Monday, 17 February 2014

Empirical evidence to support pole-ward bound pathogenic Vibrios?

An increase in Vibrio related infections pushing pole-wards has been linked to increasing temperatures, however Baker-Austin et al. (2012) have now provided some empirical evidence linking long-term temperature increases around the Baltic sea with increased incidences of Vibrio related infections, particularly V. vulnificus.  The study looked at multi-decadal sea surface temperature (SST) data sets of the Baltic sea, which suggested that this area is warming at unprecedented rates, more so than virtually any other body of water on Earth.  Authors suggest that this study provides evidence that anthropogenic change is driving the increased incidents of Vibrio disease in temperate regions and in turn, affecting the distribution of human pathogenic bacteria globally.

V. vulnificus abundance peaks at temperatures > 19 °C and recent studies suggest that the pathogenic properties of some Vibrios is mediated by temperature and certainly emergence of increased Vibrio outbreaks in temperate regions such as Chile, Peru and NW Spain have been linked with warming patterns.  Many human pathogens, such as Vibrios, grow and replicate in warm (>15 °C), low salinity (< 25 ppt) marine environments, with the Baltic being semi-enclosed and predicted rising and warming seas, the authors predict increased Vibrio related incidents.  The study analysed the potential risk and presence of V. vulnificus in the Baltic area using a model based on low salinity (<25 ppt as is common in the region) and observed SST of >19 °C.  This was compared to models with temperature projections for 2050 and results suggest that temperature can be used as a predictor of Vibrio outbreaks.  Moreover, results predict significant expansion of waters capable of hosting large populations of pathogenic Vibrios. 

Whilst I am in agreement that changing temperatures are undoubtedly linked to increases in Vibrio clinical cases (Fig. 1), I am not convinced that this study provides the conclusive evidence that it proclaims to, in relation to humans driving the change.  Don’t get me wrong, I think humans are driving the temperature increases at unprecedented rates, but this paper doesn’t offer any evidence in this regard.  Other limitations of this study included the sourcing of epidemiological data, which was collated from several different sources, including ‘grey’ literature, and it was not clear whether all cases included in the study were actually from the Baltic region (eg, may have been infected abroad), potentially confounding results.   Furthermore, the models used offer some possible projections of pathogenic Vibrio distribution around the Baltic region however they are only applicable if SSTs continue along the same trend.   What may be beneficial however, as suggested by the authors, is the use of near-real-time remote sensing data reporting SSTs, which may serve as an indicator of the risk of pathogen outbreaks, allowing local authorities to inform bathers accordingly.  


Figure 1. Vibrio cases and SST. a, The relationship between Vibrio infections reported around the Baltic Sea area and maximum annual SST. Stars show observed data, dashed line shows GLM model predictions (based on the influence of SST alone). b, Time series of Baltic Sea Vibrio cases. Solid line shows observed cases and dotted line shows GLM model predictions based on the influence of maximum SST and time.

Baker-Austin, C., Trinanes, J. A., Taylor, N. G., Hartnell, R., Siitonen, A., & Martinez-Urtaza, J. (2012). Emerging Vibrio risk at high latitudes in response to ocean warming. Nature Climate Change, 3(1), 73-77.


  1. Do you think that the increased outbreaks of V. vulnificus with temperature increase could be due to an increase in the number of active bacteria? From lectures and other blog posts we have seen that unfavourable environmental conditions, such as temperature extremes, can cause bacteria to enter a VBNC state as a mechanism to protect them from damage. It is plausible that the increase in temperature is enough to alter the environmental conditions from unfavourable to favourable, increases the amount of active bacteria thus increasing the proportion of bacteria capable of causing infection.

    1. Yes, it is quite plausible that vibrios may exit a VBNC state with an increase in temperature - V. vulnificus enters VBNC state at temperatures below ~15 degrees C. It is also highly likely that different biotypes of Vibrio species become more or less pathogenistic under different environmental conditions. Chase & Harwood (2011) reported temperature had the greatest effect on specific growth rate of biotype 1 (mainly human pathogens), with optimum temperature at ~37 degrees C but salinity had the greatest effect on specific growth rates of biotypes 2 (mainly eel pathogens) & 3 (fish & human wound pathogens). The greatest risk of human infection may therefore depend on a specific combination of temperature and salinity, where conditions are optimum for different biotypes. I'll look into this paper further, with a view to submitting a blog to give further insight.

  2. I just had a quick look for the paper and couldn’t access it right away but I did find a presentation by Baker-Austin ( in which he says “Where available, information regarding the timing, geographical location and agent responsible was subsequently gathered. Cases that were not domestically acquired were removed and where suspected (e.g vibrio cases reported during the winter) were subsequently removed from final analysis.”
    I think this sounds a bit suspicious – I agree that if you are unsure of data you shouldn’t use it but am worried that this means they were deliberately leaving out data points for Vibro infections in winter. Was there anything in the paper about leaving out data points in this manner?
    I think it would be a nice direction to use this same technique for different geographical locations – perhaps more tropical regions? I wonder if temperate areas would be more at risk due to their lower natural temperatures (hence a small temperature increase may produce more dramatic changes when compared to similar temperature increases in tropical regions) but that’s just speculation on my part.

    1. In response to your concerns, I think the authors intentions were to compare data sets across multi-decadal summers for the Baltic region, so including winter figures would confound their results. As mentioned in my post, I was also a little wary of their methods in obtaining data, however I think they were forthcoming in their shortcomings and attempted to account for these in their analyses. Agreed, further work in different locations would be very interesting, although a set of standards should be developed to ensure comparable data sets.
      You can find the paper by copy and pasting the following link
      DOI: 10.1038/NCLIMATE1628