Article Text
PDF
Abstract
Objective The Emergency Surgery Score (ESS) is a predictive tool used to assess morbidity and mortality rates in patients undergoing emergent surgery. This study explores the ESS’s predictive ability and reliability in the Jordanian surgical population.
Design A retrospective validation study.
Setting A tertiary hospital in Jordan.
Participants A database was created including patients who underwent emergent surgery in King Abdullah University Hospital from January 2017 to June 2021.
Primary and secondary outcome measures Relevant preoperative, intraoperative and postoperative variables were retrospectively and systematically gathered, and the ESS was calculated for each patient accordingly. In addition, a multivariable logistic regression analysis was performed to assess the correlations between the ESS and postoperative mortality and morbidity along with intensive care unit (ICU) admissions.
Results Out of total of 1452 patients evaluated, 1322 patients were enrolled based on inclusion and exclusion criteria. The mean age of the population was 47.9 years old. 91.9% of the patients were admitted to the surgical ward through the emergency department, while the rest were referred from inpatient and outpatient facilities. The mortality and postoperative complication rates were 3.9% and 13.5%, respectively. Mortality rates increased as the ESS score gradually increased, and the ESS was evaluated as a strong predictor with a c-statistic value of 0.842 (95% CI 0.743 to 0.896). The postoperative complication and ICU admission rate also increased with reciprocal rises in the ESS. They were also evaluated as accurate predictors with a c-statistic value of 0.724 (95% CI 0.682 to 0.765) and a c-statistic value of 0.825 (95% CI 0.784 to 0.866), respectively.
Conclusion The ESS is a robust, accurate predictor of postoperative mortality and morbidity of emergency general surgery patients. Furthermore, it is an all-important tool to enhance emergency general surgery practices, in terms of mitigating risk, quality of care measures and patient counselling.
- accident & emergency medicine
- surgery
- adult surgery
Data availability statement
Data are available upon reasonable request. The database created and analysed during this study is not publicly available as per patient privacy regulations of our hospital but is available from the corresponding author on reasonable request after blinding patient identifying information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Strengths and limitations of this study
An external validation of the Emergency Surgery Score in Jordan.
The data sample is well distributed in terms of patients’ demographic characteristics and surgical approach.
An insight into the nature of emergency general surgery in a hospital that receives referrals from the region and whole of Jordan.
The retrospective nature of the data collected for the study is a limitation.
Our definition of emergency surgery and practice protocols may differ from other healthcare providing establishment given the nature of resources available.
Introduction
Emergency general surgery (EGS) includes an assortment of pathologies linked by their urgency and accounts for an increasingly large share of overall patient populations.1 Patients requiring EGS have a substantial rise in 30-day mortality and postoperative complication rates when compared with elective surgery patients.2 Such upsurge has been credited to clinical and patient factors at the time of surgery such as age, comorbidities and acute physiologic disturbances. Thirty-nine per cent of EGS patients are probably going to die within 30 days postoperation, with EGS being found to be an independent risk factor.3 With the increase in EGS, public health and financial repercussions may result, highlighting its burden on healthcare systems.4 5 The Acute Care Surgery (ACS) model has been created to benefit safe and efficient care of trauma and EGS, with an initiative joining trauma, critical care and EGS demonstrating better results in the management of appendicitis and biliary disease.6 7 The model’s success relies on guidelines, benchmarks and quality improvement processes. Nevertheless, EGS is deficient in benchmarking and quality enhancement, and in light of its healthcare burden, there is a need for standardisation of patient care.8
It is important to be able to predict postoperative results in EGS patients in order to facilitate proper bedside decision-making and optimise postoperative care. The Emergency Surgery Score (ESS) has been derived to predict mortality and postoperative complications after EGS, calculated by summing specific points to 22 distinct demographic, comorbidity and preoperative laboratory variables for a maximum of 29 points.9 10 The ESS has been retrospectively validated that has confirmed its efficacy as a pre-operative risk score.11–13 The score has also been reported to accurately predict the development of postoperative infectious complications and can be used as a preoperative tool to assess the suitable level of care after surgery.14 15 Moreover, it has been ratified prospectively, reinforcing its role as a bedside tool to help surgeons counsel patients and their families, and as an intensive care unit (ICU) triaging tool if critical care capacity is limited.16 As ESS is specific to EGS, the previously mentioned studies help to verify its use as a tool to benchmark the quality of service and care.13–16
External score validations are extremely valuable to assess differentiation and standardisation in other populations, as well as increase global awareness of applicable predictive models.17 To date, only a couple of such validations exist, focusing on different domains within the ESS predictive capability.12 18
In low and middle-income countries, surgical emergencies define most emergencies and have been estimated to cause a tremendous health burden.19 Variance in access to surgical care and a growing volume of unmet surgical needs delineate the broad difference between low and high-income countries. In Jordan, emergency surgery is actively stressing our capacity to care for our patients and are adding to the financial load in our healthcare system. As the COVID-19 pandemic plays an active and disruptive role in the change in current surgical practice around the world, the effect is prominent in Jordan.20 Validating the ESS in our population is a step towards improving bedside decision-making and postoperative care, as well as a tool to enhance triaging EGS patients and help curtail the onus of emergent surgery. Our study will allow us to estimate the postoperative mortality and complication rates, in addition to assess the ability to predict them through ESS.
Methods
A database was created using the electronic medical record system at King Abdullah University Hospital (KAUH), a tertiary hospital serving northern regions in Jordan. We identified all patients that were admitted to the general surgery ward for emergency surgery from January 2017 to June 2021. All adult patients 16 years of age and older presenting to the ER for general surgery and operated on were included. This includes patients operated on for thoracic, hepatobiliary, upper gastrointestinal, colorectal, vascular, breast and endocrine, trauma and soft tissue pathologies. Patients under 16 years old, pregnant, patients that we treated conservatively, patients that voluntarily refused treatment against medical advice, patients referred from clinic visits or admitted electively were excluded from the study. In addition, an emergent surgery was defined in line with the ACS-NSQIP (Amercian College of Surgeons- National Surgical Quality Improvement Program) definition as an emergency case that is ‘performed within a short interval of time between patient diagnosis or the onset of related preoperative symptomatology’.21
We calculated the ESS for each patient as described by Sanji et al9 (table 1). For each patient, demographic, preoperative laboratory and comorbidities were retrieved from the electronic medical system. Data collection included the following preoperative variables: age, nationality, whether the patient was transferred from an outside emergency department or transferred from a primary hospital inpatient facility, body mass index (BMI), history of disseminated cancer, hypertension, diabetes mellitus, steroid use, ascites, dyspnoea, ascites, history of chronic obstructive pulmonary disease (COPD), functional dependence, ventilator requirement within 48 hours preoperatively, weight loss in preceding 6 months. In addition, we also recorded the following lab values: albumin, alkaline phosphatase (ALP), blood urea nitrogen, creatinine, international normalised ratio (INR), platelets, aspartate transaminase (AST; SGOT (spartate aminotransferase)), sodium, white blood count. Other variables included diagnosis leading to the admission, the incidence of postoperative complications, the need for ICU admission, and hospital length of stay if the patient was reoperated for the same cause, and outcome of the admission including whether the surgical intervention led to resolution of the acute condition.
Emergency Surgery Score (ESS) variables
All included patients identified as Jordanian and thereby considered as non-white and received a corresponding score of zero. For any missing variable, the null score value was used, and zero points were given, as previously described by Naar et al, in which the predictive capability of the ESS was not affected by the inclusion of patients with any missing data.22 The lowest capture rate per variable was albumin, INR/prothrombin time (INR/PT), ALP and AST (41.53%, 60.28%, 41.81%, 41.25%, respectively).
Our validation aimed to study the score’s ability to predict mortality, the incidence of any complication after surgery, admission to ICUs, and whether the patient was reoperated for the same admitting diagnosis. The aforementioned postoperative outcomes were observed over a 30-day period beginning at the time of surgery.
All the variables were recorded and presented as frequencies and percentages. The Shapiro-Wilk test was used to assess the distribution of the data across the age groups. The score validity as a predictor of mortality, postoperative complication and ICU admission was evaluated by performing multivariant logistic regression that included all ESS variables. After determining mortality and complication rate per score, the receiver operator characteristic and area under the curve was calculated and compared with the American Society of Anesthesiology physical status classification system (ASA), which was previously determined for each patient as it is a part of the preoperative assessment protocol at our facility. The statistical analysis was two-tailed, and the significance threshold was set at 0.05 or less. The IBM SPSS Statistics V.26 was used, and the study was ethically approved by the institutional review board before the data collection commenced, waiving the need for patient consent while maintaining patient privacy
Patient and public involvement
None
Results
A total of 1452 patients were admitted to the surgery ward in our institution for emergent surgery, and after excluding patients that did not meet our inclusion criteria, 1322 patients were enrolled. 52.6% of the patients were identified as men and 47.4% as women. The mean age of the population is 47.9 years old, with the youngest being 16 years old and the oldest patient admitted being 97 years old. 24.3% of the patients were over 60 years old. The distribution of the population based on age was assessed and was well distributed in terms of gender (P value:0.968), ASA score (P value: 0.558), surgical approach (P value: 0.094), mortality (P value: 0.085) and postoperative complication rate (0.200) and ICU admission (P value: 0.578). A vast majority of the patients were admitted through our emergency department, accounting for 91.9% of included patients, followed by 6.8% of patients who were transferred from other hospitals, and 17 patients (1.3%) who were transferred to the surgery ward from other inpatient departments within KAUH. Less than half of the population underwent laparoscopic operations, and complication (95% CI: 0.793 to 0.858; P value: 0.0001) and mortality rates (95% CI: 0.922 to 0.959; P value: 0.001) were significantly lower compared with patients who underwent open surgical techniques or laparotomies (table 2); 27.2% of the population had been previously diagnosed with hypertension, and 20.7% had diabetes, while 5.1% had diagnosed malignancies.
Patient demographics
More than half of the population were evaluated and given an ASA score of one, 24.1% had a score of two, 16.8% had a score of three and 7.4% of the population had a score of four or more. The diagnoses leading to admissions are described in table 3.
Patient outcome data
The most common diagnoses were acute cholecystitis (26.2%), diabetic foot infection (8.2%) and obstructive jaundice (7.4%). The average length of hospital stay was 7 days; a median of 5 days (IQR. 3–9 days), with 36.2% of the patients spending less than 3 days, and 26.9% of patients spending more than 9 days. In addition, 93 (7%) patients were admitted to the ICU postoperatively, and 115 (8.7%) patients were reoperated for the same admitting diagnosis, or a complication of the surgery performed within 30 days of their first encounter (table 4).
Emergency surgery outcomes, length of hospital stay, postoperative complications
3.9% of the population were pronounced dead within 30 days of the emergent surgery, and 13.5% of the population experienced postoperative complications. The most prevalent postoperative complication was superficial surgical wound infections (32 patients; 2.4%), followed by deep surgical wound infections (20 patients; 1.5%) and bleeding leading to blood transfusions postoperatively (17 patients;1.3%) . Mortality and complication rates increased when older age groups were compared with younger ones, as shown in table 2 (figure 1).
Mortality and morbidity according to age.
There was a gradual increase in laparotomies and open surgery techniques as the ESS increased. Patients with scores below three had relatively higher utilisation of minimally invasive techniques compared with higher ESS score points, and patients with scores greater than 11 all underwent open versus laparoscopic surgeries. (figure 2) Higher mortality and morbidity rates were associated with specific ESS variables, for instance, patients that presented with ascites, history of ascites, functional dependence, high INR, or elevated creatinine (table 5). It was also noticed that higher mortality and morbidity rates were found in patients transferred from other facilities.
ESS variables with respect to death rate and complications
Surgical approach comparison per ESS. ESS, Emergency Surgery Score.
Mortality rates increased as the ESS score increased (table 6). We observed that patients who scored two or less points had a mortality rate of 0.7%, while patients who scored more than eleven had a mortality rate 57%. The ESS was evaluated as a predictor of 30-day mortality based on a c-statistic of 0.842 (95% CI: 0.743 to 0.896) compared with ASA grade with a c-statistic of 0.670 (95% CI: 0.597 to 0.743) (figure 3).
Regression result of Emergency Surgery Score (ESS)
(A) Postoperative death according to ESS, (B) comparison of receiver operator curves (ROC) between the ESS and ASA regarding postoperative death. ASA, American Society of Anesthesiology; ESS, Emergency Surgery Score.
Complication rates also increased with reciprocal raises in ESS points scored. It is noted that patients who scored less than three points had a complications rate of 8.3%, while patients that scored more than ten points had a complication rate of 83.3% (figure 4A,B). The ESS had a c statistic value of 0.724 (95% CI: 0.682 to 0.765), and the ASA had 0.670 (95% CI: 0.597 to 0.743) We also compared the ability of the ESS to predict specific complications and found that the ESS is a strong predictor of sepsis; C-statistic value of 0.928 (95% CI:0.876 to 0.981), cardiac arrhythmia; C-statistic value of 0.876 (95% CI: 0.765 to 0.988) and cardiac arrest; C-statistic value of 0.859 (95% CI: 0.797 to 0.920).
(A) Postoperative complications according to ESS, (B) comparison of receiver operator curves (ROC) between the ESS and ASA regarding postoperative complication. ASA, American Society of Anesthesiology; ESS, Emergency Surgery Score.
An increase in ICU admission was observed as the points scored on the ESS increased. The ICU admission rate for patients who scored below three was 2.4%, while for patients who scored higher than eight, the admission rate was 43.4% (figure 5A,B). The ESS was also evaluated as a predictor of ICU admission postoperatively with a C-statistic value of 0.825 (95% CI: 0.784 to 0.866) compared with ASA with C-statistic value of 0.703 (95% CI: 0.647 to 0.759). It is also noted that the frequency of patients who were reoperated on within 30-days increased as their respective scores increased. Patients who scored three points or less had a reoperation rate of 6.5%, while patients who scored between four and nine had reoperation rates of 13.5% compared with patients who scored more than 10 who had a reoperation rate of 16.7%. In addition, the patient’s length of hospital stay increased as the ESS score increased. The majority of patients who scored zero and one spent less than 3 days in the hospital, while a majority of patients who scored more than nine points were admitted for more than 9 days (figure 6).
(A) Postoperative intensive care unit (ICU) admission per ESS, (B) comparison of receiver operator curves (ROC) between the ESS and ASA regarding postoperative ICU admission. ASA, American Society of Anesthesiology; ESS, Emergency Surgery Score.
Duration of hospital stay comparison per ESS. ESS, Emergency Surgery Score.
Discussion
The results of this study show that the ESS has a high predictive value for EGS 30-day mortality, morbidity, need for ICU admission and risk of postoperative complications. These results appear to corroborate the findings of previous studies that have attempted to validate the ESS. The overall trend being that a higher ESS correlates to a higher mortality rate, this risk becomes critical in our study, with an ESS score of seven or higher, where the mortality rate for an ESS between 7 and 10 and an ESS of 11 or higher are 21.6% and 57%, respectively. The incidence of ICU admissions correlated with the ESS similarly to the mortality rate, as a significant cut-off was found in our study at an ESS of eight, with ICU admissions for patients with an ESS less than eight and an ESS of eight or higher being 6.8% and 43.4%, respectively. These findings are congruent with the findings in the study by Kongkaewpaisan et al, where they proposed a cut-off ESS of seven or higher as grounds for ICU admission, as they found that only 6.2% of patients with ESS of seven or less had an ICU need, while 58.2% of patients with an ESS higher than seven had an ICU admission.15
It is important to note, however, that the ESS has a prognostic value even in low-risk patients with lower ESS values, as it was found to accurately predict the risk of re-operation, increased length of hospital stay and postoperative complications. The 30-day reoperation rate for patients with an ESS above three was double that of those with an ESS less than three. The length of hospital stay was also significantly longer for patients with an ESS of only two to four compared with those with an ESS of zero to one. The study by Alburakan et al found that the length of hospital stay was expected to increase by 2.7 days for each increase in ESS value of one.12 We also found that the rate of complications for patients with an ESS of four to nine was also more than three times that of patients with an ESS of less than three. These findings are in line with the results in the study by Han et al, where postoperative infections developed in 7% and 24% of patients with an ESS of one and five, respectively.14 This provides an added benefit for both the physician and patient, as it shows that even low-risk patients with low mortality risk and no need for ICU admission can be counselled objectively regarding their expected clinical course based on their ESS score.
Interestingly, we found that EGS procedures done for patients with an ESS of three or higher were predominantly done with a laparotomy incision, while EGS procedures for patients with an ESS of zero to two were well distributed between both approaches with a preponderance towards laparoscopy. While these findings are merely correlations, these results indicate that it may be possible to use the ESS to decide which EGS patients are fit for the use of laparoscopic technique, allowing the surgeon to avoid unnecessary laparotomies in a subgroup of EGS cases where minimally invasive alternatives may be used. Still, further evidence from future studies is needed to support or disprove this observation. The use of Laparoscopic techniques in EGS, when appropriate, leads to decreased postoperative pain, earlier mobilisation and reduces the need for negative laparotomies,23 while also providing effective management in EGS cases that are hemodynamically stable and not in septic or haemorrhagic shock.24
We also compared the ESS to the ASA and found that it performed better than the ASA score in all parameters analysed. The ESS was found to be a strong predictor of 30-day mortality, morbidity and need for ICU admission. This is not to say that the ASA score is not predictive of patient outcomes, as our results clearly show that it is. Rather, our findings show the superiority of the ESS as a predictive tool when used in the setting of EGS. This may be due to the subjective nature of the ASA Score parameters, while in contrast the ESS takes more objective factors into account.
Compared with non-emergency general surgery, EGS has been shown to have a higher mortality rate, complication rate, and length of hospital stay.3 The cost associated with emergency surgery is also more significant due to the more extended average hospital stay of patients, the need for ICU admission and the increased cost of management.25 The increased cost can often be a hindrance to timely patient care, especially in underprivileged communities where the increased cost might result in the need for transfers to other facilities with better resources. This shows the critical nature of emergency surgery and the importance of efficient triaging to designate priority in terms of surgical interventions and the allocation of ICU beds, especially in underserved hospitals. The ESS being able to accurately predict the incidence of 30-day postoperative complications and the need for ICU admissions, in addition to the mortality rates, supports the validity of the ESS as an accessible and relevant tool for risk assessment in acute EGS settings. Furthermore, using the ESS may provide a more time-efficient manner by which surgical intervention priorities are assessed, and as a result, improved overall patient outcomes are achieved. In Jordan, studies on the availability of CT imaging according to the region and hospital are not available, but to our knowledge, there is a lack of medical imaging services required to meet the demand of emergent medical and surgical needs. Therefore, we believe the ESS will also prove invaluable to smaller centres in underprivileged populations who may not have timely access to accurate imaging services, thereby improving overall outcomes.
Few studies have attempted to validate the ESS outside of the USA, this one being the first to provide insight on EGS in Jordan. The results of this study support the validity of the ESS in the Jordanian population and further underscores the benefit of the ESS as a tool in EGS. KAUH is one of a small number of tertiary centres in Jordan and receives referrals from the whole region. This data sample is also well distributed between male and female patients, as well as between both laparoscopic and open laparotomy approaches. Therefore, the data in this study are relatively representative of the outcome of using the ESS in the general population.
This study has a few limitations. First, the data for this study were collected retrospectively. Second, missing values were equated instead of excluding the patients from the data set. This was done to avoid excluding otherwise eligible patients because of missing values that were not collected due to them being low-risk patients and not requiring all of the investigations included in the ESS, thereby increasing the power of the study and avoiding bias. Thirdly, the data set collected had a small number of patients with an ESS above 11; therefore, findings in that subgroup of two patients in this study need further corroboration with larger cohorts to establish confidence in the generalisability of these findings. Fourthly, the heterogeneity of the diagnoses included in the study shows the extensive utility of the ESS. However, this heterogeneity might affect the perceived significance of the statistical results. Finally, our definition of emergency surgery and its protocols might differ from that of previous studies or what is practiced in other countries.
Conclusion
The ESS outperformed the ASA score in the emergency surgery setting and was able to accurately predict the 30-day mortality rate, need for ICU admission, 30-day reoperation rate and length of hospital stay in a Jordanian population. The ESS, therefore, appears to be a valuable and accessible tool for the preoperative assessment of EGS outcomes at the bedside.
Data availability statement
Data are available upon reasonable request. The database created and analysed during this study is not publicly available as per patient privacy regulations of our hospital but is available from the corresponding author on reasonable request after blinding patient identifying information.
Ethics statements
Patient consent for publication
Ethics approval
The Institutional Research Board (IRB) at King Abdullah University Hospital and Jordan University of Science and Technology approved this study (decision reference number: 224-2021).The data were collected retrospectively and can not be traced to any patient as it has been completely de-identified and thereby informed consent was not required.
References
Footnotes
Contributors LR, SAA, TSM, KA were responsible for the conceptualisation and design of the study. LR, TSM, KA were responsible for project administration. SAA, NBA, MAAH, HAA, FSA did data collection. SAA was responsible for data analysis, tables’ and figures’ creation. SAA and NBA wrote and drafted the initial manuscript. LR, TSM, KA, GC and FI critically read the manuscript. All authors read, edited, proofread and approved the final manuscript before submission. LR and SAA both equally share the responsibility of a guarantor.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.