US · guidance
CMS SOM App. H, Tag V207
ANSI/AAMI RD52:2004 Requirements as Adopted by Reference 42 CFR 494.40(a)
5.2.9 Ultrafiltration (effective/opaque housing)
6.2.9 Ultrafiltration (monitoring)
Refer to RD62:2001, 4.3.12 Ultrafilters: When used in a water purification system for
hemodialysis applications, an ultrafilter [should] be shown to reduce the concentrations of
bacteria and endotoxin in the feed water to the ultrafilter by factors at least as great as
those specified in the manufacturer's labeling.
5.2.9 Ultrafiltration
Refer to RD62:2001, 4.3.12 Ultrafilters: Ultrafilters [should]have an opaque housing or
that other means be used to inhibit proliferation of algae.
Ultrafilters should be included in routine disinfection procedures to prevent uncontrolled
proliferation of bacteria in the feed water compartment of the filter.
6.2.9 Ultrafiltration
The pressure drop across the ultrafilter (∆P) should be measured using simple inlet and
outlet pressure gauges. Ultrafilters operated in the cross-flow mode should also be
monitored in terms of the flow rate of water being directed to drain (concentrate).
Results of pressure measurements and bacteria and endotoxin levels should be recorded in
a log.
Interpretive Guidance § 494.40(a)
ANSI/AAMI RD52:2004
5.2.9 Ultrafiltration
Ultrafilters are membrane-based separation devices that may be used to remove particles as
small as 1,000 daltons and are thus well suited to remove both bacteria and endotoxins.
If bacterial proliferation is not controlled, bacteria may “grow through” the membrane and
contaminate the product water compartment of the filter.
6.2.9 Ultrafiltration
Such monitoring will indicate when membrane fouling has progressed to the point that
membrane replacement or cleaning is needed.
Monitoring is also necessary to ensure that the device is being operated in accordance with the
manufacturer’s instructions.
Page 70 of 420
Monitoring guidelines from ANSI/AAMI RD52, Table 4 include: For ultrafilters, monitor
pressure drops across the filter daily for pressure drop less than a value set by the facility.
Additional Guidance:
Test results drawn pre and post ultrafilter should be used to determine if the concentrations of
bacteria and endotoxin are reduced as stated by the manufacturer’s label. These samples can be
drawn at usual testing points: e.g., the post RO sample would function as the pre-UF sample,
and the sample from the first water distribution outlet would serve as the post-UF sample. Such
testing should be performed whenever an UF is originally installed, and whenever the specific
type of UF is changed. Low levels of bacteria and endotoxin in the feed water to the ultrafilter
may prevent the user from being able to show the manufacturer’s specified level of reduction; an
acceptable test result should be defined by facility policy, and in all cases must be within the
AAMI standards for microbial contamination.
Disposable ultrafilters do not require disinfection when used within the manufacturer’s
parameters and pressure monitored to detect fouling.
Every filter has an initial pressure drop when installed new. Users should monitor the change in
pressure and validate the acceptable pressure change each time the filter is replaced.
Ultrafilters are not required in every facility; the source water should determine the water
treatment components needed. As referenced in V204, an ultrafilter is required if DI is the last
water treatment component in the system. If ultrafilters are in use, the facility must follow these
requirements. The typical range of micron filter size is .001 to .05 microns.
History
Rev.
Provenance
- Source
- cms.gov
- Retrieved
- 2026-07-22
- Edition
- som-2026-07-22
- Content hash
10ab49b18799e9d729b9491c9f230aedaafc7a59c406082eaab6625a3ae47256
The link goes to the issuing authority’s own document — the one we read to produce this record. Where a source publishes whole titles rather than sections, your browser may need a moment to jump to the provision.
Unofficial copy of government-published law, reproduced from official sources with full provenance. Not an official publication; verify against official sources before relying on it in a filing. Records in the 'guidance' corpus, and only that corpus, are sub-regulatory (interpretive guidelines, survey procedures) and are not binding law. Validity bounds follow each jurisdiction's declared temporalBasis.