Relationship of Protein Calorie Malnutrition to Alcoholic Liver Disease: A Reexamination of Data from Two Veterans Administration Cooperative Studies

Authors

  • Charles Mendenhall,

    Corresponding author
    1. Department of Veterans Affairs Medical Center, Cincinnati, Ohio; the Department of Biostatistics, University of Cincinnati Medical Center, and the VA Cooperative Studies Coordinating Program Center, Hines, Illinois.
    Search for more papers by this author
  • Gary A. Roselle,

    1. Department of Veterans Affairs Medical Center, Cincinnati, Ohio; the Department of Biostatistics, University of Cincinnati Medical Center, and the VA Cooperative Studies Coordinating Program Center, Hines, Illinois.
    Search for more papers by this author
  • Peter Gartside,

    1. Department of Veterans Affairs Medical Center, Cincinnati, Ohio; the Department of Biostatistics, University of Cincinnati Medical Center, and the VA Cooperative Studies Coordinating Program Center, Hines, Illinois.
    Search for more papers by this author
  • Thomas Moritz,

    1. Department of Veterans Affairs Medical Center, Cincinnati, Ohio; the Department of Biostatistics, University of Cincinnati Medical Center, and the VA Cooperative Studies Coordinating Program Center, Hines, Illinois.
    Search for more papers by this author
  • The Veterans Administration Cooperative Study Groups 119 and 275

    1. Department of Veterans Affairs Medical Center, Cincinnati, Ohio; the Department of Biostatistics, University of Cincinnati Medical Center, and the VA Cooperative Studies Coordinating Program Center, Hines, Illinois.
    Search for more papers by this author

  • This study was supported by the Department of Veterans Affairs, Cooperative Studies Program.

Reprint requests: Charles L. Mendenhall, M.D., Ph.D., Department of Veterans Affairs, Medical Center (151F), 3200 Vine Street, Cincinnati, OH 45220.

Abstract

The relationship of protein calorie malnutrition (PCM) to alcoholic liver disease was studied in 666 patients enrolled in two Veterans Administration Cooperative Studies. Some findings of malnutrition could be detected early in 62% of the comparison patients (43 subjects who were alcoholic, but had not yet developed clinical or laboratory evidence of liver injury). In those who had progressed to the stage of liver injury sufficient to manifest clinical jaundice (536 patients), some findings of malnutrition were present in every patient (100%). The degree of malnutrition correlated closely with the development of all the serious complications of the liver disease (ascites, encephalopathy, and hepatorenal syndrome), as well as the overall mortality.

The degree of malnutrition was also important in predicting response to some forms of treatment. When prednisolone, a catabolic adrenal steroid, was used, efficacy was independent of the level of malnutrition. However, a relationship was observed with the severity of the liver injury [quantified by the level of jaundice and coagulopathy, i.e., Maddrey's discriminant function (DF(Maddrey))]. For prednisolone, the response was seen only when the DF was 81-100 reducing mortality 45%.

When oxandrolone, an androgenic anabolic steroid treatment was given, efficacy was observed only in those with moderate malnutrition (PCM score 60–79% of normal) and maximized with adequate caloric intake reducing mortality 86%. To simplify the method of calculating the PCM score for predicting response to anabolic therapy, a multiple logistic regression model was developed from the parameters used to assess nutritional status: DF(PCM)= 0.098 (peripheral blood lymphocytes) + 0.078 (creatinine height index). Using the DF(PCM), oxandrolone improved survival 42% when the DF(PCM) exceeded 6.5, but was < 11.0. These results suggest that the two DFs should be determined in patients with life-threatening alcoholic liver injury and the appropriate therapy administered based on the observed results.

Ancillary