Clinical Evaluation of Doppler Blood Pressure Measurement in Continuous-Flow LVAD Patients: Implications for Postoperative Management.

Background

Continuous-flow left ventricular assist devices (LVADs) generate non-pulsatile circulation, rendering conventional oscillometric blood pressure measurements unreliable. Accurate monitoring is critical to prevent complications including stroke, pump thrombosis, and aortic regurgitation. Doppler-based measurement is widely used as a non-invasive alternative, yet its accuracy relative to invasive arterial pressure remains insufficiently characterized.

Methods

In this prospective single-centre study, 32 adult continuous-flow LVAD patients underwent simultaneous invasive radial artery and Doppler blood pressure measurements twice daily over three consecutive days (192 paired readings; Day 3: n = 27 due to technical recording issues). Pulsatility was assessed by means of peripheral pulse palpation and transthoracic echocardiography. Spearman's rho, Wilcoxon signed-rank test, and Bland-Altman analysis were applied.

Results

Median invasive MAP was 73.0 [IQR 66-80] mmHg and median Doppler pressure was 75.0 [IQR 70-80] mmHg. Doppler measurements demonstrated strong-to-excellent correlation with invasive MAP across all time points (r = 0.78-0.91, p < 0.001), with no significant paired differences (all p > 0.05). Bland-Altman analysis revealed a bias of -0.35 mmHg with limits of agreement of -9.10 to +8.40 mmHg, within the accepted ±10 mmHg threshold. Correlation with systolic pressure was lower (r = 0.66-0.89, p < 0.001), with a positive bias of +13.47 mmHg and wide limits of agreement (+1.28 to +25.67 mmHg), indicating clinically unacceptable agreement.

Conclusions

Doppler-derived blood pressure may provide a reliable estimate of invasive MAP in continuous-flow LVAD patients, whereas its utility for systolic pressure estimation appears limited. Doppler measurement represents a practical, non-invasive tool for routine MAP monitoring in both inpatient and outpatient settings.