Abdominal wall hernia repair in cirrhotic patients: outcomes seen at a tertiary care hospital in a developing country

Introduction

The prevalence of hepatitis B and C in Pakistan is among the highest in the world. ^1,2 Chronic liver disease (CLD) and cirrhosis as a result of these infections is becoming an important cause of morbidity and mortality in these patients. ^1,3 A number of factors in cirrhosis, including raised intra-abdominal pressure from ascites, weakness of abdominal wall from poor nutritional status and enlargement of the supra-umbilical facial opening from dilated umbilical veins, contribute to the development of abdominal wall hernia. ⁴ The prevalence of umbilical hernia has been reported as up to 20% in patients with long standing cirrhosis and 40% in patients with ascites in previous studies. ⁵ However, no significant data is available about the prevalence of inguinal herniation in the presence of ascites. ⁶

Despite the common recommendation to repair the hernia in all patients, procedural difficulties associated with complications of cirrhosis in such patients, such as coagulopathy, thrombocytopenia, varices and ascites, increase the risks involved in surgical procedures, resulting in higher mortality. ⁷ Although elective repair of abdominal wall hernia in such patients has been considered safe in some studies, ^5,8 authors have still highlighted the higher rate of complications in cirrhotic patients as compared to noncirrhotic patients. ⁹ Weighing out the risks and benefits of taking the decision to operate becomes a predicament for the operating surgeon. At this standpoint we aim to evaluate the outcome of hernia repair in patients with longstanding liver disease in terms of early complications and subsequent recurrence at Aga Khan University Hospital, a tertiary care hospital in a developing country with reasonable medical facilities, with the exception of liver transplant.

Materials and methods

A retrospective cohort study was conducted on 61 CLD patients who underwent abdominal wall hernia repair between January 2001 and December 2007 at our institution. A list of cirrhotic patients undergoing hernia repair was generated from a Health Information Management System database using the ICD-9-CM coding system. The data recorded included patient's demographics, type of hernia, complexity of hernia, aetiology of cirrhosis, Child-Turcotte-Pugh (CTP) class and type of admission, i.e. elective or emergency.

The severity of liver disease was quantified both pre- and postoperatively by CTP classification to objectively assess the outcomes of hernia repair in CLD patients. ^10,11 This takes into account serum bilirubin, serum albumin, coagulation profile, presence and severity of ascites, and porto-systemic encephalopathy in patients with cirrhosis, and each patient is assigned a score, as outlined in Table 1. The presence of co-morbid conditions like diabetes, hypertension and renal disease were also considered. Procedural details including use of prolene mesh, type of anaesthesia and duration of surgery were also recorded.

Outcomes were assessed over a period of 30 days after surgery for early complications including mortality. The period of follow-up for all patients was calculated and any recurrence of previous hernia was duly recorded. A comparison was made between elective and emergency admissions and between Child class C and non-C. The data were analysed using SPSS version 18 (SPSS, Chicago, IL, USA). Results are given as mean ± standard deviations and percentages, where applicable. The statistical analysis was done using Chi-square test and a P value of less than 0.05 was considered significant.

Results

The details of 61 patients with CLD who underwent hernia repair during the study period including demographics, aetiology and severity of underlying liver disease, type of hernia and surgery performed are given in Table 2. A total of 43 (70.3%) patients had one or more co-morbid conditions including diabetes mellitus in 24, hypertension in 23, chronic obstructive airway disease in 11, ischaemic heart disease in 10 and chronic renal failure in 6 patients. The mean time from diagnosis of CLD to presentation was 57 ± 53 months. The mean duration of surgery was 125 ± 65 minutes. Prolene mesh was used in 55 (90%) patients to provide strength to the abdominal wall. The mean length of stay in hospital was 5 ± 4 days. A comparison of our results has been made with a similar study from Korea, as shown in Table 2.

Early postoperative complications were noted in 20 (33%) patients, including two mortalities. Both of these patients were operated on as an emergency and had longstanding hepatitis C. One patient was labelled as CTP class C who presented with obstructed umbilical hernia. After the procedure the patient developed acute liver failure with porto-systemic encephalopathy and died on postoperative day three. The other patient was CTP Class B, and required emergency repair for a strangulated incisional hernia. The patient developed significant coagulopathy after surgery requiring multiple transfusions of fresh frozen plasma and cryoprecipitate. However, bleeding could not be controlled and the patient ultimately succumbed to death.

The majority of the complications (65%) consisted of deterioration of liver function including spontaneous bacterial peritonitis in 3 (15%), porto-systemic encephalopathy in 5 (25%) and worsening of ascites in the remaining 5 (25%) patients. Postoperatively, CTP class deteriorated from A to B in one patient and from B to C in four patients. Nineteen (31%) patients required transfusion of one or more types of blood products postoperatively. Wound-related complications were observed in 5 (25%) patients, including scrotal hematoma in 2 patients and wound hematoma, ascitic leak from the wound and ascitic collection in the scrotum in one patient each.

Although there was a marked difference in the number of elective and emergency admissions who developed complications (26% versus 50%), it was not statistically significant. A noteworthy finding was made with regards to development of complications in relation to CTP class. Approximately 71% of patients with Child class C cirrhosis developed complications as compared to only 21% of those with either Child class A or B cirrhosis (P < 0.001).

The mean duration of follow-up was 11 ± 5 months. There was recurrence of the herniation in 8 (13%) patients and mean time between surgery and presentation of recurrence was 11 months. The period of recurrence was less than three months in five patients and recurrence occurred in the first few weeks in two patients.

Discussion

Despite numerous assessment tools and recent advances in the management of patients with CLD, abdominal surgery remains a major challenge in these patients. ^7,12 It has also been documented that patients with CLD are at higher risk of developing surgical problems when compared to the normal population. ^6,13 With increasing prevalence of hepatitis B and C in the developing world, ^1,2 the challenge seems to be ever increasing as there is a scarcity of transplantation facilities in these countries. The repair of abdominal wall hernia at the time of liver transplantation has been shown to be safe without any significant complications, ⁴ but out of this rare situation, the outcome of hernia repair in cirrhotic patients has been shown to be poor in numerous large size nationwide studies. ^5,9

In view of the high incidence of complications, a ‘wait and see’ policy has been recommended by some surgeons, but the natural history of abdominal wall hernia in patients with cirrhosis has not been well studied. Marsman et al. compared the outcome of conservative management of umbilical hernia with elective repair in 34 patients with CLD and ascites. ⁴ Over a five-year period of observation, the success rate of initial conservative management was only 23%; 10 out of 13 conservatively managed patients required hospital admission for complicated hernia and out of these, 6 patients required emergency hernia repair. The study showed that conservative management was associated with higher rates of morbidity (77%) and mortality (15%), when compared to elective repair where complication rates were low (18%) and no death observed. The data suggests that ‘simple observation’ of hernia patients in cirrhotic patients may have a detrimental impact on the final outcome.

Risk factors associated with poor outcomes after surgery in patients with CLD have been outlined in previous studies and include need for blood transfusion, low albumin levels, abnormal coagulation profiles, refractory ascites and history of gastrointestinal bleeding. As reflected by these factors, the most important determinant of outcome is the severity of underlying liver decompensation commonly quantified by CTP score. ^5,12 A similar conclusion can be made from our study results as the outcome of patients with CTP class A and B was appreciably better than those patients with Child Class C. Although overall incidence of complications was 33%, the proportion of patients with complications was much higher (71% versus 21%) in CTP class C including two mortalities, and the difference between the two groups was statistically significant.

A recently published study from Korea has suggested that the outcome of hernia repair in patients with CTP class C and gross ascites was similar to patients with CTP class A and B. ¹⁴ But these results need to be interpreted carefully because of many limitations of the study. The authors acknowledge that most of their patients had relatively preserved liver functions with early scores in CTP class C. The majority of the patients (80%) in this study had inguinal hernia as compared to ventral hernia. It has been suggested that outcome of inguinal hernia in cirrhotic patients may be possibly better as compared to ventral hernia. ⁶ Interestingly, the authors have not mentioned any complications in their study.

Another important observation in cirrhotic patients is that outcome of surgical intervention has been universally poor when surgery is performed in an emergency setting. ^5,12,15 Although not statistically significant, the incidence of complications in our study was definitely higher in patients who underwent emergency hernia repair as compared to an elective situation. This issue needs to be further emphasized, especially in developing countries with limited transplantation facilities, that indefinite waiting on cirrhotic patients with abdominal wall hernia may result in the need for emergency surgery with poor outcome.

The recurrence rate after hernia repair in cirrhotic patients has been shown to be higher as compared to normal population. Our rate of hernia recurrence of 13% was similar to the observations made in earlier studies. ¹⁶ The interesting point to note is that most of the patients had recurrence in the initial few weeks or months after surgery. This is despite the fact that 90% of our patients had a prolene mesh used to strengthen the abdominal musculature at the time of surgery. This reflects the innate weakness of the abdominal wall in patients with chronic liver disease and ascites.

Although our study is a single centre study with a relatively small sample size, to our knowledge, no such data on this subject is available from this part of the world. With increasing prevalence of infectious hepatitis and chronic liver disease in the developing countries, there is a need to develop a national database to devise strategies to counter such overwhelming issues and improve the ultimate outcome.

Based on our results we conclude that cirrhotic patients with CTP class A and B can be safely managed by elective repair of abdominal wall hernia after optimization. On the other hand, hernia repair in CTP class C was associated with a higher incidence of early complications and mortality. Except in emergency circumstances, surgery in such patients may either be delayed until the patient is medically optimized or performed early before liver disease progresses to Class C.