Portal de Periódicos da CAPES

Kawasaki Disease and Pertussis Epidemics

2014; Lippincott Williams & Wilkins; Volume: 25; Issue: 2 Linguagem: Inglês

10.1097/ede.0000000000000047

1531-5487

Paolo Pellegrino, Carla Carnovale, Valentina Perrone, Dionigi Salvati, Marta Gentili, Tatiana Brusadelli, Stefania Antoniazzi, Marco Pozzi, Emilio Clementi, Sonia Radice,

Mechanical Circulatory Support Devices

To the Editor: Kawasaki disease is a vasculitis of unknown etiology that affects children of all ethnic groups.1 Kawasaki disease appears in clusters and with seasonal outbreaks,1,2 suggesting that an infectious agent may trigger the abnormal immune response.1 Case reports have identified various possible causative agents (eg, Mycoplasma pneumoniae, Staphylococcus aureus, and rotaviruses),2 but none has been confirmed.1,2 We explored the association between hospitalization for various infections and hospitalization for Kawasaki disease, using the estimated number of hospitalizations in the United States reported in the Nationwide Inpatients Sample. We carried out this analysis using the Health Care Cost and Utilization online tool.3 We considered the number of estimated hospitalizations with at least one listed diagnosis of Kawasaki disease or other infectious agent in patients of all age groups, retrieved with International Classification of Disease, 9th revision (code 446.1 for Kawasaki disease). We calculated Pearson’s correlation coefficients for hospital admissions for the various pathogens and Kawasaki disease. The strongest association was between hospital admissions for pertussis (whooping cough) due to Bordetella pertussis and Kawasaki disease (r = 0.77). The peaks and troughs of hospitalizations for pertussis and Kawasaki disease coincided over time, even as ascertainment of both diseases has improved (Figure, upper panel). There was no similarly positive relationship with other pathogens previously associated with Kawasaki disease, such as M. pneumoniae (Figure; lower panel) (r = −0.35), Chlamydia pneumoniae (r = 0.06), mumps (r = −0.35), rotavirus (r = 0.11), S. aureus (r = 0.35), influenza virus (r = 0.28), scarlet fever due to Streptococcus pyogenes (r = −0.48), or Epstein-Barr virus infection (r = 0.07).2 Moreover, we found little correlation with pertussis due to Bordetella parapertussis (r = 0.02) or Bordetella bronchiseptica (r = 0.11).FIGURE: Hospitalizations in the United States for Kawasaki disease and pertussis due to B. pertussis (upper panel), and for M. pneumonia and C. pneumonia (lower panel).Pertussis is a worldwide endemic-epidemic disease, with outbreaks every 3–5 years and summer-autumn seasonality.4,5 These outbreaks continue in Western countries despite high vaccine coverage, resulting in waves of hospital admissions (Figure, upper panel). The typical course of the disease varies among patients, ranging from severe in unvaccinated children to a mild, unspecific, and undiagnosed disease in adolescents.4,5 Only a fraction of subjects exposed to B. pertussis develop severe disease requiring hospital admission; it is thus reasonable to assume that an increase in hospital admissions for B. pertussis corresponds to an increased circulation of the pathogen in the general population.4,5 To pursue this hypothesis further, we analyzed changes in the number of Kawasaki disease cases following a pertussis outbreak. Considering the 2010 outbreak of pertussis in California,4 the number of Kawasaki disease hospitalizations estimated with the State Inpatient Databases increased by 10%. Other supporting evidence comes from the first Kawasaki disease outbreak in Japan (in 1979).3 This outbreak came during a pertussis outbreak that had followed suspension of the pertussis vaccination program due to an increased number of adverse events following vaccination.6 Despite the introduction of a new acellular vaccine for pertussis, the incidence of Kawasaki disease has continued to increase over the years. These epidemiologic observations represent the first indication of a possible role of B. pertussis as a causal agent of Kawasaki disease. Although we could not ascertain the full incidence of the two diseases in the overall population, these results justify further exploration of this hypothesis. Paolo Pellegrino Carla Carnovale Valentina Perrone Dionigi Salvati Marta Gentili Tatiana Brusadelli Unit of Clinical Pharmacology Department of Biomedical and Clinical Sciences University Hospital “Luigi Sacco” Università di Milano Milan, Italy Stefania Antoniazzi IRCCS Foundation Ca’ Granda Ospedale Maggiore Policlinico Milan, Italy Marco Pozzi Scientific Institute IRCCS E. Medea Bosisio Parini, Lecco, Italy Emilio Clementi Sonia Radice Unit of Clinical Pharmacology Department of Biomedical and Clinical Sciences University Hospital “Luigi Sacco” Università di Milano Milan, Italy [email protected]