Objectives: Incisional hernias can occur after any abdominal operation, including after renal transplant. Several risk factors have been identified in nonimmunosuppressed surgical patients. We aimed to identify whether specific risk factors correlated with the development of incisional hernias after renal transplant. The existence of associations between these risk factors and postoperative complications was also reviewed.
Materials and Methods: We reviewed 969 kidney transplants performed between February 2000 and January 2011. Thirty-nine kidney transplant recipients who were treated with rapamycin were excluded. The following potential risk factors were evaluated: recipient age, sex, body mass index at transplant, delayed graft function, diabetes, albumin, postoperative platelet count, drain placement, donor body mass index, donor type, warm ischemic time, and cold ischemic time. We performed univariate and multivariate logistic regression tests.
Results: In our patient group, a total of 52 (5.4%) transplants were complicated by incisional hernia. On univariate analysis, we found that delayed graft function (P = .001) and infection (P < .001) were statistically significant predictors for development of incisional hernia. Multivariate analyses revealed that delayed graft function and length of stay remained statistically significant predictors.
Conclusions: Delayed graft function and length of stay are significant predictors of incisional hernia after kidney transplant.
Key words : Kidney, Postoperative complications
Incisional hernias represent significant, and often underestimated, complications of abdominal surgery.1-3 More than half of incisional hernias appear within 6 months after surgery, approximately 75% within 2 years, and 97% within 4 years.1-6 Transplant patients are at higher risk for incisional hernias from long-term uremia, muscle, and tissue wasting and higher prevalence of obesity, diabetes, and chronic pulmonary diseases.5-7
The effects of immunosuppression and renal transplant on wound healing have been well-documented.7-18 Corticosteroids play an important role in the development of incisional hernias by impeding wound healing.8-18 During acute rejection, when higher doses of methylprednisolone (1 g/day) are administered, steroids have been shown to have considerable effects on the development of incisional hernias.8-11
Use of mycophenolate mofetil, tacrolimus (Astellas Pharma, Chuo-ku, Japan), and sirolimus (Pfizer, Groton, CT, USA) for immunosuppression has been shown to be significant risk factors for hernias.12-18 Mycophenolate mofetil has been shown to have an antiproliferative effect that may interfere with the healing of the muscles and fascia.12-18
Surgical site infection (SSI) plays a prominent role in the development of incisional hernias. Several studies have shown that SSI is the most important risk factor for hernia development.1,6-12 The negative effects of immunosuppressive therapy after renal transplant can further contribute to the high incidence of SSI, leading to increased incidence of incisional hernias due to disturbed and delayed wound healing.
The aim of our study was to identify specific risk factors associated with the development of incisional hernias.
Materials and Methods
After approval from the Institutional Review Board, the data were retrospectively analyzed for consecutive renal transplants performed from February 2000 through January 2011. Patients whose transplants were complicated by incisional hernias (study group) were compared with the remainder of the cohort (control group). Figure 1 illustrates the patient selection process. Our institution has a single immunosuppression protocol. Thirty-nine kidney transplant recipients who were induced or maintained with rapamycin were excluded because of the known association between rapamycin use and the development of hernia.
Under the standard immunosuppression protocol at our institution, patients were immunosuppressed with tacrolimus, mycophenolate mofetil, and steroids. Antithymocyte globulin (Thymoglobulin, Genzyme Boston, MA, USA) was the most common agent used for induction therapy, followed by interleukin 2 receptor antibody. Antithymocyte globulin was administered at 1.5 mg/kg/d for 4 to 7 days starting intraoperatively through a central line and titrated for leukopenia, thrombocytopenia, or other adverse effects attributed to induction. The goal dose ranged between 5 and 6 mg/kg. All patients were started on mycophenolate mofetil with doses ranging from 1 to 2 g/d. The dosage was titrated for gastrointestinal adverse effects and leukopenia. Patients received methylprednisolone intravenously at 400 to 500 mg intraoperatively followed by a taper to 30 mg/d by day 7, 20 mg/d by day 30, and 5 mg/d by day 90 of oral prednisone. Tacrolimus was started by day 4, with target trough levels between 5 and 12 ng/mL.
Potential risk factors for hernia were evaluated with univariate and multivariate logistic regression tests. Recipient characteristics analyzed were age, sex, race, body mass index, type 2 diabetes mellitus, drain placement, albumin level, postoperative platelet count, wait time, postoperative collection (seroma, hematoma), SSI, intraperitoneal implant, length of stay, and retransplant before development of incisional hernias. Donor characteristics evaluated were body mass index, type, sex, ethnicity, and hepatitis C virus seropositivity. Graft characteristics evaluated were delayed graft function (DGF), defined as dialysis within the first 7 days after transplant, warm ischemic time, and cold ischemic time. Colinearity was tested but not observed. The calibration of the final multivariate logistic regression model was tested with Hosmer-Lemeshow test. Data were tested for normality. The categorical variables were compared with t, Mann-Whitney U, and chi-squared tests. Additionally, the survival rates of the 2 groups were compared with Kaplan-Meier statistical method, and associated hazard ratios were calculated, using COX regression with time-dependent covariate analysis. Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 22.0, IBM Corporation, Armonk, NY, USA).
To identify incisional hernias after renal transplant, the database was queried using International Classification of Diseases, Clinical Modification ICD-9 code 553.21. We then performed a chart review and removed patients with dehiscence or incisional hernias not directly related to renal transplant. To identify SSI within the cohort, the database was queried using ICD-9 code(s) associated with SSI. All patients with SSI were identified by ICD-9 code 998.5.
Generally, a Gibson incision, ranging from 7 to 10 cm in length, was employed, and the contents of the iliac fossa were exposed. The inferior epigastric artery was usually divided. In male recipients, the spermatic cord was generally preserved. The grafts were usually placed in the retroperitoneum, although a significant number of intraperitoneal placements were performed to avoid potential compartment syndrome.
Looped polydioxanone sutures, started on both ends to meet in the middle of the incision with 1-cm advancements, were generally used. A suture length-to-wound length ratio of at least 4:1 was attempted. Drain placement was left to the surgeon’s discretion.
Between February 2000 and January 2011, 969 renal transplants were performed. Thirty-nine patients who received rapamycin were excluded, and the remaining 930 renal transplants were analyzed. Forty-seven recipients (5%) of the 930 recipients analyzed were complicated by incisional hernias and comprised the study group. The remaining 883 transplants not complicated by incisional hernias comprised the control group. The mean age of the hernia cohort (standard deviation) was 54.23 ± 12.84 years. The mean age of the control group was 54.67 ± 12.75 years (P = .817). In the hernia cohort, 27 patients (57.4%) were male; 571 patients (61.4%) from the control group were male (P = .645). The average body mass index in the hernia group was 28.4 ± 6.4 kg/m2 compared with 27.7 ± 5.8 kg/m2 in the control group (P = .447). The median wait time on the transplant list was 245.5 days (range, 21-1712 d) for the hernia group compared with 310 days (range, 0-3929 d) for the control group (P = .299). Twenty-eight patients (59.6%) from the hernia group developed DGF, and 313 patients (35.4%) from the control group developed DGF (P = .001). One patient (2.1%) in the hernia group was identified as having developed a SSI, compared with 6 (0.7%) in the control group (P = .305). Full lists of recipient, donor, and graft characteristics are listed in Tables 1, 2, and 3 separately.
A univariate analysis revealed that DGF, length of stay, and extended criteria donation were statistically significant predictors of incisional hernias (Table 4). Next, we performed a multivariate analysis with these 3 variables. In multivariate analysis, DGF and length of stay stayed significant and extended criteria donation lost significance (Table 5).
We then performed Kaplan-Meier analyses and calculated the associated hazard ratios. Initially, patients with incisional hernias had higher survival (albeit not significant). However, that advantage disappeared after 1200 days, and ultimately the 2 survival curves intersected. After 1200 days the hazard ratio was 2.945, with P < .0001 (Figure 2).
The incidence of incisional hernias after abdominal surgery is approximately 20%.1-4,6,19,20 Predisposing factors are age, obesity, alcoholism, smoking, multiple operations, foreign body implantation, wound infection, hematoma, technical error, and unsuitable suture material.1-17 In addition, medical illnesses such as chronic renal failure, liver insufficiency, infection, and pulmonary diseases are known risk factors.
In transplanted patients, because of the higher prevalence of comorbidities and the administration of immunosuppressive agents, the rate of incisional hernias is higher.26 Incisional hernias mostly develop in the first 3 months after kidney transplant.21 The major predisposing surgical factors included incision type, emergency surgery, drains placed through the wound, implantation of a foreign body, technical closure faults, or inadequate suture material.21 We did not find drain placement to be a predictor of incisional hernia.
In this study, 5% of kidney transplants over an 11-year period were complicated by incisional hernias. In multivariate analysis, DGF and length of stay were the ultimate statistically significant predictors. Smith and associates retrospectively studied 2247 renal transplants, and DGF was not a statistically significant risk factor for hernia formation in their multivariate analyses. In their univariate analyses, DGF was associated with P = .003.22 A possible reason for this could be their surprisingly low rate of DGF (175/2247; 7.8%) compared with ours (36.7%). Such a low rate of DGF in the study of Smith and associates is unusual. The renal transplants included in their study were performed over a period of 11 years. Eleven years is a relatively long period. In addition, the statistical method they employed was a Cox proportional hazards model, which was different than our statistical method (logistic regression). Length of stay was not evaluated as a risk factor in the study of Smith and associates. Length of stay was considered a statistically significant risk factor in a prospective study conducted by Goodenough and associates. They followed 625 postabdominal surgery patients for a median of 41 months and introduced a risk stratification score system for development of ventral incisional hernia.23 Length of stay was not a coefficient in the final formula for score calculation that they presented. Their findings are interesting; however, the portion of renal transplant patients in their study is unclear. Because they focused on abdominal surgery in general, DGF was not included in their analyses. Other factors such as age, body mass index, wait time, cold and warm ischemic times, drain placement, and postoperative collections were not found to predict this complication.
Twenty-eight patients (59.6%) of the 47 transplants complicated by incisional hernia developed DGF. Delayed graft function is a manifestation of acute kidney injury, with attributes unique to the transplant process.24-26 Delayed graft function is defined as the requirement of dialysis within 7 days after renal transplant. Delayed graft function after renal transplant adversely affects patient quality of life, and the requirement for dialysis and additional biopsy evaluation of the transplanted kidney prolong hospital stays.1-6,9-17 The average cost of initial hospitalization for patients undergoing deceased donor renal transplant, who develop DGF, is approximately $25 000 higher per patient.17-20,24-28 When the clinical manifestations of DGF, such as increased rate of acute rejection and worse graft survival, are taken into account, the economic impact is expanded.19,20,27 Delayed graft function is thus associated with lower graft survival, longer hospital stay, higher costs, and increased psychologic and significant medical sequelae.17-21,24-30
In patients with DGF, hemodialysis itself is an additional oxidative stress. The need for even 1 dialysis session posttransplant creates a milieu that affects B and T cells, which can lead to impaired wound healing and/or wound breakdown. In a recent study, Roine and associates demonstrated that transplant recipients who were predialytic had no wound dehiscence, whereas those who received hemodialysis had a wound dehiscence rate of 13% (P = .02).26
Older donors also have a greater association with DGF, in part because older kidneys react poorly to acute kidney injury.1-21,24-30 The lower physiologic reserve tempers the ability to recover from insult.1-20,24,25,27,28 We did not see a correlation between incisional hernias and obesity. This is contrary to many studies published in the literature, which directly correlate incisional hernias with obesity.1-20,24,25,27,28
Donation after cardiac death is another well-established risk factor for DGF.1-20,24,25,27,28,31,32 Donations after cardiac death inevitably sustain a period of warm ischemia from circulatory arrest until initiation of organ preservation. This causes ischemic acute kidney injury, which results in an increased incidence of DGF compared with renal transplant from conventional donations after brain death.31,32 We were unable to find an association between patients with incisional hernias and donation after cardiac death kidneys.
In our hernia cohort, 1 patient (2.1%) developed SSI. Renal transplant is considered a clean-contaminated procedure. Therefore, this rate of infection is within the average range of 2% to 5%.1,3-9,11-20 Several studies have suggested that SSI after kidney transplant is a risk factor for incisional hernias.1-21,24-32 The association between SSI and incisional hernias was verified in this study. Surgical site infection is also considered to be an important risk factor contributing to the development of incisional hernias in nontransplanted patients.8,21,29 The negative effects of immunosuppressive therapy after kidney transplant on the patient’s immune system further contribute to the high incidence of SSI and therefore the increased incidence of incisional hernias due to disturbed and delayed wound healing in the early postoperative period after kidney transplant.3,4
Surgical site infections, the second most common cause of nosocomial infection after urinary tract infections, cause approximately 17% of all hospital-acquired infections and lead to increased costs and worse patient outcomes in hospital inpatients.25 The US Centers for Disease Control estimates that 500 000 SSIs occur annually in the United States. Costs and outcomes secondary to SSI can vary by location and surgery type. Ho and associates25 found that, in cases with SSI after renal transplant, hospital revenue increased by $20 176, hospital margin decreased by $4278, and hospital costs increased by $24 454. When adjusted for donor and recipient characteristics, SSI was independently associated with an $11 132 increase in cost. Given the high costs and adverse patient outcomes associated with SSI, the physical, emotional, and economic impacts of SSI are high.
Finally, many studies have shown that vitamin A stimulates the healing of wounds that have been delayed by cortisone.22,25,26 The antagonism between vitamin A and cortisone occurs in the wound itself. Vitamin A stimulates but does not enhance wound healing beyond a normal rate. In renal transplant patients who are experiencing difficulty with wound healing, vitamin A could be used to aid in countering the deleterious wound effects brought on by steroids.
This study is not without limitations. Our database did not allow us to examine different incision types or closure techniques. Second, we could not evaluate smoking in our database. However, the institution’s protocol strongly advises against smoking. Third, our database was not equipped to determine whether our patients had pulmonary disease. Nevertheless, those patients with significant pulmonary disease would not be allowed to pursue transplant. Our study design was not a randomized prospective clinical trial. Hence, potential bias may have been introduced in addition to relative lack of power and instability of the univariate and multivariate models. Last, the actual number of patients who had incisional hernias may have been underestimated. Our database only captured those who actually underwent hernia repair. Therefore, those patients who were not deemed surgical candidates or who had their hernias repaired at another institution would be missed.
In terms of Kaplan-Meier analyses, the survival curves for the 2 groups were not statistically different when verified with log-rank test. The two groups were not necessarily exposed to the same risk factors with constant influence. Patients with incisional hernias were sicker and had to undergo a second surgical procedure for incisional hernia repair, which implied more risks, compared with the control group (patients without incisional hernias).
In conclusion, DGF and length of stay can predict the development of incisional hernias after renal transplant. Although wound complications do not influence patient and graft survival drastically, they can result in significant morbidity, prolonged hospitalization, or hospital readmission, reoperation, and increases in overall transplant costs.
Volume : 15
Issue : 1
Pages : 27 - 33
DOI : 10.6002/ect.2015.0263
From the 1Department of Surgery, Einstein Healthcare Network,
Philadelphia, Pennsylvania, USA; and the 2Department of Surgery,
University of Toledo Medical Center, Toledo, Ohio, USA
Acknowledgements: The authors declare that they have no sources of funding for this study, and they have no conflicts of interest to disclose.
Corresponding author: Jorge Ortiz, University of Toledo Medical Center, 3065 Arlington Ave. Toledo, OH 43614, USA
Table 1. Recipient Characteristics
Table 2. Donor Characteristics
Table 3. Graft Characteristics
Table 4. Results of Univariate Logistic Regression
Table 5. Results of Multivariate Logistic Regression
Figure 1. Case Selection Process
Figure 2. Case Selection Process