Comparison of laboratory results and pain perception in self-sampled capillary blood versus venous blood sampling: a systematic review and meta-analysis
Schröder et al.
The finding, in our words
This meta-analysis found strong agreement between self-collected capillary blood and venous samples for laboratory results. Capillary sampling was associated with significantly lower pain scores than venipuncture, particularly when using upper arm devices rather than fingerprick methods.
A paraphrase to the Library’s standard, never the abstract. The source is one link away and is always the authority.
In a cohort of 62 participants evaluating home self-collection with the TassoPlus device, 38 returned samples, with 29% excluded due to insufficient plasma volume below 200 ΔL. Although viral load measurements in adequate samples correlated well with conventional testing (r = 0.800), the high failure rate and missed detection in low-volume viremic samples indicate that further technical refinements are necessary before clinical adoption.
Capillary finger prick sampling at home showed close agreement with venous sampling for measuring infliximab, vedolizumab and CRP, with no systematic bias and high patient tolerability and practicality, supporting decentralised monitoring of biologic therapy in IBD.
This review of remote HbA1c testing via capillary blood microsampling finds that both dried and liquid formats can produce results comparable to venous sampling when collection, stability and extraction are properly validated. It matters because self-collected microsamples could enable decentralised diabetes monitoring and telemedicine, but heterogeneity in study designs and methods must be resolved before reliable implementation.
In 30 adult solid organ transplant recipients, capillary blood self-collected using the Tasso+ device showed 91% concordance with venous samples for detecting CMV DNAemia above the limit of detection, with excellent quantitative correlation (Spearman R² = 0.99). Most participants preferred self-collection over venipuncture, suggesting the device could enable decentralised viral monitoring, though sensitivity was lower (95% at 308 IU/mL).
Capillary blood collected in microtubes yielded HbA1c results that agreed closely with venous sampling, while dried blood spots showed markedly poorer agreement. High patient acceptability and willingness to repeat suggest microtubes are a viable option for decentralised diabetes monitoring.