Glasgow aneurysm score in predicting outcome after ruptured abdominal aortic aneurysm

Abstract

Objective

To assess the ability of Glasgow Aneurysm Score in predicting postoperative mortality for ruptured aortic aneurysm which may assist in decision making regarding the open surgical repair of an individual patient.

Methods

A total of 121 patients diagnosed of ruptured abdominal aortic aneurysm who underwent open surgery in our hospital between 1999 and 2013 were included. The Glasgow Aneurysm Score for each patient was graded according to the Glasgow Aneurysm Score (Glasgow Aneurysm Score = age in years + 17 for shock + 7 for myocardial disease + 10 for cerebrovascular disease + 14 for renal disease). The groups were divided as Group 1 (containing the patients who died) and Group 2 (the patients who were discharged). The Glasgow Aneurysm Scores amongst the groups were compared.

Results

Out of 121 patients, 108 (89.3%) were males and 13 (10.7%) were females. The in-hospital mortality was 48 patients (39.7%). The Glasgow Aneurysm Score was 84.15 ± 15.94 in Group 1 and 75.14 ± 14.67 in Group 2 which revealed significance (p = 0.002). The most appropriate cut-off value for Glasgow Aneurysm Score was determined as 78.5 (AUC = 0.669, p = 0.002, sensitivity: 64.6%, specificity: 60.3%). Glasgow Aneurysm Score value above 78.5 is associated with almost threefold increase in mortality (p = 0.007, OR:2.76, 95% CI 1.30–5.89). In further logistic regression models, Glasgow Aneurysm Score value and preoperative hematocrit values were found to be independent predictors for mortality (p = 0.023 and p = 0.007, respectively).

Conclusion

Glasgow Aneurysm Score may have a predictive value for outcome of patients with ruptured abdominal aortic aneurysm undergoing open surgical procedure and it appears to be a useful tool in clinical decision-making of an individual patient when integrated with clinical experience.

Keywords

Risk assessment ruptured aortic aneurysm surgery mortality

Introduction

Abdominal aortic aneurysm (AAA) is a common and life-threatening disease affecting 5–9% of the population over the age of 65 years.¹ Most patients with AAA are asymptomatic unless they develop a complication. Rupture, the most feared complication of AAA, is the one of the most fatal surgical emergencies.^2,3

Open abdominal repair (OAR) is the surgical approach to replace the aneurysmal part of the abdominal aorta using a graft which can be performed with a mortality rate of 2–5% for elective surgery in low-risk patients, whereas the mortality rate of ruptured aneurysms remains remarkably high.¹ The mortality rates of patients arriving at the hospital range from 32% to 70%, and mortality rates in some studies increased to 90% if patients who died in transit or at home were included.^4,5

Scoring systems may be helpful in predicting the outcome of surgical treatment for AAAs. Glasgow Aneurysm Score (GAS) is one of the most used scoring systems derived from multi-variate analysis.⁶ An accurate prediction of outcome could be made using the scoring systems while assessing the patient's fitness for surgery. Such a score would also help even when communicating with the most unfit patients and their relatives about the high risk of surgery.⁷ We analyzed the performance of GAS in predicting the 30-day mortality in ruptured AAA patients who had undergone OAR in our hospital.

Methods

A total of 121 patients diagnosed of ruptured AAA who had undergone OAR in our hospital between 1999 and 2013 were included. This study complies with the Declaration of Helsinki, and ethical approval was granted by the local institutional review board. Informed consent was obtained from all patients.

The GAS for each patient was graded according to the GAS (GAS = age in years + 17 for shock + 7 for myocardial disease + 10 for cerebrovascular disease + 14 for renal disease). Myocardial disease included previously documented myocardial infarction or ongoing angina pectoris or both. Cerebrovascular disease included stroke and transient ischemic attack. Renal disease was defined as a serum creatine level >150 mmol/L or a urea >20 mmol/L or a history of acute or chronic renal failure or both. The GAS was calculated by summing up the elements.

The groups were divided as Group 1 containing the patients who died and Group 2 containing the patients who were discharged. The GASs amongst the groups were compared.

Statistical analysis

Continuous variables were tested for normal distribution by Kolmogorov–Smirnov test. Normally distributed continuous variables were expressed as ‘mean values ± standard deviation (SD)’ or median values with the interquartile range if not normally distributed. Categorical variables were expressed as numbers and percentages. Demographic characteristics, perioperative variables and calculated values were compared using “independent samples t-test” or “Mann–Whitney-U test” for continuous variables and “chi-square test” or “Fisher’s exact test” for categorical variables. Receiver operating characteristic curve analysis was used to determine the optimum cut-off value for the GAS. A p value <0.05 was considered statistically significant. All statistical analyses were performed using the SPSS statistical software (SPSS for Windows 15.0, Inc., Chicago, IL, USA).

Results

A total of 121 patients, diagnosed of ruptured AAA who had undergone OAR were involved in the study. Group 1 (containing the patients who died) and Group 2 (the patients who were discharged) were comparable by means of demographic data (Table 1).

Table 1.

Demographic characteristics of the patients.

Features	Group 1, in-hospital mortality (n = 48)	Group 2, alive (n = 73)	p
Age (years)	70.8 ± 10.5	67.7 ± 10.5	0.122
Male gender	43/48	65/73	0.925
Hypertension	97.9%	97.3%	0.654
Diabetes mellitus	31.3%	23.3%	0.331
Smoking	45.8%	53.4%	0.414
Chronic lung disease	16.7%	23.3%	0.379
Peripheral arterial disease	25.0%	26.0%	0.899
Cerebrovascular disease	18.2%	23.3%	0.724
Coronary artery disease	20.8%	32.9%	0.156
Chronic renal failure	16.7%	13.7%	0.887
GAS	84.15 ± 15.94	75.14 ± 14.67	0.002
Urea (mg/dL)	63.5 ± 36.4	58.0 ± 29.9	0.382
Creatinine (mg/dL)	1.44 ± 0.77	2.20 ± 7.47	0.511
Hematocrit (%)	30.7 ± 7.5	35.5 ± 8.0	0.003

A total of 94 patients (77.7%) had retroperitoneal rupture whilst 27 (22.3%) had intraperitoneal rupture. Four patients had renal vein injury during surgery; two in Group 1 and two in Group 2. Acute renal failure occured in 15 patients (12.39%) postoperatively. Operative time was 223.50 ± 105.44 in Group 1 and 226.23 ± 79.13 in Group 2 (p = 0.880) and cross-clamp time was 66.63 ± 50.66 in Group 1 and 65.23 ± 40.51 in Group 2 (p = 0.876).

The in-hospital mortality was 48 (39.7%) patients. The GAS was 84.15 ± 15.94 in Group 1 and 75.14 ± 14.67 in Group 2 which revealed significance (p = 0.002). The other significant differences among the groups were; duration of ventilation (p = 0.001), number of erythrocyte suspension transfused (p = 0.001), number of fresh frozen plasma transfused (p = 0.002), intensive care unit stay (p = 0.011), hospital stay (p = 0.001) revealing an increase in all of these parameters in Group 1 in comparison to Group 2 (Table 2).

Table 2.

Postoperative data of the patients.

Features	Group 1, in-hospital mortality (n = 48)	Group 2, alive (n = 73)	p
Duration of ventilation (h)	158 ± 204	22 ± 46	0.001
Number of erythrocyte suspension transfused (units)	8.0 ± 7.2	2.9 ± 4.3	0.001
Number of fresh frozen plasma transfused (units)	14.7 ± 14.6	5.1 ± 7.0	0.002
Intensive care unit stay (days)	9.9 ± 11.4	7.3 ± 19.0	0.011
Hospital stay (days)	9.5 ± 10.8	17.3 ±2 27.1	0.001

The area under the ROC curve was 0.669 (Figure 1). Therefore, for 64.6% sensitivity rate, the cut-off value for GAS is 78.5 with a 60.3% specificity rate (p = 0.002). The mortality in the patients with a GAS value below 78.5 was 17 patients (27.9%) and in the patients with a GAS value above 78.5 was 31 patients (51.7%). GAS value above 78.5 is associated with almost threefold increase in mortality (p = 0.007, OR:2.76, 95% CI 1.30–5.89).

Figure 1.

ROC curve analysis (AUC = 0.669, p = 0.002, sensitivity 64.6%, specificity 60.3% for cut-off value of 78.5).

Logistic regression models revealed that the GAS value was an independent predictor for mortality in ruptured AAA (OR:1.04, 95% CI: 1.01–1.07, p = 0.023). Furthermore, preoperative hematocrit values were also found to be an independent risk factor for mortality (OR: 0.93, 95% CI: 0.88–0.98, p = 0.007).

Discussion

Scoring systems may be helpful in predicting the outcome of surgical treatment for AAAs. In this study, we assessed the performance of GAS in predicting the 30-day mortality in ruptured AAA patients who had undergone OAR in our hospital.

Our study revealed that the GAS had predictive value for in-hospital mortality in ruptured AAA patients. However, there are several studies in contrary to our study. The study of Tambyraja et al.⁸ reported the GAS as a poor predictor of mortality after repair of ruptured AAA. Similarly, the authors of another study held in 2009 concluded that the scoring systems including the GAS did not help predict the outcome of ruptured AAA surgery and did not aid clinical judgement.⁹ A study titled prediction of 30-day mortality after EVAR or OAR in patients with ruptured AAAs was done by Visser et al.¹⁰ They revealed limited discriminative ability of GAS and hence updated the GAS by adding the type of procedure performed and their updated version of GAS predicted 30-day mortality for patients with ruptured AAAs treated with EVAR or OAR.¹⁰

In contrary to the studies mentioned above, the study of Korhonen et al.¹¹ assessed the predictive value of GAS in postoperative death after repair of ruptured AAAs. They assigned a best cut-off value of the GAS in predicting postoperative death as 84. The mortality rate was 28.2% in patients with a GAS of ≤84 whilst 65% with a GAS of ≥84. They concluded that the GAS predicted postoperative death after repair of ruptured AAA in their series.¹¹ In another similarly designed study, the cut-off value for GAS was determined as 85. The mortality rate among patients with a GAS >85 was 88.9% whereas in those with <85, it was 15.9%. They concluded that the GAS was a valuable predictor of immediate postoperative death after emergency OAR.¹² A lower GAS threshold value in our study for mortality compared to the study of Korhonen et al.¹¹ is due to differences in population characteristics such as age and differences regarding the ratio of patients with shock, myocardial disease, cerebrovascular disease and renal disease.

A recent study including 101 patients displayed that the GAS was found to have a significant correlation with in-hospital mortality rates. They also displayed that a high score did not necessarily correspond with a definite mortality.¹³

We determined the most appropriate cut-off value for GAS at 78.5. The mortality in the patients with a GAS <78.5 was 17 patients (27.9%) and in the patients with a GAS >78.5 was 31 patients (51.7%). As a result, the GAS value above 78.5 is associated with almost threefold increase in mortality according to our study. Since our study revealed that the GAS had predictive value for in-hospital mortality in ruptured AAA patients, we did not require to modify or update the GAS as it was done by Visser and colleagues.¹⁰

Operative time and cross-clamp time were not significant amongst the groups in our study. Two patients in Group 1 and two patients in Group 2 had renal vein injury during surgery. One of the patients in Group 1 died intraoperatively (independent of renal injury) due to hyotension and multiple organ failure (MOF) whilst the other one had an increased urea and creatine level and died in the ICU due to MOF. One of the patients in Group 2 developed a mild increase in urea and creatine levels which decreased after two weeks whilst the other one had normal levels of urea and creatine. Most of the patients with postoperative renal failure died. The hematocrit levels of cases in both groups seem to be relatively higher than normally expected; the majority of the patients were referred from other centres with a preliminary diagnosis of ruptured AAA having been fluid resuscitated. This included the administration of blood. Therefore, the hematocrit level presented in this study was that of the full blood count taken on their arrival in our centre.

Exclusion of certain intra and postoperative data such as operative time, cross-clamp time, left renal vein damage, postoperative acute renal failure and coagulopathy seem to devalue the application of GAS. Therefore, the search for more sophisticated scoring systems continues. Despite everything, its easy preoperative applicability means that it can still be used in evaluating postoperative mortality.

In conclusion, GAS identifies high-risk patients and possesses a predictive value for mortality after ruptured AAAs. GAS value above 78.5 is associated with almost threefold increase in mortality.

Footnotes

Conflict of interest

None declared.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Patterson

Holt

Hinchliffe

. Predicting risk in elective abdominal aortic aneurysm repair: a systematic review of current evidence. Eur J Vasc Endovasc Surg: Off J Eur Society Vasc Surg 2008; 36: 637–645.

Donaldson

Rosenberg

Bucknam

. Factors affecting survival after ruptured abdominal aortic aneurysm. J Vasc Surg 1985; 2: 564–570.

Budd

Finch

Carter

. A study of the mortality from ruptured abdominal aortic aneurysms in a district community. Eur J Vasc Surg 1989; 3: 351–354.

Johansson

Swedenborg

. Ruptured abdominal aortic aneurysms: a study of incidence and mortality. Brit J Surg 1986; 73: 101–103.

Lawler

Jr . Aggressive treatment of ruptured abdominal aortic aneurysm in a community hospital. Surgery 1984; 95: 38–44.

Samy

Murray

MacBain

. Glasgow aneurysm score. Cardiovasc Surg 1994; 2: 41–44.

Rix

Bates

. Pre-operative risk scores for the prediction of outcome in elderly people who require emergency surgery. World J Emerg Surg 2007; 2: 16–16.

Tambyraja

Fraser

Murie

. Validity of the Glasgow Aneurysm Score and the Hardman Index in predicting outcome after ruptured abdominal aortic aneurysm repair. Brit J Surg 2005; 92: 570–573.

Gatt

Goldsmith

Martinez

. Do scoring systems help in predicting survival following ruptured abdominal aortic aneurysm surgery? Ann R Coll Surg Engl 2009; 91: 123–127.

10.

Visser

Williams

Kievit

. Prediction of 30-day mortality after endovascular repair or open surgery in patients with ruptured abdominal aortic aneurysms. J Vasc Surg 2009; 49: 1093–1099.

11.

Korhonen

Ylonen

Biancari

. Glasgow Aneurysm Score as a predictor of immediate outcome after surgery for ruptured abdominal aortic aneurysm. Brit J Surg 2004; 91: 1449–1452.

12.

Leo

Biancari

Nesi

. Risk-scoring methods in predicting the immediate outcome after emergency open repair of ruptured abdominal aortic aneurysm. Am J Surg 2006; 192: 19–23.

13.

Kurc

Sanioglu

Ozgen

. Preoperative risk factors for in-hospital mortality and validity of the Glasgow aneurysm score and Hardman index in patients with ruptured abdominal aortic aneurysm. Vascular 2012; 20: 150–155.