Value of heart rate-blood pressure product of multiplication for identifying compensated shock in children

Abstract

Objective: To estimate the predictive value of heart rate (HR)-blood pressure (BP) products of multiplication for compensated shock in children. Methods: The study population consisted of 99 children with shock who had lactate measured before receiving vasopressor agents in Department of Critical Care Medicine of Children's Hospital, Capital Institute of Pediatrics from October 2015 to March 2021. The clinical data including the HR, BP, HR to BP ratio, HR-BP product and lactate at admission and after the correction of shock, as well as the 28-day mortality were collected. According to the outcome at the 28th day, the patients were divided into survival group and non-survival group. Comparisons between groups were performed with unpaired Student t test, or Mann-Whitney U test, or chi-square test. Pearson correlation analysis was used to analyze the correlations between lactate and HR, BP, HR to BP ratio and HR-BP product, respectively. Receiver operating characteristic (ROC) curve was analyzed to evaluate the predictive values of HR, BP, HR to BP ratio and HR-BP product for lactate greater than 2 mmol/L. Results: In these 99 children, 49 were males, and the median age was 3.8 (0.7-6.0) years. The most common type of shock was septic shock (61 cases, 62%), followed by cardiogenic shock (12 cases, 12%), hemorrhagic shock (12 cases, 12%), Kawasaki disease shock syndrome (8 cases, 8%) and anaphylactic shock (6 cases, 6%). Sixty-six patients (67%) survived, and 33 patients (33%) died. ROC curve showed that the area under curves (AUC) of lactate (optimal cutoff value 3.15 mmol/L, sensitivity 96.0%, specificity 54.4%, P<0.01) and HR to systolic blood pressure ratio (HR/SBP) (optimal cutoff value 2.0 times/(min·mmHg), sensitivity 62.5%, specificity 69.0%, P = 0.03) for predicting adverse outcome were 0.769 and 0.649, respectively. There were significant correlations between lactate and HR to diastolic blood pressure (DBP) ratio, HR to mean blood pressure (MBP) ratio, SBP, HR/SBP, MBP, DBP and HR (r= 0.476, 0.452, -0.444, 0.425,-0.410, -0.364, 0.177, all P<0.01), while no significant correlation was found between lactate and the products of HR and BP(all P>0.05). HR/SBP performed better than the other six parameters for predicting lactate>2 mmol/L, with the AUC of 0.872 and the optimal cutoff value of 1.4 bpm/mmHg (sensitivity 92.1%, specificity 70.9%, P<0.01). When MBP was greater than or equal to 65 mmHg, MBP × HR, DBP × HR, SBP × HR, HR, HR/SBP, HR/MBP and HR/DBP were significantly correlated with lactate (r= 0.706, 0.705, 0.669, 0.626, 0.555, 0.502, 0.446, all P<0.01). And MBP × HR performed better for predicting lactate>2 mmol/L than the other six parameters, with the AUC of 0.974 and the optimal cutoff value of 9446 bpm × mmHg (sensitivity 100.0%, specificity 90.9%, P<0.01). Conclusions: The product of HR and BP, especially the MBP × HR, shows higher predictive values for abnormally elevated lactate in children with compensated shock than the HR/SBP does. It is worth recommending for early identification of compensated shock in children.