Diagnosing norovirus-associated infectious intestinal disease using viral load

Abstract

BackgroundReverse transcription-polymerase chain reaction (RT-PCR) is the main method for laboratory diagnosis of norovirus-associated infectious intestinal disease (IID). However, up to 16% of healthy individuals in the community, with no recent history of IID, may be RT-PCR positive; so it is unclear whether norovirus is actually the cause of illness in an IID case when they are RT-PCR positive. It is important to identify the pathogen causing illness in sporadic IID cases, for clinical management and for community based incidence studies. The aim of this study was to investigate how faecal viral load can be used to determine when norovirus is the most likely cause of illness in an IID case.MethodsReal-time RT-PCR was used to determine the viral load in faecal specimens collected from 589 IID cases and 159 healthy controls, who were infected with genogroup II noroviruses. Cycle threshold (Ct) values from the real-time RT-PCR were used as a proxy measure of viral load. Receiver-operating characteristic (ROC) analysis was used to identify a cut-off in viral load for attributing illness to norovirus in IID cases.ResultsOne hundred and sixty-nine IID cases and 159 controls met the inclusion criteria for the ROC analysis. The optimal Ct value cut-off for attributing IID to norovirus was 31. The same cut-off was selected when using healthy controls, or IID cases who were positive by culture for bacterial pathogens, as the reference negative group. This alternative reference negative group can be identified amongst specimens routinely received in clinical virology laboratories.ConclusionWe demonstrated that ROC analysis can be used to select a cut-off for a norovirus real time RT-PCR assay, to aid clinical interpretation and diagnose when norovirus is the cause of IID. Specimens routinely received for diagnosis in clinical virology laboratories can be used to select an appropriate cut-off. Individual laboratories can use this method to define in-house cut-offs for their assays, to provide the best possible diagnostic service to clinicians and public health workers. Other clinical and epidemiological information should also be considered for patients with Ct values close to the cut-off, for the most accurate diagnosis of IID aetiology.

Prevalence and characteristics of asymptomatic norovirus infection in the community in England

“…The distribution of norovirus cycle threshold values in IID cases and controls used in this study has been described previously (8).…”

“…In a previous analysis of norovirus cycle threshold values from the IID Study, we used receiver operating characteristic (ROC) analysis to select a cutoff for attributing disease to norovirus in IID cases (8). However, standard ROC analysis does not provide confidence limits around the selected cutoff.…”

“…Existing estimates of norovirus-associated IID incidence in the community and among individuals presenting to their general practitioner in England are based on electron microscopy, which has poor diagnostic sensitivity for identifying norovirus-associated IID (6)(7)(8); it is very likely that these estimates underrepresent the burden of norovirus disease.…”

“…However, semiquantitative real-time RT-PCR testing has demonstrated a wide range of viral loads in norovirusinfected IID cases (8); many IID cases shed norovirus at the same concentration as healthy individuals, with no recent history of IID (8,9). It is therefore unlikely that all IID cases with norovirus infection detected by RT-PCR have disease caused by norovirus; another pathogen is probably causing illness in IID cases shedding norovirus at very low concentrations.…”

“…We demonstrated in previous work that viral load measurements can be used to identify IID cases with disease caused by norovirus and to exclude IID cases with ''asymptomatic'' norovirus infection concurrent with disease caused by another pathogen (8). In this study, we used viral load measurements from IID cases in the Study of Infectious Intestinal Disease in England to improve estimates of the incidence of norovirus-associated IID in the community and leading to general practice consultations.…”