Abstract
Aims
Studies in cardiogenic shock (CS) often have a heterogeneous population of patients, including those with acute myocardial infarction and acute decompensated heart failure (ADHF‐CS). The therapeutic profile of milrinone may benefit patients with ADHF‐CS. We compared the outcomes and haemodynamic trends in ADHF‐CS receiving either milrinone or dobutamine.
Methods and results
Patients presenting with ADHF‐CS (from 2014 to 2020) treated with a single inodilator (milrinone or dobutamine) were included in this study. Clinical characteristics, outcomes, and haemodynamic parameters were collected. The primary endpoint was 30 day mortality, with censoring at the time of transplant or left ventricular assist device implantation. A total of 573 patients were included, of which 366 (63.9%) received milrinone and 207 (36.1%) received dobutamine. Patients receiving milrinone were younger, had better kidney function, and lower lactate at admission. In addition, patients receiving milrinone received mechanical ventilation or vasopressors less frequently, whereas a pulmonary artery catheter was more frequently used. Milrinone use was associated with a lower adjusted risk of 30 day mortality (hazard ratio = 0.52, 95% confidence interval 0.35–0.77). After propensity‐matching, the use of milrinone remained associated with a lower mortality (hazard ratio = 0.51, 95% confidence interval 0.27–0.96). These findings were associated with improved pulmonary artery compliance, stroke volume, and right ventricular stroke work index.
Conclusions
The use of milrinone compared with dobutamine in patients with ADHF‐CS is associated with lower 30 day mortality and improved haemodynamics. These findings warrant further study in future randomized controlled trials.
Keywords: Milrinone, Dobutamine, Cardiogenic shock, Inotropes, Heart failure
Introduction
Dobutamine and milrinone are the two of the most widely used intravenous inodilators for the treatment of acute heart failure (HF) and cardiogenic shock (CS). Although milrinone improves symptoms in patients with end‐stage HF, it has also been associated with increased mortality in patients with acute HF and in patients with ambulatory New York Heart Association (NYHA) class III‐IV symptoms. 1 , 2 There is less robust data reported for dobutamine, and these data originate mostly from a non‐CS acute decompensated heart failure (ADHF) population. 3 , 4
CS is the most severe clinical presentation of patients admitted to the hospital with ADHF. Despite advances in medical therapy and the use of mechanical circulatory support, mortality rates have not improved and remain high. 5 Clinical practice guidelines propose that norepinephrine and inodilator agents be used to maintain end‐organ perfusion in CS patients. 6 , 7 However, there is some evidence to suggest more deleterious effects of vasopressors compared with inodilators in CS. 8 Both milrinone and dobutamine are cornerstone therapies in the management of CS, although data are limited with respect to clinical outcomes. 7 , 9 , 10 While observational studies point towards a trend of lower mortality with milrinone when compared with dobutamine, these data were based on CS populations from older eras. 11
Recently, the DOREMI trial showed that in a head‐to‐head comparison of a heterogeneous population of patients with CS, there was no significant difference in mortality, with limited haemodynamic data available. 12 The purpose of this study was to compare the effects of milrinone and dobutamine on clinical outcomes and haemodynamic parameters in a contemporary cohort of patients admitted with ADHF‐CS.
Methods
Study design
All adult patients admitted with CS to the cardiac intensive care unit (CICU) from 1 January 2014 to 31 December 2020, at the Peter Munk Cardiac Centre, Toronto General Hospital, University Health Network were identified from a retrospective database (n = 1070). CS was defined as per the SHOCK trial, and its severity was graded according to the modified Society for Cardiovascular Angiography and Interventions criteria based on the number of interventions required. 13 , 14 Patients excluded from the analysis included those who previously had received a heart transplant (n = 1), those who did not receive inodilators during their stay (n = 349), and those who received both milrinone and dobutamine during admission (n = 71). Acute myocardial infarction‐CS (AMI‐CS) was defined as any acute coronary syndrome leading to CS. Patients with ADHF‐CS were identified as those not meeting the criteria for AMI‐CS.
Baseline clinical characteristics, laboratory data (obtained within 24 h of admission to the CICU), pulmonary artery catheter (PAC) values, and receipt of orthotopic heart transplantation (OHT) or left ventricular assist device (LVAD) were collected. The primary outcome was all‐cause mortality at 30 days. The secondary outcome was whether patients received OHT or LVAD during their admission. Other clinical endpoints that were explored included the length of invasive mechanical ventilation (IMV), and the length of CICU and hospital stay. This study was approved by the institutional Research Ethics Board review (CAPCP study #16‐5476). The investigation conforms with the principles outlined in the Declaration of Helsinki.
PACs were inserted at the discretion of the attending physician. Variables from the PAC were collected at the time of its insertion and at 24, 48, and 72 h post‐insertion (when available). Cardiac output (CO) was estimated using thermodilution, and stroke volume was obtained as CO/heart rate. To assess right ventricular (RV) performance, right ventricular stroke work index (RVSWI, g/m/beat/m2) was calculated using the formula RVSWI = stroke volume index * (mean pulmonary artery pressure (PAP) − central venous pressure) * 0.0136. Pulmonary vascular resistance (PVR) was defined as (mean PAP − pulmonary capillary wedge pressure (PCWP))/CO, and pulmonary artery compliance (CPA) as systolic volume/(systolic PAP − diastolic PAP).
Statistical analysis
The distribution of continuous variables was assessed with a Shapiro–Wilk test. Differences in baseline characteristics and variables at admission were tested with the Mann–Whitney U test, and median and interquartile range was provided. Categorical variables are presented as proportions (percentage), and differences were assessed with the χ 2 test.
The effect of milrinone on 30 day mortality was assessed using a multivariate Cox regression adjusted for clinically relevant variables and those showing significant differences between groups, and the hazard ratio (HR) with its 95% confidence interval (CI) is provided. Patients were censored at 30 days or at the time of OHT or durable LVAD implant, and the results are shown in Kaplan–Meier plots with significance provided by the log‐rank test. The proportionality of hazards was tested with a log–log plot. We also evaluated the effect of each inotrope in a combined endpoint of death, LVAD implantation, or HT within 30 days.
To confirm the robustness of our findings, several multivariate models were developed using covariables at admission and the addition of interventions performed during CICU admission. In a separate analysis, we used a logistic regression to find the admission variables associated with receiving milrinone and estimated the predicted probability of getting the treatment (Table S1 ). The score obtained from this analysis was used as an adjustment covariate. We then performed a 1:1 matching without replacement using the nearest neighbour algorithm with a maximum calliper of 0.05. A standardized mean difference (SMD) between groups >10% was considered significant, and variables displaying SMD > 10% were used in the final multivariate analysis for adjustment within the matched cohort.
To test for differences in the probability of undergoing OHT or LVAD implantation, a Fine and Grey model was developed using death as a competing risk for receiving advanced HF therapies. The HR sub‐distribution is shown, and the difference in the cumulative incidence function was tested with the Pepe and Mori test. An exploratory subgroup analysis was performed to test for the significance of interactions between milrinone and baseline characteristics on the primary outcome, and the HR for each subgroup is provided in a forest plot. The trends in haemodynamic parameters obtained from PAC were analysed and plotted using a repeated measures mixed‐effects model with a random intercept at the individual level. A two‐tailed P value of <0.05 was considered significant for all comparisons. All analyses were performed with Stata 15.0 for Mac (StataCorp LLC, TX, USA).
Results
During the study period, 573 patients with a diagnosis of CS were treated with a single inodilator, with 207 (36.1%) receiving dobutamine and 366 (63.9%) receiving milrinone (Figure 1 ). Patients who received milrinone were younger, had a lower body mass index, and were more likely to have chronic kidney disease. Also, patients receiving milrinone had lower rates of IMV at the time of admission, less frequently received an intraaortic balloon pump (IABP) or Impella® device (Abiomed, MA, USA), and had lower admission creatinine and lactate values (Table 1 ). Patients in the milrinone group were less likely to require new IMV, had shorter duration of IMV, less frequently required renal replacement therapy and vasopressors, but were more likely to receive a PAC. This group was also more likely to receive an OHT in the first 30 days after admission. There was no difference in the rates of extracorporeal membrane oxygenation, IABP or Impella, and the length of stay in CICU was similar (Table 2 ).
Figure 1.
Flow chart of patients included in the study. AMI‐CS, acute myocardial infarction cardiogenic shock; CS, cardiogenic shock.
Table 1.
Comparison of baseline characteristics of patients receiving dobutamine or milrinone
| Total (N = 573) | Dobutamine (N = 207) | Milrinone (N = 366) | P value | |
|---|---|---|---|---|
| Demographics | ||||
| Age (years) | 61 (47–69) | 63 (50–71) | 60 (45–68) | 0.006 |
| Male sex | 406 (70.9%) | 147 (71.0%) | 259 (70.8%) | 0.950 |
| BMI (kg/m2) | 25 (22–29) | 27 (24–31) | 25 (22–29) | <0.001 |
| Co‐morbidities | ||||
| Smoking history | 0.012 | |||
| Non‐smoker | 410 (71.6%) | 148 (71.5%) | 262 (71.6%) | |
| Previous smoker | 91 (15.9%) | 42 (20.3%) | 49 (13.4%) | |
| Current smoker | 72 (12.6%) | 17 (8.2%) | 55 (15.0%) | |
| Hypertension | 236 (41.4%) | 88 (42.7%) | 148 (40.7%) | 0.632 |
| Dyslipidaemia | 188 (32.8%) | 74 (35.7%) | 114 (31.2%) | 0.260 |
| Diabetes | 167 (29.1%) | 63 (30.4%) | 104 (28.4%) | 0.609 |
| eGFR <60 mL/min/1.73 m2 | 185 (32.3%) | 52 (25.1%) | 133 (36.4%) | 0.005 |
| Dialysis | 11 (1.9%) | 6 (2.9%) | 5 (1.4%) | 0.201 |
| Cerebrovascular accident | 57 (10.0%) | 25 (12.1%) | 32 (8.7%) | 0.200 |
| Peripheral vascular disease | 31 (5.4%) | 10 (4.8%) | 21 (5.7%) | 0.645 |
| COPD | 41 (7.2%) | 13 (6.3%) | 28 (7.7%) | 0.541 |
| Chronic heart failure | 370 (64.8%) | 132 (64.1%) | 238 (65.2%) | 0.786 |
| Previous MI | 119 (20.8%) | 42 (20.3%) | 77 (21.0%) | 0.832 |
| Atrial fibrillation | 218 (38.1%) | 69 (33.3%) | 149 (40.7%) | 0.081 |
| Previous VT/VF | 72 (12.6%) | 22 (10.6%) | 50 (13.7%) | 0.293 |
| Congenital heart disease | 31 (5.4%) | 5 (2.4%) | 26 (7.1%) | 0.017 |
| Medication | ||||
| Beta‐blocker | 349 (60.9%) | 119 (57.5%) | 230 (62.8%) | 0.207 |
| ACEI/ARB | 214 (37.4%) | 85 (41.1%) | 129 (35.3%) | 0.167 |
| MRA | 247 (43.1%) | 72 (34.8%) | 175 (47.8%) | 0.002 |
| Loop diuretics | 358 (62.5%) | 122 (58.9%) | 236 (64.5%) | 0.188 |
| Anticoagulation | 166 (29.0%) | 48 (23.2%) | 118 (32.2%) | 0.022 |
| Variables at admission | ||||
| Last location before | 0.228 | |||
| CICU | 122 (21.3%) | 36 (17.4%) | 86 (23.5%) | |
| ER | 252 (44.0%) | 95 (45.9%) | 157 (42.9%) | |
| Ward | 199 (34.7%) | 76 (36.7%) | 123 (33.6%) | |
| ICU | ||||
| Heart rate (b.p.m.) | 85 (70–104) | 89 (72–105) | 85 (70–104) | 0.205 |
| Temperature (°C) | 36.6 (36.4–36.8) | 36.6 (36.4–36.8) | 36.6 (36.4–36.8) | 0.662 |
| Mean BP | 73 (67–83) | 73 (67–82) | 75 (67–83) | 0.054 |
| Haemoglobin (g/L) | 122 (105–140) | 121 (100–139) | 123 (107–140) | 0.510 |
| Platelets | 183 (140–244) | 176 (131–234) | 191 (142–249) | 0.093 |
| Sodium (mmol/L) | 135 (131–138) | 135 (130–138) | 135 (131–137) | 0.976 |
| Potassium (mmol/L) | 4.1 (3.7–4.7) | 4.2 (3.7–4.9) | 4.1 (3.7–4.6) | 0.095 |
| eGFR (mL/min/1.73 m2) | 27 (17–47) | 23 (15–41) | 29 (19–48) | 0.001 |
| Lactate (mmol/L) | 1.9 (1.3–3.4) | 2.2 (1.5–4.2) | 1.8 (1.2–2.9) | 0.003 |
| Mechanical ventilation | 94 (16.4%) | 50 (24.2%) | 44 (12.1%) | <0.001 |
| IABP or Impella | 24 (4.2%) | 15 (7.3%) | 9 (2.5%) | 0.006 |
| SCAI stage | 0.235 | |||
| C | 2 (0.4%) | 0 (0.0%) | 2 (0.6%) | |
| D | 570 (99.5%) | 206 (99.5%) | 364 (99.5%) | |
| E | 1 (0.2%) | 1 (0.5%) | 0 (0.0%) | |
Abbreviations: ACEI, angiotensin converting enzyme inhibitors; ARB, angiotensin receptor blockers; BMI, body mass index; BP, blood pressure; CICU, cardiac intensive care unit; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; ER, emergency room; IABP, intraaortic balloon pump; ICU, intensive care unit; MI, myocardial infarction; MRA, mineralocorticoids receptor antagonists; SCAI, Society for Cardiovascular Angiography and Interventions; VF, ventricular fibrillation; VT, ventricular tachycardia.
Table 2.
Resource consumption while in CICU according to the inotrope received
| Total (N = 573) | Dobutamine (N = 207) | Milrinone (N = 366) | P value | |
|---|---|---|---|---|
| Interventions | ||||
| New IABP or Impella | 24 (4.2%) | 10 (4.8%) | 14 (3.8%) | 0.564 |
| New mechanical ventilation | 47 (8.2%) | 24 (11.6%) | 23 (6.3%) | 0.026 |
| PAC insertion | 299 (52.5%) | 94 (45.6%) | 205 (56.3%) | 0.014 |
| PCI performed | 12 (2.1%) | 7 (3.4%) | 5 (1.4%) | 0.106 |
| Renal replacement therapy | 72 (12.6%) | 36 (17.4%) | 36 (9.9%) | 0.009 |
| ECMO | 25 (4.4%) | 11 (5.3%) | 14 (3.9%) | 0.418 |
| Durable LVAD | 70 (12.2%) | 20 (9.7%) | 50 (13.7%) | 0.160 |
| Heart transplant | 23 (4.0%) | 3 (1.5%) | 20 (5.5%) | 0.019 |
| Medication use | ||||
| Dopamine | 20 (3.5%) | 11 (5.3%) | 9 (2.5%) | 0.074 |
| Norepinephrine | 211 (36.8%) | 108 (52.2%) | 103 (28.1%) | <0.001 |
| Epinephrine | 57 (10.0%) | 39 (18.8%) | 18 (4.9%) | <0.001 |
| Phenylephrine | 9 (1.6%) | 7 (3.4%) | 2 (0.6%) | 0.009 |
| Vasopressin | 66 (11.5%) | 37 (17.9%) | 29 (7.9%) | <0.001 |
| Nitroprusside | 118 (20.6%) | 36 (17.4%) | 82 (22.4%) | 0.154 |
| Number of pressors | 0 (0–1) | 1 (0–2) | 0 (0–1) | <0.001 |
| Length of stay | ||||
| LOS in CICU | 6 (3–10) | 6 (3–10) | 6 (3–10) | 0.963 |
| LOS in CICU >7 days | 199 (34.7%) | 73 (35.3%) | 126 (34.4%) | 0.839 |
| LOS in hospital | 21 (11–37) | 17 (8–29) | 23 (12–39) | <0.001 |
| Length of IMV | 3 (1–5) | 3 (1–5) | 3 (1–5) | 0.952 |
| Length of IMV > 5 days | 30 (5.2%) | 16 (7.7%) | 14 (3.8%) | 0.044 |
Abbreviations: CICU, cardiac intensive care unit; ECMO, extracorporeal membrane oxygenation; IABP, intraaortic balloon pump; IMV, invasive mechanical ventilation; LOS, length of stay; LVAD, left ventricular assist device; PAC, pulmonary artery catheter; PCI, percutaneous coronary intervention.
During the first 30 days following CICU admission, there were 133 deaths: 76 (36.7%) in the dobutamine group and 57 (15.6%) in the milrinone group. Those receiving milrinone had lower rates of mortality (unadjusted HR = 0.36, 95% CI 0.25–0.52; Figure 2 ) and were more likely to receive an OHT or durable LVAD in the competing risk regression (unadjusted HR sub‐distribution = 1.79, 95% CI 1.12–2.02; Figure S1 ). The timing of LVAD or OHT in each group is shown in Figure S2 . When a combined endpoint of death, LVAD implant, or OHT was used, a greater event‐free survival was observed in the milrinone group (HR = 0.65, 95% CI 0.49 vs. 0.85; Figure S3 ). There was no interaction between the type of inodilator used and age group, previous history of HF, eGFR, or use of beta‐blocker therapy (Figure 3 ) with respect to the risk of 30 day mortality.
Figure 2.
Kaplan Meier plot for the primary endpoint, 30 day mortality, according to the treatment received. CICU, cardiac intensive care unit.
Figure 3.
Subgroup analysis – unadjusted and adjusted. eGFR, estimated glomerular filtration rate.
After adjusting for age, diabetes, chronic kidney disease, atrial fibrillation, previous history of HF, congenital heart disease, beta‐blocker use, last location before admission, mean blood pressure at admission, eGFR at admission, IMV at admission, IABP or Impella at admission, and Cardiovascular Angiography and Interventions stage, milrinone remained independently associated with reduced mortality (adjusted HR = 0.44, 95% CI 0.30–0.65). This association remained unchanged when the number of pressors, use of IMV during admission, IABP, Impella, or extracorporeal membrane oxygenation during admission, renal replacement therapy and use of a PAC were added as covariates (adjusted HR = 0.52, 95% 0.35–0.77). These results remained unchanged after introducing the logistic‐regression derived propensity score within the Cox model, showing a mortality benefit of milrinone at 30 days.
After 1:1 propensity matching, 133 patients receiving dobutamine (64% of the sample) were compared with 133 patients receiving milrinone, with no significant differences in baseline characteristics (Table 3 ) and adequate overlap of the kernel density plots (Figure S4 ) and little imbalance using SMD (Table S2 ). The unadjusted Cox regression showed a reduction in mortality for milrinone in the matched population (HR = 0.58, 95% CI 0.35–0.96; Figure S5 ), that was still seen after adjustment for covariates with a SMD > 10% (HR = 0.51, 95% CI 0.27–0.96), consistent with the adjusted results observed in the original cohort.
Table 3.
Baseline characteristics in the matched cohort
| Total (N = 266) | Dobutamine (N = 133) | Milrinone (N = 133) | P value | |
|---|---|---|---|---|
| Demographics | ||||
| Age (years) | 62 (51–69) | 61 (50–69) | 62 (53–69) | 0.956 |
| Male sex | 186 (69.9%) | 94 (70.7%) | 92 (69.2%) | 0.789 |
| BMI (kg/m2) | 26 (23–31) | 26 (24–30) | 26 (22–31) | 0.764 |
| Comorbidities | ||||
| Smoking history | 0.182 | |||
| Non‐smoker | 190 (71.4%) | 95 (71.4%) | 95 (71.4%) | |
| Previous smoker | 42 (15.8%) | 25 (18.8%) | 17 (12.8%) | |
| Current smoker | 34 (12.8%) | 13 (9.8%) | 21 (15.8%) | |
| Hypertension | 117 (44.3%) | 54 (40.9%) | 63 (47.7%) | 0.265 |
| Dyslipidaemia | 94 (35.3%) | 49 (26.8%) | 45 (33.8%) | 0.608 |
| Diabetes | 72 (27.1%) | 39 (29.3%) | 33 (24.8%) | 0.408 |
| eGFR <60 mL/min/1.73 m2 | 67 (25.2%) | 38 (28.6%) | 29 (21.8%) | 0.204 |
| Dialysis | 8 (3.0%) | 5 (3.8%) | 3 (2.3%) | 0.473 |
| Cerebrovascular accident | 28 (10.5%) | 15 (11.2%) | 13 (9.8%) | 0.689 |
| Peripheral vascular disease | 12 (4.5%) | 5 (3.8%) | 7 (5.3%) | 0.555 |
| COPD | 21 (7.9%) | 10 (7.5%) | 11 (8.3%) | 0.820 |
| Chronic heart failure | 171 (64.3%) | 87 (65.4%) | 84 (63.2%) | 0.701 |
| Previous MI | 58 (21.8%) | 30 (22.6%) | 28 (21.1%) | 0.766 |
| Atrial fibrillation | 85 (32.0%) | 46 (34.6%) | 39 (29.3%) | 0.357 |
| Previous VT/VF | 32 (12.0%) | 15 (11.3%) | 17 (12.8%) | 0.706 |
| Congenital heart disease | 2 (0.8%) | 2 (1.5%) | 0 (0.0%) | 0.156 |
| Medication | ||||
| Beta‐blocker | 149 (56.0%) | 77 (57.9%) | 72 (54.1%) | 0.537 |
| ACEI/ARB | 106 (39.8%) | 57 (43.9%) | 49 (36.8%) | 0.316 |
| MRA | 96 (36.1%) | 50 (37.6%) | 46 (34.6%) | 0.610 |
| Loop diuretics | 157 (59.0%) | 80 (60.2%) | 77 (57.9%) | 0.708 |
| Anticoagulation | 65 (24.4%) | 33 (24.8%) | 32 (24.1%) | 0.887 |
| Variables at admission | ||||
| Last location before CICU | 0.458 | |||
| ER | 48 (18.1%) | 21 (15.8%) | 27 (20.3%) | |
| Ward | 111 (41.7%) | 60 (45.1%) | 51 (38.4%) | |
| ICU | 107 (40.2%) | 52 (39.1%) | 55 (41.4%) | |
| Heart rate (b.p.m.) | 90 (72–105) | 88 (74–105) | 90 (70–108) | 0.402 |
| Temperature (°C) | 36.6 (36.4–36.9) | 36.6 (36.4–36.8) | 36.6 (36.4–36.9) | 0.527 |
| Mean BP | 75 (67–83) | 73 (69–83) | 75 (67–83) | 0.942 |
| Haemoglobin (g/L) | 123 (104–141) | 122 (102–143) | 123 (106–138) | 0.616 |
| Platelets | 196 (148–253) | 186 (144–234) | 205 (149–264) | 0.128 |
| Sodium (mmol/L) | 135 (130–138) | 135 (130–138) | 134 (131–138) | 0.754 |
| Potassium (mmol/L) | 4.1 (3.7–4.8) | 4.0 (3.7–4.8) | 4.2 (3.7–4.7) | 0.362 |
| eGFR (mL/min/1.73 m2) | 26 (17–46) | 26 (17–47) | 27 (17–44) | |
| Lactate (mmol/L) | 2.0 (1.4–3.5) | 2.1 (1–5 – 4.3) | 2.0 (1.3–3.4) | 0.254 |
| Mechanical ventilation | 59 (22.2%) | 29 (21.8%) | 30 (22.6%) | 0.883 |
| IABP or pVAD | 15 (5.6%) | 8 (6.0%) | 7 (5.3%) | 0.790 |
| SCAI stage | 0.368 | |||
| C | 1 (0.4%) | 0 (0.0%) | 1 (0.8%) | |
| D | 264 (99.3%) | 132 (99.3%) | 132 (99.3%) | |
| E | 1 (0.4%) | 1 (0.8%) | 0 (0.0%) | |
Abbreviations: ACEI, angiotensin converting enzyme inhibitors; ARB, angiotensin receptor blockers; BMI, body mass index; BP, blood pressure; CICU, cardiac intensive care unit; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; ER, emergency room; IABP, intraaortic balloon pump; ICU, intensive care unit; MI, myocardial infarction; MRA, mineralocorticoids receptor antagonists; SCAI, Society for Cardiovascular Angiography and Interventions; VF, ventricular fibrillation; VT, ventricular tachycardia.
In the cohort of 573 patients, PAC was used in 299 (52.5%) patients, of which 94 (45.6%) were in the dobutamine group and 205 (56.3%) in the milrinone group. Data at the time of insertion were available in 211 patients, in 230 patients at 24 h, 214 patients at 48 h, and 180 patients at 72 h post‐insertion. There were no differences between both groups in the number of available measurements and similar baseline values at the time of PAC insertion. In both groups, there were similar increases in cardiac index, cardiac power output, and mixed venous oxygen saturation, along with a reduction in pulmonary pressures, PCWP, and central venous pressure. However, milrinone was associated with a significant greater improvement in CPA, as well as increased stroke volume and RVSWI (Figure 4 ).
Figure 4.
Comparison of haemodynamic trends between both groups. *P < 0.05; **P < 0.01.
Discussion
This study identified a difference in outcomes in patients with ADHF‐CS receiving a single inodilator (milrinone or dobutamine) in a contemporary CICU registry. ADHF‐CS patients receiving milrinone had lower mortality and a higher likelihood of receiving advanced HF therapies (OHT and durable LVAD). Improved mortality in the milrinone group is supported by improved haemodynamics in those receiving a PAC.
Our analysis provides some insights into the use of milrinone in CS. In our registry, its use was preferred in patients with less haemodynamic compromise, as shown by the higher MAP, lower lactate, and less use of vasopressors, short‐term ventricular support devices, and IMV. This has been previously reported in other registries 15 and may contribute to the perception that milrinone is superior to dobutamine in the setting of CS. However, the survival benefit in our cohort is consistent despite adjustment for admission variables and in‐hospital interventions and is reproductible in a propensity score‐matched population with no difference in baseline characteristics even after accounting for unbalanced covariates. Also, it does not seem related to earlier use of LVAD or OHT in the milrinone population, as shown in Figure S2 . These findings differ from some of the existing literature that suggests there is no difference in short‐term mortality. 12 , 16 However, previous literature compared both inodilators in heterogeneous populations of CS (with both AMI‐CS and ADHF‐CS), which may have influenced those results. 17
A major limitation of previous CS trials and observational studies when assessing the potential benefits of interventions has been the heterogeneity of the CS population. The physiology leading to ADHF‐CS differs from that of AMI‐CS. In ADHF‐CS, CS is often a subacute phenomenon, whereby patients often present with insidious congestion and concomitant hypoperfusion resulting in end‐organ damage and haemodynamic instability. In contrast, AMI‐CS occurs abruptly after a sudden drop in CO that is poorly compensated. These patients present acutely and have worse outcomes compared with those with ADHF‐CS. 18 , 19 Patients with de novo ADHF‐CS present with mixed features of AMI‐CS and ADHF‐CS, presumably related to the absence of compensatory mechanisms. 20 In our study, two‐thirds of patients (64.8%) had long‐standing HF before admission, suggesting that ADHF‐CS from progressive clinical decline was the more likely contributor to outcomes. In a population of outpatients with advanced HF symptoms receiving palliative inotropes, a retrospective analysis showed better outcomes with milrinone when compared with dobutamine, associated with greater use of prognosis‐modifying HF medication. 21 These results were also observed in a recent metanalysis, 22 supporting the findings from our study.
RV dysfunction carries a poor prognosis in CS. In patients with ADHF‐CS (but not in those with AMI‐CS), the RV function assessed using invasive haemodynamics was associated with worse outcomes, 23 , 24 highlighting that RV dysfunction may discriminate the risk of mortality among ADHF‐CS patients better than AMI‐CS. In our cohort, haemodynamic assessments showed greater RVSWI increase with milrinone. There was also marked improvement in CPA with milrinone, related to a higher stroke volume despite a similar reduction in PCWP with both inodilators. CPA reflects the RV pulsatile load and adds prognostic information to PVR (which reflects the resistive load) in patients with left‐sided HF. In patients with CS, it has been demonstrated that CPA is strongly associated with mortality independent from pulmonary pressures or PVR, and the increase in CPA in the first 24 h was associated with better survival. 25 The milrinone group parallels this finding, with greater CPA improvement in the first 24 h post‐PAC insertion that may have led to greater RVSWI.
Intravascular congestion, reflected by stressed blood volume, has been proven to be closely linked to mortality in ADHF‐CS but not in AMI‐CS. 26 Milrinone increases the splanchnic blood flow after cardiac surgery 27 , 28 , 29 and decreases renal vascular resistance by vasodilating both the efferent and afferent arterioles, thus improving renal oxygen delivery and perfusion to a greater extent than dobutamine. 30 , 31 These effects on the splanchnic and renal vascular beds may provide the biological plausibility for the lower mortality observed in our cohort of ADHF‐CS patients.
Levosimendan is a non‐adrenergic calcium sensitizer capable of increasing cardiac contractility without increasing oxygen consumption and can also decrease both pulmonary and vascular resistance by acting on KATP channels. 32 It has shown a reduction in HF admissions, improved quality of life and biomarkers in patients with ADHF with advanced functional class in the outpatient setting. 33 , 34 Although its use in all‐comers with LVEF <35% undergoing cardiac surgery was not useful to prevent organ dysfunction in randomized clinical trials, 35 , 36 its effectiveness in patients with CS is currently being explored (NCT04020263). Unfortunately, we could not include in our registry patients treated with levosimendan, as it is not commercially available in Canada.
Limitations
As this is a single centre retrospective analysis of patients with CS, this analysis may be subjected to the inherent bias associated with local practice patterns. Selection bias is of concern, given the initial imbalances between groups. However, we obtained different models using multiple adjusting techniques incorporating variables known at admission and those collected during CICU stay to account for confounding and were able to reproduce the same results after a 1:1 matching, providing more robust evidence in support of the findings. Additionally, insertion of a PAC was dependent on the CICU team, and therefore, not all patients had haemodynamic measurements available.
Conclusions
In patients with ADHF‐CS, milrinone use as a single inodilator was associated with a reduction in 30 day mortality. Haemodynamic assessment demonstrated a greater increase in CPA, increased stroke volume, and RVSWI in these patients. Further studies are required to better understand the effects and associated outcomes of milrinone in patients with ADHF‐CS.
Funding
Eduard Rodenas‐Alesina has received funding from the Spanish Society of Cardiology (Magda Heras grant, SEC/MHE‐MOV‐INT 21/001). Darshan H. Brahmbhatt is supported by TRANSFORM HF. Adriana Luk is supported by the Heart and Stroke Foundation/University of Toronto Polo Chair in Cardiology Young Investigator Award.
Conflict of interest
Eduard Rodenas‐Alesina has received non‐conditioned grants from Biotronik, Microport, Johnson and Johnson, and Sanofi, outside of the submitted work. Darshan H. Brahmbhatt has received travel support from Abbott and Biotronik, and honoraria, travel support, and a grant from Boston Scientific outside of the submitted work.
Supporting information
Table S1. Variables included in logistic regression to perform the 1:1 matching.
Table S2. Standardized mean differences before and after the propensity score matching.
Figure S1. Cumulative incidence function for receiving an LVAD or heart transplant, accounting for death as a competing risk.
Figure S2. Histogram showing the distribution of patients undergoing LVAD implantation or OHT in both milrinone and dobutamine groups.
Figure S3. Event‐free survival at 30 days for the combined endpoint of death, LVAD implant or heart transplantation (log‐rank P = 0.001).
Figure S4. Kernel density plots of the score distribution before and after matching.
Figure S5. Survival analysis in the propensity‐matched population
Rodenas‐Alesina, E. , Luis Scolari, F. , Wang, V. N. , Brahmbhatt, D. H. , Mihajlovic, V. , Fung, N. L. , Otsuki, M. , Billia, F. , Overgaard, C. B. , and Luk, A. (2023) Improved mortality and haemodynamics with milrinone in cardiogenic shock due to acute decompensated heart failure. ESC Heart Failure, 10: 2577–2587. 10.1002/ehf2.14379.
References
- 1. Cuffe MS, Califf RM, Adams KFJ, Benza R, Bourge R, Colucci WS, Massie BM, O'Connor CM, Pina I, Quigg R, Silver MA, Gheorghiade M, Outcomes of a Prospective Trial of Intravenous Milrinone for Exacerbations of Chronic Heart Failure (OPTIME‐CHF) Investigators . Short‐term intravenous milrinone for acute exacerbation of chronic heart failure: a randomized controlled trial. JAMA. 2002; 287: 1541–1547. [DOI] [PubMed] [Google Scholar]
- 2. Packer M, Carver JR, Rodeheffer RJ, Ivanhoe RJ, DiBianco R, Zeldis SM, Hendrix GH, Bommer WJ, Elkayam U, Kukin ML, Mallis GI, Sollano JA, Shannon J, Tandon PK, DeMets DL. Effect of oral milrinone on mortality in severe chronic heart failure. The PROMISE study research group. N Engl J Med. 1991; 325: 1468–1475. [DOI] [PubMed] [Google Scholar]
- 3. Costanzo MR, Johannes RS, Pine M, Gupta V, Saltzberg M, Hay J, Yancy CW, Fonarow GC. The safety of intravenous diuretics alone versus diuretics plus parenteral vasoactive therapies in hospitalized patients with acutely decompensated heart failure: a propensity score and instrumental variable analysis using the acutely decompensated heart. Am Heart J. 2007; 154: 267–277. [DOI] [PubMed] [Google Scholar]
- 4. Abraham WT, Adams KF, Fonarow GC, Costanzo MR, Berkowitz RL, LeJemtel TH, Cheng ML, Wynne J, ADHERE Scientific Advisory Committee and Investigators, ADHERE Study Group . In‐hospital mortality in patients with acute decompensated heart failure requiring intravenous vasoactive medications: an analysis from the Acute Decompensated Heart Failure National Registry (ADHERE). J Am Coll Cardiol. 2005; 46: 57–64. [DOI] [PubMed] [Google Scholar]
- 5. van Diepen S, Katz JN, Albert NM, Henry TD, Jacobs AK, Kapur NK, Kilic A, Menon V, Ohman EM, Sweitzer NK, Thiele H, Washam JB, Cohen MG, American Heart Association Council on Clinical Cardiology; Council on Cardiovascular and Stroke Nursing; Council on Quality of Care and Outcomes Research; and Mission: Lifeline . Contemporary management of cardiogenic shock: a scientific statement from the American Heart Association. Circulation. 2017; 136: e232–e268. [DOI] [PubMed] [Google Scholar]
- 6. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, Burri H, Butler J, Čelutkienė J, Chioncel O, Cleland JGF, Coats AJS, Crespo‐Leiro MG, Farmakis D, Gilard M, Heymans S, Hoes AW, Jaarsma T, Jankowska EA, Lainscak M, Lam CSP, Lyon AR, McMurray JJV, Mebazaa A, Mindham R, Muneretto C, Francesco Piepoli M, Price S, Rosano GMC, Ruschitzka F, Kathrine Skibelund A, ESC Scientific Document Group . 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2021; 42: 3599–3726. [DOI] [PubMed] [Google Scholar]
- 7. De Backer D, Biston P, Devriendt J, Madl C, Chochrad D, Aldecoa C, Brasseur A, Defrance P, Gottignies P, Vincent JL, SOAP II Investigators . Comparison of dopamine and norepinephrine in the treatment of shock. N Engl J Med. 2010; 362: 779–789. [DOI] [PubMed] [Google Scholar]
- 8. Pirracchio R, Parenica J, Resche Rigon M, Chevret S, Spinar J, Jarkovsky J, Zannad F, Alla F, Mebazaa A, for the GREAT network . The effectiveness of inodilators in reducing short term mortality among patient with severe cardiogenic shock: a propensity‐based analysis. PLoS ONE. 2013; 8: e71659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Levy B, Perez P, Perny J, Thivilier C, Gerard A. Comparison of norepinephrine‐dobutamine to epinephrine for hemodynamics, lactate metabolism, and organ function variables in cardiogenic shock. A prospective, randomized pilot study. Crit Care Med. 2011; 39: 450–455. [DOI] [PubMed] [Google Scholar]
- 10. Levy B, Clere‐Jehl R, Legras A, Morichau‐Beauchant T, Leone M, Frederique G, Quenot JP, Kimmoun A, Cariou A, Lassus J, Harjola VP, Meziani F, Louis G, Rossignol P, Duarte K, Girerd N, Mebazaa A, Vignon P, Mattei M, Thivilier C, Perez P, Auchet T, Fritz C, Boisrame‐Helme J, Mercier E, Garot D, Perny J, Gette S, Hammad E, Vigne C, Dargent A, Andreu P, Guiot P. Epinephrine versus norepinephrine for cardiogenic shock after acute myocardial infarction. J Am Coll Cardiol. 2018; 72: 173–182. [DOI] [PubMed] [Google Scholar]
- 11. Mathew R, Visintini SM, Ramirez FD, DiSanto P, Simard T, Labinaz M, Hibbert BM. Efficacy of milrinone and dobutamine in low cardiac output states: systematic review and meta‐analysis. Clin Invest Med. 2019; 42: E26–E32. [DOI] [PubMed] [Google Scholar]
- 12. Mathew R, Di Santo P, Jung RG, Marbach JA, Hutson J, Simard T, Ramirez FD, Harnett DT, Merdad A, Almufleh A, Weng W, Abdel‐Razek O, Fernando SM, Kyeremanteng K, Bernick J, Wells GA, Chan V, Froeschl M, Labinaz M, Le May MR, Russo JJ, Hibbert B. Milrinone as compared with dobutamine in the treatment of cardiogenic shock. N Engl J Med. 2021; 385: 516–525. [DOI] [PubMed] [Google Scholar]
- 13. Thayer KL, Zweck E, Ayouty M, Garan AR, Hernandez‐Montfort J, Mahr C, Morine KJ, Newman S, Jorde L, Haywood JL, Harwani NM, Esposito ML, Davila CD, Wencker D, Sinha SS, Vorovich E, Abraham J, O’Neill W, Udelson J, Burkhoff D, Kapur NK. Invasive hemodynamic assessment and classification of in‐hospital mortality risk among patients with cardiogenic shock. Circ Heart Fail. 2020; 13: e007099. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Hochman JS, Sleeper LA, Webb JG, Sanborn TA, White HD, Talley JD, Buller CE, Jacobs AK, Slater JN, Col J, McKinlay SM, Picard MH, Menegus MA, Boland J, Dzavik V, Thompson CR, Wong SC, Steingart R, Forman R, Aylward PE, Godfrey E, Desvigne‐Nickens P, LeJemtel TH. Early revascularization in acute myocardial infarction complicated by cardiogenic SHOCK. SHOCK investigators. Should we emergently revascularize occluded coronaries for cardiogenic shock. N Engl J Med. 1999; 341: 625–634. [DOI] [PubMed] [Google Scholar]
- 15. Gao F, Zhang Y. Inotrope use and intensive care unit mortality in patients with cardiogenic shock: an analysis of a large electronic intensive care unit database. Front Cardiovasc Med. 2021; 8: 696138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Uhlig K, Efremov L, Tongers J, Frantz S, Mikolajczyk R, Sedding D, Schumann J. Inotropic agents and vasodilator strategies for the treatment of cardiogenic shock or low cardiac output syndrome. Cochrane Database Syst Rev. 2020; 11: CD009669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Jung RG, Di Santo P, Mathew R, Abdel‐Razek O, Parlow S, Simard T, Marbach JA, Gillmore T, Mao B, Bernick J, Theriault‐Lauzier P, Fu A, Lau L, Motazedian P, Russo JJ, Labinaz M, Hibbert B. Implications of myocardial infarction on management and outcome in cardiogenic shock. J Am Heart Assoc. 2021; 10: e021570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Abraham J, Blumer V, Burkhoff D, Pahuja M, Sinha SS, Rosner C, Vorovich E, Grafton G, Bagnola A, Hernandez‐Montfort JA, Kapur NK. Heart failure‐related cardiogenic shock: pathophysiology, evaluation and management considerations: review of heart failure‐related cardiogenic shock. J Card Fail. 2021; 27: 1126–1140. [DOI] [PubMed] [Google Scholar]
- 19. Arrigo M, Jessup M, Mullens W, Reza N, Shah AM, Sliwa K, Mebazaa A. Acute heart failure. Nat Rev Dis Primers. 2020; 6: 16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Bhatt AS, Berg DD, Bohula EA, Alviar CL, Baird‐Zars VM, Barnett CF, Burke JA, Carnicelli AP, Chaudhry SP, Daniels LB, Fang JC, Fordyce CB, Gerber DA, Guo J, Jentzer JC, Katz JN, Keller N, Kontos MC, Lawler PR, Menon V, Metkus TS, Nativi‐Nicolau J, Phreaner N, Roswell RO, Sinha SS, Jeffrey Snell R, Solomon MA, van Diepen S, Morrow DA. De novo vs acute‐on‐chronic presentations of heart failure‐related cardiogenic shock: insights from the Critical Care Cardiology Trials Network Registry. J Card Fail. 2021; 27: 1073–1081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Sami F, Acharya P, Noonan G, Maurides S, Al‐Masry AA, Bajwa S, Parimi N, Boda I, Tran C, Goyal A, Mastoris I. Palliative inotropes in advanced heart failure: comparing outcomes between milrinone and dobutamine. J Card Fail. 2022; 28: 1683–1691. [DOI] [PubMed] [Google Scholar]
- 22. Szarpak L, Szwed P, Gasecka A, Rafique Z, Pruc M, Filipiak KJ, Jaguszewski MJ. Milrinone or dobutamine in patients with heart failure: evidence from meta‐analysis. ESC Heart Failure. 2022; 9: 2049–2050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Jain P, Thayer KL, Abraham J, Everett KD, Pahuja M, Whitehead EH, Schwartz BP, Anuradha LA, Sinha SS, Kanwar MK, Garan AR. Right ventricular dysfunction is common and identifies patients at risk of dying in cardiogenic shock. J Card Fail. 2021; 27: 1061–1072. [DOI] [PubMed] [Google Scholar]
- 24. Josiassen J, Helgestad OKL, Møller JE, Schmidt H, Jensen LO, Holmvang L, Ravn HB, Hassager C. Cardiogenic shock due to predominantly right ventricular failure complicating acute myocardial infarction. Eur Hear journal Acute Cardiovasc care. 2021; 10: 33–39. [DOI] [PubMed] [Google Scholar]
- 25. Zorzi MF, Cancelli E, Rusca M, Kirsch M, Yerly P, Liaudet L. The prognostic value of pulmonary artery compliance in cardiogenic shock. Pulm Circ. 2019; 9: 2045894019877161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Whitehead EH, Thayer KL, Sunagawa K, Hernandez‐Montfort J, Garan AR, Kanwar MK, Sinha SS, Mahr C, Kapur NK, Burkhoff D. Estimation of stressed blood volume in patients with cardiogenic shock from acute myocardial infarction and decompensated heart failure. J Card Fail. 2021; 27: 1141–1145. [DOI] [PubMed] [Google Scholar]
- 27. Iribe G, Yamada H, Matsunaga A, Yoshimura N. Effects of the phosphodiesterase III inhibitors olprinone, milrinone, and amrinone on hepatosplanchnic oxygen metabolism. Crit Care Med. 2000; 28: 743–748. [DOI] [PubMed] [Google Scholar]
- 28. Möllhoff T, Loick HM, Van Aken H, Schmidt C, Rolf N, Tjan TD, Asfour B, Berendes E. Milrinone modulates endotoxemia, systemic inflammation, and subsequent acute phase response after cardiopulmonary bypass (CPB). Anesthesiology. 1999; 90: 72–80. [DOI] [PubMed] [Google Scholar]
- 29. Yamaura K, Okamoto H, Akiyoshi K, Irita K, Taniyama T, Takahashi S. Effect of low‐dose milrinone on gastric intramucosal pH and systemic inflammation after hypothermic cardiopulmonary bypass. J Cardiothorac Vasc Anesth. 2001; 15: 197–203. [DOI] [PubMed] [Google Scholar]
- 30. Lannemyr L, Bragadottir G, Redfors B, Ricksten S‐E. Effects of milrinone on renal perfusion, filtration and oxygenation in patients with acute heart failure and low cardiac output early after cardiac surgery. J Crit Care. 2020; 57: 225–230. [DOI] [PubMed] [Google Scholar]
- 31. Liang CS, Thomas A, Imai N, Stone CK, Kawashima S, Hood WBJ. Effects of milrinone on systemic hemodynamics and regional circulations in dogs with congestive heart failure: comparison with dobutamine. J Cardiovasc Pharmacol. 1987; 10: 509–516. [DOI] [PubMed] [Google Scholar]
- 32. Gustafsson F, Guarracino F, Schwinger RHG. The inodilator levosimendan as a treatment for acute heart failure in various settings. Eur Hear J Suppl J Eur Soc Cardiol. 2017; 19: C2–C7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Altenberger J, Parissis JT, Costard‐Jaeckle A, Winter A, Ebner C, Karavidas A, Sihorsch K, Avgeropoulou E, Weber T, Dimopoulos L, Ulmer H, Poelzl G. Efficacy and safety of the pulsed infusions of levosimendan in outpatients with advanced heart failure (LevoRep) study: a multicentre randomized trial. Eur J Heart Fail. 2014; 16: 898–906. [DOI] [PubMed] [Google Scholar]
- 34. Comín‐Colet J, Manito N, Segovia‐Cubero J, Delgado J, García Pinilla JM, Almenar L, Crespo‐Leiro MG, Sionis A, Blasco T, Pascual‐Figal D, Gonzalez‐Vilchez F, Lambert‐Rodríguez JL, Grau M, Bruguera J, on behalf of the LION‐HEART Study Investigators . Efficacy and safety of intermittent intravenous outpatient administration of levosimendan in patients with advanced heart failure: the LION‐HEART multicentre randomised trial. Eur J Heart Fail. 2018; 20: 1128–1136. [DOI] [PubMed] [Google Scholar]
- 35. Landoni G, Lomivorotov VV, Alvaro G, Lobreglio R, Pisano A, Guarracino F, Calabrò MG, Grigoryev EV, Likhvantsev VV, Salgado‐Filho MF, Bianchi A, Pasyuga VV, Baiocchi M, Pappalardo F, Monaco F, Boboshko VA, Abubakirov MN, Amantea B, Lembo R, Brazzi L, Verniero L, Bertini P, Scandroglio AM, Bove T, Belletti A, Michienzi MG, Shukevich DL, Zabelina TS, Bellomo R, Zangrillo A, CHEETAH Study Group . Levosimendan for hemodynamic support after cardiac surgery. N Engl J Med. 2017; 376: 2021–2031. [DOI] [PubMed] [Google Scholar]
- 36. Mehta RH, Leimberger JD, van Diepen S, Meza J, Wang A, Jankowich R, Harrison RW, Hay D, Fremes S, Duncan A, Soltesz EG, Luber J, Park S, Argenziano M, Murphy E, Marcel R, Kalavrouziotis D, Nagpal D, Bozinovski J, Toller W, Heringlake M, Goodman SG, Levy JH, Harrington RA, Anstrom KJ, Alexander JH, LEVO‐CTS Investigators . Levosimendan in patients with left ventricular dysfunction undergoing cardiac surgery. N Engl J Med. 2017; 376: 2032–2042. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. Variables included in logistic regression to perform the 1:1 matching.
Table S2. Standardized mean differences before and after the propensity score matching.
Figure S1. Cumulative incidence function for receiving an LVAD or heart transplant, accounting for death as a competing risk.
Figure S2. Histogram showing the distribution of patients undergoing LVAD implantation or OHT in both milrinone and dobutamine groups.
Figure S3. Event‐free survival at 30 days for the combined endpoint of death, LVAD implant or heart transplantation (log‐rank P = 0.001).
Figure S4. Kernel density plots of the score distribution before and after matching.
Figure S5. Survival analysis in the propensity‐matched population