Perfusion Index Variations in Children With Septic Shock: Single-Center Observational Cohort Study in India

Abstract

Objectives: To study in children with septic shock: 1) variation in peripheral perfusion index (PI), which is a derived variable from pulse oximetry; 2) correlation between PI and lactate concentration; and 3) exploratory diagnostic evaluation between mortality and PI.

Design: Prospective observational study (from October 2018 to March 2020).

Setting: Pediatric emergency department and PICU of a tertiary hospital in India.

Patients: Children (1 mo to 16 yr old) with septic shock.

Interventions: None.

Measurements and main results: Data collected included demographic, clinical, laboratory, and outcome-related variables. Hemodynamic variables like heart rate, mean arterial pressure, and PI, along with serum lactate were recorded at specified intervals. A total of 112 children with septic shock were recruited, with median (interquartile range [IQR]) age of 50 (IQR 12,118.5) months and 65 of 112 (58%) were male children. Overall mortality was 25 of 112 (22%). At admission, the median PI was 0.6 (IQR -0.30, 0.93), and we used PI less than or equal to 0.6 to define a "critical PI." Of 61 children with critical PI at admission, 26 of 61 increased above this threshold by 6 hours. We observed a negative correlation between PI and lactate, at admission ( r = -0.27; 95% CI, -0.44 to -0.08; p = 0.006) and at 6 hours ( r = -0.21; 95% CI, -0.39 to -0.02; p = 0.03). In the exploratory analysis, a PI cutoff of less than or equal to 0.6 at 6 hours had area under the receiver operating curve of 0.74 (95% CI, 0.60-0.88). That is, with a 70% sensitivity and 81% specificity for mortality, the performance of such a test in our population (pre-to-post-test probability) for mortality would be 0.22-0.51.

Conclusions: We have used pulse oximetry-derived PI in children presenting with septic shock and found that the value is negatively correlated with a rise in serum lactate concentration. However, the utility of using a critical threshold value in PI (≤ 0.6) after 6 hours of treatment to be indicative of later mortality has considerable uncertainty.