To the Editor—Every year, millions of people are infected with hepatitis B virus (HBV) and hepatitis C virus (HCV). HBV and HCV share common modes of transmission. The exact route of transmission is unknown in a third of patients with acute HBV infections; similarly, 30% of HCV patients do not have an identifiable risk factor. According to current HBV and HCV practice guidelines, including those from the United States Centers for Disease Control and Prevention, ^{1 , 2} and the American Association for the Study of Liver Diseases,Reference Lok and McMahon ^{3 ,} Reference Ghany, Strader and Thomas ⁴ the sharing of toothbrushes is a risk factor for HBV and HCV transmission, but kissing is not. This seems to be a contradiction because it implies that indirect contact through the sharing of toothbrushes provides a more effective route of transmission than direct contact through kissing.

A toothbrush may induce gum bleeding, thereby facilitating oral infection. However, oral lesions and bleeding occur at any time and may be caused by traumas or multiple oral diseases (eg, ulcers, inflammation, cysts, oral tumors, jaw deformities, tooth impaction, etc.). These oral lesions and the associated bleeding may also facilitate HBV and HCV transmission during kissing.

A toothbrush is usually rinsed after use; therefore, it is relatively clean prior to its potential use by a second person. Further reduction in the number of available viral particles occurs when the potential second person rinse his/her mouth while tooth brushing. Thus, the amount of viral particles transferred should be at a trace level. In contrast, kissing directly transfers a greater amount of saliva between individuals. These facts do not support the current guidelines, which say that toothbrush sharing is a risk factor for HBV and HCV transmission but that kissing is not.

Investigations of infection routes have often focused on risk factors identified in practice guidelines, and patients also tend to associate their infections with risk factors they are aware of. These present blind spots in the study of risk factors. For example, sexual intercourse is considered a potential transmission route for HBV and HCV. ^{1 –} Reference Ghany, Strader and Thomas ⁴ However, oral–oral kissing typically occurs simultaneously with sexual intercourse, confounding the analysis of whether the infection originated from genital sex, oral-genital sex, oral-oral kissing, or a combination of these. Few studies have attempted to control for oral-oral infection during sexual intercourse. The density of HBV (10Reference Jenison, Lemon, Baker and Newbold ^{5 –} Reference Rey, Fritsch, Schmitt, Meyer, Lang and Stoll-Keller ⁷ virions/mL) in saliva is actually similar to that in semen,Reference Jenison, Lemon, Baker and Newbold ^{5 ,} Reference Karayiannis, Novick, Lok, Fowler, Monjardino and Thomas ⁶ and the density of HCV (10⁶ genome equivalents/mL) in saliva is nearly 10% of that in serum.Reference Rey, Fritsch, Schmitt, Meyer, Lang and Stoll-Keller ⁷ Furthermore, oral lesions are the most common form of lesions, and oral bleeding is the most common form of bleeding. Unfortunately, all these have been largely neglected.

Kissing directly transfers saliva and pathogens (if present) as does premastication. Reports have suggested that premastication may be associated with HBV transmission. For example, Huang reported that children fed by premastication had twice the prevalence of HBV infection.Reference Huang ⁸

A study in Japan reported a case of acute HBV infection and suggested that this infection was caused by kissing.Reference Kubo, Furusyo and Sawayama ⁹ The patient had a steady partner infected with HBV and the sexual relationship between them only included deep kissing, with no sexual intercourse, oral–genital sex, or anal–genital sex, because the patient knew his partner was also infected with HIV. After the diagnosis of acute HBV infection, direct sequencing of the full HBV DNA genome indicated identical sequences in the patient and his partner.

A valid hypothesis that describes the transmission routes of a pathogen should be able to explain various epidemiological aspects of the diseases. We recently proposed that oral wounds can be a route of transmission for HBV, and this hypothesis explains various observations regarding HBV epidemiology.Reference He, Chen and Sun ¹⁰

Clinical practice guidelines play an important role in preventing the transmission of infectious diseases. Here, we present a striking inconsistency in the current HBV and HCV clinical practice guidelines regarding oral transmission. This inconsistency indicates that our understanding of HBV and HCV transmission is incomplete, especially with regard to the potential for oral transmission, and it suggests future directions for exploring potential risk factors.