Discrimination Efficacy of Fecal Pollution Detection in Different Aquatic Habitats of a High-Altitude Tropical Country, Using Presumptive Coliforms, Escherichia coli, and Clostridium perfringens Spores
by Denis Byamukama, Robert L. Mach, Frank Kansiime, Mohamad Manafi, and Andreas H. Farnleitner
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 2005, p. 65–71
The Main Issue: Detecting fecal contamination is critical to public health measures. Water quality depends on many things, of course, but human fecal contamination is the chief mechanism by which waterborne epidemics spread.
Classically, detecting the contamination itself was less important than detecting the potential to spread pathogens: that is, nobody cares about contamination by sterile feces or contamination that does not result in the spread of potentially pathogenic bacteria. Potential pathogens were most often the coliform bacteria (Salmonella, E. coli, Shigella, and friends). Detecting them meant filtering the water onto a membrane and plating it on selective differential media, counting colonies of particular colors to get coliform counts (all very nicely written up by the government in various standards).
This was not only a very direct measure of a relevant variable (fecal coliform contamination) but also a proxy for fecal contamination as a whole, since coliforms are not all pathogenic, nor are all intestinal pathogens coliform in nature. Finally, it served as a proxy for risk - the risk that someone in the upstream contaminating community would get ill and start spreading pathogens.
This is all well and good but it has become clear that the correlations are often confounded. Coliform bacteria can both die and proliferate in water. This is not a problem if coliform bacteria are indeed the variable of interest, but if they are being used as a proxy for other contaminants, then the correlation is suddenly broken. In particular, coliforms proliferate at higher temperatures than most environmental bacteria - they prefer body temperature (37o C). This creates a particular problem because they die at different rates in temperate climates vs tropical climates, so the standards developed and tested in Boston don't hold in India, for example.
The current article is one of many trying to study different alternative strategies for tracking fecal pollution. They are focusing on an underdeveloped region in Africa, using various proposed methods for tracking the feces. The have four: total coliform, fecal coliform, E. coli, and Clostridium perfringans spores. The spores are the most robust, least likely to change numbers over time ... unless they germinate in the water. The E. coli is the most specific pathogen, and fecal coliforms tend to be specific for feces while total coliforms can contain a wider range of animal borne and environmental bacteria.
Each of these counts is performed using filters and media, just as in the classical methods.
They sampled a small number of streams, springs and lakeside locations. There were places with obvious pollution and places without obvious pollution ... but it was all in a relatively populated area, so it seems most places would be polluted. They samples over an entire year, several times, but found little seasonal difference.
Chemical tests showed similarities among similar sites geologically (like lakes, streams, springs). The four microbiological tests tended to group the obvious pollution and the less polluted areas.
It seems E. coli gave the most 'false negatives' and total coliform counts the most 'false positives' in terms of pollution. Or more to the point, the E. coli is the most specific and the total coliform the most sensitive methods. Clostridial spores were sensitive in another way, having higher background but still being predictive of the more polluted sites, even when E. coli was scarce.
They conclude that combining the E. coli and spore tests makes for the most useful method of detecting water pollution. Total coliform is definitely not ideal and fecal coliform also was not as strongly predictive.
The conclusion seems sound - the statistics are thorough. All in all an interesting piece of work.
by Denis Byamukama, Robert L. Mach, Frank Kansiime, Mohamad Manafi, and Andreas H. Farnleitner
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 2005, p. 65–71
The Main Issue: Detecting fecal contamination is critical to public health measures. Water quality depends on many things, of course, but human fecal contamination is the chief mechanism by which waterborne epidemics spread.
Classically, detecting the contamination itself was less important than detecting the potential to spread pathogens: that is, nobody cares about contamination by sterile feces or contamination that does not result in the spread of potentially pathogenic bacteria. Potential pathogens were most often the coliform bacteria (Salmonella, E. coli, Shigella, and friends). Detecting them meant filtering the water onto a membrane and plating it on selective differential media, counting colonies of particular colors to get coliform counts (all very nicely written up by the government in various standards).
This was not only a very direct measure of a relevant variable (fecal coliform contamination) but also a proxy for fecal contamination as a whole, since coliforms are not all pathogenic, nor are all intestinal pathogens coliform in nature. Finally, it served as a proxy for risk - the risk that someone in the upstream contaminating community would get ill and start spreading pathogens.
This is all well and good but it has become clear that the correlations are often confounded. Coliform bacteria can both die and proliferate in water. This is not a problem if coliform bacteria are indeed the variable of interest, but if they are being used as a proxy for other contaminants, then the correlation is suddenly broken. In particular, coliforms proliferate at higher temperatures than most environmental bacteria - they prefer body temperature (37o C). This creates a particular problem because they die at different rates in temperate climates vs tropical climates, so the standards developed and tested in Boston don't hold in India, for example.
The current article is one of many trying to study different alternative strategies for tracking fecal pollution. They are focusing on an underdeveloped region in Africa, using various proposed methods for tracking the feces. The have four: total coliform, fecal coliform, E. coli, and Clostridium perfringans spores. The spores are the most robust, least likely to change numbers over time ... unless they germinate in the water. The E. coli is the most specific pathogen, and fecal coliforms tend to be specific for feces while total coliforms can contain a wider range of animal borne and environmental bacteria.
Each of these counts is performed using filters and media, just as in the classical methods.
They sampled a small number of streams, springs and lakeside locations. There were places with obvious pollution and places without obvious pollution ... but it was all in a relatively populated area, so it seems most places would be polluted. They samples over an entire year, several times, but found little seasonal difference.
Chemical tests showed similarities among similar sites geologically (like lakes, streams, springs). The four microbiological tests tended to group the obvious pollution and the less polluted areas.
It seems E. coli gave the most 'false negatives' and total coliform counts the most 'false positives' in terms of pollution. Or more to the point, the E. coli is the most specific and the total coliform the most sensitive methods. Clostridial spores were sensitive in another way, having higher background but still being predictive of the more polluted sites, even when E. coli was scarce.
They conclude that combining the E. coli and spore tests makes for the most useful method of detecting water pollution. Total coliform is definitely not ideal and fecal coliform also was not as strongly predictive.
The conclusion seems sound - the statistics are thorough. All in all an interesting piece of work.
posted by BenK at 10:07 AM
0 Comments:
Post a Comment
<< Home