--- titulo: Microbiological water testing descripcion: Collect sterile samples, a homemade H2S test and E. coli counts with incubation: classify fecal risk and verify treatment. nivel: 3 estado: borrador requisitos: [] relacionados: [01-agua-y-saneamiento/potabilizacion/analisis-basico-de-calidad-del-agua, 01-agua-y-saneamiento/potabilizacion/hervido-del-agua, 01-agua-y-saneamiento/potabilizacion/desinfeccion-solar-sodis, 01-agua-y-saneamiento/potabilizacion/desinfeccion-quimica-del-agua-con-cloro-y-yodo, 01-agua-y-saneamiento/potabilizacion/filtro-lento-de-arena-y-biofiltro-de-carbon, 01-agua-y-saneamiento/potabilizacion/filtros-de-ceramica-y-carbon] fuentes: - "Sobsey, M.D.; Pfaender, F.K. Evaluation of the H2S Method for Detection of Fecal Contamination of Drinking Water. OMS, WHO/SDE/WSH/02.08, 2002." - "Manja, K.S.; Maurya, M.S.; Rao, K.M. A simple field test for the detection of faecal pollution in drinking water. Bulletin of the WHO 60(5): 797-801, 1982." - "Pillai, J.; Mathew, K.; Gibbs, R.; Ho, G.E. H2S paper strip method: a bacteriological test for faecal coliforms in drinking water at various temperatures. Water Science and Technology 40(2): 85-90, 1999." - "OMS. Guidelines for Drinking-water Quality, 4.ª ed. con addenda 1 y 2, Ginebra, 2022. Cap. 4, 5 y 7." - "OMS. Guidelines for Drinking-water Quality, 2.ª ed., vol. 3: Surveillance and control of community supplies. Ginebra, 1997." - "WHO/UNICEF JMP y UNICEF MICS. Módulo de análisis de calidad del agua en encuestas de hogares. washdata.org; mics.unicef.org." - "Aquagenx. Compartment Bag Test (CBT): instrucciones e Incubation Period Guidance." - "APHA/AWWA/WEF. Standard Methods for the Examination of Water and Wastewater, 23.ª ed., 2017. Métodos 9040, 9060 A y 9222." - "Aquaya Institute. Best Practices: Sample Holding Times for Microbial Water Analysis." idioma: en traduccion: automatica --- # Microbiological water testing ## Summary This article explains how to find out whether water carries **fecal contamination** and how to check that a treatment removes it. It covers taking sterile samples and preserving them, the **H2S presence-absence test** with a medium prepared in a workshop with basic chemistry (N3), and the **count of *E. coli*** in CFU/100 mL with a kit or membrane filtration (N4). The result is a risk classification by source (< 1, 1-10, 11-100 and > 100 *E. coli*/100 mL) and proof that the treated water comes out **negative**. It complements the [basic analysis](analisis-basico-de-calidad-del-agua.md), which covers inspection, turbidity, chlorine, pH, salts and chemistry. A negative result does not prove the water is safe: it only reduces uncertainty. ## Prerequisites - **Knowledge:** sanitary inspection of the source and turbidity ([basic analysis](analisis-basico-de-calidad-del-agua.md)); basic notions of hygiene and of handling a pressure cooker. - **Materials for sample collection (N3):** 150-250 mL glass bottles with a heat-resistant screw cap, or sterile bags; aluminum foil; 10 % sodium thiosulfate if the water is chlorinated; a portable cooler with ice packs; labels and a notebook. - **Materials for the H2S test (N3):** bacteriological peptone, dipotassium phosphate, ferric ammonium citrate, sodium thiosulfate, anionic detergent, distilled water, absorbent paper and 30-40 mL bottles. - **Materials for the count (N4):** a commercial kit (CBT bags or chromogenic plates) or membrane filtration equipment with 0.45 µm membranes, a filter holder, a manual vacuum pump, forceps and selective medium. - **Incubation:** body heat or an insulated box with a bottle of warm water (25-37 °C, H2S and CBT); a polystyrene box with a bulb and a thermometer (35-37 °C); a water bath or incubator with a ±0.5 °C thermostat (44 °C). A 0-60 °C thermometer in every case. - **Tools:** pressure cooker (N3), alcohol lamp (N2), ×5-10 magnifying glass (N2). - **Time and people:** preparing 50 H2S bottles takes one full day for 2 people. Taking a sample, 10 min. Setting up an H2S test, 5 min, and reading it, 12-48 h. An *E. coli* count, 20-48 h. ## How it works ### Fecal indicators Waterborne diseases (cholera, dysentery, typhoid fever, hepatitis A, giardiasis) almost always come from feces. Looking for every pathogen is impossible in the field, so an **indicator** is used instead: a highly abundant intestinal bacterium whose presence proves recent fecal contamination. The reference organism is *Escherichia coli*, and **thermotolerant coliforms** (which grow at 44 °C) are an acceptable approximation. WHO requires that none be detected in 100 mL of drinking water (WHO 2022, chap. 7). “Total coliforms” do not work as a fecal indicator, because many live in soil. The absence of *E. coli* does not guarantee the absence of viruses or of *Cryptosporidium* and *Giardia* cysts, which are more resistant. In addition, a sample is a snapshot of a single moment and contamination rises after rain. That is why the analysis is combined with sanitary inspection. ### H2S test Many intestinal bacteria (*Salmonella*, *Citrobacter*, *Proteus*, *Clostridium perfringens*) reduce the medium's thiosulfate to hydrogen sulfide. This reacts with the iron in the ferric ammonium citrate and forms **black iron sulfide**. The test costs on the order of 20 % of a coliform analysis and needs no incubator if it is warm, but **it is not specific**. Non-fecal sulfate-reducing bacteria, common in wells and iron pipes, also blacken it: in the laboratory, 4 of 16 cultures of these bacteria tested positive (Sobsey and Pfaender 2002, §3.3). Natural sulfides in groundwater also blacken it. WHO's review concludes that, **if the alternative is not testing at all**, it is recommended for cautious use, and that it has not been evaluated in cold climates (chap. 4). Its errors tend to be false positives, which err on the side of caution. ### Count and incubation In chromogenic media, an enzyme specific to *E. coli* (β-glucuronidase) releases a color, so each stained colony or compartment indicates *E. coli*. In membrane filtration, the bacteria from 100 mL are retained on a 0.45 µm filter and each one forms a visible colony on the medium. Temperature acts as a selector: at 44 °C only thermotolerant coliforms, mostly of fecal origin, grow well. Bacteria stressed by chlorine, sun or cold recover more poorly, hence the resuscitation phase before raising the temperature to 44 °C. ## Procedure ### A. Bottles, sample collection and preservation 1. **Prepare the bottles:** clean, with the cap left loose and the mouth covered with aluminum foil. Sterilize them in the pressure cooker for 15-20 min at full pressure (≈121 °C), or 30 min if the cooker does not reach 1 bar. Another option is the oven at 170 °C for at least 60 min (SM 9040), or 2 h if the thermostat is not reliable. If the water is chlorinated, first add **0.1 mL of 10 % sodium thiosulfate per 120 mL** of bottle (SM 9060 A). Mark a line at 100 mL. *Success criterion:* bottles closed, dry inside and with the wrapping intact. 2. **Tap or pump:** remove filters and hoses, clean the mouth of the tap and let the water run for 1-2 min. Close it and flame the mouth for 1 min with a cotton ball soaked in alcohol (if it is plastic, scrub it with diluted bleach). Open it and let it run for 1-2 min at medium flow (WHO 1997, vol. 3). *Success criterion:* continuous flow before filling. 3. **Open well, tank or river:** hold the bottle by its base with washed hands, or tie it to a rope with a clean weight. Uncap it without touching the inside of the cap, submerge it mouth-down at 20-30 cm and turn it mouth-up against the current. **Water stored at home:** take it the way the family serves it, with its ladle or its tap. *Success criterion:* bottle filled to the mark, with 2-3 cm of air, and capped in under 5 s. 4. **Label** it with the code, the source, the exact point, the date, the time, the rainfall in the last 48 h and the residual chlorine measured at that time. *Success criterion:* the same information on the bottle and in the notebook. 5. **Store** it in the dark in the portable cooler, **below 10 °C** and without freezing it. **Analyze it within 6 h**, or at most within 24 h at 4-10 °C (WHO, cited by Aquaya). A warm sample or one older than 24 h is not valid: take another one. *Success criterion:* the analysis time noted. ### B. H2S presence-absence test (N3) **Manja et al. (1982) medium**, for about 50 bottles of 20 mL: | Component | Amount | |---|---| | Bacteriological peptone | 20 g | | Dipotassium phosphate (K₂HPO₄) | 1.5 g | | Ferric ammonium citrate | 0.75 g | | Sodium thiosulfate | 1 g | | Liquid anionic detergent (Teepol in the original) | 1 mL | | Distilled water | 50 mL | 1. **Dissolve** the components in warm water (≈40 °C), stirring. If there is no Teepol, use an anionic detergent with no bleach or fragrance and validate each batch with negative and positive controls. *Success criterion:* brown liquid with no lumps. There are published variants with L-cystine to gain sensitivity. Venkobachar et al. (1994) used 0.25 g per liter of their concentrated medium, which is equivalent to ≈12.5 mg for these 50 mL. Manja et al. (2001) added 250 mg of L-cysteine to the original medium and dried it at 60 °C for 2 days (both cited by Sobsey and Pfaender 2002). 2. **Impregnate the paper:** soak a folded strip of absorbent paper of about 80 cm² with **1 mL** of medium and put it into a 30-40 mL bottle with a mark at 20 mL. *Success criterion:* paper stained evenly and with no leftover liquid. 3. **Sterilize and dry:** sterilize it with the cap loose in the pressure cooker (15-20 min at full pressure; 30 min if the cooker does not reach 1 bar) and dry it for 1-2 days at ≈50 °C. Close it, label it with batch and date, and store it in the dark. *Success criterion:* dry, brown paper. 4. **Inoculate:** add the sample up to the 20 mL mark, close it and shake for 10 s. There is a 100 mL version with 5 mL of medium that is more sensitive (Pillai et al. 1999). *Success criterion:* paper submerged. 5. **Rule out interference:** check the bottle **at 15 min and at the one-hour mark**. If it has already blackened, the color comes from sulfide or iron in the water and not from bacteria. The result is **not interpretable**: aerate another sample by shaking it for 1 min in a half-full bottle and repeat, or use the *E. coli* count. *Success criterion:* no color change in the first hour. 6. **Incubate at 25-37 °C.** Below 20 °C the result is not reliable: it has not been evaluated in cold climates (Sobsey and Pfaender 2002, chap. 4) and Pillai et al. (1999) validated it between 20 and 44 °C. Without an incubator, use body heat (bottle tightly closed inside a watertight bag, in an inner pocket) or an insulated box with a bottle of warm water, and check the temperature with the thermometer. *Success criterion:* temperature noted at the start and at every reading. 7. **Read it:** at 30-37 °C, at 12-18 h (Manja 1982) with confirmation at 24 h; at 25-30 °C, at 24 and at 48 h. **Black** (on the paper or in the liquid, even if partial) = **positive**. **Negative** = no blackening at 48 h at ≥ 25 °C. Include a negative control (water boiled for 1 min and cooled covered) and a positive control (pond water) in every batch. *Success criterion:* both controls give the expected result. **Interpretation:** Manja validated the test with waters of ≥ 10 total coliforms/100 mL (MPN), which he considered unsatisfactory. A positive result corresponds roughly to that level. A positive in a well that smells of rotten eggs or has rusty iron pipes may be a false positive: confirm it with *E. coli*. **A negative result does not mean you can stop treating** surface water or a source with risks found during inspection: when in doubt, treat it. ### C. E. coli or thermotolerant coliform count (N4) 1. **Choose the method:** - **Compartment bag (CBT) or chromogenic substrate:** the medium is added to 100 mL of sample, distributed into compartments and incubated. The compartments that change color give the most probable number (MPN) of *E. coli* per 100 mL. Incubation per the manufacturer: 35-37 °C, 20 h; 31-34 °C, 24-30 h; 25-30 °C, 40-48 h. Above 45 °C growth worsens (Aquagenx). - **Membrane filtration:** thermotolerant coliforms in lauryl sulfate broth (MLSB) at 44 ± 0.5 °C, 18-24 h (yellow colonies; Oxfam-DelAgua method, WHO 1997, vol. 3), or in m-FC at **44.5 ± 0.2 °C, 24 ± 2 h** (blue colonies; SM 9222 D). *E. coli* on chromogenic medium: 35-37 °C, 18-24 h. *Success criterion:* method, medium and incubation decided before going out to sample. 2. **Choose the volume:** 100 mL for treated water. For raw or high-risk water, also filter 10 mL and 1 mL, made up to ≥ 20 mL with sterile water. For thermotolerant coliforms, the plate counts well with **20-60 colonies** (SM 9222). *Success criterion:* at least one volume with countable colonies. 3. **Filter:** disinfect the table with alcohol, flame the forceps and set up the sterilized filter holder. Place the membrane with the grid facing up, pour in the shaken sample, apply vacuum and rinse with 20-30 mL of sterile water. Transfer the membrane onto the medium plate leaving no bubbles. *Success criterion:* the membrane flat and in contact with the medium over its whole surface. 4. **Resuscitate and incubate:** with MLSB, leave the plate **1-4 h at room temperature** before moving it to 44 °C (WHO 1997, vol. 3). Incubate it upside down. At 44 °C a ±0.5 °C control is needed: a thermostatted water bath and a field incubator. At 35-37 °C a polystyrene box with a 15-25 W incandescent bulb, adjusted while watching the thermometer, is enough. With body heat (≈30-35 °C, plate or bag tightly closed inside another watertight bag), extend the time per step 1. *Success criterion:* temperature recorded at the start, midway and at the end, within range. 5. **Count** the characteristic colonies, with the magnifying glass if needed, or the positive compartments. CFU/100 mL = colonies × 100 / mL filtered (see “Calculations”). *Success criterion:* the result noted together with the volume used. 6. **Classify** according to the WHO/UNICEF (JMP) risk categories: | *E. coli* per 100 mL | Risk | Action | |---|---|---| | < 1 (not detected) | Low | Meets the WHO guideline. Maintain treatment and monitoring | | 1-10 | Moderate | Always treat at home; review the sanitary inspection | | 11-100 | High | Mandatory, verified treatment; correct the source as a priority | | > 100 | Very high | Clarify and disinfect; look for an alternative source; warn users | *Success criterion:* every source with its category and its action. In treated water, any *E. coli* is a failure of treatment or storage: investigate it right away. ### D. Sampling frequency and points 1. **First assessment:** test during the rainy season or after heavy rain, which is the worst case. *Success criterion:* at least one result per source under those conditions. 2. **Routine** (a practical criterion from this article; WHO only sets minimums for piped networks): | Community well or spring | Household treatment | Chlorinated network | |---|---|---| | Minimum 2-4 tests a year, including the rainy season | When starting a method and then every 1-3 months in a sample of households | WHO: < 5,000 inhabitants, 12 samples a year; 5,000-100,000, 12 per every 5,000 inhabitants | 3. **Points:** the source, the treatment outlet and the point of consumption (public tap or household container, which is where the most contamination appears). *Success criterion:* every campaign covers all three levels. ## Verification - **Sterility and media:** in every batch, the blank (water boiled for 1 min and cooled covered) must be negative and the positive control (pond water) must be positive. If the blank comes out positive, discard the results of that batch. - **Of the treatment:** the test that counts is raw water positive and treated water **negative** on the same day, repeated at least 3 times in different weeks. - **Of the temperature:** record the maximum and minimum of every incubation; if it falls outside the range, repeat it. - **With more resources:** duplicate one in every 10 samples and send another to an accredited laboratory for comparison. ## Common mistakes | Symptom | Likely cause | Solution | |---|---|---| | H2S black in under 1 h | Sulfide or iron in the water itself | Not interpretable: aerate another sample or use *E. coli* | | H2S positive in a well that always gives 0 *E. coli* | Sulfate-reducing bacteria or natural sulfide | Confirm with *E. coli*; check iron pipes | | H2S negative in pond water | Cold, an early reading, expired or overheated medium | Incubate at ≥ 25 °C for up to 48 h; positive control in every batch | | Positive blank | Poorly sterilized bottles or hands | Check the pressure cooker, do not touch mouths or caps, flame the forceps | | Plate covered with uncountable colonies | Excessive volume for very contaminated water | Also filter 10 mL and 1 mL | | *E. coli* in chlorinated water with a correct residual | Contamination while taking the sample | Flame the tap, let it run and repeat with sterile technique; if it persists, review the treatment | | 0 *E. coli* in chlorinated water that arrived with active chlorine | Bottle without thiosulfate: the chlorine kept acting | Add thiosulfate to bottles for chlorinated water | | Everything negative in samples brought in the next day without cooling | Bacteria died during transport | Below 10 °C and analyze within 6 h (24 h maximum) | | Few colonies at 44 °C in chlorinated or sun-exposed water | Stressed bacteria not resuscitated | 1-4 h at room temperature before raising to 44 °C | ## Safety - **Cultures.** A black H2S bottle or a positive plate contain **millions of fecal bacteria, possibly pathogenic**. Do not open incubated bottles except to discard them: outdoors and turning your face away, add 5-10 mL of household bleach, close it again, wait 30 min and then bury it more than 30 m from any water source or burn it. Wear gloves or wash your hands with soap after every handling. Do not eat or drink in the testing area and cover any wounds on your hands. If a culture spills, cover it with paper, soak it in bleach diluted 1:10, wait 15 min and clean it up. If profuse watery diarrhea or blood in the stool appears, start oral rehydration and seek medical attention. - **Pressure cooker.** It can burst or release steam at 121 °C. Keep the valve clean, do not fill it more than 2/3 full and do not open it until the pressure has fully dropped. If there is a burn, cold water for 20 min. - **Flaming with alcohol.** Alcohol burns with a flame that is almost invisible in daylight. Keep the bottle closed and away, keep a container of water on hand to extinguish the forceps, and tie back your hair. Never refill a lit lamp. - **Sample collection.** Open wells and riverbanks cause drownings. Work in pairs, do not lean over without something to hold onto, and never go down into a well, which can accumulate CO₂ or H₂S. In swollen rivers, use a pole. - **Reagents.** Store them labeled and away from food and children. If they splash into the eyes, wash with plenty of water for 15 min. ## Variants - **With fewer resources (N2-N3):** the H2S test with body-heat incubation is the cheapest option. Without peptone or ferric citrate there is no reliable test. Verification is lost, and **all water must be treated** except cleanly collected rainwater or water from a well-protected spring. - **With more resources (N4):** CBT kits or chromogenic plates; membrane filtration with a battery-powered 44 °C field incubator, which allows 10-20 samples a day per technician; a local laboratory with MPN. - **Scaling up:** - **Family:** an H2S or CBT test when starting to use a treatment and after changes of source. - **Village:** a promoter with H2S bottles in controlled batches (blank and positive) who sends *E. coli* samples every month to the district center, plus a public board with the category of every source. - **City:** a laboratory with technicians, sampling based on population (table D) and results cross-checked with diarrhea outbreaks. ## Calculations **1. CFU per 100 mL.** 10 mL of river water are filtered and 34 blue colonies grow. CFU/100 mL = 34 × 100 / 10 = **340 CFU/100 mL** → very high risk. If 1 mL gave 3 colonies (≈300 CFU/100 mL), the two results are consistent. **2. Thiosulfate for the bottles.** At 0.1 mL per 120 mL bottle, 50 bottles need 5 mL of 10 % solution: dissolve 1 g of sodium thiosulfate in distilled water up to 10 mL. ## Sources - Sobsey, M.D.; Pfaender, F.K. *Evaluation of the H2S Method for Detection of Fecal Contamination of Drinking Water*. OMS, WHO/SDE/WSH/02.08, 2002. §3.3, §3.5 (tabla 3) y cap. 4. - Manja, K.S.; Maurya, M.S.; Rao, K.M. «A simple field test for the detection of faecal pollution in drinking water». *Bulletin of the WHO* 60(5): 797-801, 1982. - Pillai, J. et al. «H2S paper strip method: a bacteriological test for faecal coliforms in drinking water at various temperatures». *Water Science and Technology* 40(2): 85-90, 1999. - OMS. *Guidelines for Drinking-water Quality*, 4.ª ed. con addenda 1 y 2, 2022. Cap. 4, 5 y 7. - OMS. *Guidelines for Drinking-water Quality*, 2.ª ed., vol. 3: *Surveillance and control of community supplies*, 1997. - WHO/UNICEF JMP y UNICEF MICS. Módulo de calidad del agua. washdata.org; mics.unicef.org. - Aquagenx. *Compartment Bag Test* e «Incubation Period Guidance». - APHA/AWWA/WEF. *Standard Methods for the Examination of Water and Wastewater*, 23.ª ed., 2017. 9040, 9060 A y 9222. - Aquaya Institute. «Best Practices: Sample Holding Times for Microbial Water Analysis».