Risk factors and nomogram prediction of postoperative incisional hernia following liver transplantation: A retrospective cohort study

Abstract

BACKGROUND

Incisional hernia (IH) is one of the most common complications after liver transplantation, significantly affecting patients’ quality of life and long-term prognosis. However, its risk factors and predictive models remain insufficiently investigated. This study aimed to identify independent risk factors for IH using univariate and multivariate Cox regression analyses and to construct a nomogram prediction model, thereby providing evidence for clinical risk assessment and management.

AIM

To identify the independent risk factors for postoperative IH following liver transplantation and to develop a Cox regression-based nomogram for individualized risk prediction.

METHODS

This single-center retrospective cohort study included 511 liver transplant recipients, with a median follow-up duration of 24 months (from January 2019 to December 2021), including demographic characteristics, comorbidities, and preoperative laboratory parameters. Univariate and multivariate Cox regression models were used to analyze factors associated with IH, and a nomogram was developed accordingly. Model discrimination and predictive performance were assessed by the concordance index and risk stratification analysis.

RESULTS

Univariate analysis showed that age, hepatitis, chronic obstructive pulmonary disease, ascites, malignancy as the transplant indication, history of abdominal surgery, red blood cell count, white blood cell count, serum albumin, total bilirubin, alanine aminotransferase, aspartate aminotransferase, prothrombin time, and international normalized ratio were significantly associated with IH (P < 0.05). Multivariate analysis identified age [hazard ratio (HR) = 1.131, 95% confidence interval (CI): 1.075-1.191, P < 0.001], hepatitis (HR = 2.225, 95%CI: 1.058-4.682, P = 0.035), ascites (HR = 5.687, 95%CI: 1.925-16.802, P = 0.002), serum albumin (HR = 0.933, 95%CI: 0.886-0.982, P = 0.008), elevated alanine aminotransferase (HR = 1.013, 95%CI: 1.004-1.021, P = 0.003), elevated aspartate aminotransferase (HR = 0.996, 95%CI: 0.993-0.999, P = 0.023), and prolonged prothrombin time (HR = 0.903, 95%CI: 0.816-0.999, P = 0.047) as independent risk factors. The nomogram based on these variables showed good discriminative ability (concordance index = 0.874) and effectively predicted the 3- and 5-year risk of IH after transplantation. The median follow-up period and reported cumulative IH incidence: 3-year incidence 12.7%, 5-year incidence 15.1%. Risk stratification further demonstrated that patients in the high-risk group had a significantly higher incidence of IH compared with the low-risk group.

CONCLUSION

Hepatitis, ascites, hypoalbuminemia, impaired liver function, and coagulation abnormalities are key risk factors for IH following liver transplantation. The nomogram developed in this study demonstrated high accuracy and clinical utility, providing a valuable tool for individualized postoperative management and preventive strategies.

Key Words: Liver transplantation; Incisional hernia; Nomogram; Cox regression; Risk factors; Postoperative complications

Core Tip: This study developed and validated a nomogram that incorporates key factors, including age, hepatitis, ascites, hypoalbuminemia, elevated alanine aminotransferase, and prolonged prothrombin time, to accurately predict the risk of incisional hernia following liver transplantation, facilitating the identification of high-risk patients for targeted preventive management. The nomogram developed in this study provides a valuable tool for individualized postoperative management and preventive strategies.

INTRODUCTION

Liver transplantation represents a vital therapeutic option for a wide range of severe liver diseases, including hepatocellular carcinoma, primary biliary cirrhosis, and acute liver failure, and remains the only curative treatment for end-stage liver disease. With continuous advances in surgical techniques and optimization of immunosuppressive regimens, long-term survival and quality of life among liver transplant recipients have improved significantly. However, postoperative complications remain a major concern. Incisional hernia (IH) is among the most common complications, with an incidence of 8%-20%[1,2]. Although IH is not usually life-threatening, it often results in considerable discomfort, functional limitation, and increased healthcare costs, thereby substantially impairing patients’ recovery and quality of life[3].

Previous studies suggest that IH is associated with multiple factors, including age, smoking history, corticosteroid use, operative duration, and emergency surgery[4,5]. In the liver transplantation population, management of IH is more complex, and several reports have shown that outcomes of hernia repair after transplantation are inferior to those of primary hernia repair[6-8]. Nevertheless, data on the timing, risk factors, and surgical outcomes of IH in liver transplant recipients remain limited[9]. Therefore, this study sought to investigate the incidence and risk factors of IH following liver transplantation, as well as its impact on patient outcomes, in order to facilitate early identification of high-risk individuals and guide individualized intervention strategies.

However, few studies have systematically explored predictive tools for postoperative IH following liver transplantation. Existing research has primarily focused on general surgical populations, with limited application in transplant recipients. To the best of our knowledge, this is the first study to construct a nomogram specifically for predicting IH after liver transplantation. This model provides a practical tool for individualized risk assessment and may assist clinicians in optimizing postoperative management strategies.

MATERIALS AND METHODS

Study population

This retrospective cohort study included patients who underwent liver transplantation at Tianjin First Central Hospital between January 2019 and December 2021. Patients were categorized into the IH group and control group based on postoperative occurrence of IH. Patients with a preoperative history of abdominal wall hernia were excluded. The final diagnosis of IH was confirmed during tension-free hernia repair surgery.

Surgical technique

All patients underwent orthotopic liver transplantation performed by the same surgical team, using ABO-compatible grafts from deceased donors. Postoperative immunosuppressive therapy consisted of a triple regimen including tacrolimus, mycophenolate mofetil, and corticosteroids, with sirolimus added at three months postoperatively. Tacrolimus dosing was adjusted to maintain therapeutic blood concentrations. Post-transplant monitoring included routine blood tests and liver function assessments to prevent rejection, with immunosuppressive regimens modified accordingly.

Diagnosis, treatment, and follow-up of IH

The diagnosis of IH was based on physical examination and imaging evaluation, and confirmed intraoperatively during hernia repair. Follow-up was conducted through outpatient visits and telephone interviews: Every 2 months during the first 2 years post-transplant, and every 6 months thereafter. The primary endpoint was IH status within 2 years after repair. Follow-up was completed by December 31, 2024. Missing data were minimal (< 5%) and handled by complete-case analysis.

Statistical analysis

Descriptive statistics were used to summarize baseline characteristics. Categorical variables were compared using the χ² test or Fisher’s exact test, and continuous variables were analyzed with Student’s t-test. Univariate and multivariate Cox regression analyses were performed to identify risk factors for IH. A nomogram was constructed using the rms package in R software, and predictive performance was assessed using the concordance index (C-index). A two-tailed P < 0.05 was considered statistically significant.

RESULTS

Patient selection and baseline characteristics

This study included clinical data of patients who underwent liver transplantation at Tianjin First Central Hospital between January 2019 and December 2021. Both patients who developed IH after transplantation and those who did not were enrolled. The screening process of eligible patients according to the inclusion criteria is shown in Figure 1. During the study period, a total of 511 liver transplantations were performed. Among them, 435 patients (85.1%) did not develop IH during follow-up, while 76 patients (14.9%) developed IH, of whom 16 underwent hernia repair surgery. In the IH group, there were 55 male (72.4%) and 21 female (27.6%) patients, with an age range of 40-73 years. The median time to IH occurrence after liver transplantation was 12.5 months (interquartile range: 6.5-38 months).

Figure 1 Flowchart of patient selection for incisional hernia after liver transplantation. IH: Incisional hernia.

Baseline characteristics of the patients are summarized in Table 1. There were no significant differences between the groups in terms of sex distribution and smoking history (P > 0.05). Compared with patients without IH, those who developed IH were significantly older (median age 66 years vs 51 years, P < 0.001). Regarding comorbidities, patients in the IH group had a markedly higher prevalence of hepatitis and chronic obstructive pulmonary disease (COPD; both P < 0.001), as well as a higher incidence of ascites (P < 0.001). In addition, a history of prior abdominal surgery was strongly associated with the development of IH (P < 0.001), suggesting that surgical manipulation and tissue healing may play important roles in postoperative complications.

Table 1 Baseline characteristics of patients, n (%)/median (interquartile range).

Characteristics	No incisional hernia	Developed incisional hernia	P value
n	435	76
Gender			0.287
Male	339 (66.3)	55 (10.8)
Female	96 (18.8)	21 (4.1)
Age	51 (44, 57)	66 (64, 67)	< 0.001
Smoking			0.379
No	297 (58.1)	48 (9.4)
Yes	138 (27)	28 (5.5)
Hepatitis			< 0.001
No	435 (85.1)	56 (11)
Yes	0 (0)	20 (3.9)
COPD			< 0.001
No	435 (85.1)	69 (13.5)
Yes	0 (0)	7 (1.4)
Ascites			< 0.001
No	429 (84)	67 (13.1)
Yes	6 (1.2)	9 (1.8)
Other1			0.106
No	402 (78.7)	66 (12.9)
Yes	33 (6.5)	10 (2)
Abdominal surgeries history2			< 0.001
No	435 (85.1)	73 (14.3)
Yes	0 (0)	3 (0.6)
Cancer3			< 0.001
No	339 (66.3)	37 (7.2)
Yes	96 (18.8)	39 (7.6)
Red blood cell counts, 1 × 1012/L	4.42 (3.925, 4.875)	4.15 (3.4075, 4.5875)	< 0.001
White blood cell counts, 1 × 109/L	5.81 (4.31, 7.76)	5.095 (3.7025, 6.545)	0.013
PLT (125-350), 1 × 109/L	172 (120, 238.5)	181 (103.75, 234.25)	0.539
ALB, g/L	41.8 (37.7, 44.7)	35.3 (32.275, 40.375)	< 0.001
Total bilirubin, μmol/L	20 (14, 32)	25.5 (14, 50)	0.021
ALT, U/L	19.7 (14.2, 30.95)	34.9 (22.45, 53.725)	< 0.001
AST, U/L	23.7 (17.8, 33.55)	44.8 (26.175, 71.5)	< 0.001
PT (8.8-13.8), seconds	15.3 (14.2, 16.5)	12.9 (11.875, 15.05)	< 0.001
INR (0.8-1.2)	1.05 (1, 1.16)	1.16 (1.0675, 1.3425)	< 0.001

¹The presence of other comorbidities.

²A history of prior abdominal operations.

³Liver transplantation performed for malignancy.

COPD: Chronic obstructive pulmonary disease; PLT: Platelet; ALB: Albumin; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; PT: Prothrombin time; INR: International normalized ratio.

In terms of laboratory findings, patients with IH had abnormal postoperative red blood cell and white blood cell counts (P < 0.05), along with significantly lower serum albumin (ALB) levels (35.3 g/L vs 41.8 g/L, P < 0.001). Liver function markers, including total bilirubin, alanine aminotransferase (ALT), and aspartate aminotransferase (AST), were significantly elevated in the IH group (P < 0.05). Coagulation parameters revealed an increased international normalized ratio in patients with IH (P < 0.001). Overall, hepatitis, hypoalbuminemia, impaired liver function, coagulation abnormalities, and comorbidities (such as COPD, ascites, and history of abdominal surgery) were closely associated with the occurrence of IH following liver transplantation.

Cox regression analysis

Univariate Cox regression analysis: As shown in Table 2, univariate Cox regression analysis identified several significant prognostic factors: Age [hazard ratio (HR) = 1.191, 95% confidence interval (CI): 1.148-1.235, P < 0.001], hepatitis (HR = 15.240, 95%CI: 8.958-25.926, P < 0.001), COPD (HR = 5.032, 95%CI: 2.291-11.054, P < 0.001), ascites (HR = 6.538, 95%CI: 3.242-13.185, P < 0.001), prior abdominal surgery (HR = 25.832, 95%CI: 7.797-85.579, P < 0.001), tumor (HR = 2.810, 95%CI: 1.779-4.437, P < 0.001), abnormal red blood cell count (HR = 0.503, 95%CI: 0.373-0.678, P < 0.001), abnormal white blood cell count (HR = 0.887, 95%CI: 0.803-0.981, P = 0.020), ALB (HR = 0.895, 95%CI: 0.864-0.927, P < 0.001), total bilirubin (HR = 1.002, 95%CI: 1.000-1.004, P = 0.031), ALT (HR = 1.005, 95%CI: 1.003-1.007, P < 0.001), AST (HR = 1.002, 95%CI: 1.001-1.003, P < 0.001), prothrombin time (PT; HR = 0.803, 95%CI: 0.750-0.861, P < 0.001), and international normalized ratio (HR = 2.915, 95%CI: 1.616-5.257, P < 0.001).

Table 2 Univariate and multivariate Cox regression analysis of risk factors for incisional hernia after liver transplantation.

Characteristics	Total (n)	Univariate analysis		Multivariate analysis
		Hazard ratio (95%CI)	P value	Hazard ratio (95%CI)	P value
Gender	487
Male	376	Reference
Female	111	1.225 (0.736-2.037)	0.435
Age	487	1.191 (1.148-1.235)	< 0.001	1.131 (1.075-1.191)	< 0.001
Smoking	487
No	328	Reference
Yes	159	1.147 (0.711-1.852)	0.574
Hepatitis	487
No	468	Reference		Reference
Yes	19	15.240 (8.958-25.926)	< 0.001	2.225 (1.058-4.682)	0.035
COPD	487
No	480	Reference		Reference
Yes	7	5.032 (2.291-11.054)	< 0.001	0.266 (0.089-0.792)	0.017
Ascites	487
No	472	Reference		Reference
Yes	15	6.538 (3.242-13.185)	< 0.001	5.687 (1.925-16.802)	0.002
Other1	487
No	446	Reference
Yes	41	1.396 (0.691-2.818)	0.353
Abdominal surgeries history2	487
No	484	Reference		Reference
Yes	3	25.832 (7.797-85.579)	< 0.001	2.182 (0.338-14.068)	0.412
Cancer3	487
No	359	Reference		Reference
Yes	128	2.810 (1.779-4.437)	< 0.001	1.116 (0.651-1.916)	0.689
Red blood cell counts, 1 × 1012/L	487	0.503 (0.373-0.678)	< 0.001	0.829 (0.592-1.163)	0.277
White blood cell counts, 1 × 109/L	487	0.887 (0.803-0.981)	0.020	0.902 (0.796-1.023)	0.108
PLT (125-350), 1 × 109/L	487	0.999 (0.996-1.002)	0.425
ALB, g/L	487	0.895 (0.864-0.927)	< 0.001	0.933 (0.886-0.982)	0.008
Total bilirubin, μmol/L	487	1.002 (1.000-1.004)	0.031	1.000 (0.997-1.002)	0.775
ALT, U/L	487	1.005 (1.003-1.007)	< 0.001	1.013 (1.004-1.021)	0.003
AST, U/L	487	1.002 (1.001-1.003)	< 0.001	0.996 (0.993-0.999)	0.023
PT (8.8-13.8), seconds	486	0.803 (0.750-0.861)	< 0.001	0.903 (0.816-0.999)	0.047
INR (0.8-1.2)	487	2.915 (1.616-5.257)	< 0.001	2.042 (0.735-5.674)	0.171

¹The presence of other comorbidities.

²A history of prior abdominal operations.

³Liver transplantation performed for malignancy.

COPD: Chronic obstructive pulmonary disease; PLT: Platelet; ALB: Albumin; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; PT: Prothrombin time; INR: International normalized ratio; CI: Confidence interval.

Multivariate Cox regression analysis: In multivariate Cox regression analysis, several variables remained independent prognostic factors. These included age (HR = 1.131, 95%CI: 1.075-1.191, P < 0.001), hepatitis (HR = 2.225, 95%CI: 1.058-4.682, P = 0.035), ascites (HR = 5.687, 95%CI: 1.925-16.802, P = 0.002), serum ALB (HR = 0.933, 95%CI: 0.886-0.982, P = 0.008), ALT (HR = 1.013, 95%CI: 1.004-1.021, P = 0.003), AST (HR = 0.996, 95%CI: 0.993-0.999, P = 0.023), and PT (HR = 0.903, 95%CI: 0.816-0.999, P = 0.047). Among these, older age, hepatitis, ascites, and elevated ALT were identified as adverse prognostic factors, whereas higher serum ALB levels were associated with favorable prognosis. Notably, the HRs for AST and PT were less than 1, which may reflect complex interactions among hepatic function markers and other covariates in the multivariate model, rather than indicating a strictly protective effect.

Nomogram construction and predictive model performance

Based on the results of multivariate Cox regression analysis, a nomogram model for predicting the risk of IH after liver transplantation was developed (Figure 2), incorporating significantly associated variables. Each predictor was assigned a score according to its regression coefficient, and the cumulative score was used to estimate the probability of IH occurrence at 3 years and 5 years postoperatively. In addition, a risk factor heatmap was generated to visually demonstrate the distribution of different variables among patients (Figure 3A). Patients were stratified into risk groups according to the total risk scores derived from the nomogram. The results revealed significant differences in IH incidence between the high- and low-risk groups, indicating good discriminatory ability of the model. Moreover, the model demonstrated a high degree of calibration, with a C-index of 0.874 (Figure 3B), confirming excellent predictive performance.

Figure 2 Nomogram for predicting the 3- and 5-year risk of postoperative incisional hernia after liver transplantation. The nomogram integrates significant variables identified from multivariate Cox regression analysis, including age, hepatitis, ascites, serum albumin, alanine aminotransferase, aspartate aminotransferase, and prothrombin time. Each variable corresponds to a score on the top axis; the total score indicates the predicted probability of incisional hernia occurrence. RBC: Red blood cell; WBC: White blood cell; ALB: Albumin; TBIL: Total bilirubin; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; INR: International normalized ratio.

Figure 3 Nomogram construction and predictive model performance. A: Heatmap showing the correlation between clinical variables and risk of incisional hernia following liver transplantation; B: Higher risk scores correspond to patients with advanced age, presence of ascites, hepatitis, and lower serum albumin levels. INR: International normalized ratio; PT: Prothrombin time; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase; TBIL: Total bilirubin; ALB: Albumin; RBC: Red blood cell; WBC: White blood cell; COPD: Chronic obstructive pulmonary disease; C-index: Concordance index.

DISCUSSION

This study is the first to establish a validated nomogram specifically predicting postoperative IH following liver transplantation, providing an individualized approach to risk assessment. The findings demonstrated that hepatitis, ascites, abnormal serum ALB levels, liver function impairment (ALT, AST), and coagulation dysfunction (prolonged PT) were closely associated with IH occurrence. Among these, hepatitis, ascites, and hypoalbuminemia were confirmed as independent risk factors in multivariate Cox analysis, underscoring their clinical importance in risk assessment and the development of preventive strategies. The nomogram derived from these factors achieved a C-index of 0.874, indicating strong predictive accuracy and good discrimination. Risk stratification further validated its clinical utility, as high-risk patients exhibited significantly higher IH incidence compared with low-risk patients, suggesting that this model could serve as a practical tool for individualized risk assessment and follow-up management. Consistent with previous literature, IH is a common and clinically significant complication after liver transplantation, with reported incidence rates up to 40%[10]. Previous studies have reported that IH is associated with higher readmission rates, impaired graft function, and reduced overall survival. Additionally, most studies have shown that the repair rate of IH after liver transplantation remains low (about 5.2%)[11-13].

The occurrence of IH after laparotomy is influenced by multiple factors, including obesity, smoking, chronic lung disease, ascites or hyperbilirubinemia, anemia, malnutrition, connective tissue disorders, immunosuppressive therapy, age > 70 years, and male sex[14-16]. Liver transplant recipients are particularly vulnerable, not only due to these factors but also due to transplant-specific risks such as prolonged immunosuppression, hepatic dysfunction, and coagulation abnormalities. In this study, hepatitis significantly increased IH risk, potentially through mechanisms related to coagulopathy, hypoalbuminemia, and malnutrition, all of which impair wound healing. Hypoalbuminemia was also identified as an independent risk factor, consistent with its pivotal role in collagen synthesis and wound repair[17,18]. These findings highlight the need for clinicians to emphasize perioperative nutritional assessment and active correction of hypoalbuminemia. Furthermore, ascites is a well-recognized risk factor for IH, likely contributing through increased abdominal wall tension and delayed wound healing[19-21]. In other transplant fields, such as kidney transplantation, the incidence of IH is relatively lower, with major risk factors including overweight, pulmonary complications, and lymphocele[22-24], suggesting organ-specific differences in IH risk profiles.

Regarding hernia repair, the literature reports an overall IH incidence of approximately 15.1% following liver transplantation, typically occurring in the intermediate to late postoperative period, with recurrence rates around 12.4% after initial repair[25,26]. In our study, follow-up of patients who underwent hernia repair revealed favorable short-term quality of life outcomes, with no complications or recurrences observed, indicating that surgical repair in stable patients can achieve good short-term efficacy. Additionally, the nomogram established in this study provides a quantitative tool to estimate individual IH risk, aiding in early identification of high-risk patients and guiding personalized management strategies. For example, perioperative measures for high-risk patients could include reinforced incision protection, optimized immunosuppressive regimens, strict intra-abdominal pressure control, and long-term dynamic monitoring to reduce IH incidence. The strengths of this study include: (1) A relatively adequate sample size covering multiple clinical indicators, allowing for a comprehensive analysis; and (2) The first construction of a predictive nomogram for IH risk specifically in liver transplant recipients, providing an intuitive tool for clinical risk prediction. Nonetheless, several limitations should be noted. First, this was a single-center retrospective cohort study, with potential selection and information bias. Second, certain variables such as intraoperative technical details and postoperative immunosuppressive regimens were not included in the analysis, which may influence outcomes. Additionally, some potentially relevant perioperative variables, such as donor type (living or deceased), incision type, body mass index, presence of diabetes, and immunosuppressant blood concentrations, were not collected in this study. These missing variables may have introduced unmeasured confounding and should be considered in future prospective studies. Finally, the predictive model requires validation in larger, prospective, multicenter cohorts to enhance its stability and generalizability. The nomogram developed in this study provides a practical tool for individualized perioperative management in liver transplantation. By identifying high-risk patients before surgery, clinicians can optimize preoperative counseling and implement targeted interventions such as improved wound protection, nutritional support, and tailored postoperative follow-up. Early recognition and management of modifiable risk factors, including hepatic dysfunction, hypoalbuminemia, and coagulation abnormalities, may help reduce the incidence of postoperative IH and improve long-term outcomes.

CONCLUSION

In conclusion, this study identified hepatitis and hypoalbuminemia as independent risk factors for IH after liver transplantation and established a feasible predictive model. Future research should integrate prospective multicenter data to further investigate transplant-specific IH risk factors and refine risk stratification tools, thereby guiding precision prevention and targeted interventions.