Background: Impaired peripheral perfusion is among the first manifestations of shock and the last to restore in the critically ill patient. In acute circulatory failure the activation of the sympathetic nervous system leads to a redistribution of blood from non-vital organs to the core in order to conserve oxygen delivery and tissue function of vital organs. Therefore, the clinical assessment of peripheral perfusion is an early and easily applicable parameter for detecting and evaluating tissue hypoperfusion and can help guiding therapy in acute circulatory failure. The peripheral perfusion index (PPI) is a simple, non-invasive tool to assess peripheral perfusion by calculating a quotient of pulsatile versus non-pulsatile blood flow of distal extremities using simple commercial pulse-oximetry, with lower values indicating poorer perfusion.

Objectives: With this study we aim to establish the peripheral perfusion index as a non-invasive, objective and continuous measurement comparable to other clinical assessments of peripheral perfusion such as capillary refill time (CRT), mottling score and temperature gradient from core and knee to toe. Furthermore, we aim to establish the relationship between PPI and urine production, arterial lactate concentration and perfusion of solid organs measured by ultrasound.

Methods: In this prospective cohort study at a tertiary teaching hospital we analyzed repeated measurements of peripheral perfusion, urine and lactate in 59 critically ill patients admitted to the ICU after cardiac surgery or in septic shock during the first 72 hours after admission, as well as sonographic assessment of organ perfusion in septic patients. Additionally, we established normal values of PPI in 30 healthy volunteers at ambient temperature and at vasoconstriction state by cooling of the upper extremities.

Results: In healthy volunteers, patients after cardiac surgery and in septic shock patients, PPI was 4.07 (3.36-4.79), 0.88 (0.66-1.88) and 1.29 (0.86-1.80), respectively and CRT 1.28s (0.96-2.02), 5.32s (4.41-6.23) and 4.07s (3.15-7.28), respectively for initial values in patients and during cooling state in healthy volunteers (median, IQR, all p