Increasing arterial blood pressure with norepinephrine does not improve microcirculatory blood flow: a prospective study

Septic shock is characterized by severe vasodilation and hypotension refractory to aggressive fluid resuscitation [1]. Despite the normalization of cardiac output, evidence of tissue hypoperfusion is frequently present. Accordingly, organ dysfunctions usually develop despite normal or increased oxygen transport (DO₂). Microcirculatory alterations could be an underlying explanation for these findings [2]. Experimental models of resuscitated septic shock show that microvascular perfusion is altered despite the normalization of systemic and regional hemodynamics [3]. In addition, septic patients systematically exhibit severe disorders in sublingual microcirculation that are strongly associated with organ failures and outcome [4, 5]. The ability to improve sublingual microcirculation has also been related to survival [5]. Moreover, sublingual perfusion might be enhanced by different therapeutic strategies that include the use of vasoactive drugs [6, 7]. In this way, improving microcirculation might be an important goal in the resuscitation of patients with septic shock.

An approach to improve microcirculation is to increase the perfusion pressure. When the mean arterial pressure (MAP) decreases below an autoregulatory threshold of about 60 to 65 mmHg, organ perfusion becomes pressure dependent [8]. Nevertheless, intravascular thrombosis and vasoconstrictor mediators, along with regional deficiencies in nitric oxide production, could alter vascular reactivity and shift the autoregulatory threshold to higher values [9, 10]. Consequently, the increase in MAP could improve tissue perfusion. Clinical studies, however, have shown that the elevation in MAP beyond 65 mmHg fails to increase systemic oxygen metabolism, skin microcirculatory blood flow, urine output, splanchnic perfusion, or renal function [11, 12]. The experimental evidence regarding this issue, however, is controversial [13‐17].

Our goal was to assess the effects of titration of a norepinephrine infusion to increasing levels of MAP on sublingual microcirculation. We hypothesized that the increase in MAP from 65 to 75 mmHg, and then to 85 mmHg does not improve sublingual microcirculatory blood flow. At the time of submission of this manuscript, a very similar study was published [18], which reported that escalating doses of norepinephrine in septic patients increased DO₂ and tissue oxygenation, and were not associated with significant changes in preexisting sublingual microvascular alterations. The results of our study confirm these previous findings, but suggest that the individual responses are related to the basal microcirculatory condition.

The protocol was approved by the Institutional Review Boards of Sanatorio Otamendi and Clínica Bazterrica. Informed consent was obtained from the next of kin of all patients admitted to the study.

The main finding of this study is that the increase in MAP with norepinephrine failed to improve sublingual microcirculation, or any other variable related to perfusion as arterial lactate, anion gap, ΔPCO₂, and parameters of oxygen metabolism. Despite a trend to decreased total and perfused capillary density, there were considerable variations in the interindividual responses that seem to depend on the basal condition of the microcirculation.

The goal of vasopressor therapy is to improve tissue perfusion pressure, while avoiding excessive vasoconstriction. Marik and Mohedin showed that an infusion of norepinephrine titrated to increase the MAP to more than 75 mmHg improved intramucosal pH [26]. Martin and colleagues [27] and Desjars and colleagues [28] reported significant increases in urine output and improvements in renal function in septic shock. Nevertheless, in these studies [26‐28] the initial MAP was below 60 mmHg, a value that is most likely beyond the lower limit of autoregulation. On the other hand, Deruddre and colleagues showed that increasing MAP from 65 to 75 mmHg with norepinephrine in patients with septic shock increased urinary output and decreased renal vascular resistance [29].

Our study is consistent with the results from LeDoux and colleagues [11] and Bourgoin and colleagues [12]. In these studies, the lack of change in any perfusion variable over a range of 20 mmHg in MAP suggests that the patients were within their autoregulatory range. Jhanji and colleagues have recently demonstrated that increasing doses of norepinephrine resulted in an increase in global DO₂, and in cutaneous microvascular flow and tissue partial pressure of oxygen (PO₂) without significant changes in sublingual microcirculation [18]. They also showed, however, that when MAP was augmented from 70 to 90 mmHg, the MFI, proportion of perfused vessels, and perfused vessel density fell by about 10%. The remarkable similarity between the study by Jhanji and colleagues [18] and the current study emphasizes the reproducibility of the techniques and results.

In addition, our results expand previous knowledge by addressing the variation of interindividual responses. In particular, the change in the perfused capillary density was strongly dependent on the basal state of microcirculation. In this way, perfused capillary density improved in patients with an altered sublingual perfusion at baseline, and decreased in patients with preserved basal microvascular perfusion. Sakr and colleagues described a similar microvascular response to red blood cell transfusion [30]. Other studies have also shown that vasopressors could decrease sublingual microcirculation [31, 32], suggesting that excessive vasoconstriction might be deleterious to microcirculation.

Our study has several limitations. First, this observational study lacks a control group. Each patient, therefore, served as his/her own control. Second, the number of patients included in this study was small. Despite the sample size, significant changes in hemodynamic variables developed. Conversely, most parameters related to tissue perfusion and oxygenation remained unchanged or had a trend to worsen. Third, the infusion period was short. A longer period might have allowed the appearance of changes not observed in the present study. Nevertheless, most norepinephrine effects on hemodynamics and on tissue perfusion and oxygenation are expected to be evident within a few minutes. In addition, the short half-life of norepinephrine allows a new steady state to be reached in its plasmatic levels a few minutes after a change in the infusion rate [33]. In fact, several hemodynamic changes appeared shortly after each dose modification. Moreover, the infusion period was deliberately kept short, to avoid the background effects of changes in the underlying conditions of the patients. With longer periods of evaluation, a time effect with spontaneous changes of the studied variables related to the natural history of the disease might not be ruled out. Finally, as a different behaviour of microcirculatory beds is a characteristic of sepsis [3, 34], this study does not address the response of other territories to increasing MAP.

In this observational study, patients with septic shock showed severe microcirculatory alterations that failed to improve with the increases in MAP with norepinephrine. Furthermore, linear trend analysis showed reductions in the capillary and in the perfused capillary densities. Nevertheless, interindividual responses could be quite variable and dependent on the basal state of the microcirculation. Our results suggest that the increase in MAP above 65 mmHg is not a straightforward treatment to improve microvascular perfusion. It might be harmful for some patients, while benefiting others. Studies including greater numbers of patients are needed to determine the usefulness of individual titration of vasopressor therapy on sublingual microcirculation.