Preoperative Ephedrine Attenuates the Hemodynamic Responses of Propofol During Valve Surgery: A Dose Dependent Study

One hundred fifty ASA III-IV patients aged 18-55 years scheduled for elective valve surgery were included in this randomized double blinded placebo-controlled study at the author's center after obtaining of approval of the local ethical committee and a written informed consent from the participants. The sample size was determined by a prior power analysis indicated that 27 patients in each group would be a sufficiently large sample size to be adequate to detect a 20% changes in SVRI values, with a type-I error of 0.05 and a power of approximately 85%. We added 10% more patients to account for patients dropping out during the study. All operations were performed by the same surgeons. Participants were allocated randomly to five groups (n=30 for each) to receive saline [group 1] or ephedrine 0.07, 0.1 or 0.15 mg/kg [group 2, 3, and 4, respectively]and phenylephrine 1.5 ug/kg [group 5] 1 min before induction of anesthesia.

Patients with documented un-controlled hypertension, ischemic heart disease, left ventricular ejection fraction less than 45%, peripheral vascular disease, thyrotoxicosis, neurological, hepatic, and renal diseases, pregnancy, re-do or emergency surgery, allergy to the study medications, those requiring preoperative inotropic, vasopressor or mechanical circulatory or ventilatory support, and those who had electrocardiograph (ECG) characteristics that would interfere with ST segment monitoring, included baseline ST segment depression, left bundle-branch block, atrial fibrillation, left ventricular hypertrophy, digitalis effect, QRS duration >0.12 s, as well as pacemaker-dependent rhythms, were excluded from the study.

All routine medications except angiotensin-converting enzyme inhibitors were continued until the morning of the operation. All patients were premedicated with 0.03 mg/kg IV midazolam and fentanyl 1 µg/kg before invasive instrumentation. All patients were monitored with pulse oximetry, non invasive blood pressure and five leads electrocardiography (leads II and V5) (Life Scope Monitor, BSM-4113, Nihon Kohden Corp, Japan). Continuous ST segment trends were electronically measured at the J-point + 60 ms to exclude the T wave during the episodes of tachycardia. The tabulated and graphic ST segment data were reviewed by two investigators who are not involved in the study and are blinded to the patient's group for significant ischemic responses. The later were defined a reversible ST segment changes from baseline of either ≥ 1 mV ST-segment depression or ≥2 mV ST-segment elevation lasting for at least 1 minute. A radial artery catheter and a flow-directed balloon-tipped pulmonary artery catheter were placed under local anesthesia before induction. The final position of the pulmonary artery catheter tip was confirmed with portable chest x- ray film and pulmonary artery diastolic pressure > PAOP. CI was measured by thermodilution using ice cold injectate. Five measurements were performed, the lowest and highest readings were discarded, and the mean of the readings was recorded. Intravenous infusion of 5-7mL/Kg of 6% Hydroxyethyl Starch 130/0.4 (Voluven, Fresenius Kabi, Bad Hombourg, Germany) was given before induction of general anesthesia when the baseline PAOP was less than 10 mm Hg. End-tidal carbon dioxide monitoring and placement of a nasogastric tube, and rectal and nasopharyngeal temperature probes were done after induction of anesthesia.

Subjects were allocated randomly to four groups by drawing sequentially numbered sealed opaque envelopes containing a computer-generated randomization code. The subjects received intravenous injection of 0.1 mL/kg of a study solution containing either saline 0.9% solution [group 1 (n=30)], ephedrine 0.7 mg/mL [group 2 (n=30)], ephedrine 1 mg/mL [group 3 (n=30)] or ephedrine 1.5 mg/mL [group 4 (n=30)], or phenylephrine 15 mcg/mL [group 5 (n=30)]. All study solutions were injected over 1 min at 1 min before induction of anesthesia. The placebo and the ephedrine solutions were prepared in identical syringes labeled 'study drug' by the local pharmacy department before induction of anesthesia. The anesthesia providers were blinded to the study solution and were not involved in the study. All staff in the operating room were unaware of the randomization code.

Anesthesia was induced with fentanyl 5 µg/kg, propofol 2-2.5 mg/kg, and cisatracurium 0.2 mg/kg was given for muscle relaxation. After endotracheal intubation, the lungs were ventilated with a mixture of oxygen in air to maintain an arterial carbon dioxide tension at 4.5-6 kPa. Anesthesia was maintained with continuous infusions of propofol 4-6 mg/kg/ h, fentanyl 0.025 µg/kg/min, and cisatracurium 1-3 µg/Kg/ min to maintain suppression of the second twitch using a train-of-four stimulation. All patients received a slow injection of tranexamic acid 50 mg/kg before initiation of CPB. Target MAP and HR were within 20% from the mean baseline values. Hypotension (MAP ≤ 60 mm Hg ≥ 2-3 minutes) was treated with intravenous fluids; reduction of the infusion rate of propofol by 50%, or bolus doses of ephedrine 5 mg. Hypertension (MAP ≥ 20% from the mean baseline for ≥ 2-3 minutes) was treated with increasing of the infusion rate of propofol by 50%, or bolus doses of labetalol 20 mg, or nitroglycerin 0.05 mg. Tachycardia ≥20% from the baseline values for ≥1 minute was treated with the modulation of propofol infusion rate or boluses of esmolol 20 mg. Bradycardia (HR ≤ 40/min) was treated with atropine 0.5 mg.

The cardiopulmonary bypass (CPB) lines, oxygenator, and venous reservoir were primed. Heparin sodium 300 IU/kg was given after pericardiotomy to achieve celite-activated clotting time became higher than 480 s. Standard CPB technique was established with the ascending aorta cannula and the bicaval venous cannulae. During CPB, the non-pulsatile pump flow rate was 2.4 L min/ m2 using a twin roller pump and a hollow fiber membrane oxygenator, perfusion pressure was 50-80 mmHg, arterial carbon dioxide tension was 35-40 mmHg, unadjusted for temperature (α-stat), arterial oxygen tension was 150-250 mmHg, and moderate systemic hypothermia (nasopharyngeal temperature 33-34°C) was maintained. Myocardial viability was preserved with topical hypothermia and cold blood antegrade cardioplegia administered intermittently into the aortic root.

Before separation from CPB, all patients were rewarmed (nasopharyngeal temperature 37°C, bladder temperature 36°C) and epinephrine and nitroglycerine infusions were used to as needed after CPB. Heparin was neutralized after discontinuation of CPB with protamine sulfate.

After surgery propofol 1-2 mg/kg/ h was used for sedation in the ICU and morphine 0.05 µg/kg was used as needed for analgesia. Propofol infusion was discontinued and ventilator weaning was started once patients were awake and cooperative, hemodynamically stable without high doses of inotropic support, no severe arrhythmias, body core temperature >35.5°C, bleeding <100 mL/h, urine output> 0.5 mL/kg/h, and arterial oxygen tension >100 mm Hg with oxygen concentration <0.5.

Primary outcome variables include the changes in hemodynamic variables namely; HR, MAP, CI, SVRI, LVSWI, and ST segment changes. Secondary outcome variables were CVP, PAOP, RVSWI, and troponin I changes, and the need for vasoactive drugs. Anesthesia providers were not involved in the assessment of the patients. Other anesthesiologists who were blinded to the study group and were not in the operative room performed the assessment.

HR, MAP, CI, SVI, CVP, PAOP, SVRI, PVRI, LVSWI, and RVSWI changes were recorded before (baseline), and 5 min after induction, 5, 10, 15, and 30 min after endotracheal intubation; and 15 min after sternotomy. The changes in hemodynamic data were calculated as percentages of the baseline measurements. The numbers and total time of intra-operative ischemic episodes were recorded in each group. Venous blood samples were drawn before, 3, 12, 24, and 48 hours after CPB to measure cardiac troponin I. Blood samples were centrifuged at 3,000 rpm for 10 min and serum samples stored at-20°C. Two specific monoclonal antibodies were used to avoid the cross-reactivity with human skeletal muscle for the measurement of cTnI. The upper reference limits for cTnI in a control population was 0.6 µg/L. The number of patients who received rescue doses of labetalol, ephedrine, atropine and esmolol, times from induction to intubation (I-T) and to skin incision (I-S) and all major complications (hypoxemia (SaO2<90%), arrhythmias, respiratory failure, and cardiovascular events) were recorded in each group.

Data were tested for normality using the Kolmogorov-Smirnov test. Repeated-measures analysis of variance was used for analysis of serial changes in the hemodynamic and cTnI data at different times after administration of study solution. Fisher exact test was used for categorical data. Repeated measure analysis of variance (ANOVA) was used for continuous parametric variables and the differences were then corrected by post-hoc Bonferoni test. The Kruskal-Wallis one-way ANOVA was performed for intergroup comparisons for the non-parametric values and post hoc pairwise comparisons was done using the Wilcoxon rank sum t test. Univariate analyses of the preoperative risk factor, namely EuroSCORE for the frequency of significant hypotension and ST segment changes after propofol anesthesia were performed. Univariate predictors were examined in a stepwise manner into a multivariate logistic regression model, with entry and retention set at a significance level of p < 0.05 to assess the independent impact of this risk factor on the outcome. Data were expressed as mean (SD), number (%), or median [range]. A value of p < 0.05 was considered to represent statistical significance.