Comparison of membrane filter, multiple-fermentation-tube, and presence-absence techniques for detecting total coliforms in small community water systems

Methods for detecting total coliform bacteria in drinking water were compared using 1,483 different drinking water samples from 15 small community water systems in Vermont and New Hampshire. The methods included the membrane filter (MF) technique, a 10-tube fermentation tube (FT) technique, and the presence-absence (P-A) test. Each technique was evaluated using a 100-ml drinking water sample. Of the 1,483 samples tested, 336 (23%) contained coliforms as indicated by either one, two, or all three techniques. The FT detected 82%, the P-A detected 88%, and the MF detected 64% of these positives. All techniques simultaneously detected 55% of the positives. Evaluation of the confirmation efficiency of the P-A technique showed 94% of the presumptive positives confirming as coliforms. Thirteen different species of coliforms were identified from the 37 tests in which the P-A was positive but the MF and FT were negative. The P-A test was simple to inoculate and interpret and was considerably more sensitive than the MF and slightly more sensitive than the FT in detecting coliforms in this type of drinking water supply.     

Coliform species recovered from untreated surface water and drinking water by the membrane filter, standard, and modified most-probable-number techniques.

The species of total coliform bacteria isolated from drinking water and untreated surface water by the membrane filter (MF), the standard most-probable-number (S-MPN), and modified most-probable-number (M-MPN) techniques were compared. Each coliform detection technique selected for a different profile of coliform species from both types of water samples. The MF technique indicated that Citrobacter freundii was the most common coliform species in water samples. However, the fermentation tube techniques displayed selectivity towards the isolation of Escherichia coli and Klebsiella. The M-MPN technique selected for more C. freundii and Enterobacter spp. from untreated surface water samples and for more Enterobacter and Klebsiella spp. from drinking water samples than did the S-MPN technique. The lack of agreement between the number of coliforms detected in a water sample by the S-MPN, M-MPN, and MF techniques was a result of the selection for different coliform species by the various techniques.     

Characterization of indicator bacteria in municipal raw water, drinking water, and new main water samples

Municipal water samples were analyzed by membrane filter (MF) and presence-absence (P-A) tests for pollution indicator bacteria. In four years, 11 514 bacterial cultures were isolated from either raw water, drinking water, or new main water samples submitted to three environmental laboratories. The bacterial species occurring most often in all types of water samples were Escherichia coli (11.6-39.7%), Enterobacter aerogenes (18.1-26.3%), Aeromonas hydrophila (8.8-17.0%), Klebsiella pneumoniae (7.7-10.3%), and Citrobacter freundii (5.9-22.7%). A lactose - lauryl tryptose - tryptone broth was examined as an alternative medium to modified MacConkey broth in the presumptive portion of the P-A test. The intensity of acid and gas production in presumptive positive P-A bottles was compared with the types and frequencies of indicator bacteria shown by confirmatory tests. The results of detecting indicator bacteria following the analysis of 53 130 samples over a 2-year period were arranged by water source (well, lake, river, mixed) and water type (raw or drinking) to determine the influence of these parameters on the recovery of indicator bacteria. A further subdivision of the sample types into raw surface, raw ground, in-plant, plant discharge, reservoir, and distribution samples demonstrated the effect of water treatment practices.     

Pollution indicator bacteria associated with municipal raw and drinking water supplies.

A total of 3819 bacterial cultures isolated from municipal water samples were identified using a combination of Enterotubules and confirmatory media. Frequency distributions for the different genera or groups of bacteria were similar for raw water and drinking water isolations, except for Escherichia organisms which doubled their frequency in raw water. Differences between the membrane filter (MF) and presence-absence (P-A) test with regard to types of organisms isolated were limited to Klebsiella organisms which were preferentially cultured from MF plates. Members of the genus Enterobacter were isolated more than twice as frequently as any of the other coliform genera dealt with in this study. Aeromonas organisms were detected almost as often as such individual genera as Escherichia, Citrobacter, or Klebsiella. Although non-lactose fermenting colonies (false-negatives) of the coliform genera would not be detected by the MF technique, their lack of detection would likely be offset by the Aeromonas colonies (false-positives). At least 25% of the coliform isolates were either anaerogenic or non-lactose fermenters and would therefore go undetected by the most probably number (MPN) technique.     

Comparison of Clark's presence-absence test and the membrane filter method for coliform detection in potable water samples

A total of 2,601 water samples from six different water systems were tested for coliform bacteria by Clark's presence-absence (P-A) test and by the membrane filter (MF) method. There was no significant difference in the fraction of samples positive for coliform bacteria for any of the systems tested. It was concluded that the two tests are equivalent for monitoring purposes. However, 152 samples were positive for coliform bacteria by the MF method but negative by the P-A test, and 132 samples were positive by the P-A test but negative by the MF method. Many of these differences for individual samples can be explained by random dispersion of bacteria in subsamples when the coliform density is low. However, 15 samples had MF counts greater than 3 and gave negative P-A results. The only apparent explanation for most of these results is that coliform bacteria were present in the P-A test bottles but did not produce acid and gas. Two other studies have reported more samples positive by Clark's P-A test than by the MF method.     

Comparison of Clark's presence-absence test and the membrane filter method for coliform detection in potable water samples.

A total of 2,601 water samples from six different water systems were tested for coliform bacteria by Clark's presence-absence (P-A) test and by the membrane filter (MF) method. There was no significant difference in the fraction of samples positive for coliform bacteria for any of the systems tested. It was concluded that the two tests are equivalent for monitoring purposes. However, 152 samples were positive for coliform bacteria by the MF method but negative by the P-A test, and 132 samples were positive by the P-A test but negative by the MF method. Many of these differences for individual samples can be explained by random dispersion of bacteria in subsamples when the coliform density is low. However, 15 samples had MF counts greater than 3 and gave negative P-A results. The only apparent explanation for most of these results is that coliform bacteria were present in the P-A test bottles but did not produce acid and gas. Two other studies have reported more samples positive by Clark's P-A test than by the MF method.     

Evaluation of the Autoanalysis Colilert test for detection and enumeration of total coliforms

The Autoanalysis Colilert (AC) test was compared with the membrane filter (MF), 10-tube multiple-tube fermentation (MTF) technique, and the presence-absence test as described in Standard Methods for the Examination of Water and Wastewater for the detection and enumeration of total coliforms in water. The methods were evaluated with 31 samples from seven different sources. Each sample was analyzed by each of the techniques, using replicate 100-ml sample volumes. A total of 582 confirmed tubes were positive by the MTF test, and 533 tubes were positive by the AC test. Statistical analysis of the most-probable-number comparability data showed a statistically significant difference in the number of positive tubes, with the MTF test resulting in more positive tubes. There were no statistically significant differences in precision between the two methods. All the methods were comparable in detection of total coliforms. Levels of heterotrophic bacteria generally encountered in drinking water did not interfere with detection or enumeration of coliforms by the AC test.     

Evaluation of the Autoanalysis Colilert test for detection and enumeration of total coliforms.

The Autoanalysis Colilert (AC) test was compared with the membrane filter (MF), 10-tube multiple-tube fermentation (MTF) technique, and the presence-absence test as described in Standard Methods for the Examination of Water and Wastewater for the detection and enumeration of total coliforms in water. The methods were evaluated with 31 samples from seven different sources. Each sample was analyzed by each of the techniques, using replicate 100-ml sample volumes. A total of 582 confirmed tubes were positive by the MTF test, and 533 tubes were positive by the AC test. Statistical analysis of the most-probable-number comparability data showed a statistically significant difference in the number of positive tubes, with the MTF test resulting in more positive tubes. There were no statistically significant differences in precision between the two methods. All the methods were comparable in detection of total coliforms. Levels of heterotrophic bacteria generally encountered in drinking water did not interfere with detection or enumeration of coliforms by the AC test.     

Evaluation of a rapid, defined substrate technology method for enumeration of total coliforms and Escherichia coli in chlorinated drinking water

A rapid, defined substrate technology method, commercially available as Colilert, simultaneously enumerates total coliforms and Escherichia coli in drinking water samples in 24 h without the need for confirmatory tests. The ability of this method to enumerate both total coliforms and E. coli in simulated chlorine-treated drinking water samples was compared with the standard UK method (minerals-modified glutamate most probable number) which requires up to 96 h to complete including confirmation. Statistical analysis by a nonparametric matched-pair test showed the Colilert method to be less efficient at detecting down to one E. coli in these samples compared to the standard UK method. No statistically significant difference between the two methods of enumeration for total coliforms was detected.     

Effect of noncoliforms on coliform detection in potable groundwater: improved recovery with an anaerobic membrane filter technique

A total of 529 well and distribution potable water samples were analyzed for total coliforms by the most-probable-number and membrane filter (MF) techniques. Standard plate count bacteria and MF noncoliform bacteria were also enumerated. Frequency of coliform detection, turbidity in most-probable-number tubes, and extensive overgrowth by noncoliforms on MF filters were directly proportional to standard plate counts. Recovery of coliforms was greatest by the MF method at low (less than 100 CFU/ml) standard plate count densities and better by the most-probable-number method (confirming gas and turbid tube) at high (greater than 500 CFU/ml) standard plate count densities. In the latter case, overgrowth by noncoliforms on MF filters suppressed sheen development and, in turn, masked coliform detection. Of 341 atypical (no sheen) MF colonies verified by parallel inoculation of lauryl sulfate broth and billiant green-bile broth, 156 were aerogenic in the latter medium. Of atypical isolates, 84% were identified as either Citrobacter or Enterobacter species. A 4.3-fold reduction in numbers of overgrown MF filters and an 2.2-fold increase in numbers of coliforms recovered from 127 water samples was accomplished by anaerobic incubation of MF cultures. This anaerobic modification of the standard MF technique significantly reduced overgrowth and enhanced recovery of coliforms from potable groundwater. This technique is simple, cost effective, and suitable for monitoring of untreated ground water common to some small water systems and private water supplies.     

Effect of noncoliforms on coliform detection in potable groundwater: improved recovery with an anaerobic membrane filter technique.

A total of 529 well and distribution potable water samples were analyzed for total coliforms by the most-probable-number and membrane filter (MF) techniques. Standard plate count bacteria and MF noncoliform bacteria were also enumerated. Frequency of coliform detection, turbidity in most-probable-number tubes, and extensive overgrowth by noncoliforms on MF filters were directly proportional to standard plate counts. Recovery of coliforms was greatest by the MF method at low (less than 100 CFU/ml) standard plate count densities and better by the most-probable-number method (confirming gas and turbid tube) at high (greater than 500 CFU/ml) standard plate count densities. In the latter case, overgrowth by noncoliforms on MF filters suppressed sheen development and, in turn, masked coliform detection. Of 341 atypical (no sheen) MF colonies verified by parallel inoculation of lauryl sulfate broth and billiant green-bile broth, 156 were aerogenic in the latter medium. Of atypical isolates, 84% were identified as either Citrobacter or Enterobacter species. A 4.3-fold reduction in numbers of overgrown MF filters and an 2.2-fold increase in numbers of coliforms recovered from 127 water samples was accomplished by anaerobic incubation of MF cultures. This anaerobic modification of the standard MF technique significantly reduced overgrowth and enhanced recovery of coliforms from potable groundwater. This technique is simple, cost effective, and suitable for monitoring of untreated ground water common to some small water systems and private water supplies.     

Comparison of membrane filtration and Autoanalysis Colilert presence-absence techniques for analysis of total coliforms and Escherichia coli in drinking water samples.

Over a 4-month period, 950 samples of treated drinking water were analyzed for total coliforms (TC) and Escherichia coli by both membrane filtration (MF) and Autoanalysis Colilert presence-absence (AC) techniques. The two tests agreed 97% of the time on the basis of presumptive TC results and 98.5% of the time on the basis of verified TC results. Samples which produced disagreement between the two tests were most often TC positive by MF and TC negative by AC. E. coli was recovered four times: twice by MF only, and twice by AC only but without the diagnostic fluorescence reaction. In two samples, E. coli could not be isolated from fluorescence-positive AC tests. On the basis of these results, the AC test was implemented as the routine analytical procedure for TC but not for E. coli.     

Comparison of membrane filtration and Autoanalysis Colilert presence-absence techniques for analysis of total coliforms and Escherichia coli in drinking water samples

Over a 4-month period, 950 samples of treated drinking water were analyzed for total coliforms (TC) and Escherichia coli by both membrane filtration (MF) and Autoanalysis Colilert presence-absence (AC) techniques. The two tests agreed 97% of the time on the basis of presumptive TC results and 98.5% of the time on the basis of verified TC results. Samples which produced disagreement between the two tests were most often TC positive by MF and TC negative by AC. E. coli was recovered four times: twice by MF only, and twice by AC only but without the diagnostic fluorescence reaction. In two samples, E. coli could not be isolated from fluorescence-positive AC tests. On the basis of these results, the AC test was implemented as the routine analytical procedure for TC but not for E. coli.     

National field evaluation of a defined substrate method for the simultaneous detection of total coliforms and Escherichia coli from drinking water: comparison with presence-absence techniques

A defined substrate method was applied to drinking water to simultaneously enumerate total coliforms and total Escherichia coli directly from samples. After incubation at 35 degrees C for 24 h, the development of yellow in an initially colorless solution was specific for total coliforms; fluorescence at 366 nm in the same tube(s) or vessel demonstrated the presence of E. coli. No confirmatory or completed steps were necessary. Known as autoanalysis colilert (AC), this method was constituted as a presence-absence test and compared with the methods described in Standard Methods (SM) in the P-A format. Seven water utilities representing a wide geological and hydrological spectrum participated in the evaluation. A total of 702 split drinking water samples were analyzed. Of these, 358 were negative in both tests (SM- and AC-); 302 were positive (SM+ and AC+); and 42 were mixed (SM+ and AC-, 20; AC+ and SM-, 22). The overall agreement rate was 94%. Comparison of the SM and AC results by nonparametric statistics demonstrated no differences. Heterotrophic plate count bacteria exerted no discernible effect on the AC test. By subculture, each time the AC test was yellow, a total coliform was present; when the test was fluorescent, E. coli was isolated.     

National field evaluation of a defined substrate method for the simultaneous detection of total coliforms and Escherichia coli from drinking water: comparison with presence-absence techniques.

A defined substrate method was applied to drinking water to simultaneously enumerate total coliforms and total Escherichia coli directly from samples. After incubation at 35 degrees C for 24 h, the development of yellow in an initially colorless solution was specific for total coliforms; fluorescence at 366 nm in the same tube(s) or vessel demonstrated the presence of E. coli. No confirmatory or completed steps were necessary. Known as autoanalysis colilert (AC), this method was constituted as a presence-absence test and compared with the methods described in Standard Methods (SM) in the P-A format. Seven water utilities representing a wide geological and hydrological spectrum participated in the evaluation. A total of 702 split drinking water samples were analyzed. Of these, 358 were negative in both tests (SM- and AC-); 302 were positive (SM+ and AC+); and 42 were mixed (SM+ and AC-, 20; AC+ and SM-, 22). The overall agreement rate was 94%. Comparison of the SM and AC results by nonparametric statistics demonstrated no differences. Heterotrophic plate count bacteria exerted no discernible effect on the AC test. By subculture, each time the AC test was yellow, a total coliform was present; when the test was fluorescent, E. coli was isolated.     

Enumeration of Escherichia coli and coliforms in surface water by multiple tube fermentation and membrane filter methods

The current investigation was carried out in order to compare directly the multiple tube fermentation method (MTF), using standard procedures (lactose broth, LB) and the Colilert reagent, with the membrane filter method (MF) using Les Endo agar (LEA), m-faecal coliform agar (mFCA) and chromogenic coliform agar (CCA), for recovery of coliforms and Escherichia coli in 80 surface water samples. Total coliforms were isolated from 100% of samples by all methodologies. Faecal coliforms/E. coli were detected in 100% of samples by MTF methods, but only in 75.5% by MF-mFCA and in 86.2% by MF-CCA. Even if MTF-LB counts were consistently higher, the Colilert reagent accurately determined total coliforms and E. coli levels within 24 h with no additional confirmatory tests. Therefore, it could be a powerful tool for rapidly assessing possible faecal contamination and a suitable alternative to the traditional MTF and MF techniques utilized for coliform detection.     

Detection and enumeration of coliforms in drinking water: current methods and emerging approaches.

The coliform group has been used extensively as an indicator of water quality and has historically led to the public health protection concept. The aim of this review is to examine methods currently in use or which can be proposed for the monitoring of coliforms in drinking water. Actually, the need for more rapid, sensitive and specific tests is essential in the water industry. Routine and widely accepted techniques are discussed, as are methods which have emerged from recent research developments.Approved traditional methods for coliform detection include the multiple-tube fermentation (MTF) technique and the membrane filter (MF) technique using different specific media and incubation conditions. These methods have limitations, however, such as duration of incubation, antagonistic organism interference, lack of specificity and poor detection of slow-growing or viable but non-culturable (VBNC) microorganisms. Nowadays, the simple and inexpensive membrane filter technique is the most widely used method for routine enumeration of coliforms in drinking water.The detection of coliforms based on specific enzymatic activity has improved the sensitivity of these methods. The enzymes beta-D galactosidase and beta-D glucuronidase are widely used for the detection and enumeration of total coliforms and Escherichia coli, respectively. Many chromogenic and fluorogenic substrates exist for the specific detection of these enzymatic activities, and various commercial tests based on these substrates are available. Numerous comparisons have shown these tests may be a suitable alternative to the classical techniques. They are, however, more expensive, and the incubation time, even though reduced, remains too long for same-day results. More sophisticated analytical tools such as solid phase cytometry can be employed to decrease the time needed for the detection of bacterial enzymatic activities, with a low detection threshold.Detection of coliforms by molecular methods is also proposed, as these methods allow for very specific and rapid detection without the need for a cultivation step. Three molecular-based methods are evaluated here: the immunological, polymerase chain reaction (PCR) and in-situ hybridization (ISH) techniques. In the immunological approach, various antibodies against coliform bacteria have been produced, but the application of this technique often showed low antibody specificity. PCR can be used to detect coliform bacteria by means of signal amplification: DNA sequence coding for the lacZ gene (beta-galactosidase gene) and the uidA gene (beta-D glucuronidase gene) has been used to detect total coliforms and E. coli, respectively. However, quantification with PCR is still lacking in precision and necessitates extensive laboratory work. The FISH technique involves the use of oligonucleotide probes to detect complementary sequences inside specific cells. Oligonucleotide probes designed specifically for regions of the 16S RNA molecules of Enterobacteriaceae can be used for microbiological quality control of drinking water samples. FISH should be an interesting viable alternative to the conventional culture methods for the detection of coliforms in drinking water, as it provides quantitative data in a fairly short period of time (6 to 8 h), but still requires research effort.This review shows that even though many innovative bacterial detection methods have been developed, few have the potential for becoming a standardized method for the detection of coliforms in drinking water samples.     

Total coliform detection in drinking water: comparison of membrane filtration with Colilert and Coliquik

The Colilert (CL) and Coliquik (CQ) systems were compared in a presence-absence format against the Standard Methods membrane filtration (MF) technique to determine whether differences existed in total coliform detection. Approximately 750 water samples were collected from distribution systems, covered and uncovered storage reservoirs, well sites, and the influent to drinking water treatment plants. Samples were analyzed for total coliforms and heterotrophic bacteria with MF, CL, and CQ. The agreements between CL and MF and between CQ and MF were both greater than 94.8%, which indicates that both may be acceptable methods for total coliform detection. Disagreement between the CL and CQ methods was primarily due to false-negative results. Furthermore, laboratory and field inoculation methods were compared for CL, more than 98% agreement was obtained. This finding indicates that sampling and immediate field inoculation may be an alternative to the traditional laboratory inoculation.     

Total coliform detection in drinking water: comparison of membrane filtration with Colilert and Coliquik.

The Colilert (CL) and Coliquik (CQ) systems were compared in a presence-absence format against the Standard Methods membrane filtration (MF) technique to determine whether differences existed in total coliform detection. Approximately 750 water samples were collected from distribution systems, covered and uncovered storage reservoirs, well sites, and the influent to drinking water treatment plants. Samples were analyzed for total coliforms and heterotrophic bacteria with MF, CL, and CQ. The agreements between CL and MF and between CQ and MF were both greater than 94.8%, which indicates that both may be acceptable methods for total coliform detection. Disagreement between the CL and CQ methods was primarily due to false-negative results. Furthermore, laboratory and field inoculation methods were compared for CL, more than 98% agreement was obtained. This finding indicates that sampling and immediate field inoculation may be an alternative to the traditional laboratory inoculation.     

Evaluation of L-Cystine-amended H2S test for the detection of faecal pollution in potable water: comparison with standard multiple tube fermentation method

L-Cystine-amended H₂S strip test was compared with conventional multiple tube fermentation tests for detecting faecal pollution in 200 drinking water samples drawn from different water sources. Of these, 134 samples were positive for total coliforms (TC), whereas the cystine-amended H₂S test was positive with 116 samples. Faecal coliforms were present in 99 samples. Seventy-seven samples were also tested with presence–absence (P–A) strip test. By taking only the colour change (acid production) as positive indication for the presence of coliforms, 12 (47%) out of 26 water samples were graded as polluted in the absence of detectable coliforms. This indicated the critical importance of observing P–A bottle for gas production along with the colour. This aspect is important when commercially available P–A bottles are used in rural areas by unskilled personnel. Among different unpiped water sources tested, 94% samples drawn from India Mark II hand-pumps had TC counts within WHO guidelines. This revised version was published online in July 2006 with corrections to the Cover Date.