Article révisé par les pairs
Résumé : Figure 3 (Meta‑analysis of studies comparing the impact on hyponatremia occurrence of isotonic versus hypotonic solutions) published in the original version of the manuscript is incorrect [1]. A new version of Fig. 3 is provided in this erratum. (Figure presented.) Meta-analysis of studies comparing the impact on hyponatremia occurrence of isotonic versus hypotonic solutions The error arose from the reversal of the “experimental” and “control” groups during data extraction. In fact, in the included studies, the experimental and control groups corresponded to the “isotonic” and “hypotonic” groups respectively, in most studies, but not all [2, 3]. To ensure optimal homogeneity in outcome definition we have revised where possible the threshold of hyponatremia at 135 mmol/L rather than 130 mmol/L in the few studies that used a 130 mmol/L hyponatremia as the primary outcome but also provided figures for 135 mmol/L [4–6]. Finally, in the study with 3 arms, we revised the experimental and control groups to ensure better consistency in interpretation within the studies [7]. The new effect size in Fig. 3 is OR = 0.31, 95%CI [0.23; 0.42], I2 = 36%, p-value < 0.00001. The heterogeneity between studies is now low. The authors consider it important to publish this erratum to comply with good research practice. Importantly, the updated results do not alter, but rather strengthen the level of evidence for the PiCO2 recommendation: “in acutely and critically ill children, isotonic maintenance fluid should be used to reduce the risk of hyponatremia”; level of evidence A. The authors apologize for this error.

Documents en relation

DI-fusion