Inflammatory conditions play a role in recurrence of PSC after liver transplantation: An international multicentre study

Abstract

Background & aims: Liver transplantation (LT) for primary sclerosing cholangitis (PSC) is complicated by recurrence of PSC (rPSC) in up to 25% of recipients. Recurrence has been shown to be detrimental for both graft and patient survival. For both PSC and rPSC, a medical cure is not available. To predict and ideally to prevent rPSC, it is imperative to find risk factors for rPSC that can be potentially modified. Therefore, we aimed to identify such factors for rPSC in a large international multicentre study including 6 centres in PSC-prevalent countries.

Methods: In this international multicentre, retrospective cohort study, 531 patients who underwent transplantation for PSC were included. In 25% of cases (n = 131), rPSC was diagnosed after a median follow-up of 6.72 (3.29-10.11) years post-LT.

Results: In the multivariable competing risk model with time-dependent covariates, we found that factors representing an increased inflammatory state increase the risk for rPSC. Recurrent cholangitis before LT as indication for LT (hazard ratio [HR] 3.6, 95% CI 2.5-5.2), increased activity of inflammatory bowel disease after LT (HR 1.7, 95% CI 1.08-2.75), and multiple acute cellular rejections (HR: non-linear) were significantly and independently associated with an increased risk of rPSC. In contrast to the findings of previous studies, pretransplant colectomy was not found to be independently protective against the development of rPSC.

Conclusions: An increased inflammatory state before and after LT may play a causal and modifiable role in the development of rPSC. Pretransplant colectomy did not reduce the risk of rPSC per se. Recurrent cholangitis as indication for LT was associated with an increased risk of rPSC.

Impact and implications: Recurrence of PSC (rPSC) negatively affects survival after liver transplant (LT). Modifiable risk factors could guide clinical management and prevention of rPSC. We demonstrate that an increased inflammatory state both before and after LT increases the incidence of rPSC. As these are modifiable factors, they could serve as targets for future studies and therapies. We also added further evidence to the ongoing debate regarding preventive colectomy for rPSC by reporting that in our multicenter study, we could not find an independent association between colectomy and risk of rPSC.

Keywords: ACR, acute cellular rejection; CCA, cholangiocarcinoma; CD, Crohn's disease; Cholestatic liver disease; Colectomy; DBD, donation after brain death; DCD, donation after cardiac death; ELTR, European Liver Transplant Registry; HR, hazard ratio; IBD, inflammatory bowel disease; IBD-U, inflammatory bowel disease–unclassified; Inflammatory bowel disease; LD, living donor; LT, liver transplantation; Liver transplantation; MELD, model for end-stage liver disease; MRI, magnetic resonance imaging; PSC, primary sclerosing cholangitis; Primary sclerosing cholangitis; Recurrence of disease; Risk factors; SES-CD, Simple Endoscopic Score for Crohn's Disease; UC, ulcerative colitis; rPSC, recurrence of primary sclerosing cholangitis.

Graphical abstract

Fig. 1

Cumulative incidence (with 95% CI) for rPSC and recipient death. Naturally, a gradual increase of recipient death is observed over the years, whereas the cumulative incidence curve of rPSC is flattening after 15 years of follow-up. LT, liver transplantation; rPSC, recurrence of primary sclerosing cholangitis.

Fig. 2

ACR and risk of rPSC. Results are expressed in HR with respective 95% confidence intervals. The association between multiple ACRs and rPSC is non-linear and is therefore not represented by a single p value. Individuals with 1 or 2 ACRs are incremental at higher risk for rPSC, whereas 3 ACRs are not associated statistically significant with an increased risk. As 4 or more ACRs in 1 patient is scarce, this is less informative and therefore displayed in grey. This graph is derived from the multivariable competing risk Cox proportional hazards model for rPSC. ACR, acute cellular rejection; HR, hazard ratio; rPSC, recurrence of primary sclerosing cholangitis.

Fig. 3

IBD activity before and after LT in individuals with and those without rPSC. The cohort is split into patients who did and those who did not develop rPSC. Each patient’s IBD state is shown before and after LT, and a line connects these states to show pattern differences in IBD state before and after LT. In patients diagnosed with rPSC, an increase of IBD activity after LT is observed, with a noticeable decrease in individuals without an active IBD. IBD, inflammatory bowel disease; LT, liver transplantation; rPSC, recurrence of primary sclerosing cholangitis.