Abstract

Purpose

To assess the utility and accuracy of creatinine measurement using point-of-care (POC) blood gas analysis compared to standard laboratory assay and to assess the effectiveness of frequent creatinine monitoring using POC blood gas analysis with respect to early detection of acute kidney injury (AKI).

Methods

We performed a single center retrospective observational study in 398 patients admitted to the intensive care unit. We investigated the clinical concordance of creatinine values measured by POC blood gas analysis (Cr-BG) compared to those measured by laboratory assay (Cr-lab). We compared the time to reach AKI diagnosis according to the KDIGO criteria using Cr-BG (KD-BG) to the standard criteria using Cr-lab (KD-lab).

Results

Cr-BG correlated well with Cr-lab for day 1 (n = 375, R² = 0.98, p < 0.001) and during the whole study period (n = 1258, R² = 0.99, p < 0.001). The KD-BG measurement allowed the identification of 6.3% more patients (60.6%) as AKI than the KD-lab measurement. Approximately one-third of the patients were staged as AKI using the KD-BG measurement more than 6 h earlier than that using the KD-lab measurement.

Conclusions

The KDIGO criteria using POC blood gas analysis were clinically useful. Frequent creatinine monitoring using POC analysis can allow an earlier detection of AKI.