Hepatitis C virus (HCV) affects 200 million people.¹ Approximately 85% of people with HCV will develop chronic infection; of those, 10% to 30% will develop cirrhosis. The prevalence of HCV within the dialysis population is as high as 13%.² HCV is a negative prognostic indicator for survival on dialysis and after kidney transplantation. There is an increased risk of death among long-term hemodialysis patients infected with HCV.³ Importantly, overall survival in these patients is improved after kidney transplantation compared with dialysis.⁴ Liver biopsies are indicated in all HCV-positive candidates considered for kidney transplantation. Up to 12% of asymptomatic patients will have cirrhosis. Those with cirrhosis (while on dialysis) have a 35% higher death rate than their counterparts without cirrhosis.⁵

Established cirrhosis is an important predictor of death after renal transplantation and is considered a relative contraindication to isolated renal transplantation. American Association for Study of Liver Disease (AASLD) guidelines recommend end-stage renal disease patients with cirrhosis be evaluated for dual-organ transplantation. The core curriculum in nephrology and the Kidney Disease: Improving Global Outcomes (KDIGO) initiative consider HCV-related cirrhosis a contraindication to kidney transplantation alone (KTA).^6,7 Some authors consider cirrhosis a relative contraindication for KTA because the prospect of survival for graft and patient is dismal.⁸

The use of KTA in asymptomatic patients with cirrhosis has not been extensively studied. Reports often exclude patients with cirrhosis,⁹ are limited by small numbers, or combine clinically compensated patients with cirrhosis with those who have only mild fibrosis.¹⁰ The United Network for Organ Sharing (UNOS) database does not track biopsy results, so registry data cannot be mined.

We performed 18 KTAs on clinically compensated patients with cirrhosis (CCCs) and compared the results to those from a control group of HCV-positive KTA recipients without cirrhosis. We surmised that the results would be equivalent between groups.

Patients and Methods

We performed a retrospective Institutional Review Board–approved review including all deceased-donor HCV⁺ kidney transplants from January 2001 to June 2010. All of the patients were followed for 3 years after they underwent kidney transplantation. Data were collected from a computerized database, paper and electronic charts, and UNOS.

A total of 147 HCV⁺ kidney transplantations were performed. We excluded 8 combined liver and kidney transplantations (Fig. 1).

View Figure

• Download as Powerpoint
• Download Figure

Results

The donor and recipient demographics are in Tables 1 and 2, respectively. The total number of transplants analyzed was 121, as reflected in Table 1. However, because 8 patients received 2 transplants within the study period, the first transplant was removed from analysis of presurgery variables in Table 2 and patient survival outcomes in Table 3. This would alleviate analysis questions about characteristics of observations within a given sample predicting the outcomes of the other observations. Therefore, 95 recipients were in the NCG as mentioned at the bottom of Table 2 (103 − 8 = 95). In addition, 2 patients with PNF, although included for patient survival outcomes data in Table 2, were excluded from data analyzing DGF, 12-month creatinine, and graft survival, as seen in Table 3. Therefore, 102 (103 − 1) allografts were in the NCG and 17 (18 − 1) were in the CCC (as mentioned at the bottom of Table 3). Table 4 demonstrates the characteristics of the 18 patients with cirrhosis in detail. Among all tested variables, only mean donor age (P = 0.0073) and cancer rate for the cirrhosis group (P = 0.002) were significant, and the rest were nonsignificant. Both groups were followed for 3 years. We collected all of the deaths and graft losses during the long-term follow-up.

Table 1 Donor demographics

View Fullscreen

Table 2 Recipient demographics

View Fullscreen

Table 3 Outcomes

View Fullscreen

Table 4 Characteristics of the 18 patients with cirrhosis

View Fullscreen

Laboratory data for the patients with cirrhosis

Platelet and albumin levels for the CCC are listed in Table 5. Of the 18 patients, 2 lacked the time-appropriate albumin level. The average pretransplantation albumin level for the remaining 16 CCCs was 3.55 (±0.57) mg/dL. The average platelet count at transplantation was 154.72 (±87.18) per microliter. A total of 15 patients had a MELD score of 20, 1 patient had a MELD score of 21, 1 had a MELD score of 22, and 1 had a MELD score of 25. All 18 patients met Child A classification criteria.

Table 5 Recipient demographics (lab data)

View Fullscreen

Summary of outcomes analysis

Our outcomes are consistent with the hypothesis that both groups perform similarly. Twelve-month creatinine levels among functioning allografts were lower among the patients with cirrhosis but fell short of significance (1.55 versus 1.92; P = NS). The same situation was observed with 36-month creatinine levels compared between the two groups (2.03 versus 2.16; P = NS). These outcomes were then also assessed among only SCD kidneys, and the same results were found.

The 1- and 3-year cumulative patient survival rates were 91% and 82%, and 100% and 83%, for the noncirrhosis and cirrhosis groups, respectively (P = NS). The 1- and 3-year cumulative graft survival rates censored for death were 94% and 81%, and 95% and 82% for the noncirrhosis and cirrhosis groups, respectively (P = NS). Figs. 2 to 5 show the 1- and 3-year patient and graft survival curves. The median survival time for patients with cirrhosis was 50 months. For additional evidence, we compared only patients with SCD kidneys. The 1- and 3-year patient survival rates were 91% and 84%, and 100% and 84% for the noncirrhosis and cirrhosis groups, respectively (P = NS). Figs. 6 and 7 show the 1-year patient and graft survival rates for the patients with SCD kidneys.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

The one- and three-year death-censored graft survival rates were 95% and 81% for noncirrhotic patients, respectively. For cirrhotic patients, the one- and three-year death-censored graft survival rates were 94% and 74%, respectively.

Discussion

Multiple studies have demonstrated diminished outcomes in patients with cirrhosis undergoing abdominal surgery, thoracotomy, open-heart surgery, and orthopedic surgery.^11–15 The life expectancy of patients with cirrhosis is 40% of that of the general population.¹⁶ The standardized mortality ratio for patients with cirrhosis for all causes of death combined is increased by 12-fold, with a relative 1-year survival rate of 67%. There is a 5- to 22-fold increased risk of dying from infectious diseases.¹⁷ Patients with cirrhosis (while on dialysis) have a 35% higher death rate than their noncirrhosis counterparts.⁵

The median survival time of cirrhotics with a MELD score of 20 or less is 11 months.¹⁸

The National Kidney Foundation and the AASLD consider cirrhosis a relative contraindication to KTA. It has been suggested that HCV-infected patients with cirrhosis but clinically compensated liver disease be considered for kidney transplantation only in an investigational setting.⁷ There are minimal data or there is minimal literature to support this.^10,19 Some authors recommend KTA for selected patients with early-stage cirrhosis.²⁰

In many large reports of HCV⁺ recipients, the liver is not histologically monitored or patients with cirrhosis are excluded outright.^9,21 Current recommendations to evaluate for combined kidney-liver transplantations in CCCs may not be the most appropriate allocation of resources.^6,7 In addition, the benefits of combined (prophylactic) liver and kidney transplantation in CCCs would have to be weighed against the added risk of performing OLT.²²

A total of 19% of physicians surveyed would perform a KTA on an asymptomatic patient with cirrhosis.² Arango et al²³ performed KTA on 5 patients with cirrhosis. Of these, 2 were alive at the time of publication and the other 3 died 3, 9, and 10 years later. Paramesh et al²⁴ studied 37 HCV⁺ patients who underwent KTA. Their cohort was composed of 9 patients with cirrhosis and 28 patients without cirrhosis. Hepatic portal venous gradient was less than 10 mmHg for all patients. They demonstrated equivalent 1- and 3-year patient and graft survival rates. In their study the negative prognosticators included recipient age and albumin level.²⁴ They concluded that KTA may be safe in patients with HCV⁺ compensated cirrhosis. Their average follow-up was 32 months.

Our larger experience with KTA (a total of 121 KTAs and 18 CCCs) with at least a full 3-year follow-up for all patients confirms this. The 12- and 36-month creatinine levels, incidence of DGF, rejection, and patient and graft survival rates did not differ between patients with cirrhosis and HCV⁺ patients without cirrhosis. According to one study, patients with MELD scores lower than 20 had a median survival time of 11 months.¹⁸ We observed a median survival time of 50 months for the cirrhosis group. In fact, all of those patients were alive over a 12-month postoperative period. They had a median MELD score of 20.

Many investigators have found that the increased mortality in KTA for HCV⁺ patients appears to be closely related to infectious complications.^9,17,19,25 One study reported a 5- to 22-fold increased risk of dying of infectious diseases.¹⁷ In our experience, the presence of cirrhosis did not lead to an increased likelihood of death or graft loss secondary to infection compared with our HCV⁺ controls without cirrhosis (Table 2).

A high proportion of our study patients had cancer, a known risk for organ transplant recipients. Recent evidence suggests that in the next two decades, cancer will surpass cardiovascular disease as the most common cause of death.²⁶ In one study, the standardized incidence ratio was 2.0.²⁷ One of the most common cancers showing excess risk (in all solid organ transplant recipients) was liver cancer.²⁷ The incidence of HCC in individuals with chronic hepatitis is as high as 0.46% per year. The annual risk of developing HCC in patients with cirrhosis is 1% to 6%.²⁸ The 5-year cumulative risk of developing HCC in patients with both HCV and cirrhosis is 17%.¹¹ The combination of a potentially oncogenic virus, cirrhosis, and immunosuppression exposes these patients to significant oncologic risk. In our cirrhosis cohort, 3 of the 6 patients who contracted cancer had undergone transplantation before, thereby increasing their risk because of the increased immunosuppression exposure. Of the 3 patients who developed HCC after KTA, all were listed for OLT. One died while waiting in his seventies with a functioning allograft. Two underwent liver transplantation with functioning renal allografts. It is doubtful that prophylactic liver transplantation for 18 recipients could be justified, because 3 of the cohort subsequently developed HCC. Instead, close surveillance with “salvage” liver transplantation appears to be the more prudent course. It is important to note that in many regions of the country, combined liver and kidney transplantation with a MELD score in the low twenties is simply not realistic. Additionally, if it had been decided to delay transplantation until the development of HCC, many patients in our cohort would have missed out on the potential benefits of a functioning kidney allograft while they gambled on remaining suitable candidates for combined transplantation in the future.

It has been established that liver transplant recipients who experience renal failure due to calcineurin inhibitor toxicity and/or other causes have diminished survival that is improved by KTA.²⁵ About 40 to 50 of these transplantations are performed annually.²⁹ Gonwa et al³⁰ reported that patients who suffered from renal failure after liver transplantation had a 6-year survival rate of 27% if they remained on dialysis. For those who received KTA, their 6-year survival rate was 71.4%. It is unknown how many of these patients also suffered from biopsy-proven allograft cirrhosis.

In our experience with 6 patients, 5 were alive with functioning allografts at 1 year. Two patients underwent KTA with biopsy-proven cirrhosis in their liver allograft. Both patients and all allografts were alive and functioning at 1 year. One patient later died from liver decompensation with a functioning kidney. The other died from a malignancy 5 years after KTA. His allograft was functioning. Of the 4 patients who underwent KTA after OLT in the noncirrhosis comparison group, 1 died with a functioning allograft at 5 months, and 1 lost allograft function at 5 years and died in the sixth year. The remaining 2 patients are alive with functioning allografts 2 and 4 years after KTA, respectively. These numbers are too small to address with any statistical relevance. However, the 1-year results appear to be acceptable.

Donor age in the cirrhosis group was significantly lower, and all donors for the cirrhosis group were SCD. Statistical analysis did not reveal that our outcomes were solely based on the use of SCD.

The use of kidneys from HCV⁺ donors is associated with worse outcomes than those from HCV⁻ donors. Kucirka et al³¹ reported that HCV⁺ donor kidneys are 2.5 times more likely to be discarded. However, receiving an HCV⁺ donor kidney is associated with improved survival compared with remaining on dialysis.^31–33 The use of HCV⁺ donor kidneys is associated with a more than 12-month shorter wait time on dialysis. This shorter wait time may be associated with a 10% to 15% lower rate of death.

In neither of our study groups did we find HCV⁺ donor status to be a negative prognosticator for patient or allograft survival. We believe that the aggressive use of HCV⁺ donors is appropriate.

Weaknesses of this paper include its retrospective nature and relatively small patient numbers (although it is the largest study ever conducted). We note that cirrhosis was not determined entirely by biopsy, and in some instances was determined entirely radiologically. There is a possibility that abdominal images and biopsies were overread or underread by the radiologists and pathologists. Percutaneous liver biopsy only represents approximately 1/50,000 of the entire liver and may underestimate the degree of fibrosis. There is a false-negative rate of 25% to 30%, along with a discordance rate of nearly 50% on repeated liver biopsy specimens.^34–36 Imaging modalities are exceptionally sensitive and specific for cirrhosis, demonstrating statistical significance in distinguishing between chronic hepatitis and cirrhosis for specific imaging parameters on computed tomography, magnetic resonance imaging, and ultrasound.^35,36 These imaging modalities are specific enough to confirm cirrhosis. In the 19 patients with both imaging and histology, findings were consistent throughout, with no discrepancies identified.

Given the lack of histologic samples from all patients in our cohort, we could not stratify any data by fibrosis staging. Instead, the cohort was stratified by a more definitive means, the presence or absence of cirrhosis.

We did not routinely measure portal gradients.^37,38 Measuring portal pressures is a relatively new practice. Instead, consistent with the practice of this cohort's timeline, compensated hepatic function was determined clinically by the lack of encephalopathy, ascites, varices/upper gastrointestinal bleeding history.

We did not track the viral genotypes in the recipients or in the HCV⁺ donors. Genotypes have not been shown to have a prognostic impact in kidney transplantation or renal failure patients. It is a marker of IFN susceptibility, not overall prognosis in the untreated patient. There are no recommendations in the current literature to limit HCV⁺ donation to genotype 1 patients.³⁹ We also could not determine whether patients were treated with IFN before transplantation and what their responses were.

Conclusion

Current recommendations to evaluate for combined kidney-liver transplants in HCV⁺ patients with cirrhosis may not be appropriate. We have demonstrated that HCV⁺ CCCs with KTA share survival and graft survival rates comparable with their HCV⁺ counterparts without cirrhosis, without higher rates of acute rejection or delayed graft function. However, we did note a high rate of malignancy. In order to prevent malignancy we may have to consider avoiding polyclonal antibody induction. The role of active surveillance is important as well. There are no data to support the contention that CCCs should wait on dialysis until they decompensate or develop HCC. Although the numbers of patients we present are small, this study represents the largest documented experience to date with a robust control group and 3-year follow-up with standardized induction and maintenance immunosuppression. At this point we recommend that KTA be considered in CCCs who have no signs or history of ascites, encephalopathy, upper gastrointestinal bleeding, thrombocytopenia, hypoalbuminemia, or HCC. These patients should have a MELD score no higher than 22 and a Child classification no higher than A. Additionally portal pressure measurements should be performed per established guidelines. A pressure gradient higher than 10 mmHg should warrant consideration for dual organ transplantation. Nevertheless, larger multicenter studies with longer follow-up will be necessary.