Physical Exam in Chronic Liver Disease

General appearance and vital signs can reveal advanced liver disease at a glance. These findings are often present before the patient mentions any liver-specific complaints.

How much can the exam actually tell you?

The JAMA Rational Clinical Examination review of this question pooled 86 studies and 19,533 patients (4,725 with biopsy-proven cirrhosis; prevalence 24%). These are the most informative findings, expressed as likelihood ratios (Udell 2012, PMID 22357834):

Finding — raises likelihood	LR+ (95% CI)
Bonacini cirrhosis discriminant score >7	9.4 (2.6–37)
Ascites	7.2 (2.9–12)
Platelet count <160 ×10³/µL	6.3 (4.3–8.3)
Spider nevi	4.3 (2.4–6.2)

Finding — lowers likelihood	LR− (95% CI)
Lok index <0.2	0.09 (0.03–0.31)
Platelet count >160 ×10³/µL	0.29 (0.20–0.39)
Absence of hepatomegaly	0.37 (0.24–0.51)

Two lessons students consistently miss

1. Lab combinations beat the bedside for ruling cirrhosis OUT. The strongest negative finding is not a physical sign at all — it is the Lok index (LR− 0.09), built from platelets, AST, ALT, and INR. No single physical sign is sensitive enough to exclude cirrhosis, which is why a normal abdominal exam does not rule it out and why non-invasive fibrosis testing exists.

2. Your gestalt is worse than the individual findings. The review found the clinician's overall impression was less informative than the specific findings and laboratory combinations. Score the findings deliberately rather than trusting an end-of-bed impression.

Quick recall

1/2

At what serum bilirubin level does scleral icterus typically appear, and why before skin jaundice?

Tap to reveal →

The abdominal exam in liver disease requires a systematic approach: inspect, percuss, then palpate — in that order. Percussing before palpating gives you a roadmap of organ borders before your hands disturb the abdomen.

Hepatomegaly — Percussion & Palpation

1Percuss the upper border: Begin in the midclavicular line at the 3rd intercostal space (resonant/lung) and percuss downward. Dullness marks the upper liver border — typically at the 5th intercostal space.
2Percuss the lower border: Resume percussing from below the umbilicus upward in the midclavicular line. Dullness marks the inferior border. Normal liver span: 6–12 cm in the midclavicular line, 4–8 cm in the midsternal line.
3Begin palpation well below the expected inferior border: Place your right hand flat on the abdomen, lateral to the rectus abdominis, starting near the right iliac fossa. Never assume where the liver ends — a massively enlarged liver will be missed if you start too high.
4Let the liver come to your fingers: Ask the patient to take a slow, deep breath. The descending diaphragm pushes the liver toward your fingers. Avoid pressing deeply into the abdomen. Move your hand superiorly in small steps with each breath until you feel the edge.
5Characterize the edge: Note tenderness (hepatitis, congestive hepatopathy, or abscess), firmness (abnormal if blunt and hard), and nodularity (rock-hard irregular edge with discrete nodules raises concern for malignant infiltration — HCC or metastases).

Caveat: Cirrhotic livers are often small and nodular, not enlarged. A non-palpable liver in a patient with other stigmata of liver disease does not exclude cirrhosis. In COPD, a depressed diaphragm can push the liver inferiorly, mimicking hepatomegaly on palpation without a true increase in span.

Splenomegaly — Percussion & Palpation

1Traube's space percussion: Percuss the space bordered by the left 6th rib superiorly, the anterior axillary line laterally, and the left costal margin inferiorly. Tympany throughout suggests no splenomegaly. Dullness suggests enlargement, though a recent large meal or left pleural effusion can confound.
2Castell's sign: Percuss the lowest intercostal space on the left anterior axillary line. Normally tympanic at rest. A shift to dullness on deep inspiration indicates the spleen has descended into the space — positive for splenomegaly.
3Palpation — start in the right lower quadrant: Begin in the right iliac fossa and move diagonally toward the left costal margin. Starting too high risks missing a massively enlarged spleen that has descended below your starting point.
4Let the spleen descend to your fingers: With each breath, feel for the splenic edge descending toward you. Avoid aggressive upward digging — this risks splenic injury, particularly in infectious mononucleosis where the spleen is friable.
5If not palpable: Repeat with the patient in right lateral decubitus with knees flexed. This brings the spleen forward and toward the left costal margin, improving detection of mild splenomegaly.

In adults, a normal spleen cannot be palpated. Any palpable spleen tip is abnormal. Splenic enlargement from portal hypertensive congestion is a strong indicator of clinically significant portal hypertension even when other findings are absent.

Ascites Detection

1Shifting dullness (most useful): Percuss from the umbilicus laterally toward the flank. Note where dullness begins (fluid is dependent). Ask the patient to roll toward you. Wait 30–60 seconds, then percuss again — if ascites is present, the dullness shifts to the new dependent area. Flank dullness requires roughly 1500 mL of ascites to be detectable (Cattau 1982, PMID 7057606).
2Fluid wave (fluid thrill): Have an assistant (or the patient) place the ulnar edge of one hand firmly on the midline abdomen to dampen vibration transmitted through fat. The examiner sharply taps one flank and feels for a transmitted fluid impulse at the opposite flank. More specific but less sensitive — it requires a substantially larger volume than shifting dullness.
3Puddle sign (small volumes): Patient positions on hands and knees for 5 minutes — fluid pools at the most dependent point below the umbilicus. Percuss the umbilical area: dullness is said to detect smaller volumes than the other manoeuvres. Rarely used in practice and of limited proven accuracy.

Shifting dullness: fluid is dependent and shifts with patient position; gas-filled bowel floats upward.

Wolfmankurd / Wikimedia Commons / CC BY-SA 3.0

Be humble about the ascites exam

When these manoeuvres were formally tested against ultrasonography, the overall accuracy of the physical examination for ascites was only 58%, with sensitivity ranging 50–94% and specificity 29–82% across manoeuvres. The one conclusion that could be drawn with >90% accuracy was a negative one: ascites is unlikely if no flank dullness is elicited (Cattau 1982, PMID 7057606).

Practical upshot: use the exam to decide whether to scan, not to decide whether fluid is there. If the question matters — and especially before a paracentesis — get an ultrasound.

Ultrasound is far more sensitive than physical examination for detecting ascites — always use imaging when clinical findings are equivocal or when you need to confirm before paracentesis.

Abdominal Venous Patterns — Assessing Flow Direction

1Two-finger technique: Place two fingertips together on a clearly visible vein. Slide them apart (2–3 cm) to milk the segment of blood. Lift one finger — observe how quickly the vein refills. Repeat, lifting the other finger. The direction that refills faster is the direction of flow.
2Test above and below the umbilicus separately. In portal hypertension, flow radiates outward from the umbilicus in all directions. In IVC obstruction, flow in the lower abdomen is upward (opposite to normal downward flow) as collaterals bypass the blocked IVC to reach the superior vena cava.
3Correlate with JVP: An elevated jugular venous pressure (JVP) + ascites should raise suspicion for a cardiac etiology — congestive heart failure or constrictive pericarditis — rather than cirrhosis.

Abdominal venous flow patterns. Key distinction: portal hypertension flows outward in all directions; IVC obstruction drives flow upward throughout.

Quick recall

1/2

When palpating for hepatomegaly, where should you start, and why?

Tap to reveal →

Physical Exam in Chronic Liver Disease

References