risk. A well-calibrated model gives a corresponding P value >0.05. Recently, it has been claimed that a nonsignificant Hosmer–Lemeshow test meant that there was no evidence of bad calibration, but that this result did not mean that there was good calibration [4]. In our opinion, statistical results are either significant or nonsignificant, black or white; there are no grey results in statistics. However, if that is true, how has it come to pass that we needed more than 10 years (during which Hosmer–Lemeshow statistics was used in more than 95% of manuscripts to test the calibration of additive and logistic EuroSCORE in cardiac surgery) to find out that Hosmer– Lemeshow test is no longer valid to determine calibration, and that it should be replaced with the risk-adjusted mortality ratio, [RAMR = observed/predicted (expected) – O/E mortality], as it is now suggested by Nashef et al [1]. An O/E ratio of 1.0 means that the score predicts mortality perfectly. An O/E ratio > 1.0 means that the model underpredicts mortality [in EuroSCORE II, for a validation data set of 5553 patients, the O/E ratio was 1.058 (4.18%/3.95%)], while an O/E ratio < 1.0 means that the model overpredicts mortality. However, how are we going to check the statistical significance of the RAMR value? Conceptually, if the observed number of deaths is equal to the expected number of deaths (as predicted by the scoring system), the RAMR would have a value of 1.0. Thus, the statistical test for the significance of the RAMR is whether it is different from 1.0. To gauge the statistical significance of the RAMR, we must first calculate the 95% confidence interval for the RAMR. If the 95% confidence interval excludes the value ‘1.0’, it may be considered statistically significant (no matter whether it overpredicts or underpredicts mortality). On the contrary, Bhatti et al .[ 5] suggested χ 2 statistics

KCH, the German Preoperative Score for Isolated Coronary Artery Bypass Surgery: Is it Superior to the Logistic EuroSCORE?

The accuracy of the logistic EuroSCORE in different patient populations has been questioned. Using the German registry database, the KoronarCHirurgie (KCH) score was introduced as a preoperative risk stratification tool specifically for patients who undergo isolated coronary artery bypass surgery in Germany. However, no direct statistical comparison of this score with the well-established logistic EuroSCORE has been previously performed. The aim of this study was to validate the preoperative German KCH score and to compare it to the logistic EuroSCORE for all coronary artery bypass surgery patients as well as for on-pump and off-pump subgroups. We prospectively included all consecutive adult patients admitted to our department between January 1, 2007 and December 31, 2008, who underwent isolated coronary artery bypass surgery. The logistic EuroSCORE and the KCH-3.0 were calculated on admission to hospital. The outcome was defined as 30-day mortality. We performed calibration (Hosmer-Lemeshow test and Anderson-Grunkemeier Observed/Expected "O/E" mortality ratio) and discrimination (receiver operating characteristic "ROC" test) analyses of both scores. The accuracy of the scores was compared using DeLong's test. A total of 1461 patients (23.96 % females, mean age 66.94 ± 9.43 years) were included. The 30-day mortality rate was 2.87 %. The two models were comparable with regard to the prediction of an individual patient's risk of mortality in the whole study population and in the on-pump and off-pump subgroups (according to the ROC test and DeLong's test). Overall, there was no significant difference between observed and expected mortality according to the Hosmer-Lemeshow test ( P > 0.05). However, the KCH-3.0 was far less likely to overpredict mortality than the logistic EuroSCORE, as demonstrated by the observed mortality/expected mortality (O/E) ratios. The O/E ratio was 0.32 for the logistic EuroSCORE and 0.74 for the KCH-3.0. For the on-pump subgroup the O/E ratios were 0.37 and 0.80, respectively, and 0.24 and 0.63, respectively, for the off-pump subgroup. The KCH-3.0 is more reliable than the logistic EuroSCORE as a preoperative mortality prediction score for patients undergoing isolated coronary artery bypass surgery, providing predicted mortality rates that are closer to the actual mortality rates with a lower overprediction of mortality.

Performance of European system for cardiac operative risk evaluation in Veterans General Hospital Kaohsiung cardiac surgery

Background The European System for Cardiac Operative Risk Evaluation (EuroSCORE) model is a widely-used risk prediction algorithm for in-hospital or 30-day mortality in adult cardiac surgery patients. Recent studies indicated that EuroSCORE tends to overpredict mortality. The aim of our study is to evaluate the validity of EuroSCORE in Veterans General Hospital Kaohsiung (VGHKS) cardiac surgery including a number of different surgical and risk subgroups. Methods From January 2006 to December 2009, 1,240 adult patients who underwent cardiac surgery in VGHKS were included in this study. The study was followed the guidelines of the Ethics Committee of Kaohsiung Veterans General Hospital, Taiwan. Both additive and logistic score of all patients were calculated depending on the formula in the official EuroSCORE website. The entire cohort, different surgical type and risk stratification subgroups were analyzed. Model discrimination was tested by determining the area under receiver operating characteristic (ROC) curve. Model calibration was tested by the Hosmer–Lemeshow chi-square test. Clinical performance of model was assessed by comparing the observed and predicted mortality rates. Results There were significant differences between the VGHKS and European cardiac surgical populations. The additive score and logistic score for the overall group were 7.16% and 12.88%, respectively. Observed mortality was 10.72% overall, 5.68% for isolated coronary artery bypass grafting (CABG), 4.67% for the mitral valve only and 4.25% for the aortic valve only group. The discriminative ability EuroSCORE was very good in all and various surgical subgroups, with area under the ROC curve from 0.75 to 0.87. The addictive and logistic models of EuroSCORE showed excellent accuracy, 0.839 and 0.845, respectively. Good calibration power was recognized by p value higher than 0.05 for the entire cohort and all subgroups of patients except for isolated CABG. The logistic EuroSCORE model overestimated mortality to different degrees in the various subgroups, indicating that the logistic EuroSCORE needs to be recalibrated by a factor about 0.55 for uncomplicated surgery and low-risk groups, and 0.85 for high-risk patients with original additive score more than six. Conclusion EuroSCORE is simple and easy to use. In the present study, the model demonstrated excellent accuracy in all and various surgical subgroups in VGHKS cardiovascular surgery populations. Good calibration ability in all and different risk categories was identified except for isolated CABG group. Recalibration factors of 0.55 and 0.85 were suggested for the various operative subgroups and risk categories.

EuroSCORE Performance in Valve Surgery: A Meta-Analysis

The European System for Cardiac Operative Risk Evaluation (EuroSCORE) was developed to predict immediate outcomes after adult cardiac operations, but less than 30% of the cases used to develop this score were valve procedures. We studied EuroSCORE performance in valve procedures. We performed a meta-analysis of published studies reporting the assessment of discriminatory power of the EuroSCORE by receiver operating characteristics (ROC) curve analysis in adult valve operations. A comparison of observed and predicted mortality rates was also performed. A literature search identified 37 potentially eligible studies, and 12 were selected for meta-analysis comprising 26,621 patients with 1250 events (mortality rate, 4.7%). Meta-analysis of these studies provided an average area under the curve (AUC) value of 0.730 (95% confidence interval [CI], 0.717 to 0.743). The same results were obtained when meta-analyses were performed separately in studies categorized on reliability of uncertainty estimation: in the seven studies reporting reliable uncertainty estimation (8175 patients with 358 events; mortality rate, 4.4%), the ROC curve provided an average AUC value of 0.724 (95% CI, 0.699 to 0.749). The five studies not reporting reliable uncertainty estimation (18,446 patients with 892 events; mortality rate, 4.8%) had an average AUC of 0.732 (95% CI, 0.717 to 0.747). We documented a constant trend to overpredict mortality by EuroSCORE, both in the additive and especially in the logistic form. The EuroSCORE has low discrimination ability for valve surgery, and it sensibly overpredicts risk. Alternative risk scoring algorithms should be seriously considered.

External validation of the paediatric logistic organ dysfunction score

External validation of the paediatric logistic organ dysfunction (PELOD) score in two paediatric intensive care units (PICU) in South America. Prospective observational cohort study including all PICU admissions from July 2003 to December 2004 in Porto Alegre, Brazil, and from January 2004 to December 2004 in Buenos Aires, Argentina. The data collected included demographic variables, diagnosis, need for mechanical ventilation, length of PICU stay and mortality, and the 12 variables in the PELOD score. For each PELOD score variable, the worst daily value and the worst value of the whole PICU stay were used for the daily PELOD (dPELOD) and PELOD scores, respectively. A total of 1,476 admissions (51.3% from Argentina and 48.7% from Brazil) were analysed. Observed and predicted mortality were, respectively, 4.7% and 6.6%, with a standardized mortality ratio of 0.72. The score showed excellent discrimination capacity, with an area under the receiver operator characteristic (ROC) curve of 0.93 (0.88-0.98). The dPELOD score on days 1-5 also showed good discrimination capacities, with areas under the ROC curve >0.85. However, PELOD and dPELOD scores showed poor calibration with the Hosmer-Lemeshow test (chi-square 72.3, p < 0.001). This poor calibration was explained by a deficiency in the PELOD score where it fails to identify two risk intervals; 3.1-16.2% and 40-80%. The PELOD score is reproducible, has excellent discrimination, but over-predicts mortality and has poor calibration. Although the lack of calibration may not invalidate the score, the PELOD score is a discontinuous variable and we advise careful consideration when using it as a surrogate endpoint in clinical trials.

Surgical audit using the POSSUM scoring tool in vascular surgery patients

Currently, crude morbidity and mortality rates are used to assess vascular surgical outcome. However, as a high-risk specialty, this may not be representative of quality of care. We evaluated the POSSUM score as a measure of performance in our vascular unit, and also its usefulness on an individual patient basis for predicting outcome. A total of 106 patients were prospectively scored using the vascular-POSSUM score, and mean predicted and observed morbidity and mortality were compared with one-sample t tests. Receiver operator characteristic curves were used to determine if POSSUM scores were associated with the end-points of morbidity and mortality. Predicted and observed morbidity (41 and 35.8%, respectively) were not significantly different (P = 0.066). POSSUM did, however, over-predict mortality at 9.7% compared to the observed mortality of 5.7% (P = 0.021). While the discrimination for predicting morbidity was poor, POSSUM scores were significantly associated with mortality endpoints (area under ROC curve = 0.97250). POSSUM morbidity scores closely correlate with observed outcomes. Although POSSUM did over-predict mortality, high mortality scores may be useful for identifying patients at particularly high risk post-operatively. The POSSUM score is a useful adjunct to interpretation of morbidity and mortality statistics, and we would recommend its wider implementation for surgical audit.

Performance of EuroSCORE in CABG and off-pump coronary artery bypass grafting: single institution experience and meta-analysis

To assess EuroSCORE performance in predicting in-hospital mortality in on-pump coronary artery bypass grafting (CABG) and off-pump coronary artery bypass grafting (OPCAB). Additive and logistic EuroSCORE were computed for consecutive patients undergoing CABG (n = 3440, 75%) or OPCAB (n = 1140, 25%) at our hospital from 1999 to September 2007. The areas under the receiver operating characteristic (ROC) curves (AUCs) were used to describe performance and accuracy. No difference in performance between CABG and OPCAB and between additive and logistic EuroSCORE (additive EuroSCORE AUCs of 0.808 and 0.779 for CABG and OPCAB, respectively; logistic EuroSCORE AUCs of 0.813 and of 0.773 for CABG and OPCAB, respectively) was found, although a marked tendency to overpredict mortality by both models was evident. A meta-analysis of previously published data was done, and a total of eight studies representing 19 212 and 5461 patients undergoing CABG and OPCAB, respectively, met inclusion criteria. Meta-analysis confirmed similar performance of EuroSCORE in CABG and OPCAB: estimated AUCs were 0.767 and 0.766 for CABG and OPCAB, respectively, with an estimated difference of 0.001 (95% CI -0.061 to 0.063). Additive and logistic EuroSCORE algorithms performed similarly, and cumulative evidence suggests comparable performance in CABG and OPCAB procedures; both risk models, however, significantly overestimated mortality.

COMPARISON OF THE POSSUM, P‐POSSUM AND CR‐POSSUM SCORING SYSTEMS AS PREDICTORS OF POSTOPERATIVE MORTALITY IN PATIENTS UNDERGOING MAJOR COLORECTAL SURGERY

Physiologic and operative severity score for the enumeration of mortality and morbidity (POSSUM), 'Portsmouth'-physiologic and operative severity score for the enumeration of mortality and morbidity (P-POSSUM) and 'Colorectal'-physiologic and operative severity score for the enumeration of mortality and morbidity (Cr-POSSUM) are three related scoring systems, which uses individual patient parameters to predict postoperative mortality. POSSUM overpredicts mortality in low-risk patients and underpredicts mortality in elderly and emergency patients. P-POSSUM was developed to compensate for these weaknesses. Cr-POSSUM was developed specifically for colorectal surgery. We aim to establish which of these scoring systems would be most useful in an Australasian context. Data were collected for 308 patients and predicted mortality risk values were generated using each of the three systems. The Mann-Whitney U-test was then carried out on the scores for each system. Receiver-operator characteristic curves were designed to determine the relative accuracy of each approach at discriminating between death and survival. All three POSSUM scoring systems showed a statistically significant ability to predict postoperative mortality. Additionally, in each system there was a significant difference in the raw physiologic and operative severity scores between survivors and those who died. A risk-stratification model was applied to each set of data, showing a correlation between an increase in risk and an increase in mortality rate. Finally, the receiver-operator characteristic curves generated showed that in this study group POSSUM, P-POSSUM and Cr-POSSUM were all satisfactory predictive tools although the latter tended to be relatively less accurate. Physiologic and operative severity score for the enumeration of mortality and morbidity, P-POSSUM and Cr-POSSUM are all reliable predictors of postoperative mortality in the Australasian context; although there was a trend towards POSSUM and P-POSSUM being better predictors than Cr-POSSUM. However, Cr-POSSUM requires fewer individual patient parameters to be calculated and is thus easier to generate. An ideal preoperative scoring system remains to be developed for predicting mortality in patients undergoing colorectal surgery.

POSSUM scoring for laparoscopic cholecystectomy in the elderly

Physiological and operative severity score for the enumeration of mortality and morbidity (POSSUM) scoring is a validated scoring system in the audit of surgical outcomes; however, evaluation of this system has mostly been applied to open surgical techniques. The present study examines the validity of POSSUM in predicting morbidity and mortality in patients undergoing laparoscopic cholecystectomy (LC) with the recognized risk factor for postoperative mortality of advanced age. All patients aged 80 years or over undergoing LC in one surgical unit between January 1993 and December 1999 were identified from the surgical operations database of the hospital. Case-note review was used to collate data in terms of clinical and operative factors as described in POSSUM. Observed/POSSUM estimated (O/E) ratio of morbidity and 30-day mortality were calculated. Laparoscopic cholecystectomy was performed in 76 patients aged 80 years or over during the study period. Of these patients, case notes for 70 patients (92%) were available for review. Median (range) age was 83 years (80-93 years) and median (range) American Society of Anesthesiologists score was 2 (2-4). Twenty-six (34%) of 70 patients underwent cholecystectomy during an acute admission. The mean physiology severity score was 23 and operative severity score, 8. A significant postoperative morbidity was observed in 15 (22%) of 70 patients. There was no 30-day mortality. Using exponential analysis, POSSUM predicted morbidity in 15 patients and mortality in seven patients. Thus, O/E ratios for morbidity and mortality were 1 and 0, respectively. POSSUM scoring performs well in predicting morbidity, but overpredicts mortality, after LC in patients aged over 80 years. An assessment of its application to other laparoscopic procedures merits evaluation.

POSSUM scoring for patients with fractured neck of femur

POSSUM scoring is validated as an audit tool in general and orthopaedic surgery. It is also used for preoperative triage to assess perioperative risk. However its ability to predict mortality in specific surgical subgroups, such as patients with fractured neck of the femur, has not been studied. This study assessed the predictive capability of POSSUM for 30-day mortality after surgery for fractured neck of femur. A cohort study was conducted in Queen's Medical Centre, Nottingham over a period of nearly 2 yr. Complete data from 1164 patients were analysed to compare the mortality predicted by POSSUM and the observed mortality. POSSUM risk of death was calculated using the original POSSUM equation, with modifications to the operative score appropriate for orthopaedic surgery. POSSUM predicted 181 (15.6%) deaths and the observed mortality was 119 (10.2%). The area under the receiver operating characteristic curve was 0.62, indicating poor performance by the POSSUM equation. POSSUM overpredicts mortality in hip fracture patients. It should be used with caution whether as an audit tool or for preoperative triage.

A carotid POSSUM model for predicting death

Abstract Background Risk adjustment to allow for differences in case mix aids comparative audit of carotid endarterectomy (CEA). The aim of this study was to test the predictive accuracy of P-POSSUM (Physiological and Operative Severity Score for the enUmeration of Mortality and morbidity – Portsmouth predictor equation for mortality) and V-POSSUM (P-POSSUM-based model for vascular surgery) and to develop a procedure-specific model for CEA. Methods A total of 709 CEAs performed in two vascular units from 1992 to 2000 were analysed. The physiological and the operative severity scores were calculated; the outcome used was 30-day mortality. Predicted mortality was calculated using P-POSSUM, V-POSSUM and V-POSSUM (physiology only = PS) equations. The observed and the predicted mortality were compared. The new model was developed on a set of 350 CEAs and validated on 359 CEAs. The data sets of the two units were combined and split into two by the procedure date order. The predictive accuracy of the model was analysed with the χ2 test. Results P-POSSUM and V-POSSUM models showed lack of fit in predicting death. There was no evidence of any lack of fit in the new model. Conclusion P-POSSUM and V-POSSUM overpredict mortality and are unsuitable for the comparative audit of CEA. Procedure-specific models based on P-POSSUM can accurately predict death.

POSSUM and P-POSSUM Overpredict Mortality for Carotid Endarterectomy

Objectives: the aim was to test the predictive accuracy of POSSUM and P-POSSUM on patients undergoing CEA. Design: retrospective and prospective study. Materials: 499 CEAs performed by four vascular surgeons from a single unit from 1992–99. The median age was 68 (range 38–86) and 60% were men.Methods : physiological parameters, operative parameters and the 30-day mortality were collected. Predicted mortality for each patient was calculated using POSSUM and P-POSSUM equations. Patients were stratified into risk groups and observed and predicted deaths were compared. Accuracy of the prediction was assessed using chi-squared analysis. Results: the observed 30-day mortality was 1.8% (9/499). The predicted deaths using POSSUM and P-POSSUM analysis were 49 and 25 respectively compared to nine observed deaths. There was significant evidence of lack of fit of both models in predicting mortality (chi-squared analysis, p<0.05). Conclusions: POSSUM and P-POSSUM overpredict mortality and are unsuitable for comparative audit of CEA. Models developed specific for CEA might accurately predict mortality.