Metachronous reoperation for recurrent and non-recurrent inguinal hernia after primary unilateral inguinal hernia repair: propensity score matched analysis of large US claims database

Inguinal hernia is a common occurrence affecting one in four men, with up to 750,000 inguinal hernia repairs (IHR) occurring yearly in the United States [1, 2]. Recurrence is a major clinical pitfall that affects about 10% of patients [3, 4], with an increase in the risk of recurrence and postoperative complications after a revision repair [5‐7]. Level II evidence has associated laparoscopic IHR (L-IHR) with higher recurrence rates when compared to an open approach; these studies included procedures performed early in the adoption curve of minimally invasive surgery (MIS) [8, 9]. Open IHR (O-IHR) was once the standard of care and is reported to have an equivalent rate of recurrence as compared to MIS [10]. L- IHR has since consistently been associated with quicker recovery time, less pain, and numbness [11, 12]. Parallel to surgeons’ gradual increased dexterity in laparoscopic surgery, additional techniques including robotic-assisted inguinal hernia repair (R-IHR) have been adopted. The impact of these minimally invasive surgery (MIS) techniques on recurrence rates and outcomes is less clear and further confounded by robotic implementation (R-IHR) [13].

L-IHR and R-IHR have demonstrated clinically similar outcomes [14], however robotic surgery is associated with increased expenditures [15‐18]. MIS-IHR can be performed in the extraperitoneal or Transabdominal Preperitoneal (TAPP) space. The latter repair offers the advantage of a posterior view that can evaluate bilateral inguinal anatomy, capable of addressing an undetected contralateral hernia. After IHR, patients can undergo reoperation for an inguinal hernia due to recurrence or metachronous contralateral inguinal hernia (MCIH). Studies on registry-data have grouped these together under “operative recurrence” and reported up to 4.7% operative recurrence for MIS, associated with overall 1% higher recurrence as compared to open repair at 5 years [19]. MCIH, however has been reported to occur in 5% of patients at 3 years [20]. The aim of our study was to compare reoperation for recurrence or MCIH at two years across surgical approaches.

A total of 73,870 patients were included in the analysis, of whom 3,421 (4.6%) had R-IHR, 30,858 (41.8%) had L-IHR, and 39,591 (53.6%) had O-IHR (Fig. 2). All baseline demographic and clinical characteristics were balanced between each pair of surgical approaches after PSM (Fig. 3). Sample characteristics after PSM are shown in Table 1. Overall, 84.9% (n = 55,294) of the patients were younger than 65, predominantly male (90.9%; n = 59,201). Most patients were from a metropolitan area (77.5%; n = 50,474) and over one-half (52.3%; n = 34,061) had an income of above $40,000 per year. Regarding insurance, patients were most frequently on a PPO (51%; n = 33,215) or HMO (13.1%; n = 8,531) plan. Regarding comorbidities, approximately 26% (n = 16,933) of the patients had hypertension, 7.5% (n = 4,885) had uncomplicated diabetes, 6.8% (n = 4,428) had chronic obstructive pulmonary disease, 3.2% (n = 2084) had mild liver disease, and 2.2% (n = 1432) had renal disease. Only 24.5% (n = 15,956) of patients had a Charlson comorbidity score of ≥ 1. Baseline characteristics before PSM are presented as supplementary (eTable 3). The overall rate of metachronous inguinal hernia reoperation at 2-years was 2.4% (n = 1,571). Reoperation for recurrence occurred in 41% (n = 644) of the total reoperations.

Compared to O-IHR, both MIS approaches showed lower risk of metachronous inguinal hernia reoperation at 2-years. L-IHR was associated with 16% lower reoperation risk than O-IHR (2.2% [n = 691] vs. 2.7% [n = 828], HR = 0.84, 95% CI = 0.74 to 0.93, p < 0.001) and R-IHR was associated with 42% lower reoperation risk than O-IHR (1.5% [n = 52] vs. 2.6% [n = 86], HR = 0.58, 95% CI = 0.42 to 0.82, p = 0.002). Furthermore, patients who underwent R-IHR when compared to L-IHR, had 33% lower risk of reoperation at 2-years (1.5% [n = 52] vs. 2.3% [n = 77], HR = 0.67, 95% CI = 0.47 to 0.96, p = 0.027). Results for PSM adjusted association between surgical approach and the study outcomes for each matched cohort are presented in Table 2.

L-IHR had a similar risk of reoperation for recurrence as O-IHR (1.1% [n = 328] vs. 0.9% [n = 292], HR = 1.13, p = 0.14). Patients who underwent R-IHR also had a similar risk of reoperation for recurrence with O-IHR (0.7% [n = 24] vs. 1.1% [n = 38], HR = 0.63, p = 0.079). Patients that underwent R-IHR, as compared to L-IHR, had 46% lower risk of reoperation for recurrence (0.7% [n = 24] vs. 1.3% [n = 44], HR = 0.54, 95% CI = 0.33 to 0.89, p = 0.016). Among the 328 L-IHR patients who had reoperation for recurrence, 21% (n = 69) of them were again performed laparoscopically. In contrast, among the 292 O-IHR patients who had reoperation for recurrence, 47.9% (n = 140) had open reoperation.

Regarding risk of reoperation for non-recurrent inguinal hernia repair, both MIS approaches were found to have lower risk of reoperation than open approach. L-IHR was associated with 32% lower risk of reoperation than O-IHR (1.2% [n = 363] vs. 1.7% [n = 536], HR = 0.68, 95% CI = 0.59 to 0.77, p < 0.001) and R-IHR was associated with 45% lower risk of reoperation than O-IHR (0.8% [n = 28] vs. 1.5% [n = 51], HR = 0.55, 95% CI = 0.35 to 0.87, p = 0.011). There was no significant difference in reoperation for non-recurrence between R-IHR and L-IHR (0.8% [n = 28] vs. 1.0% [n = 33], HR = 0.84, p = 0.51).

We found both MIS approaches had significantly higher expenditure for the index surgery compared to the open approach. The average mean difference (AMD) of expenditure for the index repair, compared to O-IHR, was about $3,391 higher for L-IHR (95% CI = $3,264 to $3,518, p < 0.001), and $4,137 higher for R-IHR (95% CI = $3,733 to $4,541, p < 0.001). Among the MIS approaches, R-IHR had on average $615 higher total index expenditure than L-IHR (95% CI = $191 to $1,039, p = 0.004). There was no difference in healthcare expenditures between surgical approaches in the 1-year or 2-years after initial repair. Results for comparison of healthcare expenditures between surgical approaches are presented in eTable 4.

On multivariable analysis, both MIS approaches were found to be protective for overall IH reoperation and metachronous primary IH when compared to O-IHR but not for recurrent IH reoperation (eTable 5). For recurrent IHR, risk factors included having BPH (HR 1.37; 95% CI 1.10–1.70; p = 0.005), tobacco use (HR 1.43; 95% CI 1.15–1.78; p = 0.001), and being overweight or having obesity (HR 1.30; 95% CI 1.03–1.63; p = 0.027). For metachronous IHR, protective factors included being treated in the south region (HR 0.69; 95% CI 0.57–0.82; p < 0.001) and undergoing IHR with laparoscopic (HR 0.67; 95% CI 0.58–0.76; p < 0.001) or robotic approach (HR 0.48; 95% CI 0.33–0.70; p < 0.001). Risk factors identified were diagnosis of BPH (HR 1.29; 95% CI 1.08–1.54; p = 0.006) and/or prostate cancer (HR 1.43; 95% CI 1.05–1.93; p = 0.022). For overall IH reoperation, identified protective factors were treatment in the south region (HR 0.78; 95% CI 0.68–0.90; p < 0.001), L-IHR (HR 0.84; 95% CI 0.76–0.92; p < 0.001), and R-IHR (HR 0.58; 95% CI 0.44–0.78; p < 0.001). Risk factors were diagnosis of BPH (HR 1.33; 95% CI 1.16–1.52; p < 0.001), Charlson Comorbidity Index ≥ 2 (HR 1.19; 95% CI 1.03–1.38; p = 0.019), tobacco use (HR 1.30; 95% CI 1.13–1.50; p < 0.001), and overweight status/obesity (HR 1.21; 95% CI 1.04–1.40; p = 0.013).

The current study demonstrates that MIS-IHR is associated with less reoperation for IHR at 2 years compared to O-IHR in the privately insured outpatient setting. Consistent with a recent metanalysis, R-IHR had less reoperation than L-IHR [10]. Expenses related to index procedure were 36.3% and 42.6% higher in L-IHR ($13,221) and R-IHR ($13,836), respectively compared to open approach ($9,699) whereas post-index expenses were similar between approaches. To our knowledge, the current study is the first of its kind regarding MCIH, recurrence, and expenses of IHR analyzed across a national insurance claims dataset.

Clinical recurrence is higher than operative recurrence, Murphy et al. addressed this specifically for the population in the United States and calculated operative recurrence to be around 10–13% [3]. The overall reoperation rate of 2.5% at two years of our study is possibly underpowered or representative of our specific patient population. This has been previously described as the healthy worker effect [25, 26]. To that point, privately insured patients might be influencing the age of the cohort as a sector of the working population, reflecting the minority of patients above age 65. Other socioeconomic factors could also impact the low recurrence rate, reflected for example in the low obesity in our cohort compared to national prevalence (13% vs. 40% [27]). We applied previously described methodology to determine operative recurrence [28], however we found that most of reoperations were coded as primary IHR. Using “reoperation for IHR” rather than operative recurrence was thus more appropriate. R-IHR were identified by an additional CPT code to that of L-IHR. Due to gradual adoption of new codes, we believe it is likely that some of the L-IHR were in fact R-IHR, possibly explaining in part why L-IHR and R-IHR had a comparable expense.

Development of minimally invasive techniques in IHR has enabled this procedure to be performed in a mainly outpatient setting with low morbidity. Increased dexterity with robotic approaches has also impacted practice, facilitating peritoneal flap closure and intracorporeal suturing. Preperitoneal MIS-IHR allows a posterior view of both myopectineal orifices, and this enables detection and repair of contralateral undetected (by imaging) inguinal hernias [29]. Patients at a high risk of developing a metachronous contralateral inguinal hernia might benefit from bilateral reinforcement while undergoing initial unilateral IHR. Complications avoided by this approach include a reoperation in patients at high perioperative risk as well as expenses related to a second hospitalization.

A recent metanalysis has described a 5.2% occurrence of MCIH at 3 years and of 17.1% at 10 years. This further emphasized the need to study the impact of MIS on reoperation and its potential advantages preventing MCIH in select cases. Concordantly, on further analysis of reoperations coded as primary IHR, our analysis found that MIS repairs had less reoperation than Open repairs. Interestingly, for reoperations coded as recurrent IHR, R-IHR had fewer reoperations than L-IHR. A recently published single center study evaluated a PSM similar to the our study exploring recurrence and cost [30]. Operative time, a frequently cited limitation to robotic use, was found to be, on average, only 15 min longer for R-IHR than L-IHR. At a follow up of two years, the measured recurrence rate for O-IHR was 3.6% vs. < 1% for minimally invasive IHR.

Current guidelines state that operative approach should be alternated when dealing with recurrent IHR [13]. However, in our cohort we observed that close to half of open IHR received a subsequent open IHR for recurrence and 20% of L-IHR had a subsequent laparoscopic approach. Previous literature recommend against L-IHR in the context of previous pelvic procedures; this relative contraindication is less clear on R-IHR [31, 32]. Our multivariable analysis showed that prostate cancer was a risk factor for Metachronous IHR. Interestingly, BPH was a risk factor for any reoperation and has been reported elsewhere [33]. Although medical and operative management of BPH would rarely influence the appearance of an inguinal hernia, speculation might lead to associate the increased Valsalva maneuvers and periodic pelvic examinations to be associated with a subsequent IHR.

The current study has limitations associated with the patient population and retrospective nature. The dataset studied is private insurance based which can limit generalizability regarding age group, income, and other social disparities that impact clinical outcome. The data also lacks clinical depth and hernia specific characteristics such as hernia width, mesh size, position of mesh, and technique used (preperitoneal vs. extraperitoneal), and in cases where laterality of both index and recurrent hernia repair was unknown from available codes, we could not ascertain the recurrent repair was ipsilateral. To assess whether the unmeasured confounding would nullify the observed associations in our study, we calculated E-values in post-hoc (eTable 6). The lowest E-value was 1.94 for L-IHR vs. O-IHR association with reoperation for non-recurrent IH, and the highest E-value was 2.43 for R-IHR vs. L-IHR association with reoperation for recurrent IH. These values signify the magnitude of association that an unmeasured confounding variable, like larger hernia size, would need to have at minimum, and be more prevalent in the group with higher recurrence rate in order to eliminate the observed relationship. The calculated E-values in our study are higher than the estimate in literature [34]. The objective data recollection from automated electronic medical records for billing purposes has significant limitations but remains the strength of the data source, allowing interinstitutional follow up focused on the patient, not the providers. Circumstances related to active patients the workforce, not captured in the data, such as early access to medical care could also be impacting outcomes. Janjua et al. have described the robotic platform to be used in more comorbid patients albeit with a higher income, with certain application in bilateral or recurrent cases [35].

The current study demonstrates that robotic IHR is associated with a lower risk of reoperations at 2 years after an initial repair as compared to L-IHR and O-IHR in the outpatient setting for an economically active population. MIS-IHR was associated with significantly higher expenses compared to O-IHR, with no difference between L-IHR and R-IHR.