BACKGROUND: There are a number of grading and staging schemes of chronic hepatitis related to management and therapeutic outcomes of hepatitis. One of the most adopted and preferred schemes is the Knodell's Histologic Activity Index (HAI). It is the assumption of this study that the bigger the size of the sample, the higher the grade since a less than optimal size may result in the under-estimation of the actual state of the pathologic changes in the liver. We hope to determine the correlation of the size of the core biopsy specimen of the liver with the grade and stage of chronic hepatitis. METHODS: Patients of SLMC diagnosed to have chronic Hepatitis B by serology and who have undergone liver biopsy from 2003 to 2007 are included in the study. Out of 207 patients, only 58 were included in the study. Slides and histopathology report were retrieved and re-evaluated, using the the Knodell's HAI. A tabulation of the Knodell's HAI, number of portal tracts in the biopsy and size of tissue cores was made. Correlation of liver size sample with needle gauge was also done. RESULTS: There seems to be no correlation between HAI score and the number of portal tracts, HAI score and actual measurements and between HAI score and diameter of the tissue strips. The number of portal tracts was moderately correlated with actual measurement but fairly correlated with the tissue diameter. The actual measurement was also fairly correlated with the diameter. The mean knodell scores were 10.6, 8.9, and 10.7 <=1.0, 1.1-1.5, and >1.5 measurements respectively. There appears to be no sufficient evidence to conclude that there is a difference in the mean Knodell scores between measurements. The mean portal tracts were 5.9 for those with <=1.0 measurements, 6.9 for those with 1.1-1.5 measurements, and 12.8 for those with >1.5 measurements. The observed differences in portal tracts by measurements were statistically significant (p<0.001). There is no difference in number of portal tracts between <=1.0 and 1.1-1.5 measurements but the mean portal tracts of those with >1.5 measurements was significantly higher compared to those with <=1.5 measurements (p<0.050). The diameters increase as measurements increase. The corresponding mean diameters for measurements <=1.0,1.1-1.5, and >1.5 were 0.53, 0.62 and 0.67 respectively. However, the diameters were only different between those with <=1.0 and >1.5 measurements (p<0.001). The mean Knodell scores were 10.7, 8.8, and 11.4 for those with gauge 18, 20, and 21 needles respectively. There is no sufficient evidence to conclude that there is a difference in the mean Knodell scores between needle gauges (p=0.150). The actual measurements were only different between gauge 18 and 20 (p=0.003) and between gauge 18 and 21 (p=0.003). CONCLUSION: Although it may seem that the numerical value of the Knodell's HAI is not influenced by the length of the tissue strip and the number of portal tracts, it cannot be statistically concluded that there is no difference in accuracy in assigning values for the parameters of the Knodell's HAI between short and long tissue samples.