http://www.ncbi.nlm.nih.gov/pubmed/24288032

Amatya P, Shah D, Gupta N, Bhatta NK. Clinical and ultrasonographic measurement of liver size in normal children. Indian J Pediatr. 2014 May; 81(5):441-5.
Abstract
OBJECTIVES:

To measure the normal range of dimensions of liver in children of various age groups and to compare the liver measurement obtained by palpation-percussion, auscultation and ultrasonography.
METHODS:

This was a cross-sectional comparative study in which 500 normal (weight for height between ± 2 SD of WHO standards for children aged less than 5 y and BMI between ± 2 SD of WHO standards for children aged more than 5 y) children (0-15 y) divided in 5 age groups (100 in each age group). Subjects were enrolled from normal hospital delivery neonates, children visiting immunization and well-baby clinics, children visiting outpatient and inpatient department with minor illnesses and healthy school children.
RESULTS:

The normal range of dimension of liver in children were estimated and percentile tables of liver size were established. Though the measurements obtained by clinical methods were significantly (P < 0.001) lower than those obtained by ultrasonography, there was a strong correlation between clinical and ultrasonographic measurement. Palpation-percussion method could estimate the liver size within ± 1.0 cm of what was obtained by ultrasonography in 88 % of cases. In more than half of the study children (54.2 %), this estimation was within ± 0.5 cm.
CONCLUSIONS:

Clinical methods of liver span estimation strongly correlate with ultrasonographic measurement. The performance of palpation-percussion method is better than that of auscultation. Clinical methods should continue to be used for the estimation of liver size.