Accuracy of Using a Point-of-Care Glucometer for Cerebrospinal Fluid Glucose Screening in Resource-Limited Countries

Validation of glucose and lactate in cerebrospinal fluid (CSF) on a Radiometer blood gas analyzer ABL90 Flex plus

PURPOSE

Rapid determination of cerebrospinal fluid (CSF) glucose and lactate is required by emergency rooms and intensive care units. Long turnaround time (TAT) on test results negatively impacts timely diagnosis and treatment of neurological infections like meningitis.

METHODS

The CSF glucose and lactate assays were evaluated on a blood gas analyzer, Radiometer ABL90 Flex Plus. Linearity, limit of quantitation (LOQ), and precision were determined using fresh and spiked patient CSF samples. Fifty-four fresh and 49 frozen CSF samples were used to compare the method againstAbbottArchitectC16000. An inter-laboratory comparison was done across eight hospital sites having ABL90 Flex Plus. The stability of both tests was tested for 48 h at ambient and refrigerated temperatures. Results were compared between centrifuged and uncentrifuged fresh CSF samples to determine if particulate in uncentrifuged samples impacted analysis.

RESULTS

Glucose and lactate assays were linear over a broad analytical range of 1-45 mmol/L and 0-37 mmol/L, respectively, and demonstrated a good correlation with the routine chemistry laboratory method. LOQ was determined as 0.4 mmol/L for CSF glucose with a coefficient of variation (CV) of 14.7 % and 0.2 mmol/L with 0 % CV for CSF lactate respectively. Repeatability and reproducibility show small imprecision for both these assays. Glucose and lactate were stable for over 48 h at room or refrigeration temperatures. Sample particulates had no impact on the measurement. The inter-laboratory comparison was within total allowable error for glucose and lactate.

CONCLUSIONS

Acceptable performance characteristics, small sample volume, and rapid TAT make ABL90 Flex Plus an acceptable alternative analyzer for CSF glucose and lactate.

Added value of a connected glucose meter for glycorrhachia assessment

Objectives

The aim of this study was to demonstrate the performance and added value of rapid glucose determination in cerebrospinal fluid using a connected glucometer.

Design and Methods

Intra-assay and inter-assay accuracies were calculated using residual clinical samples. Accuracies were measured by comparing the results obtained with the glucometer to those from the central laboratory on a large routine chemistry platform.

Results

The intra-assay coefficients of variation were between 6.1% and 6.2% for low values (18 mg/dL) and between 5.6% and 6.8% for high values (58 mg/dL). The inter-assay coefficients of variation were between 9.4% and 16.3% for the low values (18 mg/dL) and between 5.7% and 8.7% for the high values (pool; ±75 mg/dL). The regression equation by comparison to the central laboratory was y = 4.08 + 0.82 x, with a coefficient of determination (r2) of 0.95.

Conclusions

The measurement of glycorrhachia with a connected glucometer before the analysis in the central laboratory allows a rapid orientation in the deferential diagnosis of a meningitis of viral vs bacterial origin. The response time is fast (6 s) and requires only a small amount of fluid (1.2 μL), which is important in infants, especially since lumbar puncture is an integral part of the investigation of the origin of a fever in this population.