Impact of Risk Factors on Mortalityafter Elective Open Repair of Abdominal Aortic Aneurysms

IMPACT OF RISK FACTORS ON MORTALITYAFTER ELECTIVE OPEN REPAIR OF ABDOMINAL AORTIC ANEURYSMS

Ana Luísa Pinto (), Catarina Machado (), Filipa Gomes (), Hugo Sousa (), José Henrique Messias (), Joana Teixeira (), Luís Miguel Lázaro (), Maria Filipa Figueiredo (), Miguel Quinta e Costa (), Rafaela Parreira (), Sérgio Cardoso (), Sílvia Farraposo (), Sofia Figueiredo ().

MD PhD Sérgio Sampaio, Class 5

ABSTRACT

BACKGROUND: There are several risk factorsfor the development of Abdominal Aortic Aneurysms (AAA). Despite recent advances in medical care, Elective Open Repair (EOR) remains the“gold standard” procedure for AAA repair and even in selected patientsthe overall mortality rate is roughly 5%. AIMS OF THE STUDY: This study aims to summarize risk factors, postoperative complications and mortality rates of patients with AAA undergoing EOR, and analyze the influence of risk factors on patients’ outcome, thus contributing to the improvement of outcome predictionafter EOR. METHODS: Two-hundred and threepaperswere identified regarding AAA, elective open repair/surgery and mortality in the PubMed/Medline database. From those, 50 provided data related to risk factors exposition, postoperative complications and clinical outcome. RESULTS: Males with advanced age (median 71.0years old) and with a median aneurysm diameter of 5.9 cm were the most frequent patients with AAA among all studies. The most common risk factors found were Hypertension (median 60.0%), Smoking Habits (median 54.2%), History of Cardiac Diseases (median 44.3%), and History of Pulmonary Diseases (median 23.0%). Median occurrence of morbidity events was31.0%, whilemortality rate was 4.0%. Meta-analysis revealed that Hypertension (OR=2.95; p<0.001), Chronic Renal Failure (OR=2.78; p<0.001), History of CardiacDisease (OR=1.93; p<0.001), Female Gender (OR=1.58; p<0.001), Aneurysm Diameter (OR=1.58; p<0.011) and Age (OR=1.07; p<0.001) were associated with increased risk of death after EOR, while History of Pulmonary Disease (OR=1.32; p=0.112) and Diabetes Mellitus (OR=1.28; p=0.309) did not provide strong data. Moreover, Smoking habits revealed not to influence the patients’ outcome after EOR. CONCLUSION: This study provides important data that contribute to selection of patients that are at high risk of death after EOR, and therefore prediction of patient outcome may be improved.

KEY-WORDS: AbdominalAorticAneurysm (AAA); elective open repair (EOR); risk factors; postoperative complications; mortality; morbidity; meta-analysis review.

INTRODUCTION

An Abdominal Aortic Aneurysm (AAA) is considered to be an abnormal widening of 1.5-folds the normal diameter of the aorta below the renal arteries, and its development is related to several risk factors which are associated with the weakening of the aortic wall, namely: smoking, diabetes, elevated cholesterol, family history, hypertension and atherosclerosis (1-3).

Thus, AAA is a severe health condition that can lead to death if rupture occurs. Despite the different options to prevent rupture, this condition is responsible for approximately 9,000 deaths per year (1). In the United States, 40,000 patients are submitted to elective surgical repair of an AAA each year, preoperative mortality occurring in 2-8% of properly selected patients (4-6).

The incidence of AAA increases with age, rarely developing before the age of 50, and is much more common in men (1;6). When an aneurysm is detected, clinicians may have to decide which treatment options must be used, based on patients’ life expectancy and aneurysms’ size: active surveillance, immediate/emergency repair, elective open repair (EOR) and endovascular repair (EVAR) (6;7). As an aneurysm grows progressively, rupture can occur at any moment, and even with an emergency repair the mortality remains to be roughly 50%.

EOR has been the most used approach to repair AAA, and is undertaken only when the risk of rupture is considered to be high (8). EOR may face several peri and postoperative complications, such as: venous bleeding, resulting from iliac or left renal veins injury; gastrointestinal ischemia, more likely to occur following ruptured AAA repair; cardiac complications, which constitute the most serious threat to patients after AAA repair; pulmonary complications; and also renal failure. AAA repair is, therefore a technically difficult procedure and despite late complications beinginfrequent, mortality after EOR still ranges3 – 10% (4;9).

Several studies have suggested that success rates of EOR depend onpatient’s condition and exposure todifferent risk factors (10-12). Therefore, as it is unclear what risk factors influence the mortality after EOR,we performed a systematic review followed by meta-analysis to summarize risk factors, postoperative complications and mortality rates, and to analyse the influence of risk factors on patient’s outcome after EOR.

PARTICIPANTS AND METHODS

Study participants and Sampling Methods

A total of 203 paperswere identified by searchingthe PUBMED/Medline database, on October 2007, with the following query: (("abdominal aortic aneurysm"[Text Word] OR "aortic aneurysm, abdominal"[MeSH Terms] OR aaa[Text Word]) AND elective[All Fields] AND open[All Fields]) AND ((("wound healing"[TIAB] NOT Medline[SB]) OR "wound healing"[MeSH Terms] OR repair[Text Word]) OR ("surgery"[Subheading] OR "operative surgical procedures"[Text Word] OR "surgical procedures, operative"[MeSH Terms] OR "surgery"[MeSH Terms] OR surgery[Text Word])) AND mortality[Text Word].

Study design

Papers had to fulfil the following criteria to be included in this study: 1) patients must have been diagnosed with abdominal aortic aneurysms (AAA); 2) patients must have been submitted to elective open repair/surgery; 3) papersshould provide data about risk factors and clinical outcome after elective open repair of AAA. Studies were excluded if: 1) patients had thoraco-abdominal or abdomino-iliac aneurysms; 2) patients had been submitted to emergencyrepair after ruptured AAA; 3) patients had been submitted to EVAR; 4) studies were Reviews, Systematic Reviews, Meta-analysis, Letters or Editorials; 5) publishedin other language than English, French, Spanish or Portugueseand 6) there were no data related to the study design.

Eihty paperswere selected andrequested either online, by library acquisition or by e-mail request to authors. Only 67 articles were possible to obtain and then submitted to a second triage in order to select the articles that considered the variables of interest for the study. This second triage step revealed that46 papers, which represent 50 individual studies, had information about the variables of interest for the statistical analysis (preoperative clinical variables, risk factors, postoperative complications, and patients’ outcome). These 50 studies were considered for the estimation of the frequency of all variables of interest in the study; nevertheless, 17 papers were used to summarize patients’ risk factors impact on clinical outcome.

Variables description and data acquisition method

A review was performed in order to extract data from selected papers: 1) general characteristics, such as: name of first author, year of publication, country of origin, type of study, sample size; 2) clinical variables, such as: age, gender and aneurysm diameter; 3) risk factors exposition: diabetes mellitus, history of cardiac disease(previous myocardial infarction or coronary disease), hypertension, history of pulmonary disease (pulmonary embolism or Chronic Obstructive Pulmonary Disease, COPD), chronic renal failure (according to creatinine levels), and smoking habits; 4) postoperative complications, such as: general morbidity events, gastrointestinal ischemia, multiorgan failure, renal failure, general cardiac complications, myocardial infarction, and respiratory complications; and 5) clinical outcome of the patients. The clinical outcome of interest for this study was defined as mortality measured 30 days after surgery. In cases with different outcomes per report, each outcome was analyzed as an independent study.

Statistical analysis

All collected data was inserted on a database using Statistical Package for Social Sciences (SPSS) Version 16.0 for Windows (SPSS Inc). SPSS software was used to summarize the medianproportion, range of values and standard deviations of all clinical variables, risk factors exposition and postoperative complications, as well as to summarize the mortality rates of all studies. The Epi Info™ Version 6 was used to perform a Χ2 analysis in order to determine the Odds Ratio (OR) and its 95% confidence interval, when such information was not available in the paper. Meta-analysis of data was performed with the open access program R Version 2.6.2to elaborate the Forest Plot graphics in order to combine the influence of risk factors on patients’ outcome.

RESULTS

Summary of clinical variables, risk factors and postoperative complications

The first approach to this study was performed by summarizing data about clinical variables, risk factors and postoperative complications, as well as to summarize the mortality rate within all studies (Table I).

The systematic review revealed an increased frequency of Male Gender (median 86.9%) among studies. In addition, the median of mean age of diagnosis was 71.4 years, andthe median mean aneurysm diameter was 5.9 cm. Most common risk factorswere hypertension (median 60.0%), smoking (median 54.2%), history of cardiac disease (median 44.3%), history of pulmonary disease (median 23.0%), and diabetes mellitus and chronic renal failure (both with median9.5%).

Morbidity rates varied a lot among studies, and statistical analysis revealed a median occurrence of 31.0%, with the most frequent postoperative complications being: Respiratory Complications (median 7.5%), Cardiac Complications (median 5.7%), and Multi Organ Failure (median 3.6%). The median mortality rate was 4.0%.

Meta-analysis

The meta-analysis data ofrisk factors impact on patients’ mortality after EORare available in Figure 1(8;13-28).Hypertension (HT), Chronic Renal failure (CRF), History of Cardiac Disease (HCD), Female Gender and Age revealed to be associated with increased risk of death after EOR in a statistically significant way, while History of Pulmonary Disease (HPD) and Diabetes Mellitus(DM) provided mere trends, not statistically significant. Moreover, Smoking habits revealed noinfluence onthe risk of death after EOR.

Data show that Hypertension and Chronic Renal Failure represent an almost 3-folds increased risk of death after EOR; Cardiac Disease History represent an increased risk of almost 2-folds; Female Gender and an increment of 1cm in the Aneurysm Diameter revealed a 58% increased risk; and Age has also proved to influence the risk of death after EOR, with an increment of 7% per year.

DISCUSSION

Limitations of the study

Although, the literature search was extremmely comprehensive (see study design section), which should have minimized reporting biases, it still have some limitations. We found it hard to reach the most sensitive query, mainly due to the absence of a unique term to define “elective open repair”. Furthermore, we were forced to apply the query without any reference to any risk factor or postoperative complication since there is a wide variety and many of them are expressed differently, and therefore, the application of inclusion/exclusion criteria might have not been as accurate as we aimed. Wealso observed that studies were extremely heterogeneous and the majority of them failed to statesome of the demographic and preoperative clinical variables, as well as risk factors exposition, potentially of interest to perform a more accurate systematic review. Moreover, only a few studies have stated the impact of each risk factor on mortalitywhich was extremely important to the meta-analysis study.Finally, we observed that the underlying prevalence of AAA may be the reason for between-study heterogeneity (2).

Epidemiological Data

This study combined the information concerning the frequency of the different risk factors for AAA development, revealing male gender, advanced age, hypertension, smoking and cardiac disease as the major risk factors found among all studies – Table I. Moreover, the most frequent postoperative complications found were respiratory and cardiac complications, and the mortality rate of patients with AAA attending EOR was 4.0%.

Many studies reffer that men are more frequently and earlier affected by AAA development than women, with a male:female ratio of 9:1 (29).Our results have shown that AAA was significantly more frequent in male gender (median 86.9%) and in patients with advanced age (medianof means of 71.4 years old),a similar distributionto those shown in literature which refer that AAA are more common in men with age ranging 65-75 y.o.(1, 26, 29, 30).

Despite urgent treatments that aortic aneurysms require, the guidelines refer that for those smaller than 4.5 cm, only ultrasonographic surveillance every 6 months is recommended. When aneureysms reach a size between 4.5 and 5 cm, the follow-up time showld be narrowed to every 3 months, nevertheless, in young patiens <65 y.o. with an AAA of 4 to 5 cm, elective surgery instead of conservative surveillance should considered. AAAs’ with a diameter of 5.5 cm or over, are indicated for EOR or EVAR(29). In this study, the data related to the average aneurysm size revealed that patients had a mean aneurysm diameter of 5.9 cm, which taking in consideration the advanced age of patients in agreement with the guidelines for AAA treatment.

The results show that male gender, hypertension, current cigarrette smoking and history of cardiac diseasewere the major risk factors found among all papers – Table I. The majority of review studies refer age, smoking, and gender as the most significant AAA risk factors, although, hypertension and history of cardiac disease should be also considered important (2;12;30).

Literature suggests a wide variety of postoperative complications, and most of them are correlated with the healthy condition of the patient prior to the surgery, or also to the experience of the surgical team. It is commonly accepted that at leastbetween 5-25% of patientssuffer one postoperative complication prior to 30 days after surgery (5;12). Our study showed that the median occurrence of Morbidity events was of 31.0%, although, this variable varied a lot among all studies. Therefore, we may assume that these differences might be correlated with the health condition of patients. Our data reveal that the most frequent postoperative complications found were Respiratory, Cardiac complications, and Multi Organ Failure– Table I. The main conclusion to take form these data, and according to some studies, is that patients might need to stay at an Intensive Care Unit (ICU) for a few time in order to reestablish the optimal conditions for the patient to go home and complete the recover without any more complication (10).

Another curious finding was that mortality rates also varied a lot among studies, ranging between 0-17% and with a mean mortality rate of 4.0% – Table I. Mortality rates prior to 30 days after surgery for patients undergoing EOR ranges between 1-5%, although in some surgical teams this can be 0%. Thus, our data reveals a similar trend to those found in the majority of the published studies (1, 3, 7-9, 30). Nevertheless, and despite the different condition of patients, in-hospital care conditions are extremely important to prevent higher mortality rates(10).

Influence of risk factor on patient’s outcome after EOR

Despite the peri and postoperative complications, EOR remains the gold-standard procedure for AAA repair worldwide, since it is considered as definitive, requiring only imagiologic surveillance(31).Up to our knowledge this is the first meta-analysis that combines the information related to the impact of risk factors on mortality after AAA EOR. Although there were only 17 studies providing the necessary data for the meta-analysis, it was possible to achieve interesting results considering the role of risk factors on patients’ outcome after EOR (Figure 1). The results show that hypertension, chronic renal failure, history of cardiac disease, female gender and age are correlated to increasd risk of death after EOR, and that history of pulmonary disease and diabetes mellitus require more studies to reach better conclusions. Current smoking habits revealed not to influence the risk of death after EOR.

Cigarette smoking has been consistently associated with increased risk of AAA development, moreover, several studies suggested that the risk is different according to smoking condition (ever, current, or continued smoker) (3). Therefore, we were surprised that statistical analysis revealed no impact of smoking on mortality (OR=1.00; p=0.987). Curiously, some authors refer that patients may benefit of smoking cessation after being diagnosed with AAA (3).

The analysis of the impact of epidemiological data on patients’ outcome revealed that Female gender increases the risk of death after EOR in 58% when compared to Male gender, which represent a significant risk factor for AAA. However, it is well known that when females develop AAA, usually it has more severe consequences and death can occur shortly after surgery. A similar effect was shown when we observed that an increment of 1cm of the Aneurysm Diameterrepresents a 58% increase risk of death after EOR. We have also observed that despite not statistically significant, Ageinfluences the risk of death after EOR, with an increment of 7% per year. Several authors defend that mortality rate of patients with AAA may also increease with age, which might be biased by the diminuition of patients’ healthy condition (32).

Results have also shown that when considering the impact of PDHand DM on patients’ outcome, although, the analysis point to an increase of 32% and 28%, respectively, in the risk of death after EOR, the data were not statistically significant. Interestingly, it has been suggested that DM is not consistently considered a risk factor for AAA (3) , therefore it stills required more studies to clarify its role wither on AAA development or its impact on patients’ outcome.