Do Hernias Contribute to Increased Severity of Aneurysmal Disease among Abdominal Aortic Aneurysm Patients?

Background  Connective tissue disorders could contribute to the pathogenesis of both abdominal aortic aneurysms (AAA) and hernias. We tested the hypothesis that hernias in AAA patients contribute to increased severity of the aneurysmal disease. Methods  A questionnaire was used to collect information from 195 AAA patients divided into four groups: (1) survivors ( n  = 22) of ruptured AAA, (2) patients ( n  = 90) after elective open repair, (3) patients ( n  = 43) after elective endovascular repair (EVAR), and (4) patients ( n  = 40) under surveillance of AAA. The control group consisted of 100 patients without AAA whose abdominal computed tomography (CT) scans were examined for the presence of hernias. Mann–Whitney U -test, Chi-squared ( χ 2 ) test, or Fisher's exact test (as appropriate) were used for statistical analyses. Multivariate logistic regression was used to control for potential confounding variables such as sex and age. Results  The prevalence of inguinal hernias was significantly higher in the AAA than the control group (25 vs. 9%, p  = 0.001) and did not differ between the AAA subgroups (9, 24, 35, and 23% in subgroups 1 through 4, respectively, p  = 0.15) based on univariate analysis. The prevalence of inguinal hernias did not differ ( p  = 0.15) between the two open surgery groups (groups 1 and 2), or when comparing all three operative procedures as a combined group to group 4 ( p  = 0.73). The prevalences of incisional hernias were 18 and 24% for groups 1 and 2, respectively, with no significant difference ( p  = 0.39). Inguinal hernia demonstrated a significant association with AAA on multivariate analysis ( p  = 0.006; odds ratio [OR] = 4.00; 95% confidence interval [CI] = 1.49–10.66). Conclusions  Our study confirms previous observations that patients with AAA have a high prevalence of hernias. Our results suggest that hernias do not contribute to increased severity of the aneurysmal disease.


Introduction
Abdominal aortic aneurysm (AAA) is a complex disease of the aging population that affects up to 10% of Caucasian males over 65 years of age and is a leading cause of death in the United States with approximately 7,000 deaths per year. 1 Due to improved diagnostic efforts using ultrasonography imaging, AAAs can be identified earlier. Despite improved diagnostic imaging, 15% of cases present as ruptured AAA in hospitals of industrialized nations. 2 AAA has several recognized risk factors such as smoking, male sex, age over 65 years, and positive family history. 3 Additionally, an increased prevalence of AAA was described in patients with pulmonary emphysema, 4 and patients with AAA have a 2.3-to 3-fold increased risk of developing inguinal hernia compared with patients with aortoiliac occlusive disease (AOD). 5,6 It is well known that mechanical factors, such as weakening of the abdominal wall after a transperitoneal operation and increased intra-abdominal pressure, increase a person's risk for hernias. Other known risk factors for inguinal hernias include male sex, age over 65 years, smoking, pulmonary emphysema, and family history. 7 Incisional hernias occur in 5 to 15% of patients undergoing transperitoneal operations. 8 Risk factors for incisional hernias include malnutrition, malignancy, wound infection, chronic obstructive pulmonary disease, diabetes mellitus, renal insufficiency, obesity, smoking, and older age. 8 Thus, AAA patients and patients with the most common forms of hernias of the abdominal wall share many risk factors.
Vascular surgeons are aware of the high risk of developing an incisional hernia after open transabdominal AAA surgery. Furthermore, AAA patients often present with large inguinal hernias or have a history of an inguinal hernia repair. 4,6 The prevalence of inguinal and incisional hernias among AAA patients was investigated in studies summarized in ►Tables 1 and 2. A meta-analysis on several of these studies concluded that the prevalences of inguinal and incisional hernias among AAA patients were 23 and 21%, respectively. 6 A large retrospective cohort study found that the combined prevalence of all types of abdominal wall hernias was 16.7% in AAA patients (n ¼ 939) compared with 9.9% in the control group of peripheral arterial disease (PAD) patients (n ¼ 3,465; p < 0.0001). 9 The main aim of this study was to determine if hernias contribute to increased severity of the aneurysmal disease. The underlying hypothesis was that AAA is a connectivetissue disorder demonstrating vascular weakness and therefore more advanced stages of the AAA disease (i.e., diameter > 5 cm or ruptured AAA) are associated with a higher prevalence of hernias.

Patient Groups and Data Collection
A questionnaire was designed to collect the following information: (1) presence of inguinal and incisional hernias, (2) history of hernia operations, and (3)  were evaluated for presence of hernia, and patients' clinical data were collected from the medical records as previously described. 10

Statistical Analyses
First, we performed a univariate comparison of the AAA and control group. The mean, median, standard deviation (SD), maximum, and minimum values were calculated. For categorical variables we performed a cross tabulation. To measure the statistical significance, Mann-Whitney U-test, Chisquared (χ 2 ) test, and Fisher's exact test (as appropriate) were used.
The prevalences of inguinal and incisional hernias were estimated for each treatment group separately. Differences in prevalences of inguinal and incisional hernias among the four AAA patient groups were tested using χ 2 test when the sample size for each cell in the analysis was >5 or Fisher's exact test when the sample size was 5. Multivariate logistic regression was utilized to control for confounding variables of sex and age.
Age is reported in years at the time the questionnaire was administered (when information regarding the occurrence of a hernia was obtained). Statistical Analysis Software (SAS V.9.2 for Windows, SAS Institute Inc., Cary, NC) and SPSS statistical package (v22 for Windows, IBM, Armonk, NY) were used for analyses.

Literature Search
All studies which determined the incidence of inguinal hernias and/or incisional hernias in patients with AAA were identified using a two-step search strategy. First, a PubMed search with keywords "aneurysm AND hernia" was performed for articles published between 1966 and May 2019. Second, relevant studies were identified by a manual search in the references of the identified articles.

Results
The AAA group consisted of 195 patients, with a mean age of 75.9 ( AE 8.6) years. There was no significant difference in age among males (n ¼ 178) and females (n ¼ 17; p ¼ 0.08). The control group included 100 patients with a mean age of 71.2 ( AE 8.6) years and was therefore significantly younger than the AAA group (75.9 AE 8.6; p < 0.001). The mean age was 76.2 AE 8.4 years for the rAAA, 75.5 AE 8.8 years for the eAAA, 75.2 AE 8.6 years for EVAR, and 76.8 AE 7.5 years for the sAAA group. The mean sizes of the diameters of the AAAs were 5.8 AE 1.1 cm for the eAAA, 5.8 AE 0.8 cm for EVAR, and 4.2 AE 0.5 cm for the sAAA group. Altogether 11 (6%) of the 195 patients reported positive family history for AAA. Among the four subgroups, the number of AAA patients with positive family history varied from one (2%) in the EVAR group to three (14%) in the rAAA group.
Clinical data from the AAA and control patients are presented in ►Table 3. AAA patients were more likely to suffer from hypertension, PAD, coronary artery disease (CAD), dyslipidemia and chronic obstructive pulmonary disease (COPD; all p < 0.001), as well as to be ever smokers (p < 0.001). Treatment with medications (aspirin, nitrates, βblockers, and statins) was significantly higher in the AAA group. AAA patients had significantly lower hemoglobin level and thrombocyte count but higher glucose and creatinine level. In multivariate analysis, we found a significant association of AAA with hypertension, CAD, COPD, and ever smoking (►Table 4). Clinical data for each AAA subgroup is presented in ►Table 5. We looked at the age of the patients at two time points, at the initiation of the study and at the time of operation. Age did not differ significantly between the groups for either time point. Cardiovascular risk factors between the groups were not significantly different, with the exception of dyslipidemia, which was significantly lower in the rupture group. Treatment with medications (statins and β-blockers) was significantly lower in the ruptured AAA group. Preoperative laboratory measurements for C-reactive protein were significantly higher in patients with ruptured AAA. A comparison of each AAA group with the control group did not reveal any substantial differences, as stated in ►Table 6.
A total of 48 (25%) of the 195 AAA patients reported having had an inguinal hernia and 26 (23%) of the 112 patients who had undergone an open repair (rAAA þ eAAA) reported an incisional hernia (►Table 7). Many of the patients with inguinal hernia (32/48, 67%) had had herniorrhaphy, increasing the likelihood that the diagnoses reported by the patients in the questionnaire were correct. The prevalence of inguinal hernia in the control group was significantly lower than in the AAA group (9/100, 9%, p ¼ 0.001). An incisional hernia was found in 2 (2%) of control patients. We compared the prevalences of hernias in the different treatment groups for AAA (►Tables 7 and 8). The prevalences of inguinal and incisional hernias in the rAAA group of 22 patients were 9% (two patients) and 18% (four patients), respectively. Herniorrhaphy was reported in one patient with an inguinal hernia and one patient with an incisional hernia. In the eAAA group, 22 of 90 (24%) patients had inguinal hernia, 12 (55%) of which had been repaired operatively. In addition, 22 of 90 (24%) of the eAAA group patients reported incisional hernia, 11 (50%) of which were surgically repaired. In the EVAR group, 15 (35%) patients had inguinal hernias, 12 (80%) of which had been surgically repaired. In the sAAA group, 9 of 40 (23%) patients had a history of inguinal hernias, 7 (78%) of whom reported having had a herniorrhaphy.
Comparisons between males and females for family history of AAA (p ¼ 0.79; odds ratio [     Only the surgical groups (rAAA compared with eAAA) were analyzed for incisional hernias, and no significant difference was found (p ¼ 0.39).

Discussion
Our study demonstrated a higher prevalence of inguinal hernia in AAA patients compared with the control group and a significant association between hernia and AAA in a multivariate analysis. The prevalence of inguinal hernia did not, however, differ between the AAA subgroups, suggesting that hernias do not contribute to increased severity of the aneurysmal disease.
Patients diagnosed with known connective tissue disorders such as the Marfan syndrome (OMIM ID: 154700), the Loeys-Dietz syndrome (OMIM IDs: 609192, 613795, 608967, 610380, 614816, and 610168), and the Ehlers-Danlos vascular type (OMIM ID: 130050) are prone to aortic aneurysms and dissections, as well as to rupture of other hollow organs such as uterus and bowel (http://omim.org). They are also more susceptible to inguinal and incisional hernias. These patients have mutations in genes of the structural proteins of the connective tissues, that is, fibrillin I (FBN1) in the Marfan syndrome, in the SMAD3, TGFB2, and TGFB-receptor genes (TGFBR1 and TGFBR2) in the syndromic forms of thoracic aortic aneurysms, and in Type III collagen gene (COL3A1) in the case of Ehlers-Danlos syndrome Type IV, also known as the vascular type of the Ehlers-Danlos syndrome.
It is possible that more common nonsyndromic diseases, such as AAAs also involve a connective tissue defect 3 which increases the patient's risk for abdominal wall hernias. Although AAA and abdominal wall hernias present common risk factors (e.g., smoking), there seems to be an independent association of these two diseases. 3,5 Clinical observations show that abdominal wall hernias are more common in AAA patients than in the general population. 9 Furthermore, pathobiological studies on abdominal wall hernias revealed similarities to the pathogenesis of AAA. [11][12][13][14][15] Several studies have investigated the expression of proteases responsible for degradation of the extracellular matrix (ECM). Matrix metalloproteinase 1 (MMP1) expression was increased in scar tissue of the skin from patients with incisional hernia and recurrent inguinal hernias, 14 but not in patients with primary inguinal hernia. 14 Similarly, elevated levels of MMP1 have been detected in patients with AAA. 15 Also, fibroblast cultures from the abdominal skin in direct inguinal hernia patients showed an increased matrix metalloproteinase 2 (MMP2) expression, 13 MMP2 mRNA levels were increased in incisional hernia samples, 14 and in serum of AAA patients. 15 The expression of metalloproteinase inhibitor 2 (TIMP2), one of the naturally occurring inhibitors of MMPs, was decreased in abdominal skin of hernia patients, 13 in fascia transversalis specimens of inguinal hernia patients, 11 and in the aneurysmal aortic wall. 15 These findings suggest an imbalance between the proteolytic activity of MMPs and their inhibitors in the abdominal skin prone to hernia formation and in the aortic wall with aneurysmal dilatation.
It is also plausible that hernias and AAAs develop due to defects in the structural proteins of skin, fascia, and aorta. The ratio of Type I to III collagen was shown to be reduced in skin fibroblasts, fascia transversalis, and the hernial sac in patients with inguinal and incisional hernia. 12,14 Collagens, which are   major structural components of the ECM in the aorta and provide support and strength to the aorta, have also been shown to play a role in AAA development and rupture. 3 In conclusion, the results summarized above provide strong evidence for the hypothesis that similar pathobiological mechanisms are involved in hernias and AAA. Consistent with this hypothesis are findings of two recently published studies. One of them detected an increased prevalence of AAA among 235 patients with inguinal hernia who were >55 years old compared with a control group of 203 patients without inguinal hernias. 16 Another group of investigators found an increased prevalence of abdominal wall hernia in 110 patients with AAA (45%) in comparison with 60 patients with AOD (5%), disease that shares many of the same risk factors (p-value < 0,001). 17 Three other studies with fewer patients, however, showed contradictory results. 7,18,19 It has also been suggested that surgical operations increase the patient's risk for hernia formation. One study detected higher prevalence of inguinal hernias after appendectomies and suggested changes in the muscle strength of the abdominal wall after transperitoneal operation. 20 In our study, the prevalence of hernias was not increased among patients who had been operated using midline incision compared with AAA patients who were under surveillance. Moreover, a Danish nationwide retrospective study found that AAA is an independent risk factor for incisional hernia repair in patients undergoing open elective aortic reconstructive surgery. Patients undergoing AAA repair have a 1.6-fold higher risk of a subsequent hernia repair than patients undergoing AOD repair. 21 Several previous studies showed a high prevalence of inguinal and incisional hernias among AAA patients (►Tables 1 and 2). A total of 12 studies, 11 of which presented a comparison between AAA and control groups, have been published on the prevalence of inguinal hernias. In 8 of 11 of these studies, the prevalence of inguinal hernias among AAA patients was significantly higher than in the control group, and ranged between 19 and 41% (►Table 1). In three of the studies (►Table 1), the prevalence of inguinal hernia was 14, 16 and 22% in the AAA group compared to 16, 9, and 17% in the control group, respectively. Only two previous studies investigated the prevalence of abdominal wall hernias in patients with large and small AAA (AAA < 55 vs. AAA > 55 mm). While Gindera et al 35 demonstrated a lack of association between the two diseases, which confirms the finding of our study, Pitoulias et al 17 found a significantly higher prevalence of hernia in large aneurysms (50 vs. 29.2%, p ¼ 0.03). A recent study investigated the prevalence of AAA in patients with inguinal hernias and demonstrated a 27.7% prevalence of AAA among the study group. 22 The prevalence of incisional hernias has been reported in 19 studies, 12 of which presented a comparison between AAA and control groups (►Table 2). Only 7 of 12 studies found a significant difference in the prevalence of incisional hernias between AAA (range: 10-37%) and control (range: 3-26%) groups (►Table 2).
It is possible to minimize the risk for incisional hernia by surgical techniques, such as using a nonabsorbable or slowly absorbable material, 8 or prophylactic preperitoneal mesh placement after an open AAA repair. 5,8 Results with a retroperitoneal approach in AAA repair were inconsistent. 23,24 A meta-analysis published in 2007 described a 2.8-fold higher risk for incisional hernias (after midline transperitoneal approach) and 2.9-fold higher risk for inguinal hernias in patients with AAA. 6 The mean prevalence for inguinal and incisional hernias was 25 and 21%, respectively. 6 While the current study was in progress, Antoniou et al 5 and Takagi and Umemoto 25 repeated the meta-analysis, both with similar results.

Limitations
One limitation of the current study is that we obtained information about hernias by a questionnaire sent to the AAA patients. It is well known that mailed questionnaires have potential biases, since it is not clear if the answers are based on a proper clinical evaluation, or patient's self-perception. Furthermore, physical examination of hernias is known to have a relatively poor diagnostic accuracy. 26 In the current study, information about herniorrhaphies performed in the past was also collected. Most of the patients who reported having inguinal hernias (67%), had had herniorrhaphy, suggesting that the information can be considered reliable and is not based on only patient's self-perception or physical examination. Despite the differences in the study designs, our results agree with the previous studies (25% for inguinal hernias and 23% for incisional hernias). Furthermore, another study, which also used a questionnaire, found that 31% of AAA patients had inguinal hernias, 4 a prevalence very similar to ours. Further, the prevalence of inguinal hernias in our control group was relatively low (9%) when compared with previous studies (5-27%).
Another limitation of the study is that the response rate of sAAA patients was lower (51%) than that of patients in the other groups (85-92%) who underwent operations. This could have influenced the results, since it is possible that patients with existing hernias were more motivated to answer the questionnaire than the ones without hernias. The results presented here should be considered preliminary, since the sample sizes of the study groups were small. Further, stratifying the severity of AAA disease by therapy type (rAAA vs. EVAR vs. eAAA vs. sAAA) might not be as accurate as exact measurements of the AAA size or stress mapping of the aneurysm wall tension. However, this approach gives a general idea of the disease severity without the need of further, sometimes complicated and time-consuming measurements, and can therefore be easily undertaken in clinical practice. Finally, approximately half of the patients with ruptured AAA died perioperatively and could not be studied here.

Conclusions
In conclusion, our study confirms previous observations that patients with AAA have a relatively high prevalence of hernias and showed that the prevalence of hernias among AAA patients was not correlated with the size or rupture risk of AAA. Furthermore, prevalence of hernias was not increased among patients who had been operated compared with AAA patients who were under surveillance suggesting AORTA Vol. 9 No. 1/2021 © 2021. The Author(s). that surgical repair operation is not responsible for the increased hernia prevalence among AAA patients. These results suggest that the occurrence of hernias is not associated with a more aggressive AAA disease.

Funding
None.

Conflict of Interest
The authors declare no conflict of interest related to this article.