2018 · Journal of diabetes science and technology · paywalled
Evaluation of a New Noninvasive Glucose Monitoring Device by Means of Standardized Meal Experiments
Pfützner et al.
The finding, in our words
In 36 participants undergoing a meal challenge, non-invasive optical glucose prediction achieved a mean absolute relative difference of 14.4% compared with YSI reference values, with 100% of readings in consensus error grid zones A and B. Capillary measurements using the device's invasive component showed 9.2% MARD, demonstrating the feasibility of calibrated optical monitoring for non-invasive glucose tracking.
A paraphrase to the Library’s standard, never the abstract. The source is one link away and is always the authority.
Noninvasive salivary measurements of creatinine, cystatin C and urea, and fingerstick capillary blood testing, can approximate gold‑standard laboratory results for kidney function. This could enable more frequent, home‑based monitoring to detect CKD progression and AKI episodes without the burdens of conventional venepuncture and laboratory processing.
Capillary blood collected in BD Microtainer tubes remained stable for up to 12 days at room temperature and gave accurate HbA1c results when posted back, with high return rates and patient willingness to reuse the system.
In children with type 1 diabetes, ambulatory glucose profile showed mean absolute relative difference of 9.56% versus laboratory random glucose and 15.07% versus capillary glucose, with linear regression coefficients of 0.93 and 0.89 respectively, indicating acceptable agreement for outpatient glycaemic monitoring.
In a head-to-head study of 24 adults with type 1 diabetes, the Dexcom G4 Platinum sensor had a lower overall MARD than the Medtronic Enlite sensor (13.6% vs 16.6% in the clinical research centre and 13.8% vs 17.9% overall), with both devices showing reduced accuracy in the hypoglycaemic range and the Dexcom sensor less affected by this impairment.
The homeRNA capillary blood collection and stabilization platform successfully captured lipopolysaccharide-induced inflammatory gene expression profiles. These transcriptomic responses were comparable to those obtained from traditional venous blood samples stabilized with RNAlater or PAXgene, demonstrating the platform's suitability for remote transcriptomic monitoring.