Thread by @ebtapper

How do we provide a prognosis of #cirrhosis?
A #livertwitter #tweetorial

Let's go deep to the very first prognostic tools
& see how everything old is new again.

Keywords: Child class, MELD, Frailty

First, a question:
What is the expected transplant-free survival for a person with a recent diagnosis of #cirrhosis?

Act 1: Necessity is the mother of invention

Imagine you are a patient w/a variceal bleed in 1940’s (Fig1)
No EGDs
No Banding/sclerotherapy
No hope!

And then came portosystemic shunt surgery(Fig2, I 👀U, A. Oldfather Whipple!)
Problem: Surgery⬇️rebleed but often⬇️survival (Fig3)

We needed a method 2 select patients who will benefit from shunt surgery
Options
Shaldon: Do they look & feel well?(Fig1)
Linton: Check the cephalin flocculation!(Fig2, dont ask)
McDermott: Consult the albumin & Bili, optimize nutrition!(Fig3)
Child & Turcotte: Hold my🍺(Fig4)

Child, like Karnofsky, knew: have 2 avoid harm 4 those who wouldnt benefit(Fig1)
Performance status 4 chemo(Fig2)
"hepatic reserve"4 shunt surgery, defined by albumin, bili, severity of HE & ascites, and nutritional status (think performance status or eye-ball test)(Fig4)

Enter Pugh (who liked transecting bleeding esophagi over shunt surg - u read that right!)(Fig1)
⭐️PT/INR in, nutrition out
⭐️Classes now based on 15pt scale

BAM: Liverworld changed forever!

Act 2: New tech, higher-stakes
3 main developments since Child-Pugh:
1⃣Stats goes pro. Cox describes the hazard ratio,1972. Hepatologists use it to model predictors(Fig1)
2⃣@Mayoclinic pubs series of prognostic models🔼focus on purely OBJECTIVE data(Fig2)
3⃣Hello, TIPS!(Fig3)

TIPS=game-changer,🔼access 2 effective tx 4 uncontrolled bleeding&ascites
Big Problems:
Fig1: early reports had unacceptably🔼mortality
Fig2: most patients w/bleeding are Child C; we need 2 redefine the spectrum of risk within Child C
Fig3: Enter the risk modelers
Fig4: MELD!

What makes MELD special?
Innovation 1: Started as nomogram that avoids math (if u use a ruler)
Innovation 2: Predicted mortality in ALL-COMERS w/Child B-C (not just TIPS)
Innovation 3: Put score online 4all 2 use
Come @ a time when we needed objective score 2 allocate livers(Fig3

MELD is awesome, but can be improved:
1. Creatinine doesnt fully capture hepatorenal dysfx, hence MELD-Na.(Fig1)
2. HE matters! Even early/covert HE(Fig2)
3. Extrahepatic organ failure matters: Meet "ACLF" or Acute on Chronic Liver Failure(Fig3)
But wait ... there's MORE

BACK TO THE FUTURE
Remember ‘nutritional status’ from ChildClass?
Well, rigorous ways to capture it are so hot right now
How about Karnofsky? That works (Fig1)
Ask about ADLs, any1 can do it(Fig2)
Sarcopenia, measure muscle bulk(Fig3)
Frailty:Handgrip, Chairstands, Balance (Fig4)

Act 3: Mind the gap
MELD gave us: 🔼risk stratification 4 ChildC, objective transplant allocation
Unsolved: Risk discrim 4 Child A. Median survival 12yrs w/range 10-20y!!(Fig2)
Problem: Most pts w/#cirrhosis r Child A
Future: Predicting decompensation e.g. aasldpubs.onlinelibrary.wiley.com/doi/abs/10.100…

To summarize
⭐️Portosystemic shunt surgery/TIPS spurred risk score development(Child, MELD)
⭐️Child=stable(10yrs) vs sick(1-2y)
⭐️Prognosis if decompensated=MELD
⭐️Child class had 'nutritional status' and HE; MELD doesnt. But factors like HE, disability, frailty are a/w mortality

This concludes a #tweetorial on prognosis in #cirrhosis. Thanks for reading! I hope you enjoyed it! There are lots of things that are associated with survival that I did not discuss. Please check out: journal-of-hepatology.eu/article/S0168-… for more.