|Home | About | Journals | Submit | Contact Us | Français|
We have read with close attention the comments made by Stroffolini et al on our paper on sex‐related hepatitis C virus (HCV) clearance rates published in a previous issue of this journal (Gut 2006;55:1183–7). We do agree with them that cross‐sectional studies may be subject to biases in estimating clearance rates, and have extensively discussed potential biases and their effect in our paper. Stroffolini et al raise an interesting issue by stating that females chronically infected with HCV might have had higher mortality than males infected with HCV in the village of our study, thus leading to an under‐representation of chronic infection and a falsely increased clearance rate in females compared with males. While considering this hypothesis, we were surprised by the postulate of higher mortality in infected females than in males. Indeed, the literature indicates rather the opposite—that is, a higher morbidity/mortality in infected males than in females.1,2 In our study, chronically infected males were more likely to have increased alanine aminotransferase levels—that is, increased HCV‐related morbidity, compared with infected females (75/158 (47.5%) vs 32/108 (29.6%), p=0.004). Differential mortality by sex in our study is therefore more likely to be in the direction opposite to that suggested by Stroffolini et al, hence resulting in an underestimate of the true difference in clearance rates between females and males. This in turn would explain the absence of difference observed in studies in Italy, where the increased mortality in infected males compared with females would result in falsely increased clearance rates in males, hence masking the difference in clearance rates between females and males. Differential mortality by sex may be more pronounced in Italy than in Egypt, as it may be partly related to higher alcohol use in males than in females, a feature not seen in the Muslim communities of rural Egypt, where alcohol use is almost non‐existent.
In conclusion, we agree with Stroffolini et al that differential mortality between chronically infected females and males may lead to biases in estimating differences in clearance rates in cross‐sectional surveys. However, we believe that the mortality is higher among chronically infected males than among females, rather than the reverse, thus underestimating differences in clearance rates between females and males and not the opposite.
Competing interests: None.