Reagent water is the largest reagent in a clinical laboratory by volume and the one most often taken on trust, so the testing frequency question is really a question about which failures the laboratory intends to catch. This article sets out what CLSI requires a laboratory to measure, how the monitoring points differ between inline instruments and grab samples, what drives the interval up or down, and the consumables the programme consumes — including the sampling bags that carry a grab sample from the tap to the bench, stocked as 95 kPa specimen bags, 7 x 11 inch, 100 per pack.
Quick facts
- CLSI GP40 is the guideline for preparation and testing of reagent water in the medical laboratory, and it defines Clinical Laboratory Reagent Water (CLRW) in four measurable parameters.
- The CLRW specification is resistivity greater than 10 MΩ·cm at 25 °C, total organic carbon below 500 ppb, bacteria below 10 CFU/mL and particulate filtration of 0.2 µm or finer.
- ISO 3696:1987 defines three grades of water for analytical laboratory use and its scope states that it is not applicable to organic trace analysis, analysis of surface-active agents, or biological or medical analysis — which is precisely why a medical laboratory follows CLSI rather than ISO 3696.
- No Canadian statute sets a testing interval for reagent water. The interval comes from the laboratory's own validation and quality system, which is why the written schedule is the audit document.
- Inline resistivity measures the water as it is produced; a grab sample measures the water where it is used. They answer different questions and one cannot replace the other.
- The parameter most likely to fail first in a real laboratory is bacteria, not resistivity — a system can show excellent resistivity and still grow biofilm in a storage tank or a dead leg.
What does CLSI require a laboratory to measure?
Four parameters, each aimed at a different failure mode.
| Parameter | CLRW specification | What it detects | How it is usually measured |
|---|---|---|---|
| Resistivity | Greater than 10 MΩ·cm at 25 °C | Dissolved ionic contamination, including dissolved carbon dioxide | Inline resistivity cell, with a temperature-compensated reading |
| Total organic carbon | Below 500 ppb | Organic contamination from feed water, tubing or biofilm | Inline TOC monitor or a laboratory TOC method |
| Bacteria | Below 10 CFU/mL | Viable organisms, including biofilm shedding | Culture-based enumeration of a grab sample |
| Particulate filtration | 0.2 µm or finer | Particles and organisms above the filter pore size | Specification of the filter, verified at installation and service |
The CLRW definition is the anchor, and it is quoted consistently by system manufacturers: resistivity above 10 MΩ·cm, bacteria below 10 CFU/mL, TOC below 500 ppb and 0.2 µm filtration (water system reference on CLSI water, parameter summary). The same reference material notes that a laboratory's water should, as a minimum, meet CLRW as specified by CLSI, and that the use of CLRW as feed water for a specific analyser should be confirmed with that analyser's manufacturer.
That last point is the one laboratories skip. The analyser manufacturer's feed-water specification is a separate document from CLSI, and where the two differ the stricter one governs the warranty and the service contract.
How often should each parameter be tested?
There is no single answer, and any article that gives one is describing its own laboratory rather than the standard. What CLSI provides is the parameter and the method; what the laboratory provides is the frequency, justified by its own data.
| Monitoring point | What it catches | What sets the interval | Evidence the laboratory keeps |
|---|---|---|---|
| Inline resistivity, continuous | Membrane failure, resin exhaustion, feed-water change | The system's own alarm and trend function; readings are continuous, review is scheduled | Trend chart with the review date and initials |
| Inline TOC, continuous or periodic | Organic breakthrough, biofilm in the loop | Manufacturer's monitor capability and the laboratory's history | Trend with the review date |
| Bacteria, grab sample at points of use | Biofilm, dead legs, contaminated valves | Risk of the method served, tank size, loop length, prior results | Colony counts by sampling point, with the point identified |
| Particulate filter integrity | Filter breach or bypass | Installation, cartridge change and service events | Filter part number and change date |
| System sanitisation | Biofilm regrowth in tank and loop | Manufacturer's instructions plus the laboratory's trend | Sanitisation record with agent, concentration and contact time |
The practical structure most laboratories converge on is a continuous inline reading with a scheduled review, plus a periodic grab sample for bacteria at defined points of use, plus a sanitisation record tied to the manufacturer's instructions. What varies is the period, and it should be justified by three things: the water quality history, the criticality of the methods served, and the physical system — a 20-litre bench system with a daily draw behaves differently from a 200-litre loop with a storage tank and long dead legs.
Why does ISO 3696 not cover medical laboratory water?
Because it was not written for it. The scope of ISO 3696:1987 specifies requirements for three grades of water for the analysis of inorganic chemicals, and it states explicitly that it is not applicable to organic trace analysis, analysis of surface-active agents, or biological or medical analysis (ISO 3696:1987).
That sentence is the reason a clinical laboratory cannot simply buy "Grade 1 water" and stop. Grade 1 is a specification for analytical chemistry; a clinical analyser is a biological and chemical system with its own requirements, including bacterial limits that ISO 3696 does not set. CLSI GP40 exists because the medical laboratory needed a specification of its own, and the CLRW definition is that specification (CLSI GP40).
The second layer is the laboratory's quality system. ISO 15189:2022 sets requirements for quality and competence in medical laboratories and applies to point-of-care testing as well (ISO 15189:2022). Reagents and equipment have to be fit for purpose and their performance has to be monitored; water is both a reagent and an equipment output, which is why it appears in both halves of the quality manual.
What changes the testing interval in a Canadian laboratory?
Six things, and each of them is visible in the system rather than in the regulation.
Storage. A system that produces water on demand and uses it immediately has no tank to colonise. Add a storage tank and the bacterial risk changes, because the tank is where biofilm starts.
Loop length and dead legs. Water that sits in a pipe is not being polished. Long runs and unused branches are where grab samples fail.
Feed water. Municipal supply changes seasonally. A laboratory that has never tested its feed water has an unexplained variable in its trend.
Use volume. A system that runs continuously stays cleaner than one that starts and stops, because stagnation is the mechanism.
Method criticality. Water feeding a chemistry analyser and water used to make up a stain are not the same risk. The interval can differ by use, provided the uses are separated in the record.
History. The strongest justification for a longer interval is a documented run of results inside specification, with the sampling points named and the sanitisation record complete. That is what a laboratory shows an assessor; it is not something an article can supply.
What does the water programme consume, and when should it be ordered?
Four consumable streams, and each has a different reorder logic.
Sampling containers. A bacteria grab sample needs a sterile container of the right volume, and the container has to be sterile for the method rather than merely clean. Water sampling for microbiological analysis is covered by ISO 19458, which addresses the handling that protects the sample between the tap and the bench (ISO 19458:2006); the general water sampling framework sits in ISO 5667-1 (ISO 5667-1:2020) and the preservation and handling rules in ISO 5667-3 (ISO 5667-3:2018).
Filters and cartridges. These are ordered to the manufacturer's part number and to the service interval, not to a general "filters" line. A wrong cartridge is a system shutdown.
Sanitisation chemicals. Whatever the manufacturer specifies, in the concentration the instructions require, with the contact time written into the record.
Media and disposables for the bacterial count. The enumeration method determines what is consumed: membrane filters and pads, or pour plates, or a commercial system's consumable.
The ordering logic is the same as any other laboratory consumable: count the sampling points, multiply by the sampling frequency, add a waste allowance and round up to case packs. What makes water different is that the frequency is a decision the laboratory made, so the order quantity changes when that decision changes — and a laboratory that shortens its interval without telling purchasing will discover the gap as a stock-out.
Gloves are part of the sampling step, and they are stocked as nitrile examination gloves, 4 mil, supplied as 1,000-glove cases (10 × 100). Multi-site purchasing runs through the wholesale programme, and the specimen-handling lines sit alongside the sterilization monitoring collection. Reselling? become a distributor.
Before the first sampling round, print the sheet: the printable water and equipment monitoring log records the point of use, the date, the reading and the sampler, so the trend exists as a document rather than as a set of instrument memories.
If the laboratory would rather generate the form than photocopy it, the log and checklist generator on the same page builds a list from the sampling points the laboratory actually has and prints it at clipboard size.
Ask a compliance specialist through the consultation desk and the answer comes back in writing, specific to the system and the province the laboratory operates in, with the regulation or standard it rests on cited — no account and no sign-up needed.
Sources
All links checked 24 October 2026.
- CLSI GP40, Preparation and Testing of Reagent Water in the Medical Laboratory
- ISO 3696:1987, Water for analytical laboratory use — Specification and test methods
- ISO 15189:2022, Medical laboratories — Requirements for quality and competence
- ISO 19458:2006, Water quality — Sampling for microbiological analysis
- ISO 5667-1:2020, Water quality — Sampling — Part 1: Guidance on the design of sampling programmes
- ISO 5667-3:2018, Water quality — Sampling — Part 3: Preservation and handling of water samples
- ISO 8199:2018, Water quality — General requirements and guidance for microbiological examinations by culture
- CLSI water reference: CLRW specification in practice
- Clinical laboratory water: standards, grades and compliance
- Microbiological quality of reverse osmosis water in a dialysis unit
- Bacteriological water quality and biofilm formation in a treatment system
- CliniEco Medical: sterilization monitoring collection
Related reading
- Laboratory Water Baths: How Often to Drain, Disinfect and Descale
- Laboratory Water Purification Systems in Canada vs the US: Which Class Applies?
- Temperature Monitoring in a Canadian Lab: What to Record, How Often, and What to Do
Frequently Asked Questions
How often should a laboratory test its reagent water?
There is no statutory interval in Canada. CLSI defines the parameters and the test methods; the laboratory sets the frequency from its own validation, its water system design and its quality history. A defensible schedule names each sampling point, the parameter tested there, and the review date.
What are the CLRW specifications?
Resistivity greater than 10 MΩ·cm at 25 °C, total organic carbon below 500 ppb, bacteria below 10 CFU/mL and particulate filtration of 0.2 µm or finer. Those four parameters are the definition of Clinical Laboratory Reagent Water in CLSI GP40.
Why can't I just buy ISO 3696 Grade 1 water?
Because ISO 3696 states in its scope that it does not apply to biological or medical analysis. It defines grades for analytical chemistry, and it does not set a bacterial limit of the kind a clinical analyser needs. A medical laboratory follows CLSI GP40 instead.
Is inline resistivity enough to prove water quality?
No. Inline resistivity measures the water at the point of production and says nothing about biofilm in a storage tank, a dead leg or a valve. Bacteria have to be counted from a grab sample taken at the point of use, and the sampling point has to be identified in the record.
What causes reagent water to fail bacteria limits?
Storage and stagnation. A tank, a long loop or an unused branch gives water time to sit, and biofilm forms at the surfaces where it sits. Resistivity can remain excellent while the bacterial count rises, which is why the two parameters are monitored separately.
Does a new water system need to be validated before use?
Under a quality system based on ISO 15189:2022, equipment introduced into the laboratory has to be shown to be fit for purpose before it is used on patient samples. For a water system that means documented installation, a full parameter check at each point of use, and a written schedule for what follows.
CliniEco Medical supplies infection-control and sterilization consumables in Canada under MDEL #35334. This article summarises published standards and laboratory practice and is not a substitute for the current edition of CLSI GP40 or for the laboratory's own validation; confirm limits and methods against the documents that govern the methods in use.