Guidance

Referral of influenza samples to RVU, UKHSA Colindale, 2024 to 2025

Updated 6 December 2024

Key messages

The purposes of influenza surveillance are:

• to determine UK epidemiology to inform the WHO vaccine program
• to assure influenza diagnostics performance and strategy
• to support patient care
• to contribute to pandemic signalling

Severe influenza:

All severe influenza A cases (ITU/HDU/ECMO/fatal cases) should have subtyping attempted from an influenza A positive sample. Subtyping can be performed locally, if validated testing platforms are available, or by a UKHSA laboratory.

Subtyping for surveillance:

Each influenza testing site needs a strategy to either subtype locally or to refer a proportion of positive influenza positive material for subtyping. Clinical NHS laboratories can refer samples to their regional UKHSA Public Health Laboratory for influenza A subtyping, aligned to locally agreed strategies.

Unsubtypeable influenza A:

There are a number of reasons why samples may fail subtyping attempts, including most commonly if the sample has low viral load, or due to recent viral evolution affecting assay performance but also, if the sample contains a non-seasonal or zoonotic influenza. Samples for which subtyping attempts have failed must be forwarded to the UKHSA reference laboratory for exclusion of non-seasonal  influenza. Any unsubtypeable samples where clinical risk assessment indicates potential zoonotic exposure should be referred urgently for further characterisation.

Suitability of materials:

Please send native material in viral transport medium, not in lysis buffer, for routine surveillance purposes.

Background

General principles

Influenza strain surveillance informs the global vaccination programme and provides information for empirical antiviral choice, as well as pandemic early warning systems. Surveillance of influenza depends on accurate and timely virological information indicating which types of influenza viruses (influenza A or B), which subtypes of influenza A (H1N1 and H3N2), or lineages of influenza B (B/Victoria and B/Yamagata) are circulating and how closely related these are to seasonal influenza vaccine components. Influenza type classification is dependent on the characteristics of the internal genes of the segmented genome (usually NP, M or NS1 gene) whereas influenza subtyping is dependent on the characteristics of the genes coding for the highly variable viral surface protein haemagglutinin (HA). Typically, most commercial assays for influenza detection provide influenza type information (differentiate influenza A and B), but do not provide influenza A subtyping or influenza B lineage assignment. Hence, the majority of influenza test results determined in a clinical diagnostic laboratory setting will not provide the necessary strain information as needed for public health surveillance purposes, or be able to differentiate a novel or zoonotic influenza A. Confirming the disappearance of B/Yamagata lineage since 2020 also remains critical for global vaccination strategy.

The UK Health Security Agency (UKHSA) National Respiratory Virus Reference Unit (RVU) is the World Health Organization (WHO) designated National Influenza Centre (NIC) for England. There is a statutory requirement for UK laboratories to participate in national virological surveillance for influenza as part of UK’s responsibilities agreed at the World Health Assembly. It is for this reason that both positive and negative influenza testing results must be captured in the Second Generation Surveillance System (SGSS). Detailed data on the virological characterisation of influenza viruses is reported by UKHSA RVU to WHO on a weekly basis.

In-depth virological surveillance is achieved through the detailed analysis of a proportion of influenza positive samples within the NHS hospital laboratory system, and other samples taken specifically for surveillance purposes, including community-based surveillance schemes and outbreak investigations. Comparison is made between virus strains circulating in the community, causing milder illness, and those strains associated with more severe illness requiring hospitalisation, and the degree of genetic and antigenic match between circulating strains to those included in the national vaccine programmes, informing tracking of disease burden and effectiveness of national vaccine programmes.

Surveillance periods

All clinical sample contributions are very valuable and boost the overall national picture of surveillance. UK national virological surveillance data is regarded as Official National Statistics with some of the highest viewing figures of content on the government website. Influenza positive samples should be submitted regularly throughout the season (see Samples to refer to RVU) to ensure virological data is available for accurate weekly reporting (both nationally and internationally).

Out of season, PHLs should refer all influenza A and B positive samples to RVU.

NHS labs can also refer influenza positive samples that meet criteria as Acceptable Referred Samples directly to RVU out of season, up to a maximum of 10 samples per week.

When surveillance data indicates the onset of epidemic period, the Chief Medical Officer notifies that antivirals may be prescribed in primary care.

During the epidemic period of the season (mid-season), a proportion of representative positive samples should be referred (see Figure 1).

In the 2024 to 2025 season, RVU request a maximum of 25 selected samples per week to be referred from each regional UKHSA PHL to RVU (including representative, unusual and special interest samples that have had subtyping performed – see Samples to refer to RVU. NHS labs should refer samples to PHLs for subtyping during the epidemic period, not directly to RVU.

Figure 1.  Epidemic sample referral periods

2024 to 2025 season

Influenza testing in England

This guidance focuses on the sample flow for samples from individuals infected with seasonal influenza. For suspected zoonotic influenza, please see Guidance on the diagnosis and management of avian influenza.

Over 150 laboratories provide clinical diagnostic testing services for influenza, usually as part of wider syndromic illness testing. Testing may be delivered through Point of Care (PoC) testing devices, which may provide information about influenza type (A or B) but usually no information about influenza subtype (for example: H1, H3, H5), or through laboratory PCR testing, using a variety of different commercial platforms, a relatively small proportion of which provide influenza A subtype information. Sample referral strategies are particularly important when individual hospitals are heavily reliant on testing provided by PoC devices, some of which do not leave residual materials that can be used for virological analysis. In such situations, additional samples taken from cases that test positive (particularly cases of complicated influenza or severe disease) may be necessary to provide materials for virological characterisation. As the volume of influenza testing and diversity of commercial platforms used in clinical laboratories has increased, the relative proportion of samples subtyped has reduced.

Sample Referral Pathways

It is essential that each influenza testing site has a strategy in place through the course of the epidemic season to either locally subtype or to refer a proportion of influenza A positive material to a UKHSA PHL for subtyping.

During the epidemic period, clinical NHS laboratories should refer a locally-agreed proportion of influenza A positive samples to their regional UKHSA PHL for influenza A subtyping (see Table 2). UKHSA PHLs will not charge for subtyping of influenza A positive samples performed for surveillance purposes. UKHSA PHLs will select influenza positive samples for referral to the RVU reference laboratory at Colindale in accordance with nationally agreed surveillance priorities. An overview of the sample testing and referral system in England is represented in Figure 2.

All influenza positive samples from individuals with severe influenza who require treatment in Intensive Therapy Unit (ITU)/High Dependency Unit (HDU)/Extra Corporeal Membrane Oxygenation (ECMO) and fatal cases should have subtyping attempted from an influenza A positive sample. Subtyping can be performed locally if validated testing platforms available, or by a UKHSA laboratory.

All test results (both positive and negative) for influenza must be reported to SGSS through electronic reporting from laboratory information systems.

Figure 2. Sample testing and referral pathways in England

Surveillance samples from UKHSA Public Health Laboratories

UKHSA RVU Colindale requests influenza referrals from regional UKHSA PHLs, for surveillance purposes. No charge is made for processing these. RVU will notify laboratories about changes in the sample referral periods.

Native/original samples (minimum 400µL) in viral transport medium (VTM) (not in lysis buffer) should be selected for referral to the reference laboratory, with subtyping assay results included.

Please use the current version of the E3 request form. The form collects information on detection and any subtyping assays used in the sender’s laboratory, which can help identify trends in assay performance including assay failures and is thus an important piece of information for the Reference laboratory to fulfil its national advisory role.

It is vital that the requested information, (mainly through tick boxes) is completed as best as possible to support this function. A printable checklist is provided in Appendix 1 to aid the sample referral process in laboratories.

Out of season

All influenza positive samples from hospital and community sources should be referred. Samples from returning travellers must be sent together with details of travel history, as these out of season or early season travel associated influenza strains are of particular interest for surveillance purposes.

Mid influenza season (epidemic period)

UKHSA PHLs should refer a proportion of influenza positive samples up to a maximum of 25 samples per week. In these referred samples, include all influenza positive samples from categories A to D (priority level 1) and E to F (priority level 2) shown in Samples to refer to RVU, and complete the batch with samples randomly selected from category G. If a laboratory has more than 25 positive samples per week in categories A to F, please ensure that the 25 samples referred includes all those in categories A and B below.

Samples to refer to RVU

Table 1. Priority levels

Priority level 1

A.	ALL Influenza A positive samples which can not be subtyped as H3/ (H1N1)pdm09 (or H5 where performed) therefore: unsubtypeable (see Unsubtypeable samples section for more information).
B.	Influenza A positive samples that have an unusual or unexpected PCR profile (not resulting from a known assay issue), or higher than usual Ct value difference between typing and subtyping assays (‘Ct mismatch’). Please see Influenza A subtyping and referral of unsubtypeable or Ct mismatch samples section for more information.
C.	Patients admitted to hospital with complicated influenza [note 1]. All ITU/HDU/ECMO and fatal cases should be included.
D.	All influenza positive samples in which the oseltamivir resistance mutation 275Y has been detected.

Priority level 2

E.	Influenza positive samples from influenza only co-infections (H1N1/H3N2 or influenza A and B), or influenza and SARS CoV-2 co-infections, with adequate viral load (for example: Ct value ≤31) [note 2]
F.	Influenza positive samples from all individuals with a vaccination history for current season, with adequate viral load (for example: Ct value ≤31) [note 2]

Priority level 3

G.	Samples positive for influenza B [note 3], A (H3N2), or A (H1N1)pdm09, with adequate viral load (for example: PCR Ct value ≤31) [note 2]

Note 1: Complicated influenza: Influenza requiring hospital admission and/or with symptoms and signs of lower respiratory tract infection (hypoxaemia, dyspnoea, lung infiltrate), central nervous system involvement and/or a significant exacerbation of an underlying medical condition.

Note 2: Preferentially select samples with volume >400µL, as this volume is needed for reference work including virus isolation

Note 3: Lineage assignment for influenza B positives is necessary to confirm the observed disappearance of B/Yamagata lineage since 2020.

It is recommended to send samples in regular (weeky or biweekly) batches, if possible, except for those in categories A to D, which should be sent rapidly to RVU for investigation and with the reason for sending, for example ECMO patient, clearly stated on the E3 request form. Please do not send nucleic acid extracts or samples in lysis buffer. Extracts and samples referred in lysis buffer or of inadequate volume in categories E to G may be rejected without testing if received.

Surveillance samples from the NHS

Severe cases:

Influenza positive samples from fatal and severe (ITU/HDU/ECMO) cases of influenza (where influenza is a primary cause or significant exacerbator) can be sent directly to RVU from NHS labs at any time of year (where possible after subtyping locally). It is important that this clinical information is clearly stated on the current version of the typing of influenza strains E3 request form to avoid the sample being inadvertently excluded.

Subtyping for surveillance:

During the winter epidemic period, NHS laboratories should not refer samples directly to the RVU for influenza diagnosis (for primary testing or for subtyping). Instead, NHS laboratories normally contribute to influenza surveillance through forwarding influenza positive samples to their regional UKHSA PHL, as detailed in Table 2. PHLs perform influenza subtyping and choose suitable materials from categories A to G (see Samples to refer to RVU) for referral to the RVU.

During the winter epidemic period, NHS laboratories that perform their own subtyping do not need to send their samples to their regional UKHSA PHL for surveillance purposes. These laboratories could send their characterised materials (up to 10 samples/week during the epidemic period), with subtype and Ct (or similar information) fitting into categories A to G (see Samples to refer to RVU), directly to RVU, with the reason for sending clearly stated on the E3 request form. Acceptable referred samples for surveillance purposes include native material collected in VTM (no lysis buffer) of adequate volume (minimum 400µL). Please do not send original samples collected into lysis buffer. Extracts and samples referred in lysis buffer or of inadequate volume in categories E to G may be rejected without testing if received.

Unsubtypeable samples:

NHS laboratories that perform their own H1/H3 subtyping for a case with suspected exposure to H5 avian influenza (relevant travel or unusual zoonotic exposure) and identify an unsubtypeable influenza A, should forward samples urgently to their regional PHL (Table 2) for H5 subtyping. Where possible, collection of two samples from high risk cases can help to ensure adequate material for reference testing. Samples should be collected into VTM to allow for reference work. All presumptive positive H5 samples and all unsubtypeable influenza A (H1/H3/H5 negative) associated with risk exposure must be immediately forwarded to RVU Colindale (reference lab) for characterisation (minimum 500µL volume of original non-lysed sample) as well as following Guidance on the diagnosis and management of avian influenza. Please make sure to urgently inform RVU of any H5 positive samples or unsubtypeable samples from individuals with suspected exposure to avian Influenza before sending for confirmation due to the public health significance of these results.

Subtyping for clinical purposes:

NHS laboratories that do not subtype need to send their materials to regional UKHSA PHLs or NHS collaborating laboratories if they require this information for clinical reasons (see Table 2).Do not send samples for subtyping purposes to RVU.

 Table 2. Services provided by UKHSA public health laboratories (PHLs) and NHS Collaborating Laboratories

Region/HPT	Seasonal influenza subtyping for A(H1N1)pdm09 and H3N2	Avian influenza H5 subtyping [note 1]
Midlands	Birmingham	Birmingham
South West South East	Bristol	Bristol
East of England London	Cambridge	Cambridge
North West North East Yorkshire & the Humber	Manchester Newcastle [note 2] Leeds [note 2]	Manchester

Note 1: the duty PHL virologist must be contacted to agree the sample referral pathway before sending any samples for avian influenza subtyping

Note 2: NHS collaborating laboratories providing seasonal influenza subtyping

Influenza A subtyping and referral of unsubtypeable or Ct mismatch samples

Influenza A detections may fall into one of several categories as far as subtyping information is concerned. Either:

1. Influenza A subtyping performed, and subtype successfully allocated - subtyped
2. Influenza A subtyping not performed and subtype unknown – not subtyped
3. Influenza A subtyping performed, but no subtype determined because the samples have failed subtyping assays for seasonal influenza H1N1pdm09 / H3N2 (+/- H5 where performed). These are described as unsubtypeable samples

The category UNSUBTYPEABLE is different from not subtyped (subtyping not attempted), which have separate tick-boxes on the E3 request form.

There are a number of reasons why samples may fail subtyping attempts, including, most commonly, if the sample has low viral load. However, non-seasonal influenza A detections (for example: H5, H7, H9) can also present as unsubtypeable. All samples with adequate viral load for which subtyping attempts have failed (Category C unsubtypeable) must be forwarded to the UKHSA reference laboratory for exclusion of non-seasonal (zoonotic) influenza.

Samples with unexpected discrepancies observed between typing and subtyping assays (a higher than usual Ct difference when analysed on different platforms) could also indicate the presence of unusual viruses (non seasonal/zoonotic H1/H3 viruses).

Any unsubtypeable or unexpected Ct mismatch samples where clinical risk assessment indicates potential zoonotic exposure should be urgently referred for further characterisation.

Causes of unsubtypeable and Ct mismatch samples

Non-seasonal influenza (rare, high consequence)

Unsubtypeable results can arise if the sample contains non-seasonal influenza A. It is is important to consider this possibility especially in cases of severe illness particularly if there is a history of travel or unusual zoonotic exposures and considering the ongoing worldwide epizootic of H5N1. In clinically suspect cases, H5 specific subtyping assays should be urgently performed in the appropriate UKHSA PHL/avian influenza testing laboratory, where up-to-date assay assured for detection of recently circulating H5s are available (see Table 2).

Following H5 specific testing, H5 presumptive positive samples (in VTM, minimum 500µL) must be forwarded to the Reference laboratory for confirmatory testing and further characterisation (following specific Avian Influenza testing guidance and using form E73 Avian influenza: confirmatory testing request form).

Low viral load (common)

A plausible technical explanation for failure to subtype includes a low viral load, where subtyping assays may have a lower sensitivity compared to the generic detection assay. This can be  observed where higher sample volume was used in the detection assay. It is unusual to see a Ct differential ≥5 between typing and subtyping assays where the same or a similar assay/platform is used for both. A sample with Ct ≤30 in a generic influenza A assay should subtype.

Minor virus sequence variations (uncommon but significant)

Rapid virus evolution means that continued assurance for accuracy of influenza diagnostics for clinical use is required. In some influenza seasons, influenza A samples are classified as unsubtypeable due to minor virus strain sequence variations affecting performance of subtyping platforms/assays. Such variation can reflect very recent in-season virus evolution. Samples should be forwarded for initial investigation together with information on the assay/manufacturer combination used for detection by completing the relevant sections in the E3 request form, until further investigation has identified the root cause of the issue. Viral mutations can also affect pan-influenza typing assays targeting the internal genes with loss of sensitivity compared with subtyping results, as evidenced by higher-than-expected Ct values. This may lead to false negative results and/or loss of sensitivity of surveillance and clinical detections. Such discrepant results can usually only be detected by laboratories that perform multiple assays and observe an unusual or discordant pattern of reactivity – please forward any affected samples together with details of assay used for detection for initial investigation. Concerns with diagnostic assay performance should be reported by the laboratory using the assay via the MHRA yellow card scheme.

Sample analysis at the national reference laboratory during the 2024 to 2025 season

In the 2024 to 2025 season, influenza positive samples received at the national reference laboratory will be selected for detailed analysis using influenza whole genome sequencing (WGS), subject to technical qualification, typically by ensuring samples have a high enough viral load and adequate volume. Samples that are collected in lysis buffer may be unsuitable for producing good quality influenza WGS data and preclude antigenic characterisation by culture. Original/native material collected in VTM (>400µL) should be referred where available. Please do not send original samples collected into lysis buffer (see Acceptable referred samples).

Antigenic (phenotypic) characterisation is performed on a subset of viruses which are selected for virus isolation in cell culture. Both genotyping and phenotyping pathways have stringent requirements for clinical sample quality to increase chances for completed full analysis. Hence, selection for further characterisation beyond subtype classification will be based primarily on the viral load and availability of non-lysed material, with exemption for samples associated with severe illness or death (see Samples to refer to RVU), suspected exposure to non-seasonal influenza, and those with unusual testing profile/ greater than usual Ct mismatch, as this pattern of results could indicate the presence of unusual viruses or performance issues with individual assay platforms. Good completion of the questionnaire on the Typing of influenza strains request form; E3 especially for these exemption samples is crucial to prevent them from being inadvertently excluded based on Ct value.

RVU will further investigate subtyped samples for which characterisation is required for an outbreak investigation. Subtyped samples from localised or unusual outbreaks should always be submitted and will be subject to enhanced testing including sequencing. There is no PCR Ct value cut-off for referring influenza samples from outbreaks (see Acceptable referred samples). Please ensure the samples are clearly marked as from an outbreak, with details of the outbreak setting name or location and type, and the HPZone reference if known, using the E3 request form.

Reporting of results from national reference laboratory during the 2024 to 2025 season

The information obtained from each referred sample is an important component of the UK influenza virological surveillance. For typical influenza positive samples, referring laboratories will receive a standard report, without further sample specific results, with a link to the aggregated characterisation data provided in the weekly national influenza report. The data will be reported in the weekly national joint flu/COVID reports.

Reporting of results of genome sequencing, virus isolation, antigenic typing, or genomic typing for antiviral susceptibility markers of influenza virus to referring laboratories will be performed only for those samples where a specific request for characterisation has been made for patient management or epidemiological purposes, for example as part of an outbreak investigation. Reports to referring laboratories will also be issued if clinical or epidemiological follow up is required during surveillance testing, for example: if antiviral resistance is suspected.

Testing to inform patient management

To see which samples RVU can accept for supporting clinical diagnosis and management, see section on Acceptable referred samples.

Influenza positive samples for antiviral susceptibility testing for clinical purposes

Clinical susceptibility testing is not performed in RVU, but the laboratory analyses influenza whole genomes for genetic markers that have theoretically or clinically been linked with antiviral resistance. Laboratories with genuine suspicion of antiviral resistance must contact the RVU laboratory before sending samples for sequencing with the explicit purpose of antiviral genotypic testing. The routine turnaround time for results on these samples is such that the results are unlikely to be able to impact on patient management, unless specific arrangements are made.

Antiviral (AV) genotypic testing should be considered when there is clinical concern about AV treatment failure. The greatest risk of oseltamivir resistance is currently in immunocompromised patients with influenza A(H1N1)pdm09. Further recommendations can be found in the guidance on use of antiviral agents for the treatment and prophylaxis of seasonal influenza.

The RVU Colindale laboratory can be contacted on 020 8327 7125 or 020 8327 7002. Clinical queries can be directed to Dr Anika Singanayagam (Consultant Medical Virologist) via VRD enquiries line 0208 327 7887.

Acceptable referred samples

1. Reference laboratory work requires original/native clinical material, minimum 400µL. Please do not send nucleic acid extracts or samples in lysis buffer to RVU. Extracts and samples referred in lysis buffer in categories E to G may be rejected without testing if received. This is because a range of different analytical approaches that are part of the surveillance worklflow, are incompatible with diverse lysis buffers. To prepare virus isolates and to obtain good quality virus whole genome sequence for surveillance purposes, it is necessary to use original non-lysed material, with minimum volume of 400µL.

2. Please do not refer samples with low viral load (Ct values >31), unless in categories A to D (see Samples to refer to RVU). Samples with low viral load are unlikely to be successfully sequenced or cultured.

3. Please do not refer samples which have been tested with an assay that does not generate PCR Ct/Cn values, unless in categories A to D (see Samples to refer to RVU).

A checklist to aid sample referral is provided in Appendix 1.

It is recommended to send samples in regular batches (weekly or biweekly), if possible, except for those in categories A to D, which should be sent rapidly to RVU for investigation. Samples greater than 2 weeks old may be unsuitable for virological analysis. Samples (including those in lysis buffer) stored at room temperature for extended periods may also be unsuitable for virological analysis. The current version of the typing of influenza request form (E3) should be used.