How to improve ST-analysis in fetal monitoring: Relative versus absolute t/qrs ratio rises

Brief Introduction: ST-analysis is a promising method to detect intrapartum fetal metabolic acidosis (FMA). However, many false positives lower its specificity, leading to avoidable interventions and neonatal morbidity. Reducing false positives is preferable to current strict STAN guidelines. Present ST-analysis compares absolute T/QRS ratio rises to its baseline. As fetal heart axis orientation varies, fetal vector ECG parameters vary, resulting in interpatient T/QRS baseline differences. We hypothesize that patients with a higher T/QRS baseline are more prone to exceed absolute thresholds. We suggest analyzing relative rather than absolute T/QRS ratio rises would improve diagnostic accuracy. This study explores whether relative ST-analysis has distinctive value in detecting FMA and compares both methods. Materials & Methods: A retrospective case-control study was performed in a tertiary care teaching hospital with twenty healthy human fetuses during labour at term, of which 10 had an umbilical artery pH 57.05 (cases) and 10 had an arterial pH 47.20 (controls) after birth. Fetal electrocardiogram was recorded with single helix scalp electrode (Goldtrace), a maternal skin electrode and a STAN S31 monitor (Neoventa Medical, Molndal, Sweden). Relative T/QRS values were calculated based on stored data from these registrations. Absolute ST-events were obtained from Event Log of STAN viewer (Neoventa Medical). Absolute ST-events were defined as occurrence of either baseline or episodic ST-events. A receiver operator curve (ROC) of relative ST-analysis was constructed using SPSS Statistics 22. A cut-off value was determined at optimal specificity, which was used to calculate other test characteristics. Sensitivity, specificity, positive likelihood ratio (LR+), negative likelihood ratio (LR-) and diagnostic odds ratio (OR) were calculated for both relative and absolute ST analysis. Primary outcome measures were area under the curve of the ROC curve of relative T/QRS ratio and test characteristics of both relative and absolute ST analysis. Clinical Cases or Summary Results: Relative ST-analysis showed an area under the curve of 0.990 (95% CI 0.958-1.000). The optimal cut-off value of T/QRS ratio rise was determined at 70%. Relative vs. absolute ST-analysis showed specificity of 100% vs. 40% (p = 0.031); sensitivity of 90% vs. 90%; LR+ of infinity vs. 1.5; LR- of 0.1 vs. 0.25; OR of infinity vs. 6, respectively. Conclusions: Relative ST-analysis shows excellent distinctive value in this population. Specificity of this method is better than in absolute STanalysis (without CTG) at comparable sensitivity. However, this study only included patients with extreme pH values at birth; no patients with umbilical artery pH between 7.05 and 7.20 were included. Besides, the occurrence of biphasic events and CTG interpretation were not considered in this study, as we aimed to improve T/QRS baseline dependent aspects of ST-analysis. Nevertheless, relative ST-analysis seems to be a promising method for detection of metabolic acidosis during labour, as its specificity, LR- and OR are better than in absolute ST-analysis. Further research is required to determine whether our cutoff value is also appropriate in a large population that includes the complete clinical range of pH values; followed by external validation. Once validated, this method may improve detection of compromised fetuses during labour, decreasing unnecessary artificial deliveries and improving neonatal outcome