The 2021 chronic kidney disease (CKD): assessment and management NICE guideline [NG203]1 specifically recommends the adoption of the Kidney Failure Risk Equation to stratify risk in CKD. This method of risk stratification will become the default decision support tool for clinicians counselling patients with CKD and considering referral to a secondary care nephrology service. This will require coordinated implementation on a national scale, involving clinical chemistry laboratories and professionals, with dedicated modules in information technology platforms within the laboratory (Laboratory Information Management Systems, LIMS) and/or Electronic Health Records (EHRs) within secondary and primary care.
The KFRE is validated as a risk stratification tool across a wide range of healthcare systems in many countries2. In the UK, a recalibrated version of the equation has been tested and validated3. KFRE was developed using the 2009 CKD-EPI equation for estimating GFR4. NG203 recommends use of CKD-EPI but without the correction factor for ethnicity that was included in the original 2009 method. Subsequent to the publication of NG203, a new version of CKD-EPI that has been developed was published in late 20215 which also omits the correction factor for ethnicity but has been formally recalibrated to reflect this.
KFRE has a number of sources of potential error in the real world, including analytical issues and adoption of inappropriate equations. A taskforce of professionals representing the Association of Clinical Biochemistry and Laboratory Medicine (ACB), NHS GIRFT, UKNEQAS and the UK Kidney Association (UKKA) has identified a number of potential sources of error that must be avoided as we implement KFRE in the UK:
A. It is essential to use the 2009 CKD-EPI equation for eGFR as the input to the KFRE, omitting the ethnicity correction factor as per NG203. The 2021 equation was not used in the original validation and subsequent recalibration studies for KFRE as it was published after the NG2035. The 2021 equation has also not been validated for use in the UK. UKNEQAS has demonstrated that use of CKD-EPI 2021 results in a clinically significant bias that might lead higher risk patients not being referred to nephrology services. Likewise, use of the older MDRD eGFR equation is inappropriate1.
B. It is essential to use the UK validated version of the equation1,3,6 not the “North American” and not the “Non-North America” versions. The latter two are widely available and typically found by clinicians using web searches for KFRE in the UK, increasing the likelihood of this error. The appropriate equation to be used has been published in NG203 (“Terms used in this guideline” section).
C. It is essential that laboratories are aware of analytic errors associated with their serum creatinine and eGFR methods through internal QA and participation in an external QA scheme. Specifically, an enzymatic method should be used for eGFRs being used to calculate KFRE2. UKNEQAS data demonstrates that Kinetic Jaffe method creatinine assays can underestimate risk when used in KFRE.
D. To reduce the risk of a non-standard implementation, we recommend that the LIMS is the primary site for KFRE implementation in the long term. We acknowledge that implementation within primary and secondary care EHRs is already in progress. Implementation within primary care EHRs has merit as an initial strategy because it may be delivered more quickly than in LIMS in some regions, and primary care is the most important place for early KFRE adoption.
The task force will work with NICE to seek a minor amendment so that these issues are clarified in NG203. Work to perform validations studies of the CKD-EPI 2021 equation in the UK are being planned.
Yours faithfully,
On Behalf of ACB, NHS GIRFT, UKNEQAS, UKKA
References:
1. https://www.nice.org.uk/guidance/ng203
2. https://jamanetwork.com/journals/jama/fullarticle/2481005
3. https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1...
4. https://smex-ctp.trendmicro.com:443/wis/clicktime/v1/query?url=https%3a%...
5. https://www.nejm.org/doi/full/10.1056/NEJMoa2102953
6. https://kidneyfailurerisk.co.uk/
