Letter on ‘Post-TIPS Overt Hepatic Encephalopathy Increases Long-Term but Not Short-Term Mortality in Cirrhotic Patients With Variceal Bleeding: A Large-Scale, Multicentre Real-World’

Abstract

We read with great interest the article published in Alimentary Pharmacology & Therapeutics regarding the impact of post-TIPS overt hepatic encephalopathy (OHE) on long-term mortality in a retrospective cohort of cirrhotic patients with variceal bleeding [1]. In particular, multivariable competing risk analysis, adjusting for potential confounders, showed that the onset of post-TIPS OHE was an independent predictor of all-cause mortality beyond the 24-month follow-up period, but was not associated with short-term mortality. Furthermore, from the multivariable analysis, in addition to OHE, other predictors are associated with a higher risk of mortality, including age and MELD score. Therefore, especially with such long follow-up, it becomes difficult to discriminate the role of each single predictor since the effect is probably to be understood as additive, especially if they are non-collinear predictors.

Similarly to Hartl et al. [2], competing risk regression analysis revealed that early onset of post-TIPS OHE (within the first month) was significantly associated with higher long-term mortality risk (SHR: 2.390 [2.020–2.840], p < 0.001). This result was not observed in our cohort [3]. Could this result be due, at least in part, to the high rate of stent dysfunction (27%–30% of patients) observed in the Chinese cohort? However, as previously described [2, 3] when dealing with time-sensitive variables such as the development of OHE after TIPS, authors should have used Landmark analysis to avoid immortal time bias [4] although it includes certain limitations such as the impact of when the landmark is chosen on the obtained results or reduced sample size [5].

Another point to be noted is that more than half of the patients enrolled by Xiang et al. had TIPS placed for emergency haemostasis. In our cohort, we decided to exclude these patients because preemptive TIPS is known to be associated with reduced mortality and an incidence of OHE similar to pharmacological and endoscopic combined treatments. Therefore, the inclusion of patients submitted to preemptive TIPS may have altered per se the impact of OHE on survival [6]. Moreover, regarding this, Xiang et al. observed that a higher proportion of patients with OHE had variceal bleeding, which may have been the trigger for HE and mortality.

As already observed by Kumar et al. [7], a deeper analysis of post-TIPS OHE, including time to onset, severity, precipitating factors and their correlation with mortality, would provide a clearer understanding of its impact on patient outcomes. In particular, the authors do not specify how many patients develop recurrent or persistent HE which, as already described [8], are those at highest risk of mortality and therefore deserve a separate analysis.

Nevertheless, further prospective studies considering also malnutrition, frailty and cardiac index are needed to confirm these results and intensive research to improve patient selection and enable risk stratification remains crucial to ameliorate the quality of life of patients and caregivers and to avoid undermining the positive effects of TIPS on survival.

Silvia Nardelli: conceptualization, writing – original draft, methodology. Francesco Vizzutti: validation, methodology, supervision. Filippo Schepis: validation. Manuela Merli: formal analysis, supervision. Simone Di Cola: validation. Stefania Gioia: validation. Oliviero Riggio: supervision, formal analysis.

The authors declare no conflicts of interest.

This article is linked to Xiang et al. paper. To view this article, visit https://doi.org/10.1111/apt.18509.