Comparison of capillary dried blood spot and capillary microtubes with venous immunoglobulin g levels for routine diagnostics

Abstract

BACKGROUND AND AIMS: Patients with primary antibody deficiencies receiving immunoglobulin replacement therapy require frequent monitoring of immunoglobulin G (IgG) levels. Capillary IgG measurements from dried blood spots (DBS) or microtubes offer several advantages over samples obtained by venipuncture, including facilitating remote self-sampling. However, the validity of this alternative method is still unknown. We evaluated the comparability of IgG levels measured in venous samples with capillary blood samples collected on DBS cards and in microtubes. METHODS: Paired venous and capillary finger-stick DBS and microtube samples were collected from 100 patients. IgG was extracted from DBS with phosphate buffered saline and measured with a Siemens Atellica CH. For method comparison, we performed Deming regression analysis. Absolute mean bias and limits of agreement were calculated with Bland-Altman analysis. The method comparison followed the Clinical Laboratory Improvement Amendments' (CLIA) recommended approach, but stricter limits proposed by the EFLM were applied. Relative mean differences were compared to a 10.9 % total allowable error (TEa). RESULTS: Method comparison of venous versus capillary DBS samples resulted in an R of 0.77. Mean bias was 0.23 g/L with limits of agreement of -4.06 g/L to 4.53 g/L. Method comparison of venous versus capillary microtube samples resulted in an R of 1.00. Mean bias was -0.11 g/L with -0.67 g/L to 0.46 g/L limits of agreement. Relative mean differences were 2.2 % for DBS sampling and -0.6 % for capillary sampling, both fall within 10.9 % TEa and CLIA criteria. CONCLUSION: IgG measurements from DBS demonstrated insufficient correlation and excessively broad limits of agreement, making it unsuitable for accurately determining IgG levels. This result hampers implementation of DBS in routine diagnostics. Conversely, capillary microtube samples demonstrated a strong correlation and narrow limits of agreement, which makes them a viable alternative to venipuncture.