ACE inhibitors and angiotensin receptor blockers differentially alter the response to angiotensin II treatment in vasodilatory shock

The ATHOS-3 trial has been previously described (clinicaltrials.gov identifier NCT 02338843) [8]. Briefly, adults with persistent vasodilatory shock after ≥ 25 mL/kg of volume resuscitation requiring high-dose vasopressors (i.e., norepinephrine-equivalent dose [NED] > 0.2 μg/kg/min) were randomly assigned 1:1 to receive synthetic human angiotensin-II (La Jolla Pharmaceutical Co.) or saline placebo plus standard vasopressors. Randomization was stratified by MAP at screening and Acute Physiology and Chronic Health Evaluation II (APACHE-II) score.

The present study reflects a post-hoc analysis of ATHOS-3 that aimed to answer the following questions:

The co-primary exposures of interest were exposure to ARB or ACEi therapy within the 7 days prior to randomization, which were assessed as potential effect modifiers for angiotensin-II treatment. Exposure was determined through a combination of history taken from patients or their representative and review of the electronic/patient medical record by study personnel.

As an exploratory analysis, the degree of ARB exposure was further assessed as a continuous variable by determining the equipotent dose levels in losartan equivalents (Additional file 1: Table-S1), using previously established conversion factors [10].

In the ATHOS-3 trial, study drug infusion was started at 20 ng/kg/min and titrated during hr₀-hr₃ to achieve MAP ≥ 75 mmHg while keeping other vasopressor doses constant. Thereafter, study drug and other vasopressors were titrated at treating clinicians’ discretion to maintain MAP between 65 and 75 mmHg. At hr₄₈, study drug infusion was discontinued according to a protocol-specified tapering process but could be continued for up to 7 days per clinician discretion.

The primary outcome was MAP at hr₁. This was selected because we reasoned the immediate MAP change best reflected the hemodynamic efficacy of angiotensin-II and because using a continuous variable (rather than the binary primary endpoint of the ATHOS-3 trial) would reduce the sample size needed to statistically interrogate an interaction effect. Additionally, we compared the NED and the study-drug titration level during hr_0-3 (i.e., during the titration period where background vasopressors were held constant), and during hr₄-hr₄₈ (i.e., the period where background vasopressors were not held constant). To investigate the association of prior ACEi and ARB exposure with the biomolecular RAS profile, we compared the baseline levels of renin, angiotensin-I, angiotensin-II, and the angiotensin-I/angiotensin-II ratio, and the change in renin at hr₃.

Patient-centered outcomes (e.g., 28-day mortality) were not analyzed due to insufficient sample size.

In the primary analysis, MAP and NED at hr₁-hr₃ were modeled in a linear regression that included terms for treatment and hr₀ value of the dependent variable in accordance with best statistical practice for evaluating continuous outcomes in clinical trials [11, 12]. These models were performed when stratified by exposure status (ARB, ACEi, no ACEi/ARB), and including an interaction-term for whether exposure modified the effect of treatment.

For longitudinal modeling of NED and study drug dose, a mixed-effects repeated measures model was used with a random intercept for individual patients and unstructured covariance. The model included fixed-effects for time, treatment, ACEi and ARB exposure, as well as interaction effects between these terms to assess heterogeneity of treatment effect. Models were separately constructed for the active titration period (hr_0-3) and the subsequent intervention period (hr_4-48). These models have the advantage of being robust to missing data bias, even when data are not missing at random [13]. To account for right-skewed data in NED, data were log-transformed for modeling.

Differences in baseline levels of RAS biomarkers were compared with generalized linear models with post-hoc adjustment for multiple comparisons. Again, to account for right-skewed data, biomarker levels were log-transformed for analysis. To facilitate intuitive interpretation, we report back-transformed differences and least-squares means (i.e., the geometric mean [gmean]).

We used two sets of multivariable models to account for potential confounding in ACEi/ARB exposure status. The primary analysis models used a parsimonious approach, adjusted for age, sex, baseline APACHE-II score (as a measure of overall illness severity), and baseline NED (as a measure of circulatory failure severity, specifically). These covariates were selected prior to analysis as they were felt to reflect the most important sources of potential confounding. In sensitivity analyses, we also assessed the main outcomes using models that adjusted for the previous covariates as well as baseline MAP, history of chronic kidney disease, chronic hypertension, and factors that had previously been associated with response to angiotensin-II [8, 14‐16]: baseline albumin, ARDS at baseline, and whether patients were on renal replacement therapy (RRT) at baseline. We used complete case-analysis as the first set of models had no missing data and the second set of models had only one covariate with missing data (albumin), which was missing in less than 5.0% of patients in all three exposure groups. We considered the use of inverse-probability of treatment weighting as an additional method of covariate adjustment, but the sample size was too small to facilitate effective balancing of the weighted population. As such we limit the presented analysis to traditional covariate adjustment.

For the exploratory analysis of ARB dose exposure, the analysis dataset was limited to ARB-exposed and ACEi/ARB-unexposed patients. The unexposed patients were considered to have a losartan-equivalent dose of 0 mg/day.

Of 321 patients included in the primary ATHOS-3 study, 270 (84%) were not exposed to ACEi or ARB, while 29 (9%) patients had received ACEis (n = 15; 52% for angiotensin-II and n = 14; 48% for placebo treatment), and 22 (7%) patients had received ARBs (n = 11; 50% each arm). Baseline characteristics are displayed in Table 1 and Additional file 1: Table-S2. Additional file 1: Table-S3 reports the missing data prevalence.

Figure 2 displays cardiovascular measures over time. Among ACEi/ARB-unexposed patients at hr₁, MAP increased by 11.1 mmHg [95% CI 10.0–12.3] (p < 0.001) with angiotensin-II, from 66 to 77 mmHg versus 2.1 mmHg from 65 to 67 mmHg [95% CI 0.9–3.1] (p < 0.001) in placebo-treated patients (adjusted treatment-effect: 9.1 mmHg [95% CI 7.6–10.1], p < 0.0001) (Fig. 2A, Additional file 1: Table-S4).

A similar increase was seen among ACEi-exposed patients, from 69 to 80 mmHg in the angiotensin-II group (difference:11.1 mmHg [95% CI 7.2–14.9], p < 0.0001) versus from 66 to 72 mmHg (difference: 5.8 mmHg [95% CI 1.9–9.7], p = 0.006) in the placebo group (adjusted difference in treatment-effect versus ACEi/ARB-unexposed: − 2.2 mmHg [95% CI − 7.0–2.6], p_interaction = 0.38).

In contrast, ARB exposed patients showed an attenuated increase in MAP, from 65 to 71 mmHg at hr₁ in angiotensin-treated patients (difference: 5.3 mmHg [95% CI 1.3–9.3]; p = 0.0143) versus from 66 to 68 mmHg (difference: 2.2 mmHg [95% CI − 1.7–6.1]; p = 0.24) in placebo patients (adjusted difference in treatment-effect versus ACEi/ARB-unexposed: − 6.0 mmHg [95% CI − 11.5 to − 0.6], p_interaction = 0.0299). In summary, there was no difference between ACEi/ARB-unexposed and ACEi-exposed patients in the degree of MAP increase at hr₁ with angiotensin-II versus placebo, but in ARB-exposed patients the effect of angiotensin-II to increase MAP at hr₁ was attenuated.

The average decrease in NED from baseline during hr₀-hr₃ was greater for angiotensin-II compared to placebo among both ACEi/ARB-unexposed (between-group difference: 0.04mcg/kg/min [95% CI 0.02–0.06], p < 0.001) and ACEi-exposed patients (between-group difference: 0.04 mcg/kg/min [95% CI 0.01–0.07], p = 0.012). These differences reflected decreased NED in the treatment group (Fig. 2B). For ARB-exposed patients there was a nominally similar difference for angiotensin-II versus placebo (between-group difference: 0.04 mcg/kg/min [95% CI − 0.01–0.09], p = 0.14). However, in contrast to the ACEi/ARB-unexposed and ACEi-exposed patients, this reflected increased NED in the placebo group rather than decreased NED in the angiotensin-II group. Similar results were obtained when NED was modeled longitudinally (Fig. 2B, Additional file 1: Table-S5/S6). That is, in unexposed patients, angiotensin-II treatment reduced NED over hr₄-hr₄₈ (p = 0.0037). Over the hr₄-hr₄₈ period, in ACEi patients, the effect of angiotensin-II to reduce NED was enhanced compared with ACEi/ARB-unexposed patients (p_interaction = 0.0031). In contrast, this effect was attenuated in ARB patients (p_interaction < 0.0001).

The mean study-drug titration level during hr₀-hr₃ was lower in the angiotensin-II versus placebo groups among ACEi/ARB-unexposed patients (20.7 ng/kg/min versus 86.9 ng/kg/min; difference: to − 66.2 ng/kg/min [95% CI − 74.9 to − 57.5]; p < 0.001). Among ACEi-exposed patients, the angiotensin-II group had a similarly lower study-drug dose versus placebo: 22.1 ng/kg/min versus 65.3 ng/kg/min (difference: − 43.2 ng/kg/min [95% CI − 70.6 to − 15.8]; p = 0.003). In contrast, among ARB-exposed patients, the mean study-drug titration level during hr₀-hr₃ was higher and no different between angiotensin-II and placebo groups: 68.0 ng/kg/min versus 68.6 ng/kg/min (difference: − 0.6 ng/kg/min [95% CI to − 44.9–43.6]; p = 0.98). The longitudinal models produced similar results suggesting heterogeneity of treatment effect for angiotensin-II based on RASi exposure status (Fig. 2C, Additional file 1: Table-S7/S8).

RAS biomarkers according to ACEi and ARB exposure are shown in Fig. 3. At baseline, renin was higher in ACEi-exposed patients (gmean: 335.3 pg/mL) than unexposed patients (gmean: 137.5 pg/mL) (difference: 197.8 pg/mL [41.2–491.7], p = 0.0057) (Fig. 3A). The difference was preserved after multivariable adjustment (adjusted-difference: 605.1 pg/mL [167.0–1394.1], p = 0.0016) (Additional file 1: Table-S9). In contrast, for ARB-exposed patients, renin was numerically higher (gmean: 199.7 pg/mL; difference: 62.2 pg/mL [− 37.7–262.1], p = 0.29). On multivariable adjustment, ARB-exposure was associated with higher baseline renin (adjusted-difference: 394.0 pg/mL [21.0–1113.9], p = 0.0341). Moreover, compared to ACEi/ARB-unexposed patients, ACEi patients had higher baseline angiotensin-I (difference: 412.2 pg/mL [143.2–896.0], p = 0.0002) whereas ARB patients had a nominally higher angiotensin-I (difference: 127.2 pg/mL [− 29.8–435.1], p = 0.14) (Fig. 3B). Conversely, baseline angiotensin-II was lower for ACEi (difference: − 51.5 pg/mL [− 69.2 to − 18.9], p = 0.0068) and higher for ARB (difference: 92.9 pg/mL [0.8–278.4], p = 0.0473) versus ACEi/ARB-unexposed patients (Fig. 3C). Finally, the angiotensin-I/angiotensin-II ratio was higher for ACEi patients (difference: 13.5 arbitrary units [7.1–24.2], p < 0.0001), but similar between ARB and ACEi/ARB-unexposed patients (difference: − 0.4 arbitrary units [− 1.2–1.0], p = 0.51) (Fig. 3D). Group differences in angiotensin-I and angiotensin-II and their ratio were similar to unadjusted results after multivariable adjustment (Additional file 1: Table-S10-S12).

At hr₃, in both unadjusted and adjusted analysis, compared with placebo, angiotensin-II treatment decreased renin levels (p < 0.0001) (Fig. 3E). This effect was seen among ACEi/ARB-unexposed patients (reference group) and ACEi-exposed patients (p_interaction = 0.51 versus unexposed). In contrast, angiotensin-II treatment had no effect on hr₃ renin levels among ARB-exposed patients (p_interaction = 0.0193 versus unexposed) such that renin levels did not decrease at hr₃ (Additional file 1: Table-S13).

To explore whether the association of ARB exposure with attenuated response to angiotensin-II was a dose-dependent or mechanistic phenomenon, we assessed the effect of ARB dose on key outcomes. These results are summarized in Fig. 4D. Among placebo patients ARB dose was not associated with MAP at hr₁ but among angiotensin-treated patients, each log-mg increase in losartan equivalents was associated with a 1.2 mmHg [95% CI − 2.4–0.1] (p_interaction = 0.0717) decrease in MAP (Additional file 1: Table-s14). Increasing losartan equivalent dose was similarly associated with higher NED and study drug dose among angiotensin-treated patients (Fig. 4B, C, (Additional file 1: Table-S15/S16). Higher ARB dose attenuated the decrease in renin at hr₃ associated with angiotensin-treatment (difference-in-difference: 1.4 log(pg/mL)-per-log(mg) [0.02–0.26]; p_interaction = 0.0248) (Fig. 4D) (Additional file 1: Table-S17); i.e., renin decreased less in response to angiotensin-II therapy with increasing dose of ARB exposure.

In this post-hoc analysis of a phase-III placebo-controlled randomized clinical trial of angiotensin-II in patients with catecholamine-refractory vasodilatory shock, prior exposure to ACEi increased the cardiovascular response to angiotensin-II. In contrast, exposure to ARBs reduced such responsiveness. Moreover, after adjusting for baseline illness severity, both ACEi and ARB exposure were associated with higher baseline renin levels, but whereas both ACEi-exposed and ACEi/ARB-unexposed patients showed a renin response to angiotensin-II therapy, angiotensin-II did not reduce plasma renin levels in ARB-exposed patients. Finally, in exploratory dose–response analysis, we observed exposure to higher doses of ARB medications was associated with greater attenuation of the effect of angiotensin-II treatment versus placebo.

Human synthetic angiotensin-II is a relatively new vasopressor, and identifying patients most likely to benefit from therapy is an area of active investigation. ACEi and ARB medications are prescribed to more than 15% of all U.S. adults [17], yet available data on the effect of these medications on the response to angiotensin-II treatment are limited. We note the total prevalence of pre-admission ACEi and ARB receipt in this study was 15.3%, similar to the general population prevalence. Consistent with our findings, the ARAMIS-1 study found that ARB-exposed patients had higher angiotensin-II requirements to achieve target MAP as well as higher baseline renin [18]. This study was limited by small sample size and had too few patients to assess the effect of ACEi exposure. Gleeson et al. 19] did not find an association between prior ACEi or ARB exposure and renin level but only 4 subjects were exposed to a RAS-blocking medication. Secondary analysis of the SEPSISPAM trial showed ARB-exposed, but not ACEi-exposed patients, had lower occurrence of severe AKI with higher MAP targets, which could also be congruent with our findings [20]. To our knowledge, no study has yet assessed whether the dose of ARB exposure is associated with shock outcomes or response to angiotensin-II.

Our observations imply that ACEi and ARB medications have different effects on the baseline RAS profile of patients with catecholamine-refractory vasodilatory shock. As expected, ACEi-exposed patients had greater baseline elevations in renin, angiotensin-I, and angiotensin-I/angiotensin-II ratio, and lower levels of angiotensin-II than unexposed patients. Conversely, ARB-exposed patients had higher baseline angiotensin-II, but no difference in the angiotensin-I/angiotensin-II ratio. These biomarker differences are relevant not only for understanding patient biology in vasodilatory shock, but also because several studies have found strong correlation between plasma renin level and clinical outcomes including mortality [16, 19, 21‐24].

Moreover, our findings imply that ACEi and ARB exposure had opposite, yet physiologically logical (Fig. 1), effects on the cardiovascular and RAS response to angiotensin-II treatment. ACEi recipients were sensitive to angiotensin-II treatment with respect to their cardiovascular response, while ARB patients were not. Furthermore, whereas renin decreased by hr₃ with angiotensin-II treatment versus placebo among both ACEi-exposed and ACEi/ARB-unexposed patients, neither angiotensin-II nor placebo decreased renin levels among ARB-exposed patients. These concordant clinical and biological findings are consistent with the known differences in the mechanisms of these two medication classes. Whereas ACEis inhibit angiotensin-II generation, an upstream step from exogenous angiotensin-II infusion that should not affect treatment response, ARBs antagonize AT1R which would be expected to interfere with the mechanism of action for angiotensin-II infusion. The finding that ARB exposure was dose-dependently associated with angiotensin-II treatment response may loan further credibility to the mechanistic results in this study. Given the widespread use of both medication classes, these results may have direct implications for both clinical use of angiotensin-II and future studies of angiotensin-II in shock.

We acknowledge several limitations. First, only 15% of ATHOS-3 patients had prior exposure to a RAS-blocking therapy. While our findings represent, to our knowledge, the largest study of angiotensin-II therapy in patients with RASi exposure, the sample size is still small. In particular, this analysis was not powered for patient-centered clinical endpoints. Second, as a post-hoc analysis, this study cannot confirm causality, particularly given that the primary exposure of interest was not randomly allocated. Third, fluid balance, filling pressures, and serial echocardiography were not measured, which could impact treatment response. However, this study focused on the early drug-titration period, during which time administered fluid volumes were captured and were not substantively different. Fourth, we did not capture the timing of the last administration date for ACEi or ARB medications, and the retained biological activity of these drugs is an assumption. However, chronic exposure to both medications classes can durably alter the RAS—e.g., telimesartan exposure can downregulate AT1R expression [25], while chronic ACEi exposure can induce an “angiotensin escape” phenomenon, likely in the setting of upregulation of ACE or of non-canonical angiotensin-II generating enzymes [26]. Fifth, the biomarkers assessed in this study primarily reflect the classical (ACE/Angiotensin-II/AT1R) axis of the RAS, but the non-classical arms of the RAS could also influence the interaction of prior RAS-inhibition and the response to angiotensin-II. Sixth, we did not formally adjust for multiple comparisons that accounted for the number of outcomes assessed, which could increase the risk of type-I error. However, this possibility is made less likely by the consistency of results across a variety of outcomes assessed and model specifications.

Prior exposure to RAS-inhibiting medications was associated with an altered RAS profile and cardiovascular response to angiotensin-II treatment in patients with catecholamine-refractory vasodilatory shock. ACE-inhibitor exposure was associated with greater sensitivity to angiotensin-II treatment, whereas ARB exposure was associated with a blunted response to angiotensin-II. These findings may have clinical implications and indicate that these medications cannot be considered equivalent when initiating angiotensin-II treatment in vasodilatory shock.