The epidemiological consequences of leprosy-tuberculosis co-infection

While in antiquity both leprosy and tuberculosis were prevalent in Europe, leprosy declined thereafter and, simultaneously, tuberculosis prevalence increased. Since both diseases are caused by mycobacterial infections, it has been suggested that there might be a causal relationship between both epidemics. Chaussinand observed the inverse prevalence of leprosy and tuberculosis and suggested that individuals with a latent tuberculosis infection are protected from acquiring leprosy. His cross-immunity hypothesis has been countered more recently by a co-infection hypothesis. The latter suggestion, proposed by Donoghue, states that people being infected with multi-bacillary leprosy are more susceptible to tuberculosis, which leads to increased mortality from the disease. This study utilizes mathematical modeling to explore the epidemiological consequences of the co-infection hypothesis for realistically confined parameter values. While the co-infection hypothesis appears plausible at first glance, a second thought reveals that it comprises also substantial consequences for tuberculosis epidemics: if co-infection raises the mortality rate above that of purely tuberculosis infected persons, then tuberculosis might as well be eradicated by leprosy. It is the specific interplay of both increased susceptibility towards tuberculosis and increased death rate when co-infected that determines the epidemiological fate. As a result of this analysis, it is shown that there is a large parameter region where the eventual disappearance of leprosy could indeed be explained by co-infection. This parameter region is considerably larger than that predicted by the cross-immunity hypothesis. This shows that the co-infection hypothesis should be considered a significant alternative to the cross-immunity hypothesis. The time scales at which the effects of co-infection are observed depend critically on the spatial distribution of the individuals but reach epidemiologically realistic values for rather immobile individuals with local interaction.