Association of Invasive Pneumococcal Disease with Season, Atmospheric Conditions, Air Pollution, and the Isolation of Respiratory Viruses

Abstract

We examined the relation of invasive pneumococcal disease to season, atmospheric conditions, and the rate of respiratory virus isolation in a community-wide surveillance program in Houston. Among adults, the number of cases of pneumococcal bacteremia peaked in midwinter and declined strikingly in midsummer, indicating a high degree of inverse correlation with the ambient temperature. We detected significant correlations between the occurrence of pneumococcal disease and the isolation of respiratory syncytial virus (P < .001), influenza virus (P < .001), and all viruses except influenza virus (P < .001), as well as with air pollution, as measured by SO2 levels (P < .001). In contrast, the rate of invasive pneumococcal disease among infants and children was relatively more sustained from October through May, with a notable decrease in summer months; the incidence of pneumococcal disease was therefore less strongly correlated with cold weather and less closely associated with the isolation of respiratory syncitial virus or influenza virus. However, pneumococcal disease among infants and children was associated with isolation of these viruses after a 4-week lag period as well as with isolation of adenovirus and ragweed pollen counts. The finding, with regard to children, that correlations tended to be stronger for events that occurred 1 month previously than for those that occurred contemporaneously is consistent with the concept that viral or allergic events predispose to otitis media with effusion, which becomes suppurative and leads to pneumococcal bacteremia or meningitis. For adults, a more immediate predisposition to pneumococcal pneumonia and bacteremia because of viral infection or air pollution was suggested.

Heavy metals in particulate and colloidal matter from atmospheric deposition of urban Guangzhou, South China

“…The elemental compositions of the COM samples are showed in Table 1. The very high ratio of nitrogen to carbon (N/C) is likely due to the contribution of the proteinaceous virus and bacteria (Kim et al, 1996). In addition, the abundance of NO À 3 and NH þ 4 was higher than that of dissolved organic nitrogen due to high solubility in the atmospheric wet deposition (Eklund et al, 1997).…”

Temporal cross‐correlation between influenza‐like illnesses and invasive pneumococcal disease in The Netherlands

“…This Canadian study finds that influenza notifications Granger‐cause IPD, consistent with our observed lagged relation by plain cross‐correlation analysis, but yearly phase and amplitude terms of the fitted sine waves to the two time series were not correlated19—paralleling our result that (pre‐whitened) residuals were not correlated. Actually, most studies not correcting for autocorrelation report a temporal association between either influenza and/or RSV with IPD incidence 5, 18, 20, 25, 28, 29. Strikingly, the study in children up to 16 years of age by Toschke et al., taking into account temporal autocorrelation, showed no association between influenza outbreaks and invasive pneumococcal infection 27.…”

“…RSV and IPD are associated in infants/children,5, 17, 26, 29, 30 and influenza and IPD in the elderly 26, 28, 30. The strongest associations between respiratory viruses and pneumococcal disease occur among the older age groups 18, 20, 22, 26.…”

“…It shows a seasonal pattern peaking in winter,2, 3 but little is known about the driver of this pattern 4. Contributing factors proposed include climate variables,5 air pollution6 and cocirculating respiratory viruses. In particular, influenza and respiratory syncytial virus (RSV) 7, 8 are suspected to predispose to pneumococcal infection 9, 10…”

“…The strength of the interaction between influenza or influenza‐like illness (ILI) and IPD at population level, however, varies across studies from non‐existent to significant 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27. Studies looking at specific respiratory viruses also showed inconsistent results 5, 28, 29, 30. In addition, most studies did not control for seasonality of the pathogens or used a sine/cosine function to adjust for it 27.…”