Changes in left ventricular ejection time and pulse transit time derived from finger photoplethysmogram and electrocardiogram during moderate haemorrhage.

OBJECTIVES Early identification of haemorrhage is difficult when a bleeding site is not apparent. This study explored the potential use of the finger photoplethysmographic (PPG) waveform derived left ventricular ejection time (LVET(p)) and pulse transit time (PTT) for detecting blood loss, by using blood donation as a model of controlled mild to moderate haemorrhage. METHODS This was a prospective, observational study carried out in a convenience sample of blood donors. LVET(p), PTT and R-R interval (RRi) were computed from simultaneous measurement of the electrocardiogram (ECG) and the finger infrared photoplethysmogram obtained from 43 healthy volunteers during blood donation. The blood donation process was divided into four stages: (i) Pre-donation (PRE), (ii) first half of donation (FIRST), (iii) second half of donation (SECOND), (iv) post-donation (POST). RESULTS AND CONCLUSIONS Shortening of LVET(p) from 303+/-2 to 293+/-3 ms (mean+/-SEM; P<0.01) and prolongation of PTT from 177+/-3 to 186+/-4 ms (P<0.01) were observed in 81% and 91% of subjects respectively when comparing PRE and POST. During blood donation, progressive blood loss produced falling trends in LVET(p) (P<0.01) and rising trends in PTT (P<0.01) in FIRST and SECOND, but a falling trend in RRi (P<0.01) was only observed in SECOND. Monitoring trends in timing variables derived from non-invasive ECG and finger PPG signals may facilitate detection of blood loss in the early phase.