The health economic benefits of improved diagnosis of vitamin B12 deficiency in primary care patients, associated with a new diagnostic testing algorithm

Introduction: Vitamin B12 is an essential micronutrient. Deficiency is common but can be difficult to diagnose and without early recognition and treatment can lead to significant and irreversible neurological damage. The aim of this project was to compare a new B12 deficiency diagnostic testing algorithm, where deficiency is diagnosed if the first line test, Holo-transcobalamin (Holo-TC), is <25 pmol/L and if the second line test, methylmalonic acid (MMA) (reflexed for indeterminate Holo-TC results (25-50 pmol/L) is >280 nmol/L and >360 nmol/L for ≤65 years and >65 years, respectively), to an algorithm using Total Vitamin B12 (TB12). A cost-effectiveness analysis (CEA) of the two algorithms was performed. A second CEA was performed to investigate TB12 with MMA because some laboratories do not have access to Holo-TC. Methods: This was a retrospective study at The Royal Wolverhampton NHS Trust on 1003 patients with suspected B12 deficiency. The Abbott Diagnostics Active B12 kit (CE-IVD) on Abbott Architect System i2000 analyser and the MassChrom® MMA reagent kit for liquid chromatography tandem mass spectrometry (CE-IVD), employing a Shimadzu high performance liquid chromatography system and Applied Biosystems Sciex 6500 Q-Trap tandem mass spectrometer with an Electrospray Ionisation source were implemented and verified against manufacturers pre-defined quality specifications. TB12, Holo-TC, MMA, renal function tests, thyroid stimulating hormone and serum folate were analysed on surplus serum, if not already measured routinely, on the Abbott Architect c16000 and i2000 analysers. Full Blood Count analysis was performed on Sysmex XN series analysers. Patients that were deficient for TB12 (<187 ng/L) or had results in the indeterminate range (187-300 ng/L) were followed up for one year and additional tests undertaken, and treatment received recorded. The diagnostic accuracies of TB12 and Holo-TC, were calculated using MMA as the proxy ‘gold’ standard test of deficiency from receiver operator curve (ROC) analysis. CEAs were performed using decision tree analysis (TreeAge Pro software). Results: Holo-TC (<25 pmol/L) and TB12 (<187 ng/L), had similar diagnostic accuracies for B12 deficiency, using MMA as the proxy ‘gold’ standard test, 0.78 (95% CI 0.75-0.81) and 0.75 (95% CI 0.72-0.78), respectively. The sensitivity 0.60 (95% CI 0.54-0.66) and specificity 0.99 (95% CI 0.98-0.99) of the new testing algorithm was improved compared to the TB12 algorithm (sensitivity 0.25 (95% CI 0.20-0.30), specificity 0.95 (95% CI 0.93-0.96). When the Holo-TC indeterminate range was modified to the NICE 2024 (NG239), recommendation of 25-70 pmol/L then the algorithm sensitivity further increased to 0.78 (95% CI 0.73-0.83). Haemoglobin, red cell count, mean cell volume and patient symptoms were not reliable indicators of B12 deficiency, when utilised in isolation. The proposed new B12 deficiency algorithm generated both cost and utility benefits to patients with a favourable incremental cost-effectiveness ratio (ICER) of -276.68. Sensitivity analysis, varying the costs or utilities, adjusted the ICER values however most changes still met the cost-effectiveness threshold. A secondary CEA using NG239 Holo-TC indeterminate range of 25-70 pmol/L showed that using a combination of TB12 and MMA was also cost effective. Conclusion: The new diagnostic testing algorithm utilising the combination of Holo-TC ± MMA assays was cost-effective and provided health benefits for patients. Use of this algorithm will provide earlier detection and treatment of B12 deficiency and prevention of potential significant neurological consequences benefitting patient healthcare. Alternatively for those laboratories unable to access an Holo-TC assay then the combination of TB12 and MMA is also a cost-effective option.