It’s actually possible to predict with some accuracy which patients with cirrhosis will survive.
The Child-Pugh score, to predict survival in cirrhosis, is pretty simple and is based on only five things:
Bilirubin level: what makes you look yellow jaundiced
Albumin level: the liver makes this key blood protein which acts like a sponge to hold water in the blood vessels. So, when not enough is being made, you get swollen, i.e. edema
INR: a measure of blood clotting which is often abnormal in liver disease since the clotting factors are made in the liver
Ascites: accumulation of fluid in the belly; makes you look pregnant
Encephalopathy: a particular form of brain malfunctioning occurring when the liver fails to purify the blood of certain substances toxic to the brain
In the person you describe, we know his bilirubin level is over 50 (since obvious yellowing of the skin doesn’t happen at lower levels). That gives him 3 points according to the table in the link. Further, your description of his protruding belly and his report that fluid was drained from it prove he has at least mild, and probably severe, ascites. That gives him at least another 2, and maybe 3, points. I will guess he has at least some edema (you NEVER have ascites in cirrhosis without edema), and since edema in cirrhosis almost always means that the albumin level was depressed, that gives him at least another 2 points. Let’s give him the minimum 1 point for each of the remaining two criteria. Add everything up and you get 9 or 10 points.
Consulting the table for 9 or 10 points, he would have around a 50 percent chance of living two years. But, if he also had a high INR level and/or problems with encephalopathy and thus scored a few more points, his two-year survival would be around 35 percent. Surprised?