Comparative Evaluation of Blood Gas and Biochemistry Analyzers in Lactate Measurement in Pediatric Patient Groups

Abstract

Objective:

Lactate plays an important role in clinical decision-making processes as an early biochemical marker of tissue hypoxia and hypoperfusion. The aim of this study is to compare lactate measurements obtained from blood gas and biochemistry analyzers in pediatric patients and to evaluate the clinical concordance between the methods.

Methods:

Pediatric cases with lactate measurements performed simultaneously on blood gas and biochemistry analyzers at the Ankara Etlik City Hospital Biochemistry Laboratory between January 2023 and August 2025 were retrospectively reviewed. A total of 914 paired lactate measurements were included in the analysis. Measurements were evaluated using Pearson correlation, Bland–Altman analysis, Passing–Bablok regression, and Cohen's kappa coefficient. A total allowable error (TEa) of ±0.20 mmol/L was considered the clinical acceptance criterion.

Results:

A very strong positive correlation was found between blood gas and biochemistry measurements (r = 0.95, p < 0.0001). In the Bland–Altman analysis, the mean bias was −0.13 mmol/L (95% CI: −0.18 to −0.08), the lower limit of agreement (LoA) was −1.56 mmol/L, and the upper LoA was 1.31 mmol/L; 95.1% of the results fell within these limits. The Passing–Bablok regression equation was y = 0.211 + 0.975·x (intercept 0.211, 95% CI: 0.168–0.251; slope 0.975, 95% CI: 0.954–0.996), and no deviation from linearity was observed (CUSUM p = 0.97). In the categorical analysis, the fit was good, with Cohen's kappa found to be 0.73 (95% CI: 0.70–0.76), and the misclassification rate remained below 10%.

Conclusion:

This study demonstrated that blood gas and biochemistry analyzers showed a high degree of agreement in lactate measurements in the pediatric patient group. The advantage of blood gas devices in providing rapid results is an important support element for clinicians, particularly in intensive care and emergency departments. Although minor differences were observed between the methods at low and high concentrations, it was determined that these differences were not significant enough to influence clinical decisions.