This cohort study assesses the incidence and long-term outcomes after elective parastomal hernia repair.
Key Points
Question
What are the long-term outcomes after elective parastomal hernia repair?
Findings
In this cohort study of 17 625 patients undergoing elective parastomal hernia repair, the 5-year cumulative incidence of reoperation was 21.1% and was lowest for patients who underwent ostomy reversal (18.8%). In patients whose ostomy was not reversed, the hazard of repeat parastomal hernia repair was the same for patients whose ostomy was resited vs those whose ostomy was not resited.
Meaning
Reoperative recurrence following parastomal hernia repair is common and ostomy reversal may afford the best long-term outcome; for patients with parastomal hernias not eligible for reversal, local repair provides comparable outcomes with stoma resiting.
Abstract
Importance
Parastomal hernia is a challenging complication following ostomy creation; however, the incidence and long-term outcomes after elective parastomal hernia repair are poorly characterized.
Objective
To describe the incidence and long-term outcomes after elective parastomal hernia repair.
Design, Setting, and Participants
Using 100% Medicare claims, a retrospective cohort study of adult patients who underwent elective parastomal hernia repair between January 1, 2007, and December 31, 2015, was performed. Logistic regression and Cox proportional hazards models were used to evaluate mortality, complications, readmission, and reoperation after surgery. Analysis took place between February and May 2022.
Exposures
Parastomal hernia repair without ostomy resiting, parastomal hernia repair with ostomy resiting, and parastomal hernia repair with ostomy reversal.
Main Outcomes and Measures
Mortality, complications, and readmission within 30 days of surgery and reoperation for recurrence (parastomal or incisional hernia repair) up to 5 years after surgery.
Results
A total of 17 625 patients underwent elective parastomal hernia repair (mean [SD] age, 73.3 [9.1] years; 10 059 female individuals [57.1%]). Overall, 7315 patients (41.5%) underwent parastomal hernia repair without ostomy resiting, 2744 (15.6%) underwent parastomal hernia repair with ostomy resiting, and 7566 (42.9%) underwent parastomal hernia repair with ostomy reversal. In the 30 days after surgery, 676 patients (3.8%) died, 7088 (40.2%) had a complication, and 1740 (9.9%) were readmitted. The overall adjusted 5-year cumulative incidence of reoperation was 21.1% and was highest for patients who underwent parastomal hernia repair with ostomy resiting (25.3% [95% CI, 25.2%-25.4%]) compared with patients who underwent parastomal hernia repair with ostomy reversal (18.8% [95% CI, 18.7%-18.8%]). Among patients whose ostomy was not reversed, the hazard of repeat parastomal hernia repair was the same for patients whose ostomy was resited vs those whose ostomy was not resited (adjusted hazard ratio, 0.93 [95% CI, 0.81-1.06]).
Conclusions and Relevance
In this study, more than 1 in 5 patients underwent another parastomal or incisional hernia repair within 5 years of surgery. Although this was lowest for patients who underwent ostomy reversal at their index operation, ostomy resiting was not superior to local repair. Understanding the long-term outcomes of this common elective operation may help inform decision-making between patients and surgeons regarding appropriate operative approach and timing of surgery.
Introduction
Despite being the most common complication following ostomy creation, parastomal hernias are notoriously difficult to manage.1,2 Left untreated, parastomal hernias may cause pain, functional limitations, and life-threatening complications such as obstruction and strangulation that necessitate emergency surgery.3 While intervention is often deferred due to perceived prohibitively high complication rates, many patients undergo elective parastomal hernia repair, which has more than doubled in incidence in recent years.4,5 However, there is still little agreement on the optimal management of these hernias. Both technique and outcomes vary widely.6 Weighing the risks of an operation in patients with significant coexisting comorbidities with the benefits of preventing future complications has remained a persistent clinical challenge.
Underlying this challenge is that patient outcomes following parastomal hernia repair remain understudied and poorly understood. Studies of parastomal hernia repair are typically limited to small, single-center cohorts of fewer than 100 patients.7 Many of these studies aggregate outcomes for all patients regardless of repair type, even though patient outcomes differ depending on whether the ostomy was left in situ, resited, or reversed.8,9 In the few studies that have compared repair types, outcomes are limited to the immediate postoperative period; thus, there is very little evidence regarding long-term outcomes.10,11 This overall lack of evidence has led to a lack of consensus regarding the optimal management of parastomal hernias, which has resulted in variability in management and morbidity for patients. Understanding how to best approach patients with this condition requires accurate evaluation of comparative outcomes in a large, representative patient sample.
Within this context, we used 100% Medicare data to examine a national cohort of patients undergoing elective parastomal hernia repair. We specifically investigated long-term rates of reoperation as well as rates of 30-day mortality, complications, and readmission. Our goal was to describe contemporary outcomes to better inform management decisions between surgeons and patients. We hypothesized that rates of short- and long-term complications would be high even in recent years and would vary by procedure.
Methods
Data Source and Study Cohort
We performed a retrospective review of 100% fee-for-service Medicare claims to identify patients undergoing elective parastomal hernia repair between January 1, 2007, and December 31, 2015. Three categories of eligible patients were identified using International Classification of Diseases, Ninth Revision (ICD-9) procedure codes: parastomal hernia repair without ostomy resiting (46.42), parastomal hernia repair with ostomy resiting (46.43), and parastomal hernia repair with ostomy reversal (46.51 and 46.52). Patients were excluded if they underwent a parastomal hernia repair in the 6 months prior to their index operation, if their index parastomal hernia repair was an emergency operation, or if they underwent more than 1 procedure on the day of surgery. No patients had to be excluded due to missing data. This secondary analysis of deidentified Medicare claims was deemed exempt from regulation by the institutional review board of the University of Michigan, and the requirement for informed consent was waived. This study follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.12
Outcomes and Explanatory Variables
We evaluated short- and long-term outcomes. The primary 30-day outcomes were mortality, complications, and readmission. Mortality was defined as patient death within 30 days of surgery identified in the Medicare Beneficiary Denominator file. Complications were defined based on previously described methods for identifying relevant 30-day postoperative complications in administrative claims.13 Complication categories included splenic, hemorrhagic, anastomotic, wound-related, obstructive, pulmonary, cardiac, neurological, genitourinary, thromboembolic, and shock-related complications (eTable 1 in Supplement 1). Readmission was identified as hospitalization in an acute care facility for any reason within 30 days of the index operation.
The primary long-term outcome was reoperation within 5 years of index parastomal hernia repair. Reoperation was defined as any subsequent parastomal or incisional hernia repair identified using ICD-9 codes 46.42, 46.43, 46.51, 46.51, 53.51, 53.61, and 53.62. Although index repair was limited to elective procedures, reoperations included both elective and emergency reoperations.
Patient characteristics were abstracted from Medicare data and included patient age, sex, race, and 27 Elixhauser comorbidities.14 Procedural characteristics included the type of parastomal hernia repair performed, minimally invasive repair, mesh use, myofascial release, and year of operation. Mesh use was identified using Current Procedural Terminology code 49568 for open operations and presumed for minimally invasive operations since mesh use is implied for laparoscopic and robotic hernia repair and included in the procedure code for the operation. Myofascial release was identified using Current Procedural Terminology code 15734.
Statistical Analysis
Descriptive statistics were calculated for all patient characteristics, procedure characteristics, and outcomes. Analysis of 30-day outcomes involved separate multivariable logistic regression models for mortality, complications, and readmission. Each model included patient age, sex, race, comorbidities, parastomal hernia repair type, minimally invasive repair, mesh use, myofascial release, and year of surgery as covariates. These models were then used to estimate the adjusted rate of each 30-day outcome over the 3 types of parastomal hernia repair. Analysis of long-term reoperation involved a Cox proportional hazards model that included patient age, sex, race, comorbidities, parastomal hernia repair type, minimally invasive repair, mesh use, and myofascial release. The proportional hazards assumption was tested using Schoenfeld residuals, and covariates that violated the proportional hazards assumption were interacted with a variable for the natural log of time.15 Estimates from the Cox proportional hazards model were then used to calculate the adjusted cumulative incidence of reoperation up to 5 years after the index operation. An additional analysis was performed excluding patients who underwent ostomy reversal to determine the incidence of subsequent parastomal hernia repair in patients whose ostomy was left in situ or resited at the index operation. Patients were censored if they died, disenrolled from Medicare, or reached the end of the study period.
All statistical analyses were performed using SAS version 9.4 (SAS Institute) and tests of statistical significance were 2-sided with an α of .05. Analysis took place between February and May 2022.
Results
A total of 17 625 patients underwent elective parastomal hernia repair during the study period (Table 1). The mean (SD) age was 73.3 (9.1) years, and 10 059 patients (57.1%) were female. Overall, there were 975 Black individuals (5.5%), 15 925 White individuals (90.4%), and 725 individuals (4.1%) of other race (ie, Asian, Hispanic, North American Native, other, and unknown). Comorbidities were common, with the most prevalent being hypertension (14 895 [84.5%]), fluid and electrolyte disorders (13 208 [74.9%]), deficiency anemias (9865 [56.0%]), chronic pulmonary disease (6999 [39.7%]), malnutrition/underweight (6943 [39.4%]), and diabetes (6176 [35.0%]).
Table 1. Cohort Characteristics.
| Characteristic | No. (%) | |||
|---|---|---|---|---|
| All patients | PHR | PHR with ostomy reversal | ||
| Without resiting | With resiting | |||
| No. | 17 625 | 7315 | 2744 | 7566 |
| Patient characteristics | ||||
| Age, mean (SD), y | 73.3 (9.1) | 73.8 (9.2) | 74.3 (9.5) | 72.5 (8.8) |
| Female | 10 059 (57.1) | 4213 (57.6) | 1498 (54.6) | 4348 (57.5) |
| Male | 7566 (42.9) | 3102 (42.4) | 1246 (45.4) | 3218 (42.5) |
| Race | ||||
| Black | 975 (5.5) | 314 (4.3) | 203 (7.4) | 458 (6.1) |
| White | 15 925 (90.4) | 6713 (91.8) | 2426 (88.4) | 6786 (89.7) |
| Othera | 725 (4.1) | 288 (3.9) | 115 (4.2) | 322 (4.3) |
| Comorbidities | ||||
| Hypertension | 14 895 (84.5) | 6147 (84.0) | 2316 (84.4) | 6432 (85.0) |
| Fluid and electrolyte disorders | 13 208 (74.9) | 5190 (80.0) | 2136 (77.8) | 5882 (77.7) |
| Deficiency anemias | 9865 (56.0) | 3864 (52.8) | 1680 (61.2) | 4321 (57.1) |
| Chronic pulmonary disease | 6999 (39.7) | 2816 (38.5) | 1116 (40.7) | 3067 (40.5) |
| Weight loss | 6943 (39.4) | 2433 (33.3) | 1298 (47.3) | 3212 (42.5) |
| Diabetes without chronic complications | 6176 (35.0) | 2499 (34.2) | 1034 (37.7) | 2643 (34.9) |
| Depression | 5496 (31.2) | 2207 (30.2) | 934 (34.0) | 2355 (31.1) |
| Kidney failure | 5302 (30.1) | 2097 (28.7) | 856 (31.2) | 2349 (31.1) |
| Congestive heart failure | 5096 (28.9) | 1984 (27.1) | 888 (32.4) | 2224 (29.4) |
| Solid tumor without metastasis | 4627 (26.3) | 1905 (26.0) | 908 (33.1) | 1814 (24.0) |
| Hypothyroidism | 4262 (24.2) | 1747 (23.9) | 662 (24.1) | 1853 (24.5) |
| Peripheral vascular disease | 3514 (19.9) | 1343 (18.4) | 596 (21.7) | 1575 (20.8) |
| Metastatic cancer | 3267 (18.5) | 1339 (18.3) | 675 (24.6) | 1253 (16.6) |
| Other neurological disorders | 3248 (18.4) | 1262 (17.3) | 602 (21.9) | 1384 (18.3) |
| Coagulopathy | 2833 (16.1) | 1111 (15.2) | 441 (16.1) | 1281 (16.9) |
| Valvular disease | 2750 (15.6) | 1100 (15.0) | 439 (16.0) | 1211 (16.0) |
| Pulmonary circulation disease | 2245 (12.7) | 886 (12.1) | 355 (12.9) | 1004 (13.3) |
| Diabetes with chronic complications | 1864 (10.6) | 742 (10.1) | 334 (12.2) | 788 (10.4) |
| Psychoses | 1783 (10.1) | 655 (9.0) | 318 (11.6) | 810 (10.7) |
| Chronic blood loss anemia | 1713 (9.7) | 666 (9.1) | 331 (12.1) | 716 (9.5) |
| Rheumatoid arthritis/collagen vas | 1371 (7.8) | 543 (7.4) | 181 (6.6) | 647 (8.6) |
| Paralysis | 1187 (6.7) | 462 (6.3) | 301 (11.0) | 424 (5.6) |
| Liver disease | 803 (4.6) | 335 (4.6) | 114 (4.2) | 354 (4.7) |
| Alcohol misuse | 492 (2.8) | 176 (2.4) | 78 (2.8) | 238 (3.2) |
| Drug misuse | 344 (2.0) | 133 (1.8) | 51 (1.9) | 160 (2.1) |
| Lymphoma | 310 (1.8) | 123 (1.7) | 51 (1.9) | 136 (1.8) |
| Peptic ulcer disease and bleeding | 51 (0.3) | 24 (0.3) | 7 (0.3) | 20 (0.3) |
| Acquired immune deficiency syndrome | 38 (0.2) | 7 (0.1) | 9 (0.3) | 22 (0.3) |
| Year of surgery | ||||
| 2007 | 2111 (12.0) | 900 (12.3) | 365 (13.3) | 846 (11.2) |
| 2008 | 1996 (11.3) | 854 (11.7) | 325 (11.8) | 817 (10.8) |
| 2009 | 1921(10.9) | 859 (11.7) | 305 (11.1) | 757 (10.0) |
| 2010 | 1999 (11.3) | 817 (11.2) | 290 (10.6) | 892 (11.8) |
| 2011 | 2048 (11.6) | 877 (12.0) | 322 (11.7) | 849 (11.2) |
| 2012 | 2031 (11.5) | 868 (11.9) | 296 (10.8) | 867 (11.5) |
| 2013 | 1981(11.2) | 830 (11.4) | 272 (9.9) | 879 (11.6) |
| 2014 | 1892 (10.7) | 731 (10.0) | 265 (9.7) | 896 (11.8) |
| 2015 | 1646 (9.3) | 579 (7.9) | 304 (11.1) | 763 (10.1) |
| Procedure characteristics | ||||
| Minimally invasive repair | 1215 (6.9) | 745 (10.2) | 59 (2.2) | 411 (5.4) |
| Mesh use | 3849 (21.8) | 2683 (36.7) | 173 (6.3) | 993 (13.1) |
| Myofascial release | 692 (3.9) | 366 (5.0) | 44 (1.6) | 282 (3.7) |
Abbreviation: PHR, parastomal hernia repair.
Other includes Asian, Hispanic, North American Native, other, and unknown.
Overall, 7315 patients (41.5%) underwent parastomal hernia repair without ostomy resiting, 2744 (15.6%) underwent parastomal hernia repair with ostomy resiting, and 7566 (42.9%) underwent parastomal hernia repair with ostomy reversal. In addition, 1215 procedures (6.9%) were performed minimally invasively, mesh was used in 3849 procedures (21.8%), and myofascial release was performed in 692 procedures (3.9%). The median (IQR) follow-up after surgery was 2.5 (0.9-4.8) years.
In the 30 days after surgery, 676 patients (3.8%) died, 7088 (40.2%) had a complication, and 1740 (9.9%) were readmitted. After accounting for differences in patient and operative characteristics using multivariable logistic regression, compared with parastomal hernia repair without ostomy resiting, parastomal hernia repair with ostomy resiting was associated with a higher adjusted rate of 30-day mortality (5.4% [95% CI, 4.5%-6.3%] vs 3.5% [95% CI, 3.1%-3.9%]) but a slightly lower adjusted rate of 30-day complications (35.4% [95% CI, 33.4%-37.4%] vs 40.4% [95% CI, 39.2%-41.6%]) (Figure 1). There was no difference in 30-day mortality or complications between parastomal hernia repair with ostomy reversal and parastomal hernia repair without ostomy resiting. There was also no difference in 30-day readmission between all 3 operation categories (Table 2).
Figure 1. 30-Day Outcomes After Elective Parastomal Hernia Repair (PHR).
Risk-adjusted rates estimated using multivariable logistic regression that included patient age, sex, race, Elixhauser comorbidities, operation type (PHR without ostomy resiting, PHR with ostomy resiting, PHR with ostomy reversal), surgical approach (open vs minimally invasive), mesh use, myofascial release, and year of surgery.
Table 2. Multivariable Logistic Regression Models for 30-Day Outcomes.
| Characteristic | Mortality | Complications | Readmission | |||
|---|---|---|---|---|---|---|
| Adjusted OR (95% CI) | P value | OR (95% CI) | P value | OR (95% CI) | P value | |
| Procedure characteristics | ||||||
| Parastomal hernia repair without ostomy resiting | 1 [Reference] | NA | 1 [Reference] | NA | 1 [Reference] | NA |
| Parastomal hernia repair with ostomy resiting | 1.61 (1.31-1.98) | <.001 | 1.61 (1.31-1.98) | <.001 | 0.99 (0.85-1.15) | .89 |
| Parastomal hernia repair with ostomy reversal | 0.98 (0.81-1.18) | .83 | 0.98 (0.81-1.18) | .83 | 0.95 (0.84-1.06) | .35 |
| Minimally invasive repair | 1.14 (0.72-1.82) | .57 | 1.14 (0.72-1.82) | .57 | 0.67 (0.52-0.87) | .003 |
| Mesh use | 0.52 (0.38-0.71) | <.001 | 0.52 (0.38-0.71) | <.001 | 1.20 (1.04-1.39) | .01 |
| Myofascial release | 0.81 (0.45-1.46) | .49 | 0.81 (0.45-1.46) | .49 | 1.38 (1.05-1.80) | .02 |
| Year of surgery | 1.03 (1.00-1.06) | .04 | 1.03 (1.00-1.06) | .04 | 0.95 (0.93-0.96) | <.001 |
| Patient characteristics | ||||||
| Age | 1.06 (1.05-1.07) | <.001 | 1.06 (1.05-1.07) | <.001 | 1.00 (0.99-1.00) | .15 |
| Female | 0.95 (0.80-1.12) | .51 | 0.95 (0.80-1.12) | .51 | 1.22 (1.09-1.36) | .001 |
| Race | ||||||
| Black | 1.49 (1.08-2.04) | .01 | 1.49 (1.08-2.04) | .01 | 0.99 (0.79-1.23) | .93 |
| White | 1 [Reference] | NA | 1 [Reference] | NA | 1 [Reference] | NA |
| Othera | 0.81 (0.53-1.25) | .35 | 0.81 (0.53-1.25) | .35 | 1.11 (0.86-1.42) | .42 |
| Comorbidities | ||||||
| Hypertension | 0.51 (0.41-0.62) | <.001 | 0.51 (0.41-0.62) | <.001 | 1.07 (0.91-1.26) | .41 |
| Fluid and electrolyte disorders | 1.14 (0.92-1.42) | .22 | 1.14 (0.92-1.42) | .22 | 1.30 (1.12-1.51) | .001 |
| Deficiency anemias | 0.69 (0.58-0.82) | <.001 | 0.69 (0.58-0.82) | <.001 | 1.45 (1.29-1.64) | <.001 |
| Chronic pulmonary disease | 1.20 (1.01-1.42) | .03 | 1.20 (1.01-1.42) | .03 | 1.03 (0.93-1.15) | .56 |
| Weight loss | 1.38 (1.15-1.64) | <.001 | 1.38 (1.15-1.64) | <.001 | 1.34 (1.20-1.50) | <.001 |
| Diabetes without chronic complications | 1.27 (1.06-1.52) | .009 | 1.27 (1.06-1.52) | .009 | 1.01 (0.89-1.14) | .90 |
| Depression | 0.74 (0.61-0.90) | .002 | 0.74 (0.61-0.90) | .002 | 1.13 (1.01-1.26) | .03 |
| Kidney failure | 1.07 (0.89-1.29) | .45 | 1.07 (0.89-1.29) | .45 | 1.20 (1.07-1.34) | .001 |
| Congestive heart failure | 1.26 (1.03-1.54) | .02 | 1.26 (1.03-1.54) | .02 | 1.10 (0.97-1.25) | .14 |
| Solid tumor without metastasis | 0.59 (0.48-0.73) | <.001 | 0.59 (0.48-0.73) | <.001 | 1.08 (0.96-1.22) | .20 |
| Hypothyroidism | 0.91 (0.74-1.10) | .33 | 0.91 (0.74-1.10) | .33 | 0.98 (0.87-1.10) | .74 |
| Peripheral vascular disease | 1.21 (1.00-1.48) | .06 | 1.21 (1.00-1.48) | .06 | 1.04 (0.92-1.18) | .54 |
| Metastatic cancer | 1.26 (1.01-1.57) | .04 | 1.26 (1.01-1.57) | .04 | 1.01 (0.88-1.16) | .86 |
| Other neurological disorders | 0.75 (0.59-0.95) | .02 | 0.75 (0.59-0.95) | .02 | 1.12 (0.99-1.26) | .07 |
| Coagulopathy | 1.31 (1.05-1.64) | .02 | 1.31 (1.05-1.64) | .02 | 0.96 (0.84-1.09) | .53 |
| Valvular disease | 0.86 (0.68-1.09) | .22 | 0.86 (0.68-1.09) | .22 | 1.03 (0.90-1.18) | .66 |
| Pulmonary circulation disease | 1.12 (0.87-1.43) | .39 | 1.12 (0.87-1.43) | .39 | 1.03 (0.89-1.20) | .66 |
| Diabetes with chronic complications | 0.93 (0.69-1.24) | .60 | 0.93 (0.69-1.24) | .60 | 0.98 (0.83-1.15) | .82 |
| Psychoses | 1.10 (0.82-1.48) | .53 | 1.10 (0.82-1.48) | .53 | 1.03 (0.88-1.21) | .70 |
| Chronic blood loss anemia | 0.64 (0.48-0.87) | .004 | 0.64 (0.48-0.87) | .004 | 1.15 (0.98-1.34) | .08 |
| Rheumatoid arthritis/collagen vas | 0.70 (0.50-0.99) | .04 | 0.70 (0.50-0.99) | .04 | 1.10 (0.92-1.31) | .31 |
| Paralysis | 1.22 (0.88-1.68) | .23 | 1.22 (0.88-1.68) | .23 | 1.05 (0.87-1.27) | .60 |
| Liver disease | 0.99 (0.65-1.53) | .98 | 0.99 (0.65-1.53) | .98 | 1.38 (1.11-1.70) | .003 |
| Alcohol misuse | 0.87 (0.51-1.48) | .61 | 0.87 (0.51-1.48) | .61 | 0.69 (0.50-0.96) | .03 |
| Drug misuse | 1.37 (0.72-2.61) | .34 | 1.37 (0.72-2.61) | .34 | 1.18 (0.85-1.63) | .33 |
| Lymphoma | 0.60 (0.29-1.23) | .16 | 0.60 (0.29-1.23) | .16 | 1.06 (0.73-1.54) | .75 |
| Peptic ulcer disease with bleeding | 1.82 (0.55-5.97) | .32 | 1.82 (0.55-5.97) | .32 | 1.45 (0.70-3.03) | .32 |
| Acquired immune deficiency syndrome | 1.02 (0.12-8.34) | .99 | 1.02 (0.12-8.34) | .99 | 0.97 (0.34-2.78) | .96 |
Abbreviations: NA, not applicable; OR, odds ratio.
Other includes Asian, Hispanic, North American Native, other, and unknown.
During the follow-up period, 2814 patients (16.0%) underwent another parastomal or incisional hernia repair. The overall adjusted 5-year cumulative incidence of reoperation was 21.1% (95% CI, 21.0%-21.2%). Compared with parastomal hernia repair without ostomy resiting, parastomal hernia repair with ostomy resiting was associated with a higher hazard of reoperation (adjusted hazard ratio [aHR], 1.15 [95% CI, 1.02-1.29]) while parastomal hernia repair with ostomy reversal was associated with a lower hazard of reoperation (aHR, 0.81 [95% CI, 0.74-0.90]) (Table 3). Mesh use was also associated with a lower hazard of reoperation (aHR, 0.88 [95% CI, 0.79-0.98]). Five years after surgery, the adjusted cumulative incidence of reoperation was 22.5% (95% CI, 22.5%-22.6%) for parastomal hernia repair without ostomy resiting, 25.3% (95% CI, 25.2%-25.4%) for parastomal hernia repair with ostomy resiting, and 18.8% (95% CI, 18.7%-18.8%) for parastomal hernia repair with ostomy reversal (Figure 2A).
Table 3. Cox Proportional Hazards Models for Postoperative Reoperation.
| Characteristic | Reoperation | |
|---|---|---|
| Adjusted HR (95% CI) | P value | |
| Procedure characteristics | ||
| Parastomal hernia repair without ostomy resiting | 1 [Reference] | NA |
| Parastomal hernia repair with ostomy resiting | 0.92 (0.80-1.05) | .22 |
| Parastomal hernia repair with ostomy reversal | 0.14 (0.12-0.16) | <.001 |
| Minimally invasive repair | 0.95 (0.76-1.17) | .62 |
| Mesh use | 0.84 (0.74-0.95) | .006 |
| Myofascial release | 0.64 (0.48-0.85) | .002 |
| Patient characteristics | ||
| Age | 0.98 (0.97-0.98) | <.001 |
| Female | 1.00 (0.91-1.11) | .98 |
| Race | ||
| Black | 0.71 (0.56-0.91) | .003 |
| White | 1 [Reference] | NA |
| Othera | 1.15 (0.91-1.45) | .25 |
| Comorbidities | ||
| Hypertension | 1.16 (1.00-1.35) | .050 |
| Fluid and electrolyte disorders | 1.35 (1.18-1.53) | <.001 |
| Deficiency anemias | 1.14 (1.03-1.27) | .01 |
| Chronic pulmonary disease | 1.10 (1.00-1.22) | .053 |
| Weight loss | 1.14 (1.03-1.27) | .01 |
| Diabetes without chronic complications | 0.94 (0.84-1.05) | .26 |
| Depression | 1.05 (0.94-1.18) | .37 |
| Kidney failure | 0.79 (0.70-0.89) | <.001 |
| Congestive heart failure | 0.99 (0.87-1.12) | .87 |
| Solid tumor without metastasis | 1.09 (0.97-1.22) | .13 |
| Hypothyroidism | 1.05 (0.94-1.18) | .39 |
| Peripheral vascular disease | 1.14 (1.01-1.30) | .04 |
| Metastatic cancer | 1.00 (0.87-1.15) | >.99 |
| Other neurological disorders | 0.94 (0.83-1.07) | .34 |
| Coagulopathy | 1.03 (0.90-1.17) | .69 |
| Valvular disease | 1.08 (0.94-1.24) | .26 |
| Pulmonary circulation disease | 0.87 (0.74-1.01) | .07 |
| Diabetes with chronic complications | 1.04 (0.88-1.22) | .67 |
| Psychoses | 1.08 (0.92-1.26) | .36 |
| Chronic blood loss anemia | 1.02 (0.87-1.20) | .78 |
| Rheumatoid arthritis/collagen vas | 0.94 (0.78-1.13) | .51 |
| Paralysis | 0.83 (0.69-1.00) | .06 |
| Liver disease | 0.87 (0.71-1.09) | .23 |
| Alcohol misuse | 1.18 (0.92-1.52) | .20 |
| Drug misuse | 0.85 (0.60-1.21) | .37 |
| Lymphoma | 1.00 (0.70-1.43) | >.99 |
| Peptic ulcer disease with bleeding | 0.60 (0.23-1.58) | .30 |
| Acquired immune deficiency syndrome | 1.77 (0.76-4.14) | .19 |
Abbreviations: HR, hazard ratio; NA, not applicable.
Other includes Asian, Hispanic, North American Native, other, and unknown.
Figure 2. Adjusted Cumulative Incidence of Reoperation and Repeat Parastomal Hernia Repair (PHR) Up to 5 Years After PHR.
Adjusted cumulative incidence estimated from Cox proportional hazards models that included patient age, sex, race, Elixhauser comorbidities, operation type (PHR without ostomy resiting, PHR with ostomy resiting, PHR with ostomy reversal), surgical approach (open vs minimally invasive), mesh use, and myofascial release. Patients were censored if they died, disenrolled from Medicare, or reached the end of the study period.
In a subanalysis of patients who did not undergo ostomy reversal at their index parastomal hernia repair, there was no difference in the hazard of subsequent parastomal hernia repair between patients whose ostomy was not resited and patients whose ostomy was resited (aHR, 0.93 [95% CI, 0.81-1.06]) (eTable 2 in Supplement 1). However, mesh use (aHR, 0.83 [95% CI, 0.73-0.94]) and myofascial release (aHR, 0.54 [95% CI, 0.38-0.76]) were associated with a lower hazard of subsequent parastomal hernia repair. Five years after surgery, the adjusted cumulative incidence of undergoing another parastomal hernia repair was 20.7% (95% CI, 20.6%-20.7%) overall, 20.9% (95% CI, 20.8%-20.9%) for patients who underwent parastomal hernia repair without ostomy resiting, and 19.5% (95% CI, 19.4%-19.6%) for patients who underwent parastomal hernia repair with ostomy resiting (Figure 2B).
Discussion
In this national cohort of patients undergoing elective parastomal hernia repair, we found that both short- and long-term outcomes remained generally poor. First, nearly 4% of patients died and over one-third of patients had a complication within 30 days of surgery, likely reflecting the significant comorbidities present in this cohort. Second, more than 1 in 5 patients underwent another parastomal or incisional hernia repair in the 5 years after their index operation. While the rate of reoperation was lowest for patients whose ostomy was reversed, among patients whose ostomy was not reversed, repair with and without ostomy resiting afforded relatively equivalent rates of reoperation. Importantly, reoperation for hernia recurrence underestimates the rate of true clinical recurrence by 4-fold since many patients who experience hernia recurrence do not undergo reoperation.16 Taken together, these results suggest that morbidity remains common following elective parastomal hernia repair. Differences in long-term outcomes between operation types may help inform decision-making regarding parastomal hernia management.
To our knowledge, this is the largest study to date to evaluate the long-term rate of reoperation following parastomal hernia repair. We found that patients whose ostomy was reversed had the lowest risk of long-term reoperation following surgery. This makes intuitive sense and is consistent with prior studies that have also demonstrated lower hernia recurrence rates following ostomy reversal.17,18 Accordingly, for patients who may be future candidates for ostomy reversal, waiting to perform concurrent parastomal hernia repair and ostomy reversal will likely afford the most durable long-term outcome. Gavigan et al4 previously found that the annual rate of parastomal hernia repair with ostomy reversal was increasing, suggesting that this management approach is being adopted more broadly. We also found that parastomal hernia repair with ostomy resiting was the least common operation. Although it was previously believed that this approach was superior to parastomal hernia repair without resiting, growing evidence of high recurrence rates and postoperative morbidity has resulted in this approach falling out of favor.19 Indeed, in the current study, patients who underwent ostomy resiting had the highest rates of 30-day mortality and 5-year reoperation compared with patients who underwent parastomal hernia repair without resiting or with ostomy reversal.
Elective management of parastomal hernias is challenging for a variety of reasons. First, patients who develop parastomal hernias often have a number of comorbidities that are not only risk factors for parastomal hernias development, but increase the risk of adverse perioperative outcomes.20 In the current cohort, a majority of patients had hypertension and anemia, and over one-third of patients had chronic pulmonary disease, malnutrition, and diabetes. The high prevalence of comorbid conditions in this population likely underlies the relatively high rate of 30-day mortality, complications, and readmission. These generally poor 30-day outcomes are consistent with previously reported rates of short-term mortality and morbidity after parastomal hernia repair. Using the Danish Hernia Database, Helgstrand et al21 found that 6.3% of patients died within 30 days of surgery, which is likely higher than the rate observed here since that study included patients undergoing emergency repair and excluded patients undergoing ostomy reversal. Nevertheless, a 3.8% 30-day mortality rate after an elective operation is substantial.
Regarding long-term outcomes, the results of this study support the hypothesis that, in general, patients with more comorbidities are at increased risk of hernia recurrence. Specifically, we found that hypertension, fluid and electrolyte disorders, anemias, pulmonary disease, weight loss, and vascular disease were associated with increased odds of repeat parastomal hernia repair. While these risk factors may seem nonintuitively associated with hernia recurrence, their association with worse perioperative outcomes in general likely underlies their association with hernia recurrence. Conversely, the lack of an association with more intuitive risk factors such as diabetes is likely associated with patient selection. For example, there are grade A recommendations to defer hernia repair in patients with uncontrolled diabetes.22 Finally, although high body mass index was not included in the claims used for this study, prior work has found that obesity is associated with short-term complications and long-term recurrence following parastomal hernia repair.23
In addition to patient morbidity, there is also very little consensus regarding optimal management of parastomal hernias, which is reflected in the heterogeneity of repairs in the current study. For example, although a number of small studies have suggested that minimally invasive repair may afford superior short-term outcomes, only 6.9% of patients underwent minimally invasive repair in our study cohort. Other studies have reported similarly low rates of minimally invasive repair, highlighting the overall low adoption of this technique for this condition.24 Similarly, there is compelling evidence that mesh significantly reduces the risk of postoperative hernia recurrence.25 In fact, one of the only parastomal hernia management principles to receive a strong recommendation from the European Hernia Society is the avoidance of suture repair alone.26 Despite this, mesh was used in only 21.8% of repairs in the current study. Interestingly, this stands in stark contrast to a study using data from the Abdominal Core Health Quality Collaborative in which mesh was used in 94% of parastomal hernia repairs.27 This discrepancy underscores the variability with which these hernias are managed. Ultimately, the low rates of mesh use and minimally invasive repair suggest that there are opportunities for practice change that may ultimately improve patients’ short- and long-term outcomes.
The results of this study may help inform the management of patients who develop parastomal hernia. It is challenging to weigh the risks and benefits of operative repair vs nonoperative management given the dearth of literature on patient outcomes related to parastomal hernia. In the current study, we found that operative repair was associated with relatively high morbidity. However, nonoperative management is not without drawbacks. Small studies have found that between 15% to 20% of patients initially managed nonoperatively eventually experience incarceration, obstruction, or strangulation necessitating emergency repair.28 Even among patients who do not experience life-threatening complications, one-third or more of patients with parastomal hernias develop abdominal pain, intermittent obstruction, decreased physical function, dissatisfaction with physical appearance, and difficulty using ostomy appliances.29,30 Further study is clearly needed to better describe the long-term burden and risks for patients who develop a parastomal hernia.
Limitations
This study has several limitations. First, its retrospective, observational nature introduces the possibility of selection bias. For example, patients selected for different types of hernia repair may have had distinct characteristics that influenced their likelihood of undergoing one operation vs another, although we used appropriate methods to control for observed confounding due to patient selection. Another limitation is that this study only evaluated outcomes among patients who underwent parastomal hernia repair. While this was the stated aim of the study, many patients who develop parastomal hernias are unable to or choose not to pursue surgical repair. It is equally important to understand outcomes in these patients as well. Another limitation is that because we used administrative claims data to perform our analysis, we were unable to evaluate more granular details of parastomal hernia repair. For example, we could not determine the type of ostomy patients had, which is an important limitation since rates of complications and recurrence after repair appear to be highest for colostomies and lowest for urostomies.10 Similarly, we could not identify operative details such as whether a Sugarbaker or keyhole repair was performed or what the anatomic characteristics of the hernia were, such as its size and location. Furthermore, Medicare claims contain no information regarding the type of mesh used, although prior work suggests that there may be no difference between mesh types in parastomal hernia recurrence.31 This study also did not include information regarding surgeon experience or years of training. All of these factors influence patient outcomes, and their association with long-term outcomes should be evaluated.2,32 The use of Medicare data also limits the generalizability of our findings. While these results inform our understanding of parastomal hernia repair in older adults, where ostomies are not uncommon, future studies should also investigate outcomes in younger patients with fewer comorbidities. Finally, while this study identified ostomy reversal as having the lowest rate of reoperation, there are patients with permanent ostomies who are simply not candidates for this approach. In this case, the results of the current study may still help inform patients regarding expected long-term outcomes.
Conclusions
In this national study of patients undergoing elective parastomal hernia repair, short-term complications and long-term reoperations were common. Overall, more than 1 in 5 patients underwent another parastomal or incisional hernia repair within 5 years of surgery. Although this was lowest for patients who underwent ostomy reversal at their index operation, ostomy resiting was not superior to local repair. Understanding the short- and long-term outcomes of this common elective operation may help inform decision-making between patients and surgeons regarding appropriate management and timing of surgery.
eTable 1. Complication categories and codes
eTable 2. Cox proportional hazards models for repeat parastomal hernia repair among patients who did not undergo ostomy reversal at index operation
Data Sharing Statement
References
- 1.Hotouras A, Murphy J, Thaha M, Chan CL. The persistent challenge of parastomal herniation: a review of the literature and future developments. Colorectal Dis. 2013;15(5):e202-e214. doi: 10.1111/codi.12156 [DOI] [PubMed] [Google Scholar]
- 2.Carne PW, Robertson GM, Frizelle FA. Parastomal hernia. Br J Surg. 2003;90(7):784-793. doi: 10.1002/bjs.4220 [DOI] [PubMed] [Google Scholar]
- 3.van Dijk SM, Timmermans L, Deerenberg EB, et al. Parastomal hernia: impact on quality of life? World J Surg. 2015;39(10):2595-2601. doi: 10.1007/s00268-015-3107-4 [DOI] [PubMed] [Google Scholar]
- 4.Gavigan T, Rozario N, Matthews B, Reinke C. Trends in parastomal hernia repair in the United States: a 14-y review. J Surg Res. 2017;218:78-85. doi: 10.1016/j.jss.2017.04.030 [DOI] [PubMed] [Google Scholar]
- 5.Gregg ZA, Dao HE, Schechter S, Shah N. Paracolostomy hernia repair: who and when? J Am Coll Surg. 2014;218(6):1105-1112. doi: 10.1016/j.jamcollsurg.2014.01.054 [DOI] [PubMed] [Google Scholar]
- 6.Okorji LM, Kasten KR. Diagnosis and management of parastomal hernias. Dis Colon Rectum. 2019;62(2):158-162. doi: 10.1097/DCR.0000000000001293 [DOI] [PubMed] [Google Scholar]
- 7.Pastor DM, Pauli EM, Koltun WA, Haluck RS, Shope TR, Poritz LS. Parastomal hernia repair: a single center experience. JSLS. 2009;13(2):170-175. [PMC free article] [PubMed] [Google Scholar]
- 8.Hansson BM, Slater NJ, van der Velden AS, et al. Surgical techniques for parastomal hernia repair: a systematic review of the literature. Ann Surg. 2012;255(4):685-695. doi: 10.1097/SLA.0b013e31824b44b1 [DOI] [PubMed] [Google Scholar]
- 9.DeAsis FJ, Lapin B, Gitelis ME, Ujiki MB. Current state of laparoscopic parastomal hernia repair: a meta-analysis. World J Gastroenterol. 2015;21(28):8670-8677. doi: 10.3748/wjg.v21.i28.8670 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Fox SS, Janczyk R, Warren JA, et al. An evaluation of parastomal hernia repair using the Americas Hernia Society Quality Collaborative. Am Surg. 2017;83(8):881-886. doi: 10.1177/000313481708300841 [DOI] [PubMed] [Google Scholar]
- 11.Beffa LR, Warren JA, Cobb WS, Knoedler B, Ewing JA, Carbonell AM. Open retromuscular repair of parastomal hernias with synthetic mesh. Am Surg. 2017;83(8):906-910. doi: 10.1177/000313481708300845 [DOI] [PubMed] [Google Scholar]
- 12.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative . The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453-1457. doi: 10.1016/S0140-6736(07)61602-X [DOI] [PubMed] [Google Scholar]
- 13.Dimick JB, Nicholas LH, Ryan AM, Thumma JR, Birkmeyer JD. Bariatric surgery complications before vs after implementation of a national policy restricting coverage to centers of excellence. JAMA. 2013;309(8):792-799. doi: 10.1001/jama.2013.755 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36(1):8-27. doi: 10.1097/00005650-199801000-00004 [DOI] [PubMed] [Google Scholar]
- 15.Kleinbaum DG, Klein M. Evaluating the proportional hazards assumption. Survival analysis. Springer; 2012:161-200. [Google Scholar]
- 16.Helgstrand F, Rosenberg J, Kehlet H, Strandfelt P, Bisgaard T. Reoperation versus clinical recurrence rate after ventral hernia repair. Ann Surg. 2012;256(6):955-958. doi: 10.1097/SLA.0b013e318254f5b9 [DOI] [PubMed] [Google Scholar]
- 17.van Barneveld KW, Vogels RR, Beets GL, et al. Prophylactic intraperitoneal mesh placement to prevent incisional hernia after stoma reversal: a feasibility study. Surg Endosc. 2014;28(5):1522-1527. doi: 10.1007/s00464-013-3346-0 [DOI] [PubMed] [Google Scholar]
- 18.Lorenz A, Kogler P, Kafka-Ritsch R, Öfner D, Perathoner A. Incisional hernia at the site of stoma reversal-incidence and risk factors in a retrospective observational analysis. Int J Colorectal Dis. 2019;34(7):1179-1187. doi: 10.1007/s00384-019-03310-5 [DOI] [PubMed] [Google Scholar]
- 19.Israelsson LA. Preventing and treating parastomal hernia. World J Surg. 2005;29(8):1086-1089. doi: 10.1007/s00268-005-7973-z [DOI] [PubMed] [Google Scholar]
- 20.Sohn YJ, Moon SM, Shin US, Jee SH. Incidence and risk factors of parastomal hernia. J Korean Soc Coloproctol. 2012;28(5):241-246. doi: 10.3393/jksc.2012.28.5.241 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Helgstrand F, Rosenberg J, Kehlet H, Jorgensen LN, Wara P, Bisgaard T. Risk of morbidity, mortality, and recurrence after parastomal hernia repair: a nationwide study. Dis Colon Rectum. 2013;56(11):1265-1272. doi: 10.1097/DCR.0b013e3182a0e6e2 [DOI] [PubMed] [Google Scholar]
- 22.Liang MK, Holihan JL, Itani K, et al. Ventral hernia management: expert consensus guided by systematic review. Ann Surg. 2017;265(1):80-89. doi: 10.1097/SLA.0000000000001701 [DOI] [PubMed] [Google Scholar]
- 23.Khan MTA, Patnaik R, Hausman-Cohen L, et al. Obesity stratification predicts short-term complications after parastomal hernia repair. J Surg Res. 2022;280:27-34. doi: 10.1016/j.jss.2022.07.002 [DOI] [PubMed] [Google Scholar]
- 24.Halabi WJ, Jafari MD, Carmichael JC, et al. Laparoscopic versus open repair of parastomal hernias: an ACS-NSQIP analysis of short-term outcomes. Surg Endosc. 2013;27(11):4067-4072. doi: 10.1007/s00464-013-3062-9 [DOI] [PubMed] [Google Scholar]
- 25.Israelsson LA. Parastomal hernias. Surg Clin North Am. 2008;88(1):113-125. doi: 10.1016/j.suc.2007.10.003 [DOI] [PubMed] [Google Scholar]
- 26.Antoniou SA, Agresta F, Garcia Alamino JM, et al. European Hernia Society guidelines on prevention and treatment of parastomal hernias. Hernia. 2018;22(1):183-198. doi: 10.1007/s10029-017-1697-5 [DOI] [PubMed] [Google Scholar]
- 27.Gavigan T, Stewart T, Matthews B, Reinke C; Patients Undergoing Parastomal Hernia Repair Using the Americas Hernia Society Quality Collaborative . Patients undergoing parastomal hernia repair using the Americas Hernia Society Quality Collaborative: a prospective cohort study. J Am Coll Surg. 2018;227(4):393-403.e1. doi: 10.1016/j.jamcollsurg.2018.07.658 [DOI] [PubMed] [Google Scholar]
- 28.Rubin MS, Schoetz DJ Jr, Matthews JB. Parastomal hernia: is stoma relocation superior to fascial repair? Arch Surg. 1994;129(4):413-418. doi: 10.1001/archsurg.1994.01420280091011 [DOI] [PubMed] [Google Scholar]
- 29.Harb WJ. Parastomal hernia: surgical management. Semin Colon Rectal Surg . 2012;23(1):20-25. doi: 10.1053/j.scrs.2011.10.006 [DOI] [Google Scholar]
- 30.van Dijk SM, Timmermans L, Deerenberg EB, et al. Parastomal hernia: impact on quality of life? World J Surg . 2015;39(10):2595-2601. doi: 10.1007/s00268-015-3107-4 [DOI] [PubMed] [Google Scholar]
- 31.Miller BT, Krpata DM, Petro CC, et al. Biologic vs synthetic mesh for parastomal hernia repair: post hoc analysis of a multicenter randomized controlled trial. J Am Coll Surg. 2022;235(3):401-409. doi: 10.1097/XCS.0000000000000275 [DOI] [PubMed] [Google Scholar]
- 32.Mäkäräinen-Uhlbäck E, Vironen J, Vaarala M, et al. Keyhole versus Sugarbaker techniques in parastomal hernia repair following ileal conduit urinary diversion: a retrospective nationwide cohort study. BMC Surg. 2021;21(1):231. doi: 10.1186/s12893-021-01228-w [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Complication categories and codes
eTable 2. Cox proportional hazards models for repeat parastomal hernia repair among patients who did not undergo ostomy reversal at index operation
Data Sharing Statement