Reply to Kempf et al

The Journal of Infectious Diseases® 2017;215:486–7 Nitazoxanide was commenced at a dose of 500 mg twice daily. It was well tolerated with no adverse effects. Initially, an improvement in frequency and severity of diarrhea was noted. Results of PCR analysis of stool specimens became negative for norovirus; however, the MS2 bacteriophage used as an internal control added to the patient sample was also not detected by PCR, indicating inhibition of PCR. Dilution of the sample 1:100 prior to nucleic acid extraction overcame the inhibition, and both the norovirus and internal control became detectable by PCR in the diluted patient sample. Overall, 1% of feces samples that are tested in our laboratory demonstrate PCR inhibition, but consistent PCR inhibition over a series of samples, as seen in this patient, has not been seen before. Treatment with nitazoxanide was, therefore, discontinued after 12 months because of no clinical response and persistent detection of norovirus by PCR (after dilution). We believe that nitazoxanide was not inhibiting Taq polymerase within the PCR, as PCR analysis of feces specimens was consistently inhibited after nitazoxanide therapy was discontinued. Subsequent to treatment with nitazoxanide, a 12-week trial of ribavirin and pegylated interferon alfa and, separately, a further trial of enteral immunoglobulin via a jejunostomy failed to eradicate norovirus infection. This case demonstrates the lack of efficacy of nitazoxanide in treating chronic norovirus in an immunocompromised patient, despite previous reports of benefit [1]. Further clinical studies are required to establish the place of nitazoxanide in the treatment of norovirus in immunocompromised patients. Furthermore, the mechanism of the consistent PCR inhibition in this case is unclear. If it is a disease-specific effect, then it may be important to ensure that an internal control is being used in all fecal molecular assays for patients with XLA, to prevent false-negative results due to PCR inhibition. Note Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.