Simplified method of hepatic fibrosis quantification: design of a new morphometric analysis application.

INTRODUCTION: Semiquantitative evaluation of liver specimens is considered the standard method for measuring fibrosis; however, these systems lack the precision of a quantitative technique. METHODS: We developed an image analysis application (FibroXact) that automates and simplifies color segmentation. Trichrome slides were scanned and the program was used to evaluate each pixel based upon hue, lightness, and saturation values. Percent fibrosis was automatically calculated after thresholding. RESULTS: Linear regression demonstrated a strong relationship between the FibroXact result and the semiquantitative result (P<0.001). CONCLUSION: We have developed an image analysis application to quantify liver fibrosis. Correlation between our results and a standard semiquantitative system was demonstrated.

Risk of esophageal variceal bleeding based on endoscopic ultrasound evaluation of the sum of esophageal variceal cross-sectional surface area.

OBJECTIVE: The aim of this study was to evaluate the risk of future variceal bleeding, based on the endoscopic ultrasound measurement of the sum of the cross-sectional surface area (CSA) of all of the esophageal varices in the distal esophagus. METHODS: Twenty-eight patients with portal hypertension and esophageal varices, but no prior history of variceal bleeding, were evaluated using endoscopic ultrasound (20-MHz ultrasound probe, Microvasive, Boston, MA; Olympus, Tokyo, Japan). The entire esophagus was imaged, and an image was selected at a point where the varices appeared the largest. This image was digitized, and the sum of the CSA of all of the varices was measured (Image Pro Plus, Silver Springs, MD) by an investigator blinded to the patients' clinical status. The follow-up time for each patient was calculated (time to first bleed, time to liver transplantation, time to death, or time to the end of study). The Cox Proportional Hazards Model was used to determine if there was a significant difference between the sums of the CSA in the patients who bled compared with those who did not bleed. An OR was calculated to determine the risk of future variceal bleeding based on the sum of the CSA as measured in cm(2)/month. Positive and negative predictive values were calculated for future variceal bleeding. RESULTS: Six of 28 patients (21%) experienced esophageal variceal bleeding on follow-up. The mean CSA +/- SEM of the sum of the esophageal varices in these patients was 0.77 cm(2) +/- 0.31 cm(2) (range 0.07-2.09 cm(2)). The mean time to bleeding was 15.5 months +/- 4.95 months (range 1-29 months). Twenty-two of 28 patients (79%) did not experience variceal bleeding. The mean CSA +/- SEM of the sum of the varices in these patients was 0.36 cm(2) +/- 0.08 cm(2) (range 0.02-1.19 cm(2)). The mean time to follow-up was 35.7 months +/- 6.69 months (range 1.2-103.2 months). The sum of the CSA between the patients who bleed and those who did not bleed was significantly different at the p < 0.018 level. The OR for the risk of variceal bleeding for each one cm(2) difference in the sum of the CSA per month was 6.34. Using a cutoff of 0.45 cm(2), the sensitivity and specificity for future variceal bleeding was 83% and 75%, respectively. CONCLUSIONS: There is a significant difference (p < 0.018) in the sum of the esophageal variceal CSA between those patients who will experience future variceal bleeding and those who will not. There is a 76-fold increase per year in the risk of future variceal bleeding for each one cm(2) increase in variceal CSA. Using a cutoff value for the CSA of 0.45 cm(2), the sensitivity and specificity for future variceal bleeding above and below this point is 83% and 75%, respectively.