Continual measurement of arterial dP/dt_max enables minimally invasive monitoring of left ventricular contractility in patients with acute heart failure

Left ventricular (LV) contractility is one of the most important parameters determining LV performance and cardiac function and, therefore, directly influences global hemodynamic status [1]. Clinical conditions with impaired LV contractility, such as heart failure or septic cardiomyopathy, are frequent subjects of intensive and acute cardiology care [1]. There is, therefore, an apparent clinical need for bedside measurement or even monitoring of contractility. However, current options for the assessment of LV contractility are significantly limited. The reference method for the measurement of LV contractility (i.e., LV end-systolic elastance [2]) cannot be used in routine clinical practice due to invasiveness and technical issues [3, 4]. The maximum rate of LV pressure rise during ventricular contraction (LV dP/dt_max) has been adopted as a surrogate marker of LV inotropic state and contractility [5]. This parameter is determinable in clinical settings; however, it requires direct LV pressure measurement, which is impractical and too invasive for LV contractility monitoring. LV dP/dt_max can also be estimated non-invasively using echocardiographic techniques [6, 7]. However, although this measurement can be performed repeatedly, it is not feasible for continuous monitoring and is frequently limited by low-quality signal.

Recently, a new surrogate has been proposed—arterial dP/dt_max. This parameter can be calculated from the arterial pressure waveform, obtained minimally invasively from a peripheral arterial line [8‐10] or even non-invasively [11]. Arterial dP/dt_max is, therefore, available bedside and in patients with an arterial line already used for pressure monitoring and blood gas analyses; it does not require any additional invasive access. Moreover, arterial dP/dt_max can be measured on a beat-by-beat basis and, therefore, continually monitored. On the other hand, arterial dP/dt_max is not only determined by LV contraction but is also influenced by various peripheral arterial factors and load conditions [8‐11]. Recently, several experimental studies demonstrating a significant relationship between arterial dP/dt_max and LV contractility have been published [8‐10]. To date, however, clinical studies focusing on the relationship between arterial and LV dP/dt_max in patients with acute heart failure requiring intensive care and an arterial line are lacking. The aim of our study was, therefore, to assess the relationship between arterial and LV dP/dt_max in this patient population.

Forty-eight patients were enrolled in the study; baseline characteristics of the study population are summarized in Table 1. The mean age was 70.4 years, the majority were males (65%), and the main cause of acute heart failure was ischemic cardiomyopathy (65%). Eighty-five percent of patients were treated with intravenous inotropes, and the majority required vasopressors (73%).

The values from continual arterial dP/dt_max monitoring were significantly correlated with LV dP/dt_max assessed using echocardiography (r = 0.70 [95% confidence interval (CI) 0.51–0.82]; P < 0.0001) (Fig. 1). Linear regression revealed that LV dP/dt_max = 1.25 × (arterial dP/dt_max) (P < 0.0001). Arterial dP/dt_max was significantly correlated with SV (r = 0.63 [95% CI 0.41–0.78]; P < 0.0001) and CO (r = 0.42 [95% CI 0.14–0.63]; P = 0.003). In contrast, arterial dP/dt_max was not correlated with SV variation, dynamic arterial elastance, heart rate, SVR, or mean arterial pressure (Table 2).

The correlation between arterial and LV dP/dt_max was calculated also in subgroups above and below the mean value of the recorded variables. Marked differences in the correlation between arterial and LV dP/dt_max were observed in the subgroups based on the mean SVR, CO, and SV; on the other hand, similar correlation was observed in the subgroups based on SV variation (r = 0.54 vs. r = 0.59), dynamic arterial elastance (r = 0.70 vs. r = 0.73), heart rate (0.68 vs. 0.72), and mean arterial pressure (r = 0.80 vs. r = 0.76).

Our results demonstrate that in adult patients with acute heart failure, the values of arterial dP/dt_max, which can be continuously monitored by analysis of the pressure waveform, were significantly correlated with LV dP/dt_max. An even better agreement between arterial dP/dt_max and LV dP/dt_max was observed in subgroups with higher SVR, lower CO, and lower SV. This observation is particularly important because monitoring LV contractility is most desirable in patients with heart failure with critical hemodynamic collapse, such as in cardiogenic shock, characterized by increased SVR and decreased CO and SV. In contrast, in the subgroups with lower SVR, higher CO, and higher SV, the correlation was weak or even absent.

Acute heart failure is a critical condition in which LV contractility is commonly depressed, and therapeutic strategies are frequently based on increasing inotropy [12]. Although LV contractility status can be routinely and intermittently assessed using echocardiography in patients with acute heart failure, even with the use of new techniques, the measurement can be inaccurate and interpretation difficult [13‐17]. There are currently highly limited options for continuous LV contractility monitoring. In contrast to echocardiographic techniques, the maximal rate of arterial pressure increases during systole (arterial dP/dt_max) can be easily and continuously calculated from the pressure waveform. It has been shown in animal studies that arterial dP/dt_max may correlate with LV contractility status under various hemodynamic conditions. In a porcine model of endotoxin-induced shock and catecholamine infusion, Morimont et al. [8] observed that arterial dP/dt_max was significantly correlated with LV contractility measured by LV end-systolic elastance (Ees) or LV dP/dt_max. The authors also found a better correlation when adequate vascular filling according to the arterial pulse pressure variation was achieved. These results are consistent with our observations. We have demonstrated a strong relationship between arterial and LV dP/dt_max under higher SVR, which could be also a result of increased vascular filling. Monge Garcia et al. [10] analyzed the relationship among arterial dP/dt_max, LV dP/dt_max, and Ees in sequential changes of afterload, preload, and contractility in pigs. In this study, arterial dP/dt_max enabled the tracking of Ees changes, especially during the modification of afterload and contractility, and changes in cardiac contractility (i.e., Ees) were the main determinants of arterial dP/dt_max changes. Moreover, these observations are in good agreement with our results; Monge Garcia et al. recorded higher values of LV dP/dt_max in comparison with arterial dP/dt_max, similar to our study. A good correlation between arterial dP/dt_max and LV dP/dt_max in heart failure patients was also reported in the study by Tartiere et al. [11], in which the dP/dt_max from the radial artery was assessed non-invasively using applanation tonometry and LV dP/dt_max. Our results are also consistent with the observation by Scolletta et al. [18] who reported significant correlation between arterial dP/dt_max and LV dP/dt_max in a group of critically ill patients. Furthermore, and again similar to our results, a very close linear relationship was found between arterial dP/dt_max from the femoral artery and invasively measured LV dP/dt_max (with a catheter in the LV) in patients scheduled for coronary artery bypass surgery [9]; arterial dP/dt_max also underestimated LV dP/dt_max in this study.

In contrast, Kim et al. [19] studied the relationship between arterial dP/dt_max from the radial artery, aortic dP/dt_max, and selected echocardiographic variables such as LV ejection fraction or LV fractional shortening in children undergoing congenital heart disease surgery. They did not find a significant correlation between arterial dP/dt_max and the other variables, and these observations were explained by differences between radial artery and aortic pressure waveforms. The discrepancy with our results can be elucidated by the other measured parameters in Kim et al.’s study. While we used LV dP/dt_max as a surrogate parameter of LV contractility, Kim et al. measured LV ejection fraction and LV fractional shortening, which are not the accepted markers of LV contractility, because they are influenced by many other factors (e.g., preload, mitral regurgitation). The observations in the study by Kim et al. could be also influenced by CO or SVR (or vascular filling), which were not recorded. Recently, Vaquer et al. [20] published a study focused on femoral arterial dP/dt_max in critically ill patients, predominantly with septic shock. They observed increase in arterial dP/dt_max after administration of dobutamine and norepinephrine but not after volume expansion. The changes in arterial dP/dt_max were strongly correlated with the changes in pulse pressure and systolic arterial pressure in all interventions including volume expansion. Vaquer et al. conclude that femoral arterial dP/dt_max is, therefore, an unreliable estimate of LV systolic function [20]. Our study was not designed to evaluate the relationship between dP/dt_max and LV systolic function; we focused on the comparison of the arterial dP/dt_max and the LV dP/dt_max, as a surrogate marker of LV contractility. Although the study group characteristics are different, the observation of increase in arterial dP/dt_max after administration of agents with inotropic effect (dobutamine, norepinephrine) in the study by Vaquer et al. [20] is in a good agreement with our results showing significant correlation between arterial and LV dP/dt_max. Our observation that correlation between arterial and LV dP/dt_max depends on SVR is also consistent with the results by Vaquer et al. [20] describing that femoral dP/dt_max is influenced by LV preload and afterload.

Several authors reported that the arterial dP/dt_max is significantly influenced by vascular filling conditions [8, 10, 20]. We observed in our study that the correlation between arterial and LV dP/dt_max was influenced by SVR, which reflects loading conditions; however, SV variation or dynamic arterial elastance had no effect on the relationship between arterial and LV dP/dt_max. This contrast can be at least partly explained by the fact that in our study, the measurement was done in a single time point, while in the other studies, serial measurements were performed enabling evaluation of dynamic changes. In addition, we included entirely patients with acute left heart failure, where the LV preload is usually increased that may or may not be accompanied by changes in other indirect markers of vascular filling. Our study had several limitations, the first of which was possible bias caused by the small sample size. We designed only a pilot study focusing primarily on feasibility; however, a larger trial should be performed to confirm our results. We also acknowledge that arterial dP/dt_max is not only a function of LV contractility but is influenced by many other factors, at least by arterial vessel wall characteristics (e.g., arterial elasticity and stiffness), which was not assessed in the present study. We did not record pulse pressure enabling to calculate arterial elastance. It can be assumed that there can be marked individual differences in arterial system properties in heart failure patients, who often present with other diseases and various degrees of peripheral atherosclerosis. Patients with moderate to severe aortic stenosis were ineligible for this study; however, we cannot exclude the possibility that even mild aortic stenosis may have influenced the results. Moreover, in our study, we have performed only single measurements at one time point in each patient. Our study, therefore, was not designed to evaluate the trends in arterial dP/dt_max changes. In addition, while the LV dP/dt_max values were obtained at the end of expiration, the arterial dP/dt_max were calculated as a median value from 20-s interval, therefore not at the same period of respiratory cycle. Finally, LV dP/dt_max is only a surrogate marker of LV contractility and measurement of this parameter using echocardiography can be inaccurate.

Our results suggest that in patients with acute heart failure requiring intensive care with an arterial line, continuous calculation of arterial dP/dt_max may be used for monitoring LV contractility, especially in those with higher SVR, lower CO, and lower SV, such as in patients experiencing cardiogenic shock. On the other hand, there was only a weak or no significant correlation in the subgroups with higher CO, higher SV, and lower SVR.