In this study, we assesed the performance of GAS in predicting long-term outcome in elective AAA patients who had undergone EVAR. The long term mortality rate in patients with GAS > 77.5, was almost twice that of the patients with GAS < 77.5. In addition, every 10 point increase in GAS score resulted in almost a two fold increase in risk of long-term mortality. Finally, the five year survival rate in patients with GAS > 77.5 was significantly less than in those with GAS scores < 77.5.
There are various risk scoring systems used in the prediction of mortality in patients undergoing repair for AAA. Despite the development of a few new scoring systems within the last decade, GAS still preserves its value in predicting the outcomes following EVAR. Our results also show that GAS has predictive value during long-term follow-up in patients undergoing EVAR for AAA. A systematic review of 608 studies revealed that the most frequently used scoring systems for elective AAA repair were GAS, POSSUM and VBHOM. However, GAS was found to be the most used and validated scoring system for open repair.8 Moreover, Nesi et al. . published an article in 2004, in which they compared four different scoring systems with GAS. They concluded that GAS was the most useful scoring system and the easiest one to apply in order to predict AAA surgical outcome. Previous studies conducted in our centre also support the use of GAS in patients undergoing elective and ruptured AAA surgery [10, 11].
Nevertheless, Visser et al. modified GAS by adding the type of procedure perfomed and evaluated predicted 30-day mortality in patients with ruptured AAAs treated with EVAR versus surgical AAA repair . Similarly, Choke et al. developed an in-house risk prediction model, in order to predict perioperative mortality following elective AAA repair. They believed that the new models had better performance in predicting perioperative mortality for elective open surgery and EVAR .
Most significantly, there are only a few studies in the literature on whether or not GAS predicts long-term survival following EVAR [5, 6]. The mean survival rates in our study were 90.6% for the 1st year, and 70.0% for the 5th year. Similarly, the overall 1 and 5-year survival rates were 91.7% and 76.7% according to the EUROSTAR registry, which is one of the few studies with results regarding the long-term follow-up after EVAR . Moreover, in the current study, the five year survival rate in patients with GAS > 77.5 was significantly less than in those with GAS scores < 77.5. Likewise, the EUROSTAR registry revealed a 5-year overall survival rate of 65.2% in patients with GAS above the cut-off value of > 83.6 . These results indicate that the survival rate decreases as the GAS values exceed the cut-off values. Furthermore, the EUROSTAR registry defined the 30-day mortality rate as 6.4% and 1.6% in patients with a score above and below the cut-off value respectively . However, long-term mortality rates were not presented in their study. In the present study, the long term mortality rate in patients with GAS > 77.5, was almost twice that of the patients with GAS < 77.5. Besides, every 10 point increase in GAS score resulted in almost a 2 fold increase in risk of long-term mortality.
The DREAM trial is the only randomized study comparing AAA patients treated with open repair and EVAR. In this study, Baas et al. concluded that the GAS can be used to predict 30-day and 2-year mortality for both treatment modalities . There was no correlation between the GAS values and in-hospital mortality in both the DREAM study and our study. This is most likely due to the low number of early mortalities. In addition, the DREAM trial supports the idea that GAS was superior in predicting 30-day and 2-year mortality in patients treated with EVAR compared to OR . Their cut-off value for GAS for EVAR, during 2 year follow-up was also the same as ours at 77.5%.
According to the results of our study, there were no female long-term mortalities. In contrast, the study by Tümer et al. revealed no difference in all-cause mortality between the genders during the long-term follow-up period following EVAR . We believe that the abscence of female patients in the long term mortality group of our study may have been due to the comparatively lower number of study participants compared to previous multi-center studies.
The American Association of Anestesiologists (ASA) Physical Status Classification System is one of the grading methods applied in order to categorize the preoperative status as well as to assess postoperative morbidity and mortality [14, 15]. Another finding of our study was the higher ratio of ASA 3 and 4 patients in the long term mortality group. According to a study published in 2019, ICU and hospital stay were longer in ASA 3 and 4 patients. However, 1 year mortality and morbidity rates did not differ significantly between low and high risk ASA groups . In contrast, our results showed that long-term mortality rates were higher in the high risk ASA 3 and 4 groups and in addition, the duration of ICU stay was weakly correlated with the ASA grades.
Despite a strong correlation found between the GAS and the duration of post-EVAR hospital stay, the mean post procedural hospital stay was not statistically significant in patients with GAS < 77.5 and > 77.5. The most important reason for this was that ten of the patients who underwent EVAR with a GAS < 77.5, had an unexpectedly high hospital stay: four due to renal, 3 due to cardiac and 3 due to periperal arterial complications.
Another interesting finding of our study was that the presence of malignancy resulted in a 5 fold increase in risk for long-term mortality. A long-term outcome study of AAA patients, by Ahn et al. documented that mortality was significantly higher in patients with a history of malignancy . The rate of malignancy was four times higher in the long-term mortality group compared to the survivors group in the current study. However, the cause of death was primarily cardiovascular diseases followed by aneurysm related causes in the late-term mortality group. Malignancy was the third common cause of mortality; out of 7 cancer patients in the long-term mortality group, 4 died. The presence of CHF was 9 times higher and CAD was 1.5 times higher in the long-term mortality group compared to the survivors group, which explains why cardiovascular etiologies were the leading cause of death in this study.
The results of our study support that GAS continues to be of value in the prediction of AAA outcome. The easy applicability of the GAS enables its use during long-term follow-up and may help in the education and counselling of patient peri-operatively regarding the long-term risks of EVAR.