Large scale evaluation of WHO's ultrasonographic staging system of schistosomal periportal fibrosis in Ethiopia.

OBJECTIVES

To evaluate the recent WHO's ultrasonographic diagnostic staging system of schistosomal periportal thickening/fibrosis and to assess intra/inter-observer variation associated with its use.

METHODS

Local standard of portal branch wall thickness (PBWT) for height was established using 150 healthy subjects. Intra and inter-observer variation in image pattern identification and PBWT measurements were assessed in 94 and 35 subjects, respectively, with differing stages of periportal thickening fibrosis. WHO's diagnostic criteria were evaluated in 2,451 community members (1,277 males, 1,174 females; mean age 18.8 years) with an overall Schistosoma mansoni prevalence estimate of 65.9%.

RESULTS

There were no significant inter/intra-observer variations in image pattern identification and PBWT measurements. Based on Ethiopian PBWT-for-height standard, 128/2,451 (5.2%) had insipient, 46/2,451 (1.9%) had possible/probable and 112/2451 (4.6%) had definite/advanced periportal thickening/fibrosis. Comparable figures were obtained using the Senegalese PBWT-for-height standard and there was good agreement between Ethiopian and Senegalese healthy control-based diagnostic criteria in classifying the 286 subjects into stages of periportal thickening/fibrosis (kappa = 0.87, P < 0.001).

CONCLUSIONS

With further improvement, the WHO's ultrasonographic diagnostic criteria can be used in health institutions and community surveys. Image pattern based assessment is simple and more reproducible than PBWT based assessment of periportal thickening/fibrosis. The latter is, however, more useful in clarifying the status of an individual with doubtful image pattern, and in monitoring post-treatment outcome of periportal thickening/fibrosis. Considering the comparability of PBWT-for-height standards, setting one international standard of PBWT-for-height is more practical than developing local standards for each endemic area.