Prevalence of Abdominal Wall Hernia in Participants With Abdominal Aortic Aneurysm Versus Peripheral Arterial Disease—A Population-Based Study

Introduction

An aneurysm is a permanent focal dilatation of an artery to 1.5 times its normal diameter. By convention, an infrarenal aorta 3 cm in diameter or larger is considered aneurysmal. ¹ Abdominal aortic aneurysm (AAA) is a relatively common, important and preventable cause of death. ² The primary event in the development of an AAA involves proteolytic degradation of the extracellular matrix proteins elastin and collagen. Various proteolytic enzymes, including matrix metalloproteinases, are involved in the degradation and remodeling of the aortic wall. ³ The trigger for this cascade of reactions remains unknown, but the connective tissue weakness might also be present in other tissues.

A hernia is the protrusion of an organ or the fascia of an organ through the wall of the cavity that normally contains it into an abnormal position. The common varieties of hernias through the abdominal wall, in order of frequency, are as follows: inguinal (indirect or direct), femoral, umbilical, and paraumbilical, incisional, ventral, and epigastric. ⁴ Hernias are known to occur at the site of weakness in abdominal wall. ⁴ This weakness may be congenital (eg, persistence of processus vaginalis of testicular descent giving rise to congenital inguinal hernia) or acquired (eg, following a surgical incision).

As has been described in several small studies, we notice a high prevalence of abdominal wall hernia (AWH) in patients with known AAA in our clinical practice.^5–11 This study tests the above observation in a large population-based cohort. We compare the prevalence of AWH in patients with AAA versus a similar cohort of patients with known peripheral arterial disease (PAD).

Patients and Methods

This was a retrospective electronic medical record review study that was approved by the local Institutional Review Board. The population under study was derived from Olmstead County, Minnesota. Its population is served by a largely unified medical care system that has accumulated comprehensive clinical records over a long period of time. Olmsted county, located in South East Minnesota, has an estimated population of 141 360. Approximately, 70% of the county population resides within the city limits of Rochester, Minnesota, the centrally located county seat. The 2008 estimates show the county population to be 89.4% white, 3.9% black, 5.2% Asian, and 0.3% native American. ¹² The population is largely middle class and has a high proportion of the working population employed in the health care industry.

Epidemiologic research in Olmsted County is facilitated by relative isolation from other urban centers and the fact that nearly all medical care is delivered by a small number of providers. Most of this medical care is provided by the Mayo Clinic, a tertiary care center. Within a given year, over 60% of the Rochester population is seen at one of the Mayo Clinic facilities and nearly 100% are seen within a period of 3 years. The potential use of data for population-based studies has been described earlier. ¹³

The electronic medical record system was queried for the diagnosis of AAA for a period of 10 years from 1994 through 2003. The time period chosen for the study had the most updated and complete data in the electronic medical records. Pregnant women and participants under the age of 18 years were excluded from the search. The result set was then queried for the diagnoses of ventral, inguinal, femoral, umbilical, and scrotal hernia. The same sets of queries were then repeated for the diagnosis of PAD. Contingency analysis was performed using the statistical software JMP version 8.0 (SAS Campus drive, Cary, North Carolina). Prevalence of hernia in the 2 groups was compared using the chi-square test.

Results

Of the 187 151 records queried, AAA was documented in 939 residents of Olmstead County, who were seen at Mayo Clinic between the years 1994 and 2003. Of these 157 (16.7%) were diagnosed with AWH. Peripheral arterial disease was documented in 3465 participants during the same time period; and in this group, 343 (9.9%) individuals were diagnosed with AWH. Participants with AAA were 1.7 times (95% confidence interval, 1.4-2.0) more likely to have AWH than participants with PAD (P < .0001), suggesting a very significant difference in the prevalence of hernia between the 2 groups.

Discussion

The association between AAA and AWH has both clinical and pathophysiological significance. From a clinical perspective, the prevalence of AAA ranges from 1% in men aged 55 to 60 years to 10% in those 80 years and older and has increased in recent decades.^14,15 The number of AAA detected is certain to rise, given the aging population and increasing use of imaging studies. Most AAAs are asymptomatic, and physical examination lacks sensitivity for detecting an aneurysm.¹⁶ Acute rupture of an AAA accounts for 2% of mortality in over 60-year-old men. ¹⁷ Large AAAs are associated with approximately 9000 deaths annually in the United States. ¹⁸ Emergency AAA repair at the time of rupture has a mortality ranging from 47% to 75% and elective AAA repair before rupture has a mortality of less than 5%.^19,20

Early detection and treatment may prevent death from rupture. Screening programs attempt to detect asymptomatic AAAs in order to repair them electively before rupture occurs. The cost-effectiveness of population-based screening for this condition is a matter of debate, but screening patients at higher risk may be more economically feasible. ²¹ Screening for AAA would most benefit those who have a reasonably high probability of having an AAA large enough or that will become large enough to benefit from surgery. The major risk factors for AAA include age (being 65 or older), male sex, and a history of ever smoking (at least 100 cigarettes in a person’s lifetime).^{15 –17,21,22} A first-degree family history of AAA requiring surgical repair also elevates a man’s risk for AAA; this may also be true for women but the evidence is less convincing.^19,23 An increased prevalence of AAA has also been demonstrated in patients with pulmonary emphysema. ²⁴ Current US Preventive Services Task Force (USPSTF) screening recommendations for AAA include 1-time screening by ultrasonography in men aged 65 to 75 who have ever smoked (Grade B recommendation). The USPSTF makes no recommendation for or against screening in men aged 65 to 75 who have never smoked (Grade C recommendation) or women of any age (Grade D recommendation). ²³

In this retrospective population-based study, the prevalence of hernia was almost 1.7 times higher in participants with AAA than participants with PAD. Studies in the past have demonstrated the increased prevalence of AAA in participants with known hernias but evidence for adopting AAA screening in patients with hernias was considered weak. This is most likely attributable to the fact that most of these studies were observational in nature with small sample sizes and involved single centers. In a small but important study from 1995, Pleumeekers et al compared prevalence of AAA in 156 men who underwent inguinal hernia repair and 1771 men without history of inguinal hernia repair. ¹⁴ They noted the increased incidence of AAA in men undergoing inguinal hernia repair (12.2% vs 3.7%) in those without such history, suggesting that patients with a history of inguinal hernia should be screened for AAA. ²⁵ In a systematic review of 1132 patients, Takagi et al concluded that patients with AAA were at 3-fold increased risk of both inguinal and postoperative incisional hernia compared to patients with aortoiliac occlusive disease. ¹¹ Raffetto et al examined individuals undergoing surgery at 3 different institutions and found that 177 patients with AAA had higher frequency of AWH when compared with 82 patients with atherosclerotic occlusive disease (38.4% vs 11%). Patients with AAA were at 9-fold increased risk of developing incision hernia postoperatively. ⁹ Similarly McPhail, in a 2008 publication, made a case for screening participants with diastasis recti for AAA after noticing an increased prevalence of diastasis recti in consecutive patients with AAA versus patients with PAD (66.7% vs 16.7%). ²⁶ All the above studies demonstrated a positive association between AAA and presence of hernia but had major shortcomings. The sample sizes were small with the exception of the systematic review. All the studies were retrospective, and all but 1 study (Raffetto et al ⁹ ) were single-center observations.

In 2008, Anderson and Shiralkar tested the prior findings in a prospective study. They enrolled 70 consecutive males with inguinal hernia between the ages of 65 and 88 and performed abdominal ultrasound on them. The study was projected to achieve a power of 86% if the difference in prevalence of AAA between the cohort under study and historical data was 9% (1-sided type I error of 5%). This calculation seems to have been performed post hoc. In contrast to the prior findings, they found no increase in the prevalence of AAA in patients with inguinal hernia versus historical controls. ²⁷ Analytical bias associated with the post hoc power analysis and using single-sided error rates for same was not addressed in the study. In 2009, Moesbergen et al retrospectively measured the width of linia alba on computed tomography (CT) of 75 male individuals who underwent elective AAA repair and found no difference in prevalence compared with age-matched controls. ²⁸ Preliminary data from our own institution show similar findings. However, diastasis recti may not be visible on a resting CT scan with patient in supine position because the abdominal wall is relaxed. Diastasis recti typically becomes apparent on physical examination when the patient attempts to sit up, thereby contracting abdominal muscles and stretching the linea alba. These 2 studies challenge the historical data on the subject and necessitate further study.

In the present study, we sought to take an epidemiologic approach in a large population-based cohort. We used participants with PAD as a control group because of similarities in age, risk factors (especially smoking), and gender. Our findings reaffirm the earlier observation that individuals with AAA have greater occurrence of AWH. Our study could be criticized for being retrospective in nature in a predominantly caucasian population. We acknowledge these concerns; however, this is the first study to demonstrate the association between AAA and hernia in a large population and AAA is predominantly a disease of caucasians.

Our findings make a strong argument for basic research to investigate weakness in connective tissue that might underlie these diverse clinical manifestations. This suggestion is not new. In 1981, Cannon and Read coined the term “metastatic emphysema” to describe weakening of the connective tissue in both the lungs and the inguinal canal, leading to emphysema and inguinal hernia in smokers. ²⁹ Smoking has also been associated with increased risk of incisional and recurrent groin hernia.^30,31 The association between emphysema, hernia, and AAA may partly be attributable to systemic elastin and collagen fiber degeneration which is thought to be enhanced by smoking.^6,32 Levels of hydroxyproline, a compound early in the collagen synthesis pathway, have been demonstrated to be lower in patients with inguinal hernias. ³³ Smokers produce less hydroxyproline than nonsmokers. ³⁴ These findings suggest a common causal relationship between hernias and AAA at the molecular level.

The role of atherosclerosis in the pathophysiology of AAA is unclear, and the longstanding notion of atherosclerotic aneurysm has been challenged. Abdominal aortic aneurysms have been associated with atherosclerosis and occur in patients with similar risk profiles. ³ However, atherosclerosis of the aorta is much more common than AAA. ³ Atherosclerosis is found in many vascular territories, whereas aneurysms are found in fewer vascular territories—usually the aortoiliac segment and much less frequently elsewhere. ⁴ Inflammation appears to be an integral component of both the disease processes. It may be that inflammation could lead to aneurysm formation in a susceptible individual with weak connective tissue but not in one with strong connective tissue. Proteolytic enzymes that degrade the connective tissue of the aortic wall have been the subject of considerable study. Overactivity of matrix metalloproteinases (especially types 2 and 9) has been noted in AAA and may be a target for pharmacotherapy.^35,36 The activity of these destructive enzymes is unknown in the connective tissues of the abdominal wall. Connective tissues remote from the aorta could provide an alternative avenue for the study of etiology of aneurysm formation.

Our study has several limitations. The primary limitation is that it is retrospective. This study relies on the complete and accurate recording of the pertinent information in the medical record. For some data elements (eg, diagnosis of AAA and PAD), the data may be more or less complete than for other elements (eg, asymptomatic hernia). It is reasonable to assume that the completeness of documentation would be independent of year within the chosen time period. Another limitation is imposed by the study setting. The majority of the population under study was middle class caucasians with adequate health insurance and access to the state-of-the art medical care. Extrapolating findings from this study to other demographic groups, not represented within the county, may be problematic.

Conclusions

The pathogenesis of AAA remains incompletely understood and no effective medical treatment exists. Surgical treatment is most helpful if screening detects aneurysms that are at risk of rupturing. Weakness of connective tissue seems to be an integral part of this multifactorial disease process. This population-based retrospective study suggests a strong association between AWH and AAA. This finding should prompt further basic research into the cellular mechanisms that lead to connective tissue weakness. Screening of patients with AWH for AAA might be economically feasible and of high yield, although prospective validation is required.