Septic shock, the most severe form of sepsis, consists in a dysregulated host inflammatory response to infection, including an early systemic and regional vasodilatation leading to a decrease in arterial blood pressure and the generation of organ failures (1, 2). The mean arterial pressure (MAP) is an important determinant of organ perfusion and is commonly monitored in septic shock patients. Therefore, early optimization of arterial blood pressure and its maintenance using fluid administration and vasopressors is a major goal of resuscitation therapy (3).
Among patients with chronic medical condition such as heart failure and hypertension, use of cardiovascular antihypertensive medication modifying the vascular tone (including beta-blockers (BB), angiotensin-converting enzyme inhibitor (ACEi)/angiotensin receptor blockers (ARB), calcium channel blockers (CCB), and thiazide diuretic) is common and may affect systemic vascular resistances during sepsis. Therefore, previous administration of cardiovascular antihypertensive medication in patients with septic shock is often viewed as a risk factor for increased catecholamine requirement (4–6). However, most data derive from non-septic patients and remain questionable (7, 8).
On the other hand, given their pleiotropic properties, some data suggest that these molecules might exert some beneficial effects through immunomodulatory properties. Adjunctive therapy with BB, ACEi/ARB, and CCB in experimental models of sepsis have demonstrated significant anti-inflammatory effects, reducing systemic cytokine levels and restoring endothelial barrier function (9–12). However, human studies evaluating the impact of cardiovascular antihypertensive medications on outcomes in septic shock patients have given conflicting results (6, 13–16).
Thus, the aim of this study is to investigate whether prior use of cardiovascular antihypertensive medications is associated with outcomes of septic shock patients, with a particular emphasis on early catecholamine requirements.
PATIENTS AND METHODS
Patients and setting
We performed a retrospective single-center study in a 24-bed tertiary medical intensive care unit (ICU). All adult patients (age ≥18 years old) diagnosed with septic shock within the first 48 h of ICU admission were included. Septic shock was defined as a microbiologically proven or clinically suspected infection, associated with acute circulatory failure requiring vasopressors despite adequate fluid filling (17). Initiation of vasopressors was considered the onset of septic shock. Exclusion criteria included patients admitted for non-septic cause of shock or shock from multiple origin, and patients not receiving vasopressor within the first 24 h after ICU admission. Data from patients including demographics, comorbidities, chronic medication use, ICU admission characteristics (including the Simplified Acute Physiology Score (SAPS) II score), daily physiologic measurements, severity scores (including Sequential Organ Failure Assessment (SOFA) score), cumulative vasopressor dose over the first 24 h, concomitant use of inotropes, concomitant use of steroids, and biological findings on admission were computed from medical files and extracted from the patient data management system (Clinisoft, GE Healthcare). The ethics committee of the French Intensive Care Society approved the study and waived the need for patients’ consents due to its retrospective observational design (ref. CE SRLF, #16–30).
Prior cardiovascular antihypertensive medication
We collected the chronic use of medication groups with antihypertensive pharmacological properties including ACEi/ARB, CCB, BB, thiazides, and spironolactone, therefore often used in the treatment of hypertension though they often have alternative indications in non-hypertensive patients (18). Loop diuretics were not considered cardiovascular antihypertensive medication because of their relatively modest pressure-lowering effect (19). For subgroup analyses spironolactone, ACEi and ARB were clustered as Renin-Angiotensin system blockers (RASB) because they shared common pharmacological activity. To assess the association of such medication and outcomes, we included every patient receiving cardiovascular antihypertensive medication and not only those treated for arterial hypertension. Unless reported in admission medical reports, patients were assumed to be compliant.
Patients were resuscitated according to the Surviving Sepsis Campaign guidelines (20). We used norepinephrine as first-line vasopressor. Dobutamine was considered an inotropic agent for patients with echocardiographic evidence of acute cardiac failure. The use of hydrocortisone therapy was left to physician appraisal. Patients with or without cardiovascular antihypertensive medications were resuscitated with initial objectives of MAP of 65 mm Hg for the first 24 h, regardless of underlying comorbid conditions (21). Higher MAP target to 75 to 80 mm Hg was therefore considered in previously hypertensive patients with persistent renal failure, after careful hemodynamic assessment of volemia and cardiac output.
The primary outcome was cumulative averaged and bodyweight-adjusted norepinephrine dose at 24 h. Secondary outcomes included ICU mortality, day-28 mortality, death occurring within the first 72 h following admission (defined as early death), time under vasopressors and acute kidney injury (AKI) on admission and occurring within the first 7 days of ICU stay, stratified on AKIN criteria (22).
Continuous variables were expressed as median (interquartile range) and categorical variables as numbers (percentages) and were compared according prior cardiovascular antihypertensive therapy by the Kruskal–Wallis test, the Pearson chi-square test, or the Fisher exact test as appropriate.
Because two competing events were observed at discharge from ICU (alive or death), a cause-specific proportional hazard model was used for survival analysis (23). The proportionality assumptions were tested by graphical checks. The multivariate model was built by including all risk factors for death that met the 0.20 significance threshold in univariate analysis. The final model resulted from a stepwise variable selection process using the Akaike criteria. Potential interactions were tested in the final model. The results were given as cause-specific hazards (CSH) and 95% confidence intervals.
We then constructed a propensity score, using logistic regression, including variables associated with use of each type of prior cardiovascular antihypertensive therapy (Beta blocker, calcium blocker, and RASB). Each score included age, gender, weight, comorbid condition, cardiovascular antihypertensive therapy given prior admission, SAPS 2, and source of infection. Subjects were 1:1 matched without replacement by the estimated propensity score using nearest-neighbor matching with a caliper of 0.2 SD of the logit of the propensity score. Standardized differences were determined to ascertain balance between the propensity-matched groups. Competing risk analysis in matched cohort to assess primary outcome. Analyses were carried out with the R 3.1.1 statistical software (R foundation for Statistical Computing Vienna, Austria). All tests were two-sided and P values below 0.05 indicated statistical significance.
From 2008 to 2016, 735 patients were admitted for septic shock in our ICU (see e-Figure 1 for Flow-chart description, http://links.lww.com/SHK/B37). Overall, most patients were male (63.3%) with a median age of 69 years (IQR: 58–79). A large proportion of patients had previous chronic comorbid conditions, with 21% of chronic heart failure, 34.8% of arterial hypertension, and 12% of chronic kidney disease. 36.9% of patients were immunocompromised, with a vast majority of solid cancers. Lung was the main source of primary infection (48.1%). Among all patients, 345 (47%) were previously treated with cardiovascular antihypertensive medication (Table 1). Among patient receiving cardiovascular antihypertensive medications, 23% were treated with monotherapy, 16% with biotherapy, and 8% received three or more medications (e-table 1, http://links.lww.com/SHK/B38). Patients received BB (194 patients), ACEi/ARBs inhibitors (192 patients), CCB (103 patients), thiazide diuretics (48 patients), and spironolactone (40 patients). Among patients receiving cardiovascular antihypertensive medications, 224 patients (73.6%) were primarily treated for primary arterial hypertension, 9.9% (34 patients) for cardiac arrythmia, 8.1% (28 patients) for ischemic cardiac failure, 8.1% (28 patients) for cirrhosis, and 1.7% (6 patients) had other indications. On the other side, we report only two patients suffering from primary hypertension who did not receive any specific medication.
When compared with patients without prior cardiovascular antihypertensive medication, those who were treated had higher median age (75 years vs. 63 years old; P < 0.001) and a higher prevalence of cardiovascular risk factors, including diabetes (24.1% vs. 13.3%, P < 0.001), obesity (11.6% vs. 4.6%, P = 0.001), and chronic kidney disease (16.8% vs. 8.2%, P < 0.001). They also had a higher incidence of chronic heart failure (30.7% vs. 12.3%, P < 0.0001). Whether the SOFA score at admission was similar in both groups (9 vs. 9, P = 0.419), the SAPS II at admission was higher in patients previously treated with cardiovascular antihypertensive medication (72 vs. 67, P < 0.001). Heart rate on admission was lower in the cardiovascular antihypertensive medication group compared with others (101 bpm vs. 110 bpm, P < 0.001). Median lactate level at admission was similar in both groups (2. mmol/L vs. 2 mmol/L, P = 0.086). Most patients were mechanically ventilated, with no difference between groups.
Cumulative catecholamine requirement and acute kidney injury
After adjustment for bodyweight, norepinephrine requirements within the first 24 h of shock were similar between patients previously treated with cardiovascular antihypertensive medication or not (0.28 μg/kg/min (0.13–0.74) vs. 0.26 μg/kg/min (0.10–0.63), P = 0.18 (Table 1)). Time under vasopressors was no different between groups (2 days (1–4) vs. 2 days (1–4), P = 0.82). Cumulative cardiovascular antihypertensive medication prescription was not associated with increased initial requirement during the first 24 h of shock (P = 0.55) (Fig. 1). The incidence of acute cardiogenic shock requiring inotropes during the first 24 h was not different in patients treated with prior cardiovascular antihypertensive medication when compared with those who were not.
Analysis of mortality
Overall mortality rate at day 28 was 34.2% in patients receiving prior cardiovascular antihypertensive medication and 27.7% in patients not previously treated (P = 0.06). Early death occurred in 18.0% and 15.1% of cases respectively, with no difference between groups (P = 0.34). Using univariate Cox cause-specific model, prior cardiovascular antihypertensive medication was significantly associated with increased ICU mortality (CSH = 1.34, IC95% [1.04–1.72], P = 0.02) (Table 2). However, previous cardiovascular antihypertensive medication did not remain significantly associated with a higher risk of ICU mortality in a multivariate-adjusted cause-specific proportional hazard model (CSH = 1.22 [0.84–1.76], P = 0.30 for each additional medication).
Given the differences between users and non-users of cardiovascular antihypertensive medication at baseline, we conducted a propensity-matched analysis to correct for those dissimilarities. 167/194 (86.1%) patients receiving BB, 166/192 (86.5%) patients receiving RASB, and 90/103 (87.4%) of those receiving CCB could have a matched control (e-table 2–4, http://links.lww.com/SHK/B38). Results for initial catecholamine requirements and ICU mortality analyses were similar to those previously reported within the global cohort (e-Figure 2, http://links.lww.com/SHK/B37).
We herein reported that almost half of patients with septic shock admitted in our ICU were previously treated with cardiovascular antihypertensive medication. After adjustment with potential confounders using multivariate competitive risk analysis, prior cardiovascular antihypertensive medication was neither associated with higher norepinephrine requirement during the first 24 h nor with increased mortality.
In current clinical practice, it is widely admitted that chronic cardiovascular antihypertensive medications might worsen hemodynamic failure in septic shock patients (24). However, we found only one study investigating this belief in such patients (6). Using univariate analysis, this study retrieved no difference between patients with prior-to-admission BB and/or ACEi/ARBs and untreated patients in terms of norepinephrine requirement (6). In line with our results, these data call for a reappraisal of the current paradigm, most data on the field originating from the operative field among non-septic patients which might represent different pathophysiological mechanisms (25). Moreover, several pathways are involved in blood pressure control in septic patients and therefore could generate variability in results.
In recent years, two major components of the cardiovascular system have received considerable attention in septic shock: the β-adrenergic and the renin–angiotensin systems. Their modulation using specific agonists or antagonists has provided evidence of hemodynamic effects and therefore shed light on their importance in this setting. On the one hand, short-lasting blockade of the β-adrenergic system using a continuous infusion of esmolol to maintain heart rate in septic shock patients was associated with decreased norepinephrine requirements through an improvement of diastolic function (13). On the other hand, stimulation of the renin–angiotensin system using angiotensin II allowed significantly decreasing mean doses of background vasopressors in vasodilatory shock, including almost 81% of septic shock (26). Whether these data highlight the complexity of pathophysiological relations of these systems, they also point out the importance of investigating effects of prior cardiovascular antihypertensive medication on vasopressors requirement in treatment of septic shock.
Regarding mortality, our results are not in line with previously published data and most of studies reporting decreased mortality among sepsis patients previously treated with BB (14), CCB (27) and ACEi/ARB (16, 28). Several assumptions might be made to explore this difference. First, previously reported studies have focused on sepsis, therefore including patients with different levels of severity, whereas we included septic shock patients only. In line with this hypothesis, a study evaluating the impact of previous prescription of BB and ACEi/ARB in a pure population of septic shock patients found comparable mortality in patients with chronic exposure or not to those medications (6). Second, we used a competing risk regression model whereas other studies used non-competing risk approach. Indeed, septic patients are subject to the occurrence of several events that might modify the risk of death. Therefore, use of competing risk analysis decreases the risk of bias and might confer more robustness to our results. Third, only a few studies investigated the incidence of coprescription of cardiovascular antihypertensive medications, which might account for this discrepancy. For instance, in our study, almost 51% of patients receiving cardiovascular antihypertensive medication were treated with at least two molecules. Fourth, potential beneficial effects of cardiovascular antihypertensive medications may have been blunted by vasoactive therapy withdrawal at admission. In this perspective, brutal cessation of BB has been associated with increased in-hospital mortality among patients hospitalized for acute on chronic heart failure (29) and a recent study suggested that continuation of BB might be beneficial in septic shock patients (30).
Previously published data reported beneficial effect on mortality of different cardiovascular antihypertensive therapies despite distinct pharmacological pathways. This statement might suggest that septic patients primarily benefit from a controlled blood pressure in arterial hypertension context rather than specific effects of drugs. However, we retrieved no study investigating this hypothesis. Furthermore, previously reported data failed to relate observed beneficial effects of prior cardiovascular antihypertensive medication to direct effects of the drug. For instance, in a recent study reporting reduction in mortality among septic patients previously treated with CCB, authors postulated this effect could be due to inhibition of the massive influx of extracellular calcium into cells and immunomodulation (27). However, this study reported no difference in plasma levels of various inflammatory and calcium pathways biomarkers. In the same way, despite experimental evidence of angiotensin II cytotoxicity (12, 31), several studies found no association between prior SRAB treatment and sepsis-related mortality (16, 32, 33). Thus, future studies should investigate how protracted control of blood pressure in arterial hypertension context may modulate the outcome of septic shock patients.
Taken together, such observations could also suggest that the observed benefit might rely on the underlying condition more than the molecule itself. In this perspective, several animal studies suggested a protecting effect of underlying hypertension in rodent models of sepsis (34, 35). The suggested hypothesis would be a decreased inflammatory response and a greater vasoreactivity. We found no study investigating this specific hypothesis among humans, in our cohort, a subgroup analysis targeting hypertensive patients only found no difference regarding the main outcomes (e-table 5, http://links.lww.com/SHK/B38).
This study acknowledges several limitations. First, this was a retrospective study, performed in a single center with a particular case mix of patients. Such a study can only identify associations but not causality links between variables. Second, we cannot determine the compliance of patients to their treatment. However, reduced heart rate observed in the cardiovascular antihypertensive medication group argues in favor of active β-adrenergic system blockade. Third, we analyzed the outcomes of all hypertensive and non-hypertensive patients receiving cardiovascular antihypertensive medications irrespective of their primary indications. This is in contrast with previous studies in the field that solely focused on patients with arterial hypertension. At last, norepinephrine requirements only focused on the first 24 h of shock and not the total time under vasopressor.
Prior cardiovascular antihypertensive therapy has limited impact on the presentation of acute circulatory failure, vasopressor requirements, and mortality in patients with septic shock. Moreover, our observations do not support a higher risk of acute kidney injury in septic patients treated with ACEi/ARB.
1. Burgdorff A-M, Bucher M, Schumann J. Vasoplegia in patients with sepsis and septic shock
: pathways and mechanisms. J Int Med Res
46 (4):1303–1310, 2018.
2. Kimmoun A, Ducrocq N, Levy B. Mechanisms of vascular hyporesponsiveness in septic shock
. Curr Vasc Pharmacol
11 (2):139–149, 2013.
3. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, Bellomo R, Bernard G, Chiche JD, Coopersmith C, et al. The Third International Consensus Definitions for Sepsis and Septic Shock
315 (8):801–810, 2016.
4. Boccara G, Ouattara A, Godet G, Dufresne E, Bertrand M, Riou B, Coriat P. Terlipressin versus norepinephrine to correct refractory arterial hypotension after general anesthesia in patients chronically treated with renin-angiotensin system inhibitors. Anesthesiology
98 (6):1338–1344, 2003.
5. Boerma EC, Singer M. Beta-blockers in sepsis: time to reconsider current constraints? Br J Anaesth
119 (4):560–561, 2017.
6. DeMott JM, Patel G, Lat I. Effects of chronic cardiovasculars on vasopressor dosing in septic shock
. Ann Pharmacother
52 (1):40–47, 2018.
7. Coriat P, Richer C, Douraki T, Gomez C, Hendricks K, Giudicelli JF, Viars P. Influence of chronic angiotensin-converting enzyme inhibition on anesthetic induction. Anesthesiology
81 (2):299–307, 1994.
8. Brabant SM, Eyraud D, Bertrand M, Coriat P. Refractory hypotension after induction of anesthesia in a patient chronically treated with angiotensin receptor antagonists. Anesth Analg
89 (4):887–888, 1999.
9. Rudiger A, Singer M. Decatecholaminisation during sepsis. Crit Care
20 (1):309, 2016.
10. Gullestad L, Aukrust P, Ueland T, Espevik T, Yee G, Vagelos R, Frøland S, Fowler M. Effect of high- versus low-dose angiotensin converting enzyme inhibition on cytokine levels in chronic heart failure. J Am Coll Cardiol
34 (7):2061–2067, 1999.
11. Hotchkiss RS, Karl IE. Calcium: a regulator of the inflammatory response in endotoxemia and sepsis. New Horiz
4 (1):58–71, 1996.
12. Wiel E, Pu Q, Leclerc J, Corseaux D, Bordet R, Lund N, Jude B, Vallet B. Effects of the angiotensin-converting enzyme inhibitor perindopril on endothelial injury and hemostasis in rabbit endotoxic shock. Intensive Care Med
30 (8):1652–1659, 2004.
13. Morelli A, Ertmer C, Westphal M, Rehberg S, Kampmeier T, Ligges S, Orecchioni A, D’Egidio A, D’Ippoliti F, Raffone C, et al. Effect of heart rate control with esmolol on hemodynamic and clinical outcomes in patients with septic shock
: a randomized clinical trial. JAMA
310 (16):1683–1691, 2013.
14. Macchia A, Romero M, Comignani PD, Mariani J, D’Ettorre A, Prini N, Santopinto M, Tognoni G. Previous prescription of β-blockers is associated with reduced mortality among patients hospitalized in intensive care units for sepsis. Crit Care Med
40 (10):2768–2772, 2012.
15. Ince C. To beta block or not to beta block; that is the question. Crit Care
19 (1):339, 2015.
16. Kim J, Kim Y, Hwangbo B, Kim M, Cho H, Hwangbo Y, Lee E. Effect of cardiovascular medications on sepsis-related outcomes. Crit Care Med
47 (5):e386–e393, 2019.
17. Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, Cohen J, Opal S, Vincent JL, Ramsay G. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Intensive Care Med
29 (4):530–538, 2003.
18. Williams B, Mancia G, Spiering W, Agabiti Rosei E, Azizi M, Burnier M, Clement D, Coca A, de Simone G, Dominiczak A, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J
39 (33):3021–3104, 2018.
19. Musini VM, Rezapour P, Wright JM, Bassett K, Jauca CD. Blood pressure-lowering efficacy of loop diuretics for primary hypertension. Cochrane Hypertension Group, editor. Cochrane Database Syst Rev
2015; 5 (CD003825):
20. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M, Ferrer R, Kumar A, Sevransky J, Sprung C, Nunnally N, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock
: 2016. Intensive Care Med
43 (3):304–377, 2017.
21. Asfar P, Meziani F, Hamel J-F, Grelon F, Megarbane B, Anguel N, Mira JP, Dequin PF, Gergaud S, Weiss N, et al. High versus low blood-pressure target in patients with septic shock
. N Engl J Med
370 (17):1583–1593, 2014.
22. Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care Lond Engl
11 (2):R31, 2007.
23. Latouche A, Allignol A, Beyersmann J, Labopin M, Fine JP. A competing risks analysis should report results on all cause-specific hazards and cumulative incidence functions. J Clin Epidemiol
66 (6):648–653, 2013.
24. Boulain T, Garot D, Vignon P, Lascarrou J-B, Desachy A, Botoc V, Follin A, Frat JP, Bellec F, Quenot JP, et al. Prevalence of low central venous oxygen saturation in the first hours of intensive care unit admission and associated mortality in septic shock
patients: a prospective multicentre study. Crit Care
18 (6):609, 2014.
25. Magder S. The meaning of blood pressure. Crit Care
22 (1):257, 2018.
26. Khanna A, English SW, Wang XS, Ham K, Tumlin J, Szerlip H, Busse L, Altaweel L, Albertson T, Mackey C, et al. Angiotensin II for the treatment of vasodilatory shock. N Engl J Med
377 (5):419–430, 2017.
27. Wiewel MA, van Vught LA, Scicluna BP, Hoogendijk AJ, Frencken JF, Zwinderman AH, Horn J, Cremer O, Bonten M, Schultz M, et al. Prior use of calcium channel blockers is associated with decreased mortality in critically ill patients with sepsis: a prospective observational study. Crit Care Med
45 (3):454–463, 2017.
28. Mortensen EM, Nakashima B, Cornell J, Copeland LA, Pugh MJ, Anzueto A, Good C, Restrepo M, Downs J, Frei C, et al. Population-based study of statins, angiotensin II receptor blockers, and angiotensin-converting enzyme inhibitors on pneumonia-related outcomes. Clin Infect Dis
55 (11):1466–1473, 2012.
29. Miró Ò, Müller C, Martín-Sánchez FJ, Bueno H, Mebazaa A, Herrero P, Jacob J, Gil V, Escoda R, Llorens P. BETAWIN-AHF study: effect of beta-blocker withdrawal during acute decompensation in patients with chronic heart failure. Clin Res Cardiol
105 (12):1021–1029, 2016.
30. Fuchs C, Wauschkuhn S, Scheer C, Vollmer M, Meissner K, Kuhn S-O, Hahnenkamp K, Morelli A, Gründling M, Rehberg S. Continuing chronic beta-blockade in the acute phase of severe sepsis and septic shock
is associated with decreased mortality rates up to 90 days. Br J Anaesth
119 (4):616–625, 2017.
31. Kimura S, Zhang G-X, Nishiyama A, Shokoji T, Yao L, Fan Y-Y, Rahman M, Abe Y. Mitochondria-derived reactive oxygen species and vascular MAP kinases: comparison of angiotensin II and diazoxide. Hypertens
45 (3):438–444, 2005.
32. Suberviola B, Rodrigo E, González-Castro A, Serrano M, Heras M, Castellanos-Ortega Á. Association between exposure to angiotensin-converting enzyme inhibitors and angiotensin receptor blockers prior to septic shock
and acute kidney injury. Med Intensiva
41 (1):21–27, 2017.
33. Wang AY, Bellomo R, Ninomiya T, Lo S, Cass A, Jardine M, Gallagher M. Angiotensin-converting enzyme inhibitor usage and acute kidney injury: a secondary analysis of RENAL study outcomes. Nephrology (Carlton)
19 (10):617–622, 2014.
34. Bernard C, Merval R, Esposito B, Tedgui A. Resistance to endotoxin shock in spontaneously hypertensive rats. Hypertens
31 (6):1350–1356, 1998.
35. Nunes JPL. Effects of lipopolysaccharide on vascular reactivity and mortality in rats. Auton Autacoid Pharmacol
22 (5–6):247–252, 2002.