Bittermann Therese, Makar George, Goldberg David S
Division of Gastroenterology, Department of Medicine, University of Pennsylvania, United States.
Division of Gastroenterology, Department of Medicine, University of Pennsylvania, United States; Center for Clinical Epidemiology and Biostatistics, Perelman School of Medicine at the University of Pennsylvania, United States; Leonard Davis Institute of Health Economics, University of Pennsylvania, United States.
J Hepatol. 2015 Sep;63(3):601-8. doi: 10.1016/j.jhep.2015.03.034. Epub 2015 Apr 7.
BACKGROUND & AIMS: Urgency-based allocation that relies on the MELD score prioritizes patients at the highest risk of waitlist mortality. However, identifying patients at greatest risk for short-term post-transplant mortality is needed in order to optimize the potential gains in overall survival obtained through improved long-term management of transplant recipients. There are limited data on the predictive ability of MELD score for early post-transplant mortality, and no data assessing the interaction between MELD score and hospitalization status.
We analyzed UNOS data from 2002 to 2013 on 50,838 non-status 1 single-organ liver transplant recipients and fit multivariable logistic models to evaluate the association and interaction between MELD score and pre-transplant hospitalization status on short-term post-transplant mortality.
There was a significant interaction (p<0.01) between laboratory MELD score and hospitalization status on three-, six-, and 12-month post-transplant mortality in multivariable logistic models. This interaction was most pronounced in patients with a laboratory MELD score <25 transplanted from an ICU, whose adjusted predicted three-, six-, and 12-month post-transplant mortality approximated those of patients with a MELD score ⩾30. Compared to hospitalized patients with a MELD score of 30-34, those with a MELD score ⩾35 in an ICU had significantly increased risk of three-month (OR: 1.54, 95% CI: 1.21-1.97), 6-month (OR: 1.35, 95% CI: 1.09-1.67), and 12-month (OR: 1.25, 95% CI: 1.03-1.52) post-transplant mortality.
Pre-transplant ICU status modifies the risk of early post-transplant mortality, independent of MELD score. This should be considered when determining candidacy for transplantation in order to optimize efficient use of a scarce resource.
基于终末期肝病模型（MELD）评分的紧急分配方式会优先考虑等待名单上死亡风险最高的患者。然而，为了通过改善移植受者的长期管理来优化总体生存率的潜在收益，需要识别出移植后短期死亡风险最高的患者。关于MELD评分对移植后早期死亡率的预测能力的数据有限，且尚无评估MELD评分与住院状态之间相互作用的数据。
我们分析了2002年至2013年器官共享联合网络（UNOS）关于50838例非1类单器官肝移植受者的数据，并拟合多变量逻辑模型，以评估MELD评分与移植前住院状态对移植后短期死亡率的关联及相互作用。
在多变量逻辑模型中，实验室MELD评分与住院状态在移植后3个月、6个月和12个月死亡率方面存在显著的相互作用（p<0.01）。这种相互作用在实验室MELD评分<25且从重症监护病房（ICU）进行移植的患者中最为明显，其调整后的移植后3个月、6个月和12个月预测死亡率接近MELD评分≥30的患者。与MELD评分为30 - 34的住院患者相比，在ICU中MELD评分≥35的患者移植后3个月（比值比：1.54，95%置信区间：1.21 - 1.97）、6个月（比值比：1.35，95%置信区间：1.09 - 1.67）和12个月（比值比：1.25，95%置信区间：1.03 - 1.52）死亡风险显著增加。
移植前的ICU状态会改变移植后早期死亡风险，且独立于MELD评分。在确定移植候选资格时应考虑这一点，以便优化对稀缺资源的有效利用。
