Get access

Model for End-Stage Liver Disease score versus Maddrey Discriminant Function score in assessing short-term outcome in alcoholic hepatitis

Authors

  • Monil Kadian,

    1. Department of Medicine, Himalayan Institute of Medical Sciences, Dehradun, Uttarakhand, India
    2. HIHT University, Dehradun, Uttarakhand, India
    Search for more papers by this author
  • Rajesh Kakkar,

    Corresponding author
    1. Department of Medicine, Himalayan Institute of Medical Sciences, Dehradun, Uttarakhand, India
    2. HIHT University, Dehradun, Uttarakhand, India
    • Correspondence

      Dr (Brig.) Rajesh Kakkar, Department of Medicine, Himalayan Institute of Medical Sciences, HIHT University, Swami Ram Nagar, P.O. Doiwala, Dehradun-248140, Uttarakhand, India. Email: dockakkar@gmail.com

    Search for more papers by this author
  • Minakshi Dhar,

    1. Department of Medicine, Himalayan Institute of Medical Sciences, Dehradun, Uttarakhand, India
    2. HIHT University, Dehradun, Uttarakhand, India
    Search for more papers by this author
  • Rajeev Mohan Kaushik

    1. Department of Medicine, Himalayan Institute of Medical Sciences, Dehradun, Uttarakhand, India
    2. HIHT University, Dehradun, Uttarakhand, India
    Search for more papers by this author

  • Potential conflicts of interest: The authors do not have any potential conflicts of interest to declare.

Abstract

Background and Aims

The Maddrey Discriminant Function (mDF) score and the Model for End-Stage Liver Disease (MELD) score are standard prognostic scores for predicting disease severity and mortality in alcoholic hepatitis (AH).This prospective study compared the MELD score and the mDF score as predictors of short-term outcome in AH.

Methods

The admission MELD score and the mDF score were assessed in 47 patients with a diagnosis of AH in the Himalayan Institute Hospital, Dehradun, India and the concordance (C) statistics of the two scores for 28-day mortality were determined and compared.

Results

Both the MELD score and the mDF score on day 1 were significantly higher in non-survivors than in survivors (P = 0.0001 each). The C-statistic for 28-day mortality for the MELD score was 0.91 (P < 0.0001, 95% confidence interval [CI] 0.79–0.97) and for the mDF score 0.90 (P < 0.0001, 95% CI 0.78–0.97). There was no significant difference between the C-statistics of the two scores (P = 0.83, 95% CI −0.07 to 0.09). For predicting 28-day mortality, the optimal MELD score of > 19 (sensitivity 91.6% and specificity 85.7%) corresponded to the mDF score of > 52.8 (sensitivity 91.6% and specificity 82.8%).

Conclusions

Both the MELD score and the mDF score at admission were strong and equally good predictors of 28-day mortality in patients with AH, but the optimal mDF score corresponding to optimal MELD score was higher than the conventional one. Thus, MELD score may be used as an alternative to mDF score for predicting short-term mortality in AH with an advantage.

Ancillary