To the Editor:
Dr. Rangasamy and colleagues investigated the reliability and validity of noninvasive measures of blood pressure (BP) and pulsatility in continuous flow left ventricular assist device (LVAD) patients.1 Stroke remains the major cause of morbidity and mortality in LVAD patients, and improved BP control has been shown to significantly decrease stroke risk.2,3 The authors surmise that in patients with a palpable radial pulse, Doppler BP correlated better with systolic BP (SBP) than mean arterial pressure (MAP), whereas in patients with nonpalpable pulse, Doppler BP had acceptable correlation with both SBP and MAP. Consequently, the authors proposed an algorithm to treat Doppler BP as SBP in LVAD patients with a palpable pulse and as MAP in patients without a palpable pulse.
We have several concerns regarding the validity and implications of the authors’ proposed BP algorithm. First, the algorithm is predicated on an accurate determination of the presence of a palpable radial pulse which is difficult in continuous flow LVAD patients. The authors reported that among two cardiologists and two critical care nurses, the kappa statistic of interobserver agreement regarding palpable pulse is 0.41 which is only fair to moderate.4 The authors showed graphically a considerable overlap in arterial line pulse pressure (PP) between the palpable and nonpalpable pulse groups. As such, pulse palpation is neither a reliable nor accurate measure of PP.
Secondly, all studied patients were in the ICU between postoperative days 1 and 5 and nearly two thirds had a palpable pulse. Outside the immediate postoperative period and devoid of hypervolemia and vasoactive support, LVAD patients typically have narrower PP and therefore fewer will have a palpable pulse. In fact, the presence of a newly palpable pulse should prompt a clinical evaluation for its etiology, such as uncontrolled hypertension, hypervolemia, or device malfunction. As such, in an ambulatory LVAD population with low prevalence of palpable pulse, it would be more likely to erroneously classify a nonpulsatile patient as having a palpable pulse leading to undertreatment of BP by using the Doppler BP as SBP than vice versa. This potential error will have serious consequences for patients as undertreatment of BP significantly increases stroke risk. Third, it seems implausible that patients with a consistently closed aortic valve can generate a PP of nearly 30 mm Hg as reported by the authors. Finally, the authors’ conclusion that the HeartMate II pulsatility index is a reliable noninvasive measure of pulsatility is troubling, given the inconsistency and unreliability of this measure in other studies.5
We applaud the authors’ effort to contribute to the important question of BP management in LVAD patients. Unfortunately, further studies are required to ascertain how to optimally obtain BP in this patient population. Until then, algorithms like this seem premature.
Jennifer A. Beckman
Medicine-Cardiology, University of Washington
1. Rangasamy S, Madan S, Saeed O, et al. Noninvasive measures of pulsatility and blood pressure during continuous-flow left ventricular assist device [published online ahead of print April 19, 2018]. ASAIO J 2018.
2. Li S, Mahr C. Evaluating ventricular assist device outcomes internationally with a focus on neurological events [published online ahead of print September 22, 2018]. Heart 2018.
3. Milano CA, Rogers JG, Tatooles AJ, et al.; ENDURANCE Investigators: HVAD: The ENDURANCE Supplemental Trial. JACC Heart Fail 2018.6: 792–802.
4. Viera AJ, Garrett JM. Understanding interobserver agreement: the kappa statistic. Fam Med 2005.37: 360–363.
5. Lund LH, Gabrielsen A, Tirén L, Hallberg A, El Karlsson K, Eriksson MJ. Derived and displayed power consumption, flow, and pulsatility over a range of HeartMate II left ventricular assist device settings. ASAIO J 2012.58: 183–190.