Presentation is loading. Please wait.

Presentation is loading. Please wait.

Enhancement of MELD W. Ray Kim, MD Mayo Clinic College of Medicine Rochester, MN Plummer Building, Rochester, MN.

Similar presentations


Presentation on theme: "Enhancement of MELD W. Ray Kim, MD Mayo Clinic College of Medicine Rochester, MN Plummer Building, Rochester, MN."— Presentation transcript:

1 Enhancement of MELD W. Ray Kim, MD Mayo Clinic College of Medicine Rochester, MN Plummer Building, Rochester, MN

2 Agenda How to make MELD better ̵ As an indicator of pretransplant mortality Add more variable(s): MELDNa Optimize model: Refit MELD ̵ As an indicator of posttransplant outcome Limitations

3 Risk of 3-Month Mortality (after adjusting for MELD) Na (mEq/L) Serum Sodium and Mortality Kim NEJM 2008;1018 UNOS Registrants, 2005 N=6769

4 MELDNa: Incorporating Na to MELD MELD Na (mEq/L) MELDNa = MELD - Na *MELD*(140-Na) Kim NEJM 2008;1018

5 Validation: MELDNa vs MELD < MELD Observed/Expected Based on MELDBased on MELDNa Kim NEJM 2008;1018

6 Refitting Coefficients Methods Waitlist data obtained from OPTN from : ̵ Model derivation set (05-06): n=14,214 ̵ Model validation set (07-08): n=13,945 All adult, primary LTx candidates with end stage liver disease included (HCC and status 1 excluded) The proportional hazards regression analysis predicting mortality within 90 days of listing ̵ Define lower and upper bounds ̵ Optimize coefficients ̵ Bilirubin, Creatinine, INR, and Na

7 Bilirubin data

8 Bilirubin data LB=1.0 Current MELD

9 Bilirubin data LB=1.0 Refit

10 Creatinine data

11 Creatinine data UB=4.0LB=1.0 Current MELD

12 Creatinine data LB=0.8UB=3.0 Refit

13 INR data

14 INR data LB=1.0 Current MELD

15 INR data Refit UB=3.0 LB=

16 Sodium data

17 Sodium data Refit LB=125UB=140

18 Validation: Bilirubin data * *Sharma et al. Gastro. 2008;1575

19 Validation: Creatinine data *Sharma et al. Gastro. 2008;1575 *

20 Validation: INR data *Sharma et al. Gastro. 2008;1575 *

21 Validation: MELD data LB (6) and UB (40) ignored for comparison purpose

22 Validation: SRTR MELD data

23 Validation: Refit MELD(Na) data

24 The Bottom Line Validation Set ( ) Model Chi-SquareConcordance (SE) MELD (0.0062) SRTR MELD2055 (-8) (0.0063) MELDNa2370 (+7) (0.0059) Refit MELDNa2374 (+11) (0.0058)

25 Kitchen Sink Crystal Ball Unsupervised incorporation of all potentially relevant variables * includes: refit MELDNa, age, BMI, albumin, HCV, diabetes, prior malignancies ** includes refit MELDNa, age, sex, race, albumin, HCV, HBV, diabetes, prior malignancies, life support, encephalopathy, ascites, previous abdominal surgery, portal vein thrombosis C-statistics

26 MELD and Resource Utilization pRBC LOS Kim ATC. 2004

27 Transplant Benefit Survival w/o LTx Survival after LTx Days Survival C-statistics Refit MELDNa Deciles Mean MELD

28 Conclusions Ways to improve prediction of waitlist mortality ̵ Add more variable: MELD Na Represents a meaningful improvement Big change in mortality in a subgroup of patients ̵ Optimize model coeffcients: Refit MELDNa New upper and lower bounds: Creatinine: INR: ̵ Limit to predictability MELD and post-transplant outcome ̵ Statistically significant impact on survival and resource utilization ̵ Limited accuracy


Download ppt "Enhancement of MELD W. Ray Kim, MD Mayo Clinic College of Medicine Rochester, MN Plummer Building, Rochester, MN."

Similar presentations


Ads by Google