Introduction

Abdominal wall hernias, including ventral and groin hernias, are among the most common diseases encountered by the clinician with a prevalence of 2–20% following abdominal surgery [1,2,3]. The most common reasons for repair are pain, functional limitations, and poor cosmesis due to bulging. There has subsequently been an increasing focus on patient-centered outcomes (PCO) and quality of life (QOL) in hernia treatment [4,5,6,7].

Many abdominal wall-specific surveys have been developed to assess abdominal wall QOL (AW-QOL) and have been shown to have greater utility than general health surveys to evaluate patients with hernias [8]. The most widely utilized and studied surveys are the Activities Assessment Scale (AAS) and its derivatives. The AAS is a 13-question AW-QOL survey previously developed and validated in 2164 patients before and after hernia repair surgery [9]. The questionnaire investigates patient psychosocial QOL and abdominal wall function through questions covering mood, lifestyle, and physical activity [9,10,11]. Similar alternative versions of this survey have been developed, most recently with the modified AAS [8, 12], which is endorsed and utilized by the Americas Hernia Society and Ventral Hernia Outcomes Collaborative [13]. The modified AAS is an abdominal wall-specific survey and therefore has been shown to discriminate between abdominal wall-related pathology (i.e., hernias) and intraabdominal complaints, such as colorectal cancer and diverticulitis [8].

Multiple studies have demonstrated that the AAS and its derivatives are valid for evaluating AW-QOL among patients with hernias [8]. In addition, studies have shown that modified AAS scores improve following successful hernia repair [5, 8]. However, there have been no studies using these surveys to evaluate AW-QOL among individuals without abdominal wall pathology. Furthermore, it is unknown what the minimal clinically important difference (MCID) is in terms of improvement and its implications for the threshold at which ventral hernia repair (VHR) should be offered. MCID is defined as “the smallest difference in score in the domain of interest which patients perceive as beneficial and which would mandate, in the absence of troublesome side effects and excessive cost, a change in the patient’s management” [14]. The MCID constitutes a threshold for patient-reported outcome scores at which a patient would consider a specific change in score to be meaningful. Distribution-based methods of assessing the MCID are variable and include calculations determined from the standard error of the mean (SEM) or standard deviation (SD) among longitudinal AW-QOL scores of patients undergoing no major intervention [14, 15].

This is a prospective study of patients presenting to surgery clinics at a safety-net hospital. Our aims were (1) to establish the AW-QOL of patients with a clinical abdominal wall hernia versus those with no physical or radiologic evidence of hernia, (2) to determine the MCID with an abdominal wall-specific health survey (the modified AAS), and (3) to identify the baseline AW-QOL score at which patients might be expected to derive little or no AW-QOL benefit with VHR.

Materials and methods

Patient-centered outcomes data for patients with abdominal wall- and non-abdominal wall-related pathology were collected via a prospective, cross-sectional observational study design. Strengthening the Reporting of Observational Studies in Epidemiology Reporting Guidelines were followed [16] (Online Resource 1). All patients presenting to General Surgery Clinics (Breast, Colorectal, Surgical Oncology, and Hernia) from October 2016 to December 2016 at a single safety-net healthcare system hospital who had had a CT scan of their abdomen/pelvis within the last year and no intervening abdominal surgery were approached to complete the modified AAS as well as to undergo a standardized physical exam to assess for ventral and inguinal hernias. Patients completed the modified AAS surveys independently unless they claimed illiteracy, following which the survey was read to them by a research assistant. Patients were compensated with a $10 gift certificate for participation. Longitudinal data of patients with known abdominal wall pathology were also prospectively collected from Hernia Clinic patients from June 2014 to December 2016. The methodology of these databases has previously been reported and is available online [17,18,19] (Clinical Trials #NCT02457364, #NCT02365194, #NCT02363790). All of the latter patients presented initially to the Hernia Clinic for abdominal wall-related pathology, attended multiple clinic visits, and completed the modified AAS survey during each visit. Follow-up was encouraged through the administration of $10 gift certificates. All scores were normalized to a 1 to 100 score in which 1 was considered poor and 100 perfect. Modified AAS surveys missing over 2 responses were omitted from analysis. Other variables (demographics and comorbidities) were obtained through standardized intake forms based upon definitions utilized by the American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP). The local Institutional Review Boards approved all studies.

Of note, the longitudinal data were collected from all patients presenting to the Hernia Clinic who satisfied the inclusion criteria for the referenced protocols, not just those who had a CT. To determine the AW-QOL of the control population, we needed to verify the absence of any abdominal wall hernia. Current publications estimate that clinical exam alone misses 23–31% of abdominal wall hernias [3, 20]. We therefore mandated that these patients had a CT scan to ensure the absence of any hernia.

The primary outcome was AW-QOL measured using the modified AAS. The control population was comprised of general surgery patients presenting to clinic with no clinical or radiographic abdominal wall hernia. Because these individuals presented with no abdominal wall pathology, they were considered appropriate controls.

Secondary outcome was MCID. MCID was determined by 2 distribution-based methods. For calculation of the MCID, the equation frequently utilized in previous literature is SEM = SD X SQRT (1-test retest reliability coefficient) [21]. To determine a “slight” change in AW-QOL, we used a required difference of at least 1 SEM. To determine a “definite” change in AW-QOL, we utilized a mandatory difference of at least 1.96 SEM to reflect a 95% CI, in keeping with the recommendations of prior AW-QOL literature [14, 15, 22]. The MCID was determined only from the data of patients for whom modified AAS scores were obtained at multiple time intervals in the longitudinal studies and who had not undergone a major hospital-based intervention between testing intervals. Testing intervals ranged from 1 to 18 months.

In order to estimate the mean modified AAS score of each population (clinical hernia versus no hernia) with a standard deviation of 10 points, an acceptable error of 2 points, and a confidence level of 0.95, we determined that 97 individuals were needed per population to complete a modified AAS survey.

Modified AAS scores were reported as median (interquartile range). Controls, or patients with no clinical or radiologic hernia, were compared to study groups that included (1) patients with clinical and radiologic hernia, (2) patients 1 year following VHR, and (3) any patient with a recurrent ventral hernia (on initial presentation or following repair during the study period). Scores were compared using Kruskal–Wallis. Matched scores were compared using Wilcoxon matched-pairs sign-rank test. A p value less than 0.05 was considered statistically significant. All statistical analysis was performed using STATA version 14.1®.

Results

A total of 648 patients were enrolled (Fig. 1). Relevant patient demographics are presented in Table 1. Reasons for non-abdominal wall hernia-related presentation to surgery clinics included anal and colorectal (n = 82), breast (n = 3), miscellaneous (n = 10, included lipomas and trauma follow-ups), foregut (n = 6), hepatobiliary and pancreas (n = 33), and non-colorectal oncology (n = 6).

Fig. 1
figure 1

Flowsheet of all patients enrolled in general surgery and hernia clinics

Full size image
Table 1 Patient variables
Full size table

Patients with no hernia (n = 114) had modified AAS scores of 81.6 (50.4–94.4). In comparison, patients with a clinically apparent hernia (n = 534) had significantly lower modified AAS scores of 31.4 (12.6–58.7) (p < 0.001). The majority of clinical hernias were ventral as opposed to inguinal. Patients presenting with either a recurrent ventral or inguinal hernia (n = 90) had a baseline AAS score of 29.8 (14.7–55.0). There was no clinically or statistically significant (p = 0.550) difference in modified AAS scores between individuals having clinically palpable ventral and clinically palpable inguinal hernias.

Patients who presented with a hernia, underwent VHR, and completed 1-year follow-up after repair (n = 71) experienced significant improvements in AW-QOL [baseline 36.3 (15.6–63.3) versus post-VHR 73.3 (46.8–88.0), p < 0.001]. Scores following VHR were not statistically different than patients who initially presented with no hernia [81.6 (50.4–94.4), p = 0.123].

Mcid

A total of n = 150 patients with ventral hernias completed 2 to 6 sequential modified AAS surveys prior to undergoing any operative intervention. Baseline scores [28.2 (10.1–52.7)] did not significantly differ from follow-up scores [27.3 (9.2–64.2), p = 0.700] obtained up to 4–18 months later, prior to any operative intervention. On repeat surveys, patients showed a median maximal point difference in modified AAS score of 22.9 (9.9–42.9) when tested at repeated intervals. The median of the standard deviation for all patients was 11.7, and the individual intra-class correlation for all patients was 0.580, providing an SEM value of 7.6 for a “slight” change threshold of MCID and 14.9 for “definite” change threshold of MCID.

Who benefits from surgery?

Of the 71 patients for whom 1-year follow-up was performed following hernia repair, 62 (87.3%) demonstrated an improvement in modified AAS score. Above a modified AAS score of 81, the risk of worsening a patient’s QOL by surgery is higher than the chances of improvement (Table 2).

Table 2 Probability of improved abdominal wall quality of life 1 year following VHR
Full size table

Discussion

The AW-QOL among patients with a hernia is significantly worse compared to patients with no hernia. Surgical repair can improve AW-QOL to levels similar to that of the general population. However, not all patients achieve identical benefits with hernia surgery. In particular, patients with modified AAS scores of 81 or greater were unlikely to experience any AW-QOL improvements with surgery, and more than half had a worsened AW-QOL afterward.

These findings provide baseline reference data for abdominal wall studies assessing patient AW-QOL using the modified AAS or other similar derivatives. Currently, the Americas Hernia Society Quality Collaborative and Ventral Hernia Outcomes Collaborative utilize the modified AAS score [13]. Surprisingly, patients with no hernia had less than perfect modified AAS scores, suggesting that other factors unrelated to abdominal wall fascial defects may account for decreases in score. One reason for these lower than expected scores could be that only those individuals already seeking medical care in our surgical clinic were approached. These patients may have had other gastrointestinal complaints or prior abdominal surgeries that affected their AW-QOL. Further studies should (1) delineate the modified AAS scores of healthy individuals not currently seeking any surgery-related medical care, (2) determine what factors affect modified AAS scores in patients without hernias, and (3) identify the MCID of the modified AAS scores in patients without hernias. The median modified AAS scores of patients with a clinical hernia presenting to the General Surgery Clinics also differed from those of patients presenting to the Hernia Clinic, which may reflect severity of hernia (including hernia type, size, symptoms).

In addition, this work provides important information on the MCID in modified AAS scores that should be considered in designing future patient-centered trials. Previous work with modified AAS scores among a different cohort of patients, 90 consecutive individuals presenting to 1 surgeon at a single academic institution for elective VHR, has demonstrated a change in mean score of 14.3 points 6 months following hernia repair (47.2 ± 15.6 points at baseline versus 61.5 ± 13.7 6 months postoperative)5. In combination with the results of our study, this would suggest a slight but clinically significant improvement following hernia repair. In our study cohort, patients undergoing surgery experienced a 37-point improvement in modified AAS scores, suggesting a definite improvement in the majority of our patients.

Our data reflect patient-reported outcomes data in hernia repair and highlight the questionable benefit a patient with a high pre-surgical QOL may receive from surgical intervention. We have provided high-quality information to guide management decisions for patients with abdominal wall hernias looking to improve their QOL through operative repair. Decisions concerning VHR require consideration informed by each patient’s preexisting QOL. Patients already having high QOL should take into account the impact of surgery and postoperative complications on their QOL.

There are a number of limitations of this study. First, the generalizability of these findings is limited as surveys were obtained from patients seeking care at a safety-net hospital. Validation of these findings in other patients and populations is needed. However, results from multiple studies published in our patient population have been consistent with other results in the literature. In addition, although our patient population is largely underserved individuals (minority and/or low socioeconomic status), at least 1/3 to 1/2 of the American population are underserved. We also utilized distributional methods of approaching the MCID of patients with hernias. Anchor-based methods exist. It is unclear which technique is considered superior as both are widely used and reported. However, future studies should determine MCID utilizing different techniques in patients with and without hernias. Third, our study was limited to the modified AAS scale, which is a widely used AW-QOL survey endorsed by national hernia organizations. It is unclear if these findings are applicable to other abdominal wall-specific surveys.

Conclusions

Abdominal wall hernias are common pathologies, and patients have substantially worse abdominal wall QOL compared to patients with no hernia. The modified AAS is important for future clinical studies on hernia repair, but its clinical relevance and practicability require further elucidation. This is the first study to demonstrate that VHR can improve 1-year postsurgical abdominal wall QOL to levels similar to that of the general population. In addition, this study has demonstrated that the MCID when using the modified AAS survey is 7.6 points. Based on this, patients with a high baseline score should be carefully counseled about the lack of potential benefit in abdominal wall QOL from elective VHR.