Septic cardiomyopathy refers to myocardial injury with or without lowered cardiac output in patients with sepsis 12. In contrast to earlier beliefs concerning the frequency of septic cardiomyopathy, a recent prospective trial in 67 adult septic shock patients without previous cardiac disease reported an overall hypokinesia rate (left ventricular ejection fraction of less than 45%) of 60% 3. As compared with patients able to maintain hyperdynamic circulation, survival is significantly compromised in septic shock patients with low systemic blood flow 4. Even if cardiac output can be preserved, myocardial injury as indicated by increased plasma levels of troponin 5 or natriuretic peptides 678 is associated with poor outcome in septic shock.

The etiology of septic cardiomyopathy is multifactorial. Throughout the last decades, several pathogenetic mechanisms, including bacterial toxins, cytokines, nitric oxide, and reactive oxygen species, were identified 29. Recently, the contributory role of adrenergic stress and catecholamine-induced toxicity has been suggested 2. Similarities have been drawn between catecholamine-induced myocardial stunning 1011 and septic cardiomyopathy 12. Sepsis was found to be an important risk factor for development of the left ventricular apical ballooning syndrome 13, originally known as Takotsubo cardiomyopathy 14.

In view of the growing evidence for an association between beta adrenergic stress and the pathogenesis of septic cardiomyopathy 15, the administration of beta-blocking agents could be beneficial. Although at first glance it appears counterproductive to administer a potentially negative inotropic drug to a patient with myocardial depression, beta-blocker therapy improved myocardial oxygen utilization, decreased tumor necrosis factor-alpha production, and preserved cardiac function in a septic animal model 16. Similarly, Gore and Wolfe 17 found that a continuous esmolol infusion reduced heart rate by 20% but did not compromise systemic oxygen delivery or organ blood flow in six hemodynamically stable patients with sepsis.

Apart from these studies, an increasing number of reports have been published suggesting advantageous effects of beta blockers in acute critical illness. Though recently challenged 1819, perioperative beta blockade has repeatedly been shown to reduce cardiac complications and improve survival in high-risk surgery patients 2021. Similarly, preliminary data on the use of beta blockers in critically ill patients with severe trauma 22, traumatic brain injury 23, or burns 24 indicate a beneficial influence on morbidity and mortality.

In an effort to reduce tachycardia in patients with septic shock requiring inotropic therapy, we have cautiously started to use beta blockers. First, this therapeutic intervention was restricted to patients with chronic beta-blocker therapy in order to attenuate rebound tachycardia and decrease the risk of perioperative myocardial ischemia but later was also used in patients without chronic beta-blocker treatment in an attempt to decrease high heart rate and economize cardiac function. This retrospective analysis summarizes our preliminary clinical experience with the combined use of milrinone and enteral metoprolol therapy in 40 patients with septic shock and cardiac depression. Our hypothesis was that metoprolol would reduce heart rate without destabilizing cardiovascular function.

The retrospective protocol was approved by the Ethics Committee of the Krankenhaus der Barmherzigen Schwestern in Ried im Innkreis. In view of the retrospective study design, written informed consent was waived. From 1 January 2005 to 28 February 2008, all medical records of an eight-bed multidisciplinary intensive care unit (ICU) were reviewed for patients with the admission diagnosis of septic shock as defined by the American College of Chest Physicians and the Society of Critical Care Medicine 25. All patients with septic shock and cardiac depression who were treated with enteral metoprolol within 48 hours after the onset of shock or admission to the ICU were included in the analysis. Cardiac depression was defined as a central venous oxygen saturation (ScvO₂) of less than 65% despite adequate fluid resuscitation, oxygenation, and hematocrit, and/or a cardiac index (CI) of less than 2.5 L/minute per m² requiring inotropic therapy. Patients younger than 18 years, patients with any cause of low cardiac output other than sepsis (for example, myocardial ischemia), patients with pre-existent decompensated congestive heart failure (New York Heart Association classification III and IV), patients with septic shock who did not require inotropic support or in whom cardiac output was not measured, and patients who first received beta blockers more than 48 hours after the onset of shock or ICU admission were excluded.

During the review period, 174 patients with septic shock were treated in the ICU. Forty of those patients were treated with a milrinone infusion and received enteral metoprolol during the first 48 hours after shock onset or ICU admission (17.7 ± 15.5 hours) (Table 1). Seven patients died during the observation period. Heart rate was reduced to the targeted range of 65 to 95 bpm in 97.5% of the patients (n = 39) within 12.2 ± 12.4 hours. Of the 36 patients treated with AVP, 31 received AVP before baseline measurements whereas 5 patients were started on AVP therapy during the observation period (within 6 hours, n = 4; within 24 hours, n = 1). While heart rate and central venous pressure significantly decreased, SVI increased during the observation period. At the same time, norepinephrine, AVP, and milrinone dosages were significantly reduced (Table 2). A significant increase in pH as well as decreases in arterial lactate, serum creatinine, and C-reactive protein levels were seen during the observation period (Table 3). Metoprolol therapy was discontinued in two patients because asymptomatic bradycardia occurred and heart rate remained within the lower targeted limits after one and two metoprolol dosages, respectively. The incidence of adverse events during the observation period is presented in Table 4.

After cardiovascular stabilization, heart rate and central venous pressure decreased and SVI increased in this study population during combined milrinone and enteral metoprolol therapy. Simultaneously, vasopressor and inotropic drug support was reduced. Except for an increase in pH as well as decreases in arterial lactate, serum creatinine, and C-reactive protein levels, organ function variables remained unchanged.

Pathophysiologically, septic cardiomyopathy is defined as an inadequately increased cardiac output in relation to the lowered systemic vascular resistance in sepsis and does not necessarily imply that cardiac output is absolutely decreased 12. Indeed, overt cardiac failure as known from patients with cardiogenic shock is rare and has been observed in a maximum of 10% to 15% of septic shock patients 30. More commonly, the clinical picture of septic cardiomyopathy is characterized by a variable degree of myocardial depression which can be detected echocardiographically or biochemically through elevated troponin levels in greater than or equal to 50% 331 and greater than 40% 5, respectively, of sepsis patients. Independently of the presence of overt cardiac failure, the grade of myocardial depression correlates with poor prognosis in sepsis 24. In our analysis, all patients suffered from septic shock with considerably impaired cardiac pump function requiring infusion of an inotropic agent. Even though moderately elevated troponin I serum concentrations in 92.5% of the study patients (n = 37) further underline the presence of septic cardiomyopathy, the lack of echocardiography data limits the detailed investigation of cardiac dysfunction in our analysis.

Despite the growing evidence that beta blockers can be safely and probably beneficially administered in acute critical illness, current use of beta-blocking agents in patients with septic shock and cardiomyopathy must definitely be considered experimental. In an attempt to reduce tachycardia in patients with chronic beta-blocker therapy in whom rebound tachycardia was suspected 32, we have initiated enteral metoprolol therapy for the first time. The results of a study of the favorable effects of perioperative beta blockade 33 then prompted us to employ beta blockers in patients without previous chronic beta-blocker therapy. Nonetheless, in all patients, this was done compassionately and as considered indicated by the treating physician, starting cautiously with low metoprolol dosages and under tight control of cardiac output and ScvO₂.

A selective beta-1, instead of a non-selective, beta blocker was chosen to prevent inhibition of potentially beneficial beta-2 effects 34. Since in our clinical experience esmolol infusion was associated with frequent and rapid decreases in heart rate and cardiac output, metoprolol was applied via the enteral route. Accordingly, Gore and Wolfe 17 observed a 20% decrease in CI during esmolol infusion in hemodynamically stable sepsis patients. Since heart rate is a major determinant of myocardial oxygen consumption 35, it was used to dose metoprolol. In cardiovascular high-risk patients, a heart rate of 95 bpm was shown to be the critical threshold at which myocardial oxygen demand outstripped coronary supply and myocardial ischemia was likely to occur 3637.

Instead of beta-agonists, a phosphodiesterase III inhibitor was applied as an inotropic agent in all of our patients. Although this does not correspond to current recommendations 38, milrinone has been used in patients with septic shock at our institution throughout the last decade. Positive inotropic effects of milrinone are mediated through inhibition of the breakdown of cAMP by phosphodiesterases 39 and act independently of beta-1 receptors. In view of differences in cAMP-independent actions 40 and compartmentation of cAMP-mediated signaling 41, the combination of milrinone and metoprolol may hold potential benefits for myocardial function 39.

A decrease in heart rate together with an increase in SVI given an unchanged CI can be interpreted as an economization of cardiac work and oxygen consumption. Reduced heart rates lower the risk of myocardial ischemia 213637, particularly in patients with obstructive coronary artery disease 42. In light of diminishing milrinone support, these observations may even reflect improved cardiac pump function in our study patients. Moreover, a decrease in central venous pressure as observed during metoprolol therapy often follows amelioration of myocardial performance 28. The observation that organ function variables remained unchanged during beta-blocker therapy strengthens the assumption that metoprolol therapy did not reduce systemic blood flow or limit organ oxygen supply. Similarly, organ blood flow (extremity and hepatic blood flow) was not overtly affected during esmolol infusion in six hemodynamically stable sepsis patients 17. Since metoprolol therapy was commenced 17.7 ± 15.5 hours after the onset of shock and initiation of standard hemodynamic therapy, it is unlikely that cardiovascular changes simply resulted from fluid therapy, vasopressor, or milrinone infusion. However, because of the uncontrolled design, our study cannot prove a causative relationship between the observed hemodynamic changes and metoprolol therapy.

In view of the preference of milrinone over dobutamine and the frequent use of a supplementary AVP infusion, the hemodynamic effect during enteral metoprolol therapy can be interpreted only in the context of our institutional hemodynamic protocol. For example, infusion of AVP in 95.6% of the study patients could have interfered with the hemodynamic effects of metoprolol therapy. Likewise, a decrease in heart rate as well as a mild increase in CI was reported during supplementary AVP infusion in patients with advanced vasodilatory shock and hypodynamic circulation 43. Although in most patients (89%) AVP was started before metoprolol therapy, we cannot determine the extent to which this influenced the hemodynamic course during the observation period.

Response to metoprolol in this study population was not entirely homogeneous. Whereas overall MAP and CI did not decrease, nine and seven patients exhibited a decrease in MAP and CI, respectively, requiring an increase in norepinephrine or milrinone dosages during the observation period. It cannot be proven that the observed changes reflect beneficial or adverse effects of metoprolol. It is conceivable that without beta-blocker therapy SVI would have increased even more and milrinone infusion could have been withdrawn earlier. Similarly, the decrease in MAP and CI in some patients may have resulted from the course of the underlying disease process instead of being related to metoprolol therapy.

Data from experimental and clinical studies suggest that several beta-blocker effects such as heart rate control 44, antagonization of catecholamine-induced stunning of the myocardium 114546, and reduction of myocardial inflammation 4748 may be beneficial in patients with septic myocardial depression. Although data are still conflicting 49, beneficial effects of beta blockers were reported to also include attenuation of an overshooting immune response. Adult trauma patients treated with a continuous beta-blocker infusion exhibited lower serum interleukin-6 levels than did controls receiving standard of care 22. Reduced proinflammatory cytokine production was also illustrated during esmolol infusion in rats with septic cardiomyopathy 16. Interestingly, serum C-reactive protein levels decreased during metoprolol therapy in our study. Although it could be hypothesized that metoprolol reduced interleukin-6 levels and thus C-reactive protein levels 50, many other factors such as focus control 38, adequate antibiotic therapy 38, and hemodynamic stabilization 51 are likely to have caused the decrease of C-reactive protein levels in our analysis.

In addition to the uncontrolled study design, other important limitations need to be noted when interpreting the results of our analysis. First, this is a retrospective study and it entails potential difficulties because of missing values in individual patients. Furthermore, patient enrolment was not performed according to a strict protocol as in a prospective trial but at the discretion of the attending physician. Therefore, some patients who would have been eligible for metoprolol therapy according to our treatment scheme may have been missed. Second, although a hemodynamic protocol that served as a recommendation for resuscitation of septic shock patients was available, we cannot be sure whether the attending physicians strictly adhered to the protocol during resuscitation of all study patients. Third, a population of 40 patients is too small for adequately evaluating the safety profile of metoprolol therapy in patients with septic shock and cardiac depression.

Low doses of enteral metoprolol in combination with phosphodiesterase inhibitors are feasible in patients with septic shock and cardiac depression but no overt heart failure. Future prospective controlled trials on the use of beta blockers for septic cardiomyopathy and their influence on proinflammatory cytokines are warranted.

• Heart rate significantly decreased during combined milrinone infusion and enteral metoprolol therapy in patients with septic shock and cardiac depression.

• In 97.5% of patients, targeted heart rates of 65 to 95 beats per minute were achieved.

• Enteral metoprolol therapy appears to have no major adverse effects on cardiovascular or organ function.

• Mean arterial blood pressure increased despite decreasing norepinephrine, arginine vasopressin, and milrinone dosages.

• Cardiac function economized, resulting in a maintained cardiac index with a lower heart rate and a higher stroke volume index.

AVP = arginine vasopressin; bpm = beats per minute; CI = cardiac index; CPI = cardiac power index; ICU = intensive care unit; MAP = mean arterial blood pressure; ScvO₂ = central venous oxygen saturation; SVI = stroke volume index.

The authors declare that they have no competing interests.

CAS made substantial contributions to the acquisition, analysis, and interpretation of data and was involved in drafting the manuscript. MWD made substantial contributions to the concept and design of the study and the acquisition, analysis, and interpretation of data and was involved in drafting the manuscript. MH, GL, CT, and SJ made substantial contributions to the acquisition of data and critically revised the manuscript for important intellectual content. HU performed the statistical analysis and critically revised the manuscript for important intellectual content. WRH made substantial contributions to the concept and design of the study and was involved in drafting the manuscript. All authors read and approved the final manuscript.