Legionella detection in recirculating water by the membrane filtration method is the internationally recognized culture-based technique for quantifying Legionella bacteria in cooling towers, evaporative condensers and building water systems. Performed in accordance with ISO 11731 and equivalent national standards, membrane filtration concentrates bacteria from relatively large sample volumes and transfers them onto selective culture media, delivering sensitive and legally defensible results that support Legionella risk management and regulatory compliance. This article explains the technical principle, sample types, testing items, analytical procedure and practical applications of membrane filtration-based Legionella testing for recirculating water systems.
Technical Overview
Legionella is a genus of Gram-negative, rod-shaped bacteria that occurs naturally in freshwater environments but can proliferate to hazardous concentrations in engineered water systems. Among the more than 60 recognized species, Legionella pneumophila is responsible for the vast majority of disease cases, causing both Legionnaires' disease, a severe form of pneumonia, and Pontiac fever, a milder flu-like illness. Infection occurs when contaminated water aerosols are inhaled, making recirculating water systems that generate aerosols—such as cooling towers and evaporative condensers—the highest-priority monitoring targets.
Recirculating water circuits provide ideal conditions for Legionella growth: warm water temperatures (approximately 25 °C to 45 °C), stagnant zones, biofilms, corrosion products, sediment and amoebae that harbor intracellular bacteria. Because Legionella concentrations in large water volumes are often low and unevenly distributed, direct plating of small aliquots frequently fails to detect the organism. The membrane filtration method overcomes this limitation by passing a measured volume of water (typically 100 mL to 1,000 mL) through a membrane filter with a pore size of 0.2 µm to 0.45 µm, physically concentrating the bacteria on the filter surface for subsequent culture on selective media.
The membrane filtration approach is a core procedure within ISO 11731 (Water quality — Detection and enumeration of Legionella), and it is also referenced in related frameworks such as ISO 11731-2 for waters with low bacterial counts, ISO/TS 12869 for rapid qPCR screening, and various GB/T and WS/T national method standards. Accredited third-party laboratories routinely apply this technique to produce quantitative results expressed as colony-forming units per liter (CFU/L), which form the basis for risk classification and corrective-action decisions.
Testing Samples
Cooling Tower and Evaporative Condenser Water
Cooling tower basins, fill packs, distribution decks and return lines are sampled because they combine warm water, high nutrient availability and continuous aerosol generation. Samples are typically collected from the basin at a point representative of the bulk recirculating flow, or from the make-up and blow-down lines to characterize incoming and concentrated water chemistry. Sediment and sludge accumulated at the basin floor are often collected as supplementary samples, as they can carry Legionella loads one to two orders of magnitude higher than the overlying water.
Warm and Hot Water Recirculating Systems
Domestic hot water loops, recirculation returns, calorifier drains and outlets distant from the heat source are sampled in hospitals, hotels, residential buildings and industrial facilities. Temperature gradients in these systems frequently create zones between 25 °C and 45 °C where Legionella multiplies despite an overall disinfection program. Both first-draw samples and post-flush samples may be collected to distinguish point-of-use contamination from systemic contamination of the recirculating loop.
Spa Pools, Fountains and Decorative Water Features
Whirlpool spas, therapy pools, misting systems, decorative fountains and-therapy equipment recirculate warm, aerated water, which constitutes a significant aerosol exposure pathway. Samples are collected from the basin, balance tank and filter backwash water, since filter media and biofilm in these systems are common bacterial reservoirs.
Biofilm Swabs and Surface Samples
Where water samples alone are insufficient, swabs of internal pipe surfaces, cooling tower fill, showerheads, faucet aerators and gaskets can be taken. Biofilm samples are particularly valuable for verifying the effectiveness of cleaning and disinfection, because free-floating Legionella in the water phase often reappears if the biofilm reservoir remains intact. Swabs are transported in neutralizing buffer and processed by elution followed by membrane filtration or direct plating.
Correct sample handling is critical for valid results:
- Samples must be collected into sterile containers containing sodium thiosulfate to neutralize any residual chlorine or oxidizing biocide.
- Containers should be filled leaving a small headspace and labeled with location, time, date and system operating conditions.
- Samples should be transported in the dark, protected from temperature extremes, and analyzed as soon as practicable—preferably within 24 hours.
- Chain-of-custody documentation must accompany every sample to ensure traceability and legal defensibility.
Testing Items
Membrane filtration testing can be scoped to different levels of identification depending on the risk assessment objective:
- Total Legionella spp. enumeration—quantification of all culturable Legionella species, reported in CFU/L.
- Legionella pneumophila confirmation—differentiation of L. pneumophila from non-pneumophila species such as L. bozemanii, L. dumoffii, L. longbeachae and L. micdadei.
- Serogroup-level typing of L. pneumophila, distinguishing serogroup 1 (the most clinically significant type) from serogroups 2 through 14.
- Non-culture adjuncts where required, including qPCR screening for Legionella DNA and sequence-based identification of unusual isolates.
Results are reported quantitatively above the detection limit, semi-quantitatively when high background interference is present, or qualitatively as "detected / not detected" when only low volumes or swab samples are analyzed. The practical quantification limit typically ranges from 10 CFU/L to 250 CFU/L, depending on the volume filtered, the elution scheme and the aliquot volumes plated.
Testing Methods
The membrane filtration procedure follows the sequence defined in ISO 11731 and comparable national methods. Each step is designed to maximize Legionella recovery while suppressing the competing heterotrophic flora that often dominates recirculating water.
- Filtration and concentration: A measured sample volume is passed through a sterile 47 mm membrane filter (0.2 µm or 0.45 µm pore size, mixed cellulose ester or polycarbonate). Turbid samples may be split across multiple membranes to prevent clogging.
- Direct transfer or elution: The membrane is either placed directly onto the surface of selective agar, or the retained bacteria are eluted into a small volume of sterile diluent (such as Ringer's solution or sterile water) by vortexing or mechanical rubbing, after which aliquots of the concentrate are plated.
- Pre-treatment for contaminated matrices: For samples with heavy background flora, aliquots of the eluate are subjected to acid treatment (pH 2.2 HCl–KCl buffer for a short contact time) or heat treatment (50 °C ± 1 °C for 30 minutes). These treatments exploit the relative acid and heat tolerance of Legionella compared with many non-target organisms.
- Selective culture: The membrane or aliquots are applied to GVPC agar (buffered charcoal yeast extract agar supplemented with glycine, vancomycin, polymyxin B and cycloheximide), which inhibits most non-Legionella bacteria and fungi while supporting Legionella growth.
- Incubation: Plates are incubated at 36 °C ± 2 °C in a humidified atmosphere, with elevated CO₂ beneficial for many strains, for up to 10 days. Plates are examined at 3 to 4 days and again at 10 days, since different species and serogroups grow at different rates.
- Presumptive identification: Legionella colonies on GVPC exhibit characteristic morphologies—ground-glass texture, grey-white, greenish or blue-green coloring, and whole-colloy autofluorescence under 365 nm UV light for certain species. Presumptive counts are recorded at each examination.
- Confirmation: Representative colonies are subcultured onto BCYE agar with and without L-cysteine (and onto a cysteine-free non-selective medium). Growth only on the cysteine-containing medium confirms Legionella, because the bacterium cannot synthesize this essential amino acid. Confirmatory identification and serogrouping are then performed using latex agglutination, direct fluorescent antibody testing or molecular methods.
- Calculation and reporting: Confirmed colony counts are back-calculated through the dilution, elution and filtration factors to yield a final result in CFU/L, together with the method detection limit and any qualifying observations.
Quality assurance is integral to the method: laboratories run positive controls (reference strains such as L. pneumophila serogroup 1), sterility controls on media and diluents, duplicate plating of selected aliquots, and media growth-promotion and selectivity checks. Participation in interlaboratory proficiency testing schemes further demonstrates ongoing competence. As a rapid adjunct, qPCR in accordance with ISO/TS 12869 can deliver same-day screening results, but because it detects DNA from both viable and non-viable cells, culture by membrane filtration remains the reference method for regulatory enumeration and action-level decisions.
Applications
Membrane filtration-based Legionella testing supports a wide range of risk management and compliance objectives:
- Routine monitoring of cooling towers, evaporative condensers and HVAC water circuits under municipal, national or international regulatory programs.
- Verification of water safety plans and building water management programs, in which test results validate hazard analysis, control measures and corrective actions.
- Healthcare and long-term-care facility surveillance, where vulnerable populations justify more stringent testing frequency and lower action thresholds.
- Hotels, resorts, cruise vessels and public buildings, where guest safety, brand protection and outbreak prevention are priorities.
- Outbreak investigation and epidemiological tracing, combining culture with isolate typing to link environmental and clinical strains.
- Post-remediation verification after disinfection events such as hyperchlorination, thermal disinfection, copper-silver ionization adjustment or biocide program optimization.
- Commissioning of new or recommissioned systems, confirming that construction, stagnation and start-up have not created unacceptable bacterial loads.
FAQ
What sample volume is needed for membrane filtration testing?
Standard volumes range from 100 mL to 1,000 mL. Larger volumes improve sensitivity but may be limited by turbidity, since suspended solids clog the membrane. For heavily contaminated or turbid cooling tower water, laboratory staff often filter several smaller aliquots or apply dilution combined with pre-treatment to manage background flora.
How long does the test take?
Presumptive results are often available after 3 to 4 days of incubation, but confirmed results require subculture and identification, so the full turnaround is typically 7 to 10 working days after sample receipt. Rapid qPCR screening can provide same-day or next-day indicative results where an early risk signal is needed.
What Legionella concentration is considered acceptable?
Acceptance criteria vary by jurisdiction and system type. Many regulatory frameworks treat concentrations exceeding 1,000 CFU/L in cooling towers as an action level requiring immediate remediation, while healthcare and other high-risk settings frequently apply stricter internal targets such as detection limits of 10 to 100 CFU/L. Any detectable level in a high-risk setting warrants investigation of the underlying causes, including temperature control, biocide dosing and biofilm management.
Why are acid or heat pre-treatments sometimes applied?
Recirculating water often contains large populations of heterotrophic bacteria that overgrow the selective agar and mask slow-growing Legionella colonies. Short acid exposure at pH 2.2 or heat treatment at 50 °C for 30 minutes reduces competing organisms while preserving a sufficient fraction of Legionella, thereby improving detection. Because some Legionella strains are sensitive to these treatments, laboratories commonly plate both pre-treated and untreated aliquots in parallel.
Can qPCR replace membrane filtration culture?
No. qPCR delivers speed and high sensitivity, but it enumerates DNA from both living and dead cells and may therefore overestimate viable risk, particularly after disinfection. Culture-based membrane filtration remains the reference method for quantitative regulatory reporting, while qPCR serves as a valuable rapid screening and trend-monitoring tool within a combined testing strategy.
How often should recirculating water systems be tested?
Testing frequency depends on system risk classification, prior results and local regulations. High-risk systems such as healthcare cooling towers and spa pools are often tested monthly or quarterly, lower-risk open recirculating systems at least quarterly to semi-annually, and additional testing should always follow remediation, major maintenance, changes in water treatment or any suspected case of legionellosis.