Evaluating intravascular hypovolaemia by manually assessing peripheral limb coolness compared with the capillary refill time in children aged <5 years in a tropical environment: a prospective cohort study of diagnostic accuracy.

Objective: It is important to detect reduced peripheral perfusion due to hypovolaemia early before it progresses to shock. Capillary refill time (CRT) and manual palpation for peripheral coolness are used widely to detect this, while the central-peripheral temperature gradient is used as a measure of peripheral perfusion in intensive care settings. Here, our aim was to compare the precision of these three clinical signs to detect hypovolaemia in the tropics.

Design, setting and patients: Trainee paediatricians measured the CRT, graded the coolness of the toes manually and measured temperature gradients with inexpensive infrared thermometers in healthy control children and in sick children admitted to two hospitals in Ghana over 1 year, and compared these data to the intravascular volume status measured by the inferior vena cava diameter to height ratio (IVCd/ht) Z-scores.

Results: We studied 1383 well and 787 hospitalised children aged 2 months to 5 years, of which 99 had hypovolaemia. The hospitalised children had a median age of 1.63 years and weight of 8.7 kg. The most useful method to predict hypovolaemia was to detect mild peripheral coolness (sensitivity 44%, specificity 88.8%). The measured temperature gradient was less sensitive at 26%, perhaps because the ambient temperature (average 30.2°C) was close to body temperature. The CRT also had low sensitivity at all values (best 29% at ≥2 s).

Conclusions: Manual palpation for detecting mild coolness predicts hypovolaemia more effectively than the CRT in small children, and measuring temperature gradients with a thermometer is relatively ineffective in a tropical setting.