Under Pressure: Do We "Dare Change Our Way of Caring" for Patients With Shock?

Abstract

Circulatory shock is a common condition leading to ICU admission which occurs when there is a mismatch between oxygen supply and demand resulting in altered tissue perfusion, deficient oxygen, and nutrient delivery to tissues and ultimately to cellular and organ dysfunction (1). In addition to treating the underlying cause of the shock state (e.g., infection in septic shock), the initial hemodynamic goal of therapy is to provide adequate perfusion to the tissues and organs and normalize cellular metabolism (2). In patients with shock requiring vasopressor therapy, the mean arterial pressure (MAP) has been a long-standing hemodynamic variable targeted as a surrogate measure of tissue perfusion. A MAP goal is typically included in most clinical vasopressor order sets. Physiologic studies suggest that vascular bed autoregulation is compromised below a MAP of 60 mm Hg leading to a linear dependence of regional blood flow on MAP (2). These critical autoregulation MAP set points may vary by organ and depend on patient factors such as age and comorbidity (2). The most recent Surviving Sepsis Campaign (SSC) treatment guidelines recommend targeting an initial MAP of 65 mm Hg versus higher MAP targets in patients with septic shock requiring vasopressor support (3). These recommendations are based upon three recent multicenter prospective clinical trials in distributive shock looking at associations between MAP target goals and outcomes (4–6). The Sepsis and Mean Arterial Pressure (SEPSISPAM) (4) and Optimal Vasopressor Titration (OVATION) (5) studies randomized patients to target MAP goals of 65–70 mm Hg versus 80–85 mm Hg and 60–65 mm Hg versus 75–80 mm Hg, respectively, and found no significant difference in mortality between groups. Even though the SEPISPAM study found that targeting a higher MAP was associated with a higher occurrence rate of arrhythmia and more vasopressor exposure, it also reported that targeting a higher MAP was associated with a lower occurrence rate of renal impairment in the subgroup of patients with chronic hypertension. More recently, the 65 Trial (6) did not show a difference in mortality when comparing treatment aiming for a specified MAP target of 60–65 mm Hg versus “usual care” that was based on the SSC treatment guidelines. Similar to the SEPSISPAM study, the 65 Trial found that targeting a higher MAP was associated with more vasopressor exposure. A number of unanswered questions remain regarding the optimal MAP target and the balance between maintaining tissue perfusion and avoiding vasopressor-induced adverse effects. To that end, two independent systematic reviews and meta-analyses (SRMAs) in this issue of Critical Care Medicine provide further insight regarding optimal MAP targets for the treatment of shock (7,8). Richards-Belle et al (7) pooled patients (n = 3,496) from the aforementioned vasodilatory shock studies and explored whether higher versus lower vasopressor exposure was associated with mortality. The authors assumed that the MAP goals targeted in the clinical trials equated to vasopressor exposure, achieving a higher or lower MAP target equated to higher or lower vasopressor exposure, respectively. In contrast, Carayannopoulos et al (8) pooled patients from six studies (n = 3,690) and explored whether a higher versus lower targeted MAP goal was also associated with mortality in a variety of shock types. In addition to the three vasodilatory shock studies included in the study by Richards-Belle et al (7), this SRMA included three more shock studies (two cardiogenic, one hemorrhagic) (8). Although Carayannopoulos et al (8) found no difference between groups in the primary outcome of all-cause mortality at longest follow-up (relative risk [RR] 1.06; 95% CI 0.98–1.15; p = 0.12; moderate certainty), Richards-Belle et al (7) concluded that lower vasopressor exposure probably lowers 90-day mortality compared with higher vasopressor exposure (RR 0.94; 95% CI 0.87–1.02; moderate certainty). Secondary outcomes were not different between groups in either SRMA with the exception that Richards-Belle et al (7) found that lower vasopressor exposure may decrease the risk of supra-ventricular tachycardia (SVT) (odds ratio 0.55; 95% CI 0.36–0.86; low certainty). Furthermore, both SRMAs reported that the duration of vasopressor therapy was consistently decreased in the lower targeted MAP groups. Although the strengths of both of these analyses are that they are rigorous SRMAs with trial sequential analysis, there are several notable limitations. The generalizability and certainty of outcomes in both SRMAs are hindered by the heterogeneity of the pooled populations studied and the differences in the targeted MAP goals. The SRMA by Carayannopoulos et al (8) was the most varied comparing three shock types, whereas Richards-Belle et al (7) only included vasodilatory shock. Indeed, although the overall treatment goals of circulatory shock are similar across shock types, optimal MAP targets may vary. For example, data suggest that targeting a lower MAP in hemorrhagic shock as compared to the septic shock target of 65 mm Hg may be associated with better outcomes, whereas patients with cardiogenic shock may do better with a higher MAP target (9,10). Despite a clear separation between higher and lower MAP targets in each SRMA, the actual MAP values were consistently higher than the prespecified study targets. Therefore, the conclusions are based on the higher MAPs obtained, not the prespecified targets of interest. It remains unclear if adequate protocol adherence to targeted outcomes in the individual studies would have changed outcomes. Richards-Belle et al (7) interpreted that mortality is probably lower with lower vasopressor exposure; we feel the effect of MAP target on reducing mortality is minimal at best and must be interpreted with caution. Indeed, it must be noted that the certainty of this finding was lowered for imprecision and that the CI cannot exclude the possibility of harm, and even then, the findings are limited to the actual MAP targets achieved. Additionally, they concluded that even a modest a priori 20% RR reduction is not likely. Even though Richards-Belle et al (7) concluded that the risk for SVT may be decreased with lower vasopressor exposure, that outcome was underpowered, and that the sensitivity to capture arrhythmia incidence was low. Previous studies using systematic monitoring for arrhythmias during septic shock report much higher incidences of SVT (up to 40%) as compared to the incidence in both SRMAs (up to 5%) (11). Furthermore, subgroup analysis such as in patients with chronic hypertension was underpowered to make appropriate inferences and can only be seen as hypothesis-generating. So, what can we take away from these two rigorous SRMAs? Although MAP targets most likely vary between shock states, the targeting of higher (e.g., > 70 mm Hg) versus lower (e.g., 60–65 mm Hg) MAP goals does not seem to add additional benefit. For vasodilatory shock, the best target estimate is from the lowest MAPs obtained in the permissive hypotension group of the 65 Trial (6), suggesting that an initial MAP target of 65 mm Hg is reasonable. Other than reducing possible vasopressor-induced adverse effects, lower vasopressor exposure may have other benefits such as quicker vasopressor infusion discontinuation and more efficient critical care de-escalation. Given the protocol deviations in attaining the lower MAP target goals, these SRMAs also highlight that even within these rigorous randomized-controlled trial settings, clinicians overshot targeted MAP goals. One of the major tasks we face is to implement a lower MAP goal in our usual clinical practice (12). If we can set a lower MAP goal at 60–65 mm Hg instead of greater than 65 mm Hg, we are more likely to achieve an overall lower average MAP. Standardization of vasopressor algorithms and leveraging automated systems can also assist in implementation (13). We would argue that strict targeting of lower MAPs must be balanced by preventing recurring and prolonged episodes of MAPs below 60 mm Hg to the potential complications of lower MAP. Even though there was most likely a low capture of arrhythmia incidence in Richards-Belle et al (7), the reported 45% lower risk of arrhythmia in the lower vasopressor exposure group may have importance for long-term patient-centered outcomes. Therefore, future studies of MAP targets should focus on improved surveillance of arrhythmias (and other important adverse events). Last and most important, the treatment goal of shock is to restore perfusion. Targeting a defined MAP does not always equal restoration of tissue and organ perfusion as the MAP is just one piece of the perfusion restoration puzzle (14). The most likely key to successful treatment of shock is to personalize perfusion monitoring and integrate MAP along with several other macro- and microcirculation hemodynamic markers (e.g., capillary refill time) (15). So, do we dare to change the way we care for our patients in shock? The two SRMAs (7,8) presented in this issue of Critical Care Medicine suggest that targeting of lower MAPs is an appropriate treatment goal. With lower MAP targets and in turn, lower vasopressor exposure, we may be able to more rapidly de-escalate from critical care. However, changing our MAP target is easier said than done. An initial target of 60–65 mm Hg for vasodilatory shock is reasonable while also considering a more personalized approach to perfusion restoration monitoring. Further studies are certainly needed to help determine optimal MAP targets in different shock types and of equal importance, ways to implement those targets.