We read with interest the article by Hoare et al. 1 published recently in HEPATOLOGY. Hoare et al. reported that patients who are seropositive aviremic for hepatitis C can be considered likely positive for the virus on the basis of histological changes. The retrospective study initially involved a review of 172 patients selected over an 8-year period spanning 1992-2000. After the exclusion of 106 patients, a final study cohort of 66 patients remained. The grade and stage of the available liver biopsies were reported, with only 7.5% (n ϭ 5) of individuals having zero scores for inflammation and fibrosis.
Of note, nine of the original 70 individuals who met the initial inclusion criteria were found to be polymerase chain reaction (PCR)-positive for hepatitis C virus (HCV). These nine individuals can be divided into two categories: (1) those who yielded a single PCR-positive sample (n ϭ 5) and ( 2) those who yielded an unspecified number of positive samples during the review period (n ϭ 4). It is a concern that the authors excluded the first group of PCR-positive patients as the likely result of a "false positive" and stated that the second group represented true recrudescence. Is it not entirely feasible that the "false positive" results were in fact isolated occasions when the viral load within these specimens passed the lower limit of detection of reverse-transcription polymerase chain reaction (RT-PCR)? The use of relatively simple diagnostic tools would have provided some confidence in the authors' assertions. 2 No molecular data are provided concerning the likely genotype of these two groups of specimens or whether all the HCV RT-PCR-positive samples were the result of contamination. Quasispecies analysis of hypervariable region 1 would have supported the authors' suppositions.
No information related to the temporal spread of these nine HCV RNApositive individuals is provided; this would have informed the reliability of the inclusion and exclusion of these HCV RNA-positive samples. All these specimens may well represent true HCV RNA-positive samples in which viral breakthrough was serendipitously identified; however, the supporting evidence is lacking. Additionally, the authors should have tested liver biopsy material to support the supposition made in the article title.
The existence of seropositive and seronegative aviremic occult HCV has been proposed by Carreno et al. [3][4][5] These patients often have periodic elevations of alanine aminotransferase. Hoare et al. 1 reported that 15.2% of the study group had alanine aminotransferase elevations; perhaps these latter individuals would have been a more appropriate study group. If 5.7% (n ϭ 4/70) is a true representation of the frequency of recrudescence, it is to be expected that this level of recrudescence would be reflected in other seropositive aviremic defined populations. The iatrogenic infection of Irish women with HCV 1bcontaminated anti-D immunoglobulin resulted in approximately 50% of seropositive individuals exhibiting aviremia. 6 To date, despite prospective follow-up of more than 100 individuals from this cohort, since 1994, we have not observed recrudescence in this population, notwithstanding improvements in the lower limits of detection with the commercially available assays to 10 IU/mL (95% confidence interval).
The authors should acknowledge these limitations.