Non-invasive and minimally invasive glucose monitoring devices: a systematic review and meta-analysis on diagnostic accuracy of hypoglycaemia detection

Abstract

Background: The use of minimally and non-invasive monitoring systems (including continuous glucose monitoring) has increased rapidly over recent years. Up to now, it remains unclear how accurate devices can detect hypoglycaemic episodes. In this systematic review and meta-analysis, we assessed the diagnostic accuracy of minimally and non-invasive hypoglycaemia detection in comparison to capillary or venous blood glucose in patients with type 1 or type 2 diabetes.

Methods: Clinical Trials.gov, Cochrane Library, Embase, PubMed, ProQuest, Scopus and Web of Science were systematically searched. Two authors independently screened the articles, extracted data using a standardised extraction form and assessed methodological quality using a review-tailored quality assessment tool for diagnostic accuracy studies (QUADAS-2). The diagnostic accuracy of hypoglycaemia detection was analysed via meta-analysis using a bivariate random effects model and meta-regression with regard to pre-specified covariates.

Results: We identified 3416 nonduplicate articles. Finally, 15 studies with a total of 733 patients were included. Different thresholds for hypoglycaemia detection ranging from 40 to 100 mg/dl were used. Pooled analysis revealed a mean sensitivity of 69.3% [95% CI: 56.8 to 79.4] and a mean specificity of 93.3% [95% CI: 88.2 to 96.3]. Meta-regression analyses showed a better hypoglycaemia detection in studies indicating a higher overall accuracy, whereas year of publication did not significantly influence diagnostic accuracy. An additional analysis shows the absence of evidence for a better performance of the most recent generation of devices.

Conclusion: Overall, the present data suggest that minimally and non-invasive monitoring systems are not sufficiently accurate for detecting hypoglycaemia in routine use.

Systematic review registration: PROSPERO 2018 CRD42018104812.

Keywords: Blood glucose self-monitoring; Diabetes mellitus; Diagnostic accuracy; Hypoglycemia; Meta-analysis.

Fig. 1

PRISMA flow diagram showing results of the screening. A total of 3416 nonduplicate results were identified and the full text of 502 was assessed. This led to an inclusion of 14 articles containing 15 studies

Fig. 2

a Risk of bias graph. b Risk of bias summary. Methodological quality was assessed on four key domains (1. patient selection, 2. index test, 3. reference standard and 4. flow and timing). Therein none of the studies was assessed as low risk of bias in all of the four key domains. Applicability concerns were assessed in three key domains (1. patient selection, 2. index test, 3. reference standard) with the QUADAS-2 tool. Applicability concerns were generally lower

Fig. 3

a Forest plot of sensitivity and specificity with 95% confidence interval in brackets of minimally invasive devices (MID) and non-invasive devices (NID) for detection of hypoglycaemia in each study. Pooling all of the studies resulted in a mean sensitivity of 69.3% [95% CI: 56.8 to 79.4] and a mean specificity of 93.3% [95% CI: 88.2 to 96.3], n number of participants, threshold in mg/dl. b Summary receiver operating characteristic (SROC) curve for overall diagnostic accuracy to detect hypoglycaemia of MID and NID. SROC, solid curve; individual studies, circles; summary estimate, triangle; 95% confidence region, contour ellipsoid

Fig. 4

a Summary ROC for MID and NID to detect hypoglycaemia (SROC curve, solid curve; study data, circles; summary estimate, triangle; 95% confidence region, contour ellipsoid; venous blood as the reference standard, grey; capillary blood as the reference standard, black). Pooled sensitivity was significantly higher in trials using venous blood as the reference standard, whereas the influence on pooled specificity was not significant. b Summary receiver operating characteristic (SROC) curve for overall diagnostic accuracy to detect hypoglycaemia of MID and NID. SROC, solid curve; study data, circles; summary estimate, triangle; 95% confidence region, contour ellipsoid; high number of participants (> 50), black; low number of participants (≤50), grey. The pooled sensitivity was significantly higher in studies investigating a larger study cohort