MELD scores do not predict patient morbidity while on the liver transplant waiting list

T. Cowling, L. W. Jennings, R. M. Goldstein, E. Q. Sanchez, S. Chinnakotla, S. Dawson, H. B. Randall, G. B. Klintmalm, M. F. Levy

Research output: Contribution to journalArticlepeer-review


The goals of this study were to assess waitlist morbidity in terms of the frequency of health care services utilized by patients while on the liver transplant (LTX) waiting list and to determine whether that utilization can be predicted by the Model for End-Stage Liver Disease (MELD). Sixty-three noncomatose subjects were followed from waitlist placement until death, change in status, LTX, or study discontinuance. Health care events included doctor/clinic visits, labs, outpatient/inpatient tests and procedures, and hospital/intensive care unit days. Listing MELD scores and LTX MELD scores were examined against the number of health care event occurrences within 60 days of listing and 60 days of LTX, respectively, as were changes in MELD scores between listing and LTX and differences in the number of occurrences between the two time points. The only significant correlations noted were between LTX MELD scores and number of hospital days near LTX (r = .360, P = .046) and between LTX MELD scores and the sum total number of occurrences near LTX (r = .370, P = .044). These results suggest that MELD scores do not appear to predict morbidity in terms of health care utilization in patients awaiting LTX. Developing a system capable of predicting waitlist morbidity may lead to the implementation of medical interventions aimed at circumventing foreseeable complications and/or crises in patients awaiting LTX.

Original languageEnglish (US)
Pages (from-to)2174-2178
Number of pages5
JournalTransplantation Proceedings
Issue number5
StatePublished - Jun 2005
Externally publishedYes

ASJC Scopus subject areas

  • Surgery
  • Transplantation


Dive into the research topics of 'MELD scores do not predict patient morbidity while on the liver transplant waiting list'. Together they form a unique fingerprint.

Cite this