Liver Transplantation for HIV/HCV Coinfection

Where Is the Controversy?

Emily Dannhorn; James P O'Beirne


Future Virology. 2013;8(7):639-648. 

In This Article

Recent Data: What Have We Learnt?

The two recent large studies outlined above have identified a number of pre- and peri-transplant variables that impact on mortality, and have suggested a possible subgroup of coinfected patients that, in the absence of these high-risk features, can do well with LT (Box 2).

The detailed analysis afforded by these studies gives interesting insight into potential areas of improvement in the selection of LT recipients and their subsequent management.

Coinfected patients on the LT waiting list often have lower BMIs at the time of transplant compared with monoinfected controls. There are many potential reasons for this observation, but as malnutrition is a common feature of advanced liver disease and has prognostic importance, it is possible that the coinfected patients had more advanced liver disease at the time of listing, the severity of which was not reflected by the MELD score, as has previously been described.[21,37] BMI is a reproducible and reliable marker, which could be incorporated into future selection criteria for this cohort of patients in order to improve prognostication over the MELD score alone. Following on from this, it seems logical that patients at risk of malnutrition should receive aggressive nutritional support in order to maintain BMI >21. This would be facilitated by early referral to the transplant center for patients who are losing weight independently of the MELD score. The importance of renal function is highlighted by the outcomes reported in the Terrault et al. study.[35] Combined liver–kidney transplantation is utilized in patients with renal dysfunction expected not to recover following LT. The poor outcome in those patients receiving combined grafts as well as recent data showing worse renal function post-LT in coinfected patients receiving a liver alone suggest that renal impairment is an ominous prognostic sign in potential LT recipients, and highlights once again the importance of early referral. This finding is of major relevance to coinfected patients seeking LT in systems that use the MELD allocation system. As creatinine is a component of the MELD score, patients with renal dysfunction will have a higher MELD score and therefore are more likely to be transplanted. For the coinfected patient this means that they will only have a chance of being transplanted when they are at risk of a poor outcome due to renal impairment.

Donor selection has also been reviewed, and several factors affecting mortality and graft loss have been described. For patients with coinfection the use of non-HCV-positive donors was associated with increased rates of graft loss; a finding not seen in patients post-LT with HCV monoinfection. This finding is unexplained and merits further work, as these grafts are a valuable source of organs in the era of organ shortage. The finding of increased rates of graft loss in patients receiving livers from older donors is perhaps unsurprising as this is a recognized risk factor for severe HCV recurrence in monoinfected patients.[38] DCD donors are associated with increased risk of primary nonfunction and biliary complications.[39,40] The effect of DCD on the severity of HCV recurrence is debated, but it may be relevant in coinfected patients, given that this was a significant factor for graft loss in the multivariate analysis.[41,42] DCD donors represent an important and increasing source of grafts, and so to suggest that DCD donors should be avoided for coinfected recipients risks prejudicing their outcome by increasing the risk of dying on the waiting list while waiting for a liver. Certainly it seems logical that if DCD doses are used in coinfected patients they should be only those from younger donors with shorter ischemic times.

Higher rates of cellular rejection were also described in both of the aforementioned recent studies, which were associated with worse HCV recurrence. Although the definite reasons for this have yet to be ascertained, increased rejection has been attributed to the known immunomodulatory effects of HIV, as well as the difficulties in achieving optimal immunosuppression post-LT.[43] Treatment for rejection has been identified as a risk factor for severe HCV recurrence and graft loss, emphasizing the importance of preventing acute rejection in the coinfected cohort.[33] It has been suggested that overly cautious immunosuppressive use post-LT in these patients, due to concerns about worsening HIV-related diseases, has resulted in increased episodes of rejection. Similarly, the complex drug interactions have made reaching optimal levels of immunosuppression very challenging.[44]

The major issue with medication after LT in HIV/HCV-coinfected patients is the complex range of drug interactions between cART and immunosuppressive drugs. The majority of immunosuppressive medications used are metabolized by the cytochrome P450 system. Protease inhibitors, including ritonavir, can reduce the metabolism of calcineurin inhibitors, such as cyclosporine and tacrolimus, which make up the mainstay of immunosuppression regimens post-LT. The therapeutic range for tacrolimus is narrow and significant side effects such as nephrotoxicity and neurotoxicity often occur at levels above >15 ng/ml.[45] The usual therapeutic range for tacrolimus is 5–10 ng/ml, which is close to the toxic level; therefore, medications that could increase tacrolimus levels should be used with caution. Combined prescription of tacrolimus and ritonavir-boosted PI regimens requires a marked decrease in dose of tacrolimus and close drug monitoring to avoid toxicity. Published case reports have used single doses of 0.5–1 mg of tacrolimus every 1–3 weeks when combined with ritonavir-containing cART, and patients have required frequent therapeutic drug monitoring to maintain constant drug levels within the therapeutic range.[46] This is very challenging to maintain in clinical practice, and while this approach is successful in preventing overt toxicity, less is known about the actual drug exposure during the extended dosing period, leading to concerns that patients may be underimmunosuppressed. The complex interplay between levels of immunosuppression, maintaining HIV negativity with cART and avoiding toxicity is a challenge, and it is perhaps no surprise that center experience is a major factor in transplant outcome. In fact, mortality is almost three-times higher in coinfected patients undergoing LT at centers with low transplant activity (<1 LT/year for HIV) compared with more experienced centers.[32]

Miro et al. factored center experience into a risk score for mortality using other pretransplant variables using the formula: Exp([0.81966 if genotype = 1] + [0.05748 MELD pre-LT] + [1.03540 if center does <1 orthotopic LT in HIV-infected patients/year]).[32] A risk score cutoff of 1.07795 allowed classification of the 84 LT recipients as having a low risk or a high risk of death. Sixty patients in the study were calculated to be low risk and 5-year survival in this cohort was excellent, at 70%.