Effect of Intravenous Levosimendan or Milrinone on Left Atrial Pressure in Patients Undergoing Off-Pump Coronary Artery bypass Grafting—A Prospective Double-Blind, Randomized Controlled Trial

ABSTRACT

Background:

Maintaining a low left atrial pressure (LAP) in off-pump coronary artery bypass grafting (OPCAB) is desirable. This study was done to compare the effects of intravenous levosimendan or milrinone on LAP at different stages of OPCAB.

Materials and Methods:

After institutional ethics committee clearance, this two-arm double-blind randomized control trial was done in 44 adult patients with triple vessel coronary artery disease undergoing OPCAB at cardiac OT of IPGME&R, Kolkata. The patients were randomly allocated into two groups receiving intraoperative either levosimendan or milrinone. Pulmonary capillary wedge pressure (PCWP) was compared as the primary outcome parameter, whereas other echocardiographic and hemodynamic parameters were also assessed during six stages of OPCAB, that is, after sternotomy, proximal(s), left anterior descending artery (LAD), obtuse marginal (OM), posterior descending artery (PDA) grafting, and before sternal closure. Numerical parameters were compared using Student’s unpaired two-tailed t-test.

Results:

PCWP was found to be significantly lower (P < 0.05) in the levosimendan group during proximal (P = 0.047), LAD (P = 0.018), OM (P < 0.0001), PDA grafting (P = 0.028), and before sternal closure (P = 0.015). Other parameters indicate LAP, that is, from mitral early diastolic inflow velocity to mitral annular early diastolic velocity ratio (E/e’), which indicated significantly lower LAP in levosimendan group during LAD, OM, and PDA grafting and before sternal closure.

Conclusion:

Levosimendan may be used as a primary inotrope in terms of better reduction in left atrial pressure during different stages of OPCAB, translating to a decrease in left ventricular end-diastolic pressure, therefore maintaining optimum coronary perfusion pressure, which is the primary goal of the surgery.

INTRODUCTION

Off-pump coronary artery bypass grafting (OPCAB) requires diverse, non-physiological cardiac positioning to graft different culprit vessels. It also utilizes a coronary stabilizer, stay sutures which alter the cardiac morphology, and sizes of the cardiac chambers, thereby increasing left ventricular end-diastolic pressure (LVEDP) and jeopardizing coronary perfusion pressure and global cardiac output. Published data suggest that the mitral annulus is distorted primarily at the atrioventricular groove during intraoperative manipulation, giving rise to either functional mitral stenosis or worsening mitral regurgitation resulting in raised left atrial pressure.[1-3] Due to these physiological, positional, and anatomical perturbations, LVEDP rises significantly which is undesirable as it increases myocardial wall stress and oxygen consumption. Left atrial pressure (LAP) is reported to be synonymous with left ventricular end-diastolic pressure (LVEDP) in the absence of mitral valve disease.[4] The left atrium serves as a contractile chamber for augmenting left ventricular filling in late diastole.[5] Therefore, pharmacological interventions are employed to decrease the elevated LAP and LVEDP encountered during the grafting of culprit coronary arteries which are undesirable and may occur either due to adverse anatomical positioning or utilization of stabilization devices.

Levosimendan, a novel inotrope, binds to cardiac troponin C, enhances myofilament responsiveness to calcium, and prolongs the duration of actin-myosin overlap, thereby increasing myocardial contractility, but, without increasing intracellular calcium concentration and myocardial oxygen consumption.[6,7] It also possesses lusitropic actions[8] and exerts peripheral vasodilatory and potential preconditioning effects on the myocardium[9,10] by virtue of its action on mitochondrial ATP-sensitive potassium channels. In addition, it provides beneficial immunomodulatory, cardioprotective,[11] anti-stunning,[12] anti-ischemic,[13] anti-inflammatory, and antioxidant effects[14] to improve cardiac performance in the presence of ischemia.

Milrinone is selective for phosphodiesterase III inhibitor at low doses and nonselective phosphodiesterase inhibitor at high doses. Phosphodiesterase is an enzyme that hydrolyzes the second messenger cyclic adenosine monophosphate (cAMP) and guanosine monophosphate (cGMP) and prevents their breakdown, increasing protein kinase A activity, leading to phosphorylation of calcium ion channels in the sarcoplasmic reticulum and increasing calcium availability in myocyte sarcomere and increased calcium reuptake into the sarcoplasmic reticulum. This increased calcium availability manifests in increased cardiac inotropy and chronotropy and reuptake results in enhanced myocardial relaxation (lusitropy) translating into improved diastolic function as well as systolic function.[15-18] In the vasculature, PDE III inhibition prevents cGMP metabolism in the smooth musculature and results in vasodilation in both arteries and veins.

Hence, the current study compared the impact of levosimendan (myofilament calcium sensitizer) and milrinone (phosphodiesterase III inhibitor and inodilator) on LAP, if any.

MATERIALS AND METHODS

All adult patients of both gender with triple vessel coronary artery disease with left ventricular ejection fraction (LVEF) 35–55% (having mild to moderate LV systolic dysfunction) posted for OPCAB were enrolled in the study. Patients with mitral regurgitation, severe left ventricular systolic dysfunction (LVEF <35%), underwent conversion to on-pump CABG surgery, critical left main coronary artery disease, redo surgery, atrial fibrillation, regional wall motion abnormality, mitral annular disorders, and other known contraindications for the insertion of TEE were excluded from the study.

The sample size was calculated as 22 subjects per group with an assumption that standard deviation of 3 mm Hg of pulmonary capillary wedge pressure (PCWP) on the basis of observations in normal volunteers and two-sided testing with 80% power and 5% probability of type I error and keeping 20% for the margin for dropouts.

This two-arm double-blind randomized control trial was conducted after obtaining institutional ethical committee clearance and written informed consent from all the patients with CTRI registration wide no. CTRI/2022/02/040134. Patients were randomly allocated by computer-generated randomization into group 1 (received inj. levosimendan 12 mcg/kg over 10 minutes after induction followed by 0.1 mcg/kg/min intraoperatively) and group 2 (received inj. milrinone 50 mcg/kg over 10 minutes after induction followed by 0.5 mcg/kg/min in the intraoperative period) by sealed envelope technique [Figure 1].

Figure 1.

CONSORT diagram

After pre-anesthetic check-up, informed consent, age, body weight, height, and baseline vital parameters of all patients were recorded. The operating room was prepared with a primed cardiopulmonary bypass (CPB) machine. A pulse oximeter and 5-lead surface ECG with automated ST segment assessment were attached. A wide bore (16 G) peripheral venous access was done under local anesthesia. Loading dose of preoperative inj. Meropenem 30 mg/kg was infused after a proper skin test, as per institutional protocol. Radial artery catheterization was done under local anesthesia. The patients were pre-medicated with inj. midazolam 0.05 mg/kg intravenously. The patients were pre-oxygenated for 3 minutes with 100% oxygen. Inj. fentanyl citrate was injected at a dose of 5 mcg/kg followed by slow inj. thiopentone 2–3 mg/kg. After mask ventilation was ensured by capnography, neuromuscular blockade was achieved with inj. rocuronium bromide at the dose of 1.2 mg/kg, followed by mask ventilation for 1 minute. The patients were then intubated with cuffed endotracheal PVC tube of adequate size. After induction, triple lumen 7.5 Fr central venous catheterization in the internal jugular vein and 8.5 Fr sheath for Swan-Ganz pulmonary artery catheter (Edwards Lifesciences Corporation, Irvine, CA) were inserted in the internal jugular vein followed by pulmonary artery catheter (PAC) insertion. Transesophageal echocardiographic probe was inserted with a bite blocker placed in between incisors. Nasopharyngeal temperature probe insertion and Foley catheterization were done. Baseline arterial blood gas (ABG), blood sugar, and activated clotting time (ACT) were recorded.

Anesthesia was maintained with O₂: N₂O (1:1) with isoflurane at the rate of 1 MAC, midazolam, fentanyl, and vecuronium bromide. As a part of goal-guided fluid therapy, the target mean arterial pressure (MAP) of >70 mm of Hg, urine output >1.0 ml/kg/h, SpO₂ >95%, hematocrit >30, cardiac index >2.5 L/min/m², stroke volume variation <10%, and systemic vascular resistance index >1500–2500 dynes/s/cm⁵/m² were maintained. Inj. unfractionated heparin was injected at a dose of 100 international units (I.U.)/kg before grafting to maintain ACT 250–300 seconds. Inj. heparin was injected with half of the initial dose every hour to maintain optimum anticoagulation. Following grafting, anticoagulation was reversed with inj. protamine sulfate at a dose of 1.3 mg per 100 units of heparin.

After 10 minutes of induction, study drugs in both groups were initiated at a predetermined infusion rate. PCWP was noted intermittently by inflating the balloon of PAC and wedging at six predetermined stages, that is, after sternotomy, during grafting of reverse saphenous venous graft (RSVG) in the proximal aorta, grafting of the left anterior descending artery (LAD), obtuse marginal artery (OM), posterior descending artery (PDA), and before sternal closure. Transesophageal echocardiography was done at all six stages.

Mitral early diastolic inflow velocity (E) and early diastolic mitral annular velocity (e’) were recorded with pulse wave (PW) Doppler at mid-esophageal long axis view and mitral annular tissue Doppler at mid-esophageal four-chamber view. “E” wave velocity was obtained by aligning the PW cursor parallel to the mitral valve inflow and placing the sample volume at the mitral valve leaflet tips. The peak of the first negative wave in the mitral inflow velocity waveform in the diastolic phase was taken as “E.” Early diastolic mitral annular tissue Doppler velocity (e’) is obtained by putting pulse wave tissue Doppler cursor at the mitral annulus. The peak of the first positive deflection in the mitral annular velocity waveform was taken as e’. E/e’_avg was calculated using the average of both septal and lateral mitral annular e’ as the denominator. LAP was estimated using the Nagueh formula: PCWP/LAP = 1.24 × E/e’_avg + 1.9.[19]

Protocol for additional inotrope was decided as to start inj. adrenaline 0.05 mcg/kg/min during MAP less than 60 mm Hg, cardiac index (CI) less than 1.5 liter/min/m², and systemic vascular resistance index (SVRI) more than 1200 dyne-sec-m²/sec⁵ and inj. noradrenaline 0.05 mcg/kg/min when MAP less than 60 mm Hg, CI more than 1.5 liter/min/m², and systemic vascular resistance index (SVRI) less than 1200 dyne-sec-m²/sec⁵ along with fluid bolus, if needed. The study drug was stopped temporarily when MAP went below 50 mm of Hg and restarted again when MAP was raised to more than 60 mm Hg. Hypotension was managed by trendelenburg positioning along with a bolus of crystalloids or colloids.

PCWP and LAP from E/e’ were recorded in six distinct stages of OPCAB in both groups. Data were summarized by the mean and standard deviation for normally distributed numerical variables and counts and percentages for categorical variables by normality testing through Kolmogorov–Smirnov goodness-of-fit test using MedCalc version 19.6 software (MedCalc Software Limited, 2020). Numerical variables were compared between the groups by Student’s unpaired two-tailed t-test. Pearson’s Chi-square test was employed for the inter-group comparison of categorical variables. The statistical significance level was set at P < 0.05 for all comparisons.

Results and analyses

All patients were grafted in LAD with the left internal mammary artery (LIMA), OM, and PDA with RSVG. PCWP and MV E/e’ could be recorded in all patients.

The mean and standard deviation of the demographic profile and baseline clinical and hemodynamic characteristics are summarized in Table 1A and B.

Table 1.

Comparison of demographic profile and baseline clinical and hemodynamic parameters between two groups

Patient demographics
Variables	Group 1 (Mean±SD)	Group 2 (Mean±SD)	Significance (P)
Age (years)	58±4.61	60.05±4.61	0.105
Sex (Proportion)	17/5 (M/F)	18/4 (M/F)	0.708
Body Weight (kgs)	68±82	67.32±4.29	0.449
Height (meters)	1.67±0.09	1.69±0.06	0.268
BSA (m2)	1.78±0.13	1.77±0.08	0.909
BMI (Kg/m2)	24.88±3.39	23.52±1.55	0.094
Beta Blockers use	22/22	22/22	1.000
METS	8.05±2.44	6.81±2.63	0.0546
Baseline parameters
Parameters	Group 1 (Mean±SD)	Group 2 (Mean±SD)	Significance (P)
Heart Rate (bpm)	67.68±5.94	69.95±3.79	0.138
Systolic BP (mm Hg)	139.41±8.90	142.46±6.47	0.201
Diastolic BP (mm Hg)	83.09±7.60	84.41±5.40	0.511
Mean Pressure (mm Hg)	94.68±8.43	97.09±6.66	0.299
Saturation (%)	99.55±0.67	99.82±0.50	0.134
Respiratory rate (per min)	15.77±2.39	17.18±2.67	0.072
Temperature (°C)	36.51±0.45	36.50±0.31	0.891
Blood sugar (mg/dl)	95.27±7.60	99.18±8.52	0.116
Ejection Fraction (%)	43.48±10.72	44.90±10.95	0.403
MPAP (mm Hg)	24.29±5.48	23.33±5.69	0.204
PCWP (mm Hg)	13.33±3.99	15.05±4.38	0.068
Grade of DD	1.57±0.60	1.29±0.528	0.064
E/e’	11.16±2.42	10.87±2.56	0.271

Group 1=Levosimendan; Group 2=Milrinone; SD=Standard deviation; BSA=Body surface area; BMI=Body mass index; METS=Mechanical equivalent of task score; BP=Blood pressure; MPAP=Mean pulmonary artery pressure; PCWP=Pulmonary capillary wedge pressure; DD=Diastolic dysfunction; E=Mitral early diastolic inflow velocity; e’=Mitral annular (average) early diastolic velocity

After analyses of data with Student’s unpaired, two-tailed t-test, both the groups were comparable in terms of demographic profile and baseline preoperative hemodynamic parameters as given in Table 1A and B.

The patients in group 1 (levosimendan) demonstrated a statistically significant decrease in LAP at most of the stages of OPCAB during the grafting of coronary vessels. However, no statistically significant difference in group 1 and group 2 was observed after the sternotomy. During proximal RSVG grafting, group 1 showed a statistically significant decrease in PCWP, the primary outcome parameter (P = 0.047), but differences in echocardiographic parameters in both groups were not statistically significant. [Tables 2 and 3; Figures 2 and 3] There was no statistical difference between the groups in terms of duration of grafting, blood loss, intake, and output and conversion to CPB [Table 2].

Table 2.

Comparison of intraoperative variables between two groups

Variables	Group 1 (Mean±SD)	Group 2 (Mean±SD)	Significance P
LIMA-LAD duration (min)	18.99±5.41	19.74±5.62	0.519
Proximal RSVG duration (min)	23.84±5.56	22.72±4.98	0.057
RSVG-OM duration (min)	20.25±5.13	19.8±5.04	0.610
RSVG-PDA duration (min)	20.48±5.03	21.60±5.03	0.124
Total grafting duration (min)	83.55±18.63	83.87±18.46	0.842
Blood loss (ml)	744.62±227.99	692.14±202.25	0.282
Total intake (ml)	1107.38±244.95	1139.81±253.80	0.147
Total Output (ml)	1271.71±274.51	1299.24±281.79	0.075
Conversion to CPB	0	0	1

LIMA=Left internal mammary artery; LAD=Left anterior descending artery; RSVG=Reverse saphenous venous grafting; OM=Obtuse marginal artery; PDA=Posterior descending artery, CPB=cardio pulmonary bypass; SD=Standard deviation

Table 3.

PCWP and LAP changes over stages of surgery

Stages of OPCAB	Mean PCWP (mm Hg)		Mean LAP from MV E/e’ (mm Hg)
After sternotomy	Group 1	13.09±3.02	Group 1	13.2±0.45
	Group 2	14.22±2.98	Group 2	13.66±1.28
	P=0.216		P=0.118
LAD grafting	Group 1	12.14±1.86	Group 1	13.19±1.30
	Group 2	13.55±1.95	Group 2	14.60±1.49
	P=0.018		P=0.002
Proximal RSVG	Group 1	12.68±1.94	Group 1	13.21±0.97
	Group 2	13.86±1.88	Group 2	13.81±2.66
	P=0.047		P=0.329
OM grafting	Group 1	13.09±3.02	Group 1	13.2±0.45
	Group 2	13.14±1.75	Group 2	13.19±1.30
	P=<0.001		P=0.002
PDA grafting	Group 1	15.59±3.46	Group 1	15.15±2.73
	Group 2	18.41±0.03	Group 2	16.60±1.92
	P=0.028		P=0.048
Before sternal closure	Group 1	10.73±2.93	Group 1	11.52±0.87
	Group 2	12.5±1.47	Group 2	12.18±0.92
	P=0.015		P=0.019

Figure 2.

Figure showing change in PCWP over time. Group 1 = levosimendan, Group 2 = milrinone, PCWP = pulmonary capillary wedge pressure, LIMA = left internal mammary artery, LAD = left anterior descending artery, RSVG = reverse saphenous venous grafting, OM = obtuse marginal artery, PDA = posterior descending artery

Figure 3.

Figure showing change in LAP (E/e’) over time. Group 1 = levosimendan, Group 2 = milrinone, LAP = left atrial pressure, E = mitral early diastolic inflow velocity, e’ = mitral annular early diastolic velocity, LIMA = left internal mammary artery, LAD = left anterior descending artery, RSVG = reverse saphenous venous grafting, OM = obtuse marginal artery, PDA = posterior descending artery

DISCUSSION

The current study is novel in the aspect that it has compared levosimendan with milrinone at different stages of grafting in OPCAB, and in different patient positions to achieve this grafting, the authors have found a statistically significant decrease in LAP at all stages of OPCAB except immediately after sternotomy. [Table 3] Positioning during OPCAB to graft different culprit vessels causes significant distortion of cardiac chambers and valves. The use of coronary stabilizers and stay sutures aggravate these changes further. Increased LVEDP jeopardizes coronary perfusion pressure and cardiac output. Levosimendan and milrinone were studied which could maintain favorable hemodynamics and decreased LAP at different stages of grafting in OPCAB.

Desai et al.[20] found that preoperative levosimendan in OPCAB reduced the incidence of low cardiac output syndrome, postoperative atrial fibrillation, conversion to CPB, and requirement of an intra-aortic balloon pump. A study conducted by Kandasamy et al.[21] compared levosimendan and dobutamine in OPCAB which showed levosimendan maintained better cardiac contractility indices and decreased PCWP, in comparison with dobutamine. Similarly, the authors have found a statistically significant decrease in LAP with levosimendan compared with milrinone.

Apart from increased inotropy and lusitropy, milrinone has a direct vasodilating effect on arterial coronary bypass grafts[22] and increased the grafted mammary artery flow after coronary artery bypass grafting.[23] When performing OPCAB, the administration of milrinone during the operation has been shown to minimize the fall in cardiac index and mitral regurgitation.[24] Jo et al.[25] studied prophylactic milrinone in OPCAB and found improved cardiac index, stroke volume index, and mixed venous oxygen saturation during OM anastomosis and reduced need for high-dose dopamine in patients receiving atenolol. But the study conducted by Song et al.[26] did not find any statistically significant increase in contractility and hemodynamics in OPCAB.

There was a statistically significant decrease in LAP at almost all stages of OPCAB except immediately after sternotomy in both the study groups but levosimendan stands out to be a superior choice as primary inotrope than milrinone to maintain decreased LAP which is desirable to maintain coronary blood flow dynamics even at adverse positions of grafting of culprit coronary vessels.

The decrease in PCWP is always desirable in cardiac patients as it indirectly reflects the LVEDP and therefore the left ventricular wall stress, jeopardizing subendocardial perfusion. In OPCAB, decreased wall stress aids in better coronary perfusion. This study shows the importance of use of levosimendan and milrinone to decrease the LAP in OPCAB.

Limitations

In OPCAB, the handling of the heart by the surgeon also determines the LAP which might have confounded the results of the present study. In this study, all the surgeries were not done by a single surgeon.

CONCLUSION

Levosimendan may be used as a primary inotrope in terms of better reduction in left atrial pressure during different stages of OPCAB, translating to a decrease in left ventricular end-diastolic pressure, therefore maintaining optimum coronary perfusion pressure, which is the primary goal of the surgery.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.