Pediatric shock has a high mortality rate because many of the early clinical signs are subtle and have poor sensitivity and specificity. Pediatric shock was categorized either: compensated with normal blood pressure, poor skin perfusion (CRT >2 s, mottled, cool peripheries, peripheral cyanosis), weak peripheral pulse, age specific tachycardia, tachypnoea, and oliguria or decompensated with hypotension (SBP < 70 + (2× age in years) mm Hg and decreased mental status. The perfusion index is a non-invasive method for assessing peripheral perfusion and may be a useful marker for identifying shock early in pediatric patients.
This prospective cohort study (November 2019 to August 2020) evaluated whether the perfusion index, lactate, and/or lactate clearance could predict mortality among pediatric shock patients.
Fifty children (68% male) with shock underwent assessments at presentation to the emergency room to evaluate their heart rate, blood pressure, capillary refill time, central venous pressure, perfusion index, cardiac index, systemic vascular resistance, central venous oxygen saturation, and lactate clearance.
The perfusion index range was 0.03 to 2.2 and ≤0.18 as the cut-off for mortality prediction providing 74% sensitivity and 78% specificity. The serum lactate concentration range was 0 to 16 mmol/L and >5.7 mmol/L as the cut-off for mortality prediction provided 70% sensitivity and 96% specificity at presentation to the emergency room. The lactate clearance range was 3% to 75% and >10% as the cut-off for survival prediction after resuscitation and at 6 h later.
Perfusion index (PI), lactate, and lactate clearance provided comparable sensitivity and specificity for predicting outcomes among pediatric patients with shock Therefore, we suggest that the PI is an inexpensive, rapid, and non-invasive tool that can be used to predict illness severity and mortality in busy pediatric intensive care units and emergency departments. This tool may guide better patient triage and an earlier diagnosis of shock in this setting.