How this instrument works
The Child-Pugh score adds five components, each scored 1 (mildest) to 3 (most severe): bilirubin level, albumin level, INR (a measure of blood clotting), the presence and degree of ascites (fluid in the abdomen), and the presence and grade of hepatic encephalopathy (confusion linked to liver dysfunction). The five scores sum to a total between 5 and 15, which sorts into three bands — Class A (5-6), Class B (7-9), and Class C (10-15) — used as a rough marker of how advanced a patient's cirrhosis has become.
The score is adapted from a 1973 paper by Pugh and colleagues in the British Journal of Surgery, itself building on an earlier 1964 classification by Child and Turcotte that used nutritional status among its criteria rather than a coagulation measure. Pugh's original version did not use INR at all — INR wasn't standardized as a laboratory measure until the early 1980s. Instead, Pugh scored prothrombin time in seconds prolonged over a control value: under 4 seconds scored 1, 4 to 6 seconds scored 2, and over 6 seconds scored 3. The INR bands used today — under 1.7, 1.7 to 2.3, and over 2.3 — were mapped onto that same three-point structure later, once INR testing replaced raw prothrombin-time seconds in routine labs.
Ascites and encephalopathy are clinical judgments a physician makes on exam, not lab printouts, so two of this instrument's five inputs still depend on that assessment rather than a number a machine reports. The total is a severity marker used alongside imaging, a patient's full history, and other lab work — not a stand-alone diagnosis of liver disease, and not, by itself, a treatment decision. Many centers now also use the MELD score for transplant-allocation purposes, but Child-Pugh remains widely used for a quick severity read.
- Select the Bilirubin band that matches the lab result: under 2, 2-3, or over 3 mg/dL.
- Select the Albumin band: over 3.5, 2.8-3.5, or under 2.8 g/dL.
- Select the INR band: under 1.7, 1.7-2.3, or over 2.3.
- Select the Ascites finding on exam or imaging: none, mild, or moderate-severe.
- Select the Hepatic encephalopathy grade: none, grade 1-2, or grade 3-4, then read the total.
Worked example — mildest, middle, and most severe totals
All five components at their mildest value — bilirubin under 2, albumin over 3.5, INR under 1.7, no ascites, no encephalopathy — each scores 1: 1+1+1+1+1 = 5, the lowest possible total, sitting at the bottom of the Class A band (5-6).
All five components at their middle value — each scores 2: 2+2+2+2+2 = 10. Checked against the standard three-way split — Class A 5-6, Class B 7-9, Class C 10-15 — a total of 10 sits past the Class B band and into the Class C band, showing how a handful of moderate findings across five separate components can add up to a severity range that no single mid-level finding would suggest by itself.
All five components at their most severe value — bilirubin over 3, albumin under 2.8, INR over 2.3, moderate-severe ascites, grade 3-4 encephalopathy — each scores 3: 3+3+3+3+3 = 15, the maximum possible total, at the top of the Class C band (10-15) and associated with the poorest prognosis of the three classes.
Questions
What do the three Child-Pugh classes mean?
They sort the 5-to-15 total into three bands: Class A (5-6) reflects the best-compensated liver function of the three, Class B (7-9) reflects moderate impairment, and Class C (10-15) reflects the most severe, decompensated disease. The classes are historically tied to surgical risk and prognosis in cirrhosis, and remain a quick severity read used alongside other lab work and imaging.
Did the original Child-Pugh score use INR?
No — this is a frequently misunderstood point. The 1973 Pugh paper scored prothrombin time in seconds prolonged over a control value, not INR, because INR did not exist as a standardized lab measure until the early 1980s. The INR bands used today (under 1.7, 1.7-2.3, over 2.3) were mapped onto Pugh's original 1-to-3 point structure afterward, once INR replaced raw prothrombin-time seconds in routine clinical labs.
Where did the Child-Pugh score come from before 1973?
Pugh's 1973 paper was itself an adaptation of an earlier classification published in 1964 by Child and Turcotte, which graded cirrhosis severity using nutritional status among its criteria rather than a coagulation measure. Pugh's group replaced nutritional status with prothrombin time and refined the cutoffs, producing the version most later work built on.
Why do two of the five components rely on a clinician's judgment?
Ascites and hepatic encephalopathy are graded by physical exam and clinical assessment — how much fluid is detectable in the abdomen, how confused or altered a patient appears — rather than read off a lab printout. That means this calculator's output depends on the accuracy of those two judgment calls as much as on the three lab-based components, which is part of why the score works best alongside a full clinical picture rather than in isolation.
Is a total of 10 Class B or Class C?
Under the standard three-way split — Class A 5-6, Class B 7-9, Class C 10-15 — a total of 10 falls into the Class C band, not Class B, even though it can feel close to the boundary. Because the bands come from adding five separately rounded 1-to-3 scores, totals near a cutoff are worth reading as a number on a 5-to-15 scale first, and a class label second.
Does Child-Pugh replace the MELD score?
No, the two coexist and answer slightly different questions. Child-Pugh, dating to 1973 with roots in a 1964 classification, gives a fast severity read from five inputs, two of them clinical judgments. MELD is a newer, purely lab-based formula that many transplant programs use instead for organ-allocation decisions, in part because it avoids the subjectivity of grading ascites and encephalopathy by exam.
References
- Pugh et al. 1973, Br J Surg — original scoring system (PubMed)
- StatPearls — Use of the Child-Pugh Score in Liver Disease (NCBI)
Read this first: This instrument computes a screening figure from population formulas — it is not a diagnosis, and it cannot see the whole picture a clinician can. Use it to inform a conversation, not to replace one.