Author + information
- Received October 7, 1996
- Revision received March 7, 1997
- Accepted March 12, 1997
- Published online July 1, 1997.
- Alan M. Mendelsohn, MD, FACCAB,* (, )
- Cary E. Johnson, PharmDB,
- Catherine E. Brown, RNC, MSNB,
- William T. Chance, PhDA and
- Robert H. Beekman III, MD, FACCAB
- ↵*Dr. Alan M. Mendelsohn, Division of Pediatric Cardiology, Children’s Hospital Medical Center, 3333 Burnet Avenue, OSB-4, Cincinnati, Ohio 45229-3039.
Objectives. This study was undertaken to evaluate the safety, efficacy and pharmacodynamic variables of oral levodopa in pediatric patients with congestive heart failure refractory to standard therapy.
Background. Therapeutic options for children with congestive cardiomyopathies are limited to digoxin, diuretic agents and angiotensin-converting enzyme inhibitors. Previous work in adults with congestive heart failure has shown a short-term effectiveness of levodopa and improvement of cardiac function.
Methods. Baseline two-dimensional and M-mode echocardiography, surface electrocardiography, Holter monitoring and exercise testing, when applicable, were performed. Levodopa was administered in a dose escalation scale from 8 mg/kg body weight per dose to 20 mg/kg per dose over 3 days with concomitant metoclopramide and pyridoxine. Catecholamine levels at initiation of the trial and throughout dose escalation were measured, with echocardiographic and electrocardiographic correlation. After 24-h drug washout, cardiac catheterization was performed both before and after administration of levodopa.
Results. Between February 1992 and December 1995, nine children (age 10 ± 1.7 years, weight 27.8 ± 4.3 kg) were enrolled in this study. At cardiac catheterization, serum dopamine levels rose from 108.5 ± 59.2 pg/ml to 1,375.8 ± 567.9 pg/ml (p = 0.03) at 100 ± 14.8 min after levodopa administration without a significant change in serum norepinephrine or epinephrine levels. Paralleling these increases, there were significant changes in the cardiac index (1.7 ± 0.3 to 3.2 ± 0.7 liters/min per m2), stroke volume index (16.1 ± 3.2 to 31.2 ± 7.0 ml/m2per min), oxygen consumption (138.6 ± 24.4 to 188.3 ± 30.8 ml/min per m2) and systemic vascular resistance (36.8 ± 8 to 21.9 ± 5.5 indexed Wood’s units; all p < 0.01). There was a significant reversal of the daily fluid volume output/input ratio from 0.8 ± 0.1 to 1.2 ± 0.1 (p < 0.01). Levodopa administration was complicated by hypertension or tachycardia, or both, requiring a dose reduction in three patients, and by significant gastrointestinal distress in one. There was sustained symptomatic improvement a median of 19.5 months after drug initiation in seven of the patients.
Conclusions. These preliminary data support the hemodynamic value of oral levodopa in the treatment of severe congestive heart failure in children.
(J Am Coll Cardiol 1997;30:237–42)
The therapeutic options for children with a congestive cardiomyopathy and congestive heart failure are limited. The mainstays of treatment include digoxin, preload reduction with diuretic agents and afterload reduction with angiotensin-converting enzyme inhibitors. In many cases, hospital admission is necessary to provide intravenous inotropic therapy such as dopamine, dobutamine and amrinone; some patients require intravenous therapy at home. Clearly, improved oral medical therapy would be beneficial for such patients.
Levodopa (l-3,4-dihydroxyphenylalanine) is a pharmacologically inert form of dopamine that is converted to its parent compound through pyridoxine-dependent aromatic amino acid decarboxylation. Previous reports have documented some utility in adults with congestive heart failure ([1–4]). Similar studies have not been attempted in the pediatric population, nor have the pharmacodynamic variables of levodopa been delineated in children. The purposes of this study were to evaluate the safety and efficacy of oral levodopa in children with congestive heart failure refractory to standard therapy and to examine its pharmacodynamics.
Patients were recruited for this study from the pediatric cardiology services of Children’s Hospital Medical Center, Cincinnati, Ohio and C. S. Mott Children’s Hospital, Ann Arbor, Michigan. The study was approved by the Institutional Review Boards of both institutions, and written, informed consent was obtained from the patients’ guardians and the patients themselves, when applicable. All patients demonstrated evidence of severe congestive heart failure clinically and echocardiographically (ventricular end-diastolic dimension >95th percentile for age, shortening fraction ≤20%). Underlying diagnoses were idiopathic congestive cardiomyopathy or ventricular dysfunction after repair of congenital cardiac lesions. In all patients, congestive heart failure was refractory to standard anticongestive therapy (e.g., digoxin, diuretic agents and afterload reduction). All patients had been hospitalized for short-term intermittent intravenous inotropic therapy or were dependent on continuous intravenous inotropic support in the hospital.
Initial two-dimensional and M-mode echocardiograms were taken to evaluate ventricular function. Surface 12-lead electrocardiograms (ECGs) and 24-h Holter monitoring were performed to investigate subclinical dysrhythmias. In patients older than 6 years of age, exercise testing (using the Naughton protocol) was performed to evaluate changes in maximal exercise tolerance times, cardiac output and ECG changes with exercise. A complete blood count, serum electrolytes, liver function studies, digoxin levels, growth hormone and baseline serum catecholamine levels were measured after assuming a rest supine position for 15 min. Concentrations of norepinephrine, epinephrine, dopamine and 3-hydroxytyrosine were determined in plasma as previously reported (). Briefly, samples (0.5 to 1.0 ml) of plasma containing N-methyldopamine as an internal standard (40 mg/ml) were passed through 100 mg of activated alumina. Catecholamines were eluted from the alumina with 250 μl of 0.1 N HCl O4, and 50 μl of each sample was injected into the high performance liquid chromatography column (Altex, reverse phase, C-18, ultrasphere). High performance liquid chromatography mobile-phase buffer (8% acetonitrile, 92% 0.1 mol/liter KH2PO4, 0.185 mmol/liter sodium octylsulfate, 0.195 mmol/liter EDTA, pH 2.95) was pumped (model 110A, Beckman-Altex, Inc.) through the system at 1.0 ml/min. The catecholamines were quantified by amperometric detection (Bioanalytical Sciences, model LC-4 Controller), with areas of catecholamine peaks being compared with known standards and quantified by integration (Hewlett-Packard, model 3396A).
1.1 Levodopa Protocol.
No dosing data for oral levodopa are available for children. In previous studies of adults with congestive heart failure, 4 to 6 g of levodopa was administered in divided doses every 6 h ([1, 3]). We therefore used a maximal dose of 80 mg/kg body weight per day, with a maximal single dose of 1 g every 6 h. Oral levodopa therapy was initiated using a 3-day dose escalation schedule to reduce gastrointestinal symptoms and to monitor hemodynamic and pharmacologic responses to intermediate doses. Pyridoxine, a necessary cofactor for peripheral conversion of levodopa to dopamine, was administered at 0.7 mg/kg per day (maximal dosage 25 mg). Metoclopramide was administered prophylactically to diminish gastrointestinal and neuropsychiatric symptoms (0.1 mg/kg per dose 30 to 60 min before each levodopa dose).
Patients were admitted to the Clinical Research Center of each institution to begin the levodopa trial. Doses were increased from 40 mg/kg per day (“initial dose”) in four divided doses to 60 mg/kg per day (“moderate dose”) on the second day, and finally to 80 mg/kg per day (“maximal dose”) on the third day, as tolerated. Patients who could not tolerate moderate dosages were maintained on the initial dose and those who could not tolerate the maximal dosage were maintained on moderate dosages. After the fourth dose of each dosing schedule, serum catecholamine levels were obtained 0, 30, 60, 90, 120, 240 and 360 min after levodopa administration. At these same time points, heart rate, blood pressure, cardiac output (acetylene rebreathing) and echocardiographic data were obtained.
1.2 Invasive Studies.
After the dose escalation phase, levodopa therapy was withdrawn for a 24-h washout period before cardiac catheterization. Patients continued to receive pyridoxine and metoclopramide. Hemodynamic studies were performed using conscious sedation (morphine sulfate, midazolam, fentanyl). Cardiac output measurements were made by thermodilution or the Fick technique, or both, using measured oxygen consumption (MRM2 Oxygen Consumption monitor, Waters Instruments). Epinephrine, norepinephrine and dopamine concentrations were measured at baseline. Levodopa was then administered orally in a single dose at the maximal amount tolerated in the dose escalation phase (8 to 10, 15 or 20 mg/kg, maximum of 1 g). Hemodynamic measurements and blood catecholamine levels were obtained 30, 60, 90, 120 and 150 min after drug administration. Angiography was performed as indicated after the hemodynamic study. Catecholamine concentrations were also obtained after cardiac catheterization 240 and 360 min after drug administration. In patients with hemodynamic improvement demonstrated in the early part of the study (defined by an increase in cardiac index or stroke volume of ≥10%), maintenance oral levodopa therapy was reinstituted at the maximal dose tolerated in the dose escalation phase. The patients were monitored for significant changes in heart rate, blood pressure, clinical status and arrhythmias by repeat Holter monitoring for 24 h. They were sent home on oral levodopa, pyridoxine and metoclopramide, and asked to return to the Pediatric Cardiology Clinic in 1 month.
1.3 Data Analysis.
Data are presented as mean value ± SEM. Comparative data were analyzed by repeated measures analysis of variance, and statistical significance was accepted at p ≤ 0.05.
Between February 1992 and December 1995, nine children were enrolled in this study (Table 1). Patient age was 10.0 ± 1.7 years (range 2 to 18) and patient weight was 27.8 ± 4.3 kg. Four patients had underlying diagnoses of idiopathic dilated cardiomyopathy and five had ventricular dysfunction after surgical treatment for congenital heart disease.
2.1 Baseline Variables.
All nine patients were maintained on digoxin (9 to 10 μg/kg per day), furosemide (3 to 6 mg/kg per day) and spironolactone and captopril therapy (2.5 to 5 mg/kg per day). Echocardiographic study revealed diminished shortening fractions of 12.6 ± 1.8% (6% to 20%). The cardiac index at baseline was 1.2 ± 0.3 liters/min per m2(range 1.0 to 1.8). Eight patients were in supraventricular rhythm at study initiation; one patient was ventricularly paced at 100 beats/min due to complete heart block. One patient had frequent multiform premature ventricular contractions. Total exercise time in six patients was 1.5 ± 1.1 min (range 0 to 3.6).
2.2 Levodopa Administration (Table 1).
Eight of the nine patients tolerated initiation of levodopa at 40 mg/kg per day without gastrointestinal or neurologic symptoms. In one patient (Patient 2), increasing ventricular arrhythmias and hypertension led to dose reduction to 32 mg/kg per day, which was tolerated well. Levodopa was subsequently increased to 50 to 60 mg/kg per day in the remaining eight patients, and further increased to 75 to 80 mg/kg per day in six patients.
Initial clinical responses are outlined in Table 2. The time to maximal clinical response (reflected by maximal cardiac output) varied between 78 ± 15.3 min (at 15 mg/kg per dose) and 90 ± 11 min (at 10 mg/kg per dose) after drug administration. There was no significant increase in heart rate or systolic blood pressure at any dosage level when compared with baseline. There was a marked increase in the acetylene rebreathing cardiac index from 1.2 ± 0.3 liters/min per m2to 4.2 ± 0.9 liters/min per m2(p = 0.05). Ventricular shortening fraction measured by echocardiography did not change significantly. Total exercise tolerance time increased from 1.5 ± 1.1 to 10 ± 2.4 min (p < 0.05). There was also a reversal in the urine output/fluid intake ratio, particularly at higher doses. There was an increase in serum dopamine (7.3 ± 3.8 vs. 289.4 ± 93.6 pg/ml, p < 0.02) without a significant change in epinephrine (4.6 ± 1.9 vs. 2.7 ± 1.7 pg/ml) or norepinephrine (3.3 ± 1.3 vs. 2.6 ± 1.0 pg/ml). There were no changes in serum electrolytes, liver function tests or growth hormone levels.
2.3 Untoward Clinical Effects.
During the dose escalation phase, three patients developed systemic systolic and diastolic hypertension, sinus tachycardia, palpitations and diaphoresis 30 to 60 min after levodopa administration. Symptoms resolved within 30 min. These episodes occurred in one patient after initiation of the 20-mg/kg dose, in one patient after the 15-mg/kg dose and in one patient after the 10-mg/kg dose requiring diminution of the dose to 8 mg/kg. In one patient, significant gastrointestinal discomfort and emesis required diminution in the dosage from 20 mg/kg per dose to 15 mg/kg per dose. There were no observed neuropsychiatric changes.
2.4 Cardiac Catheterization Data.
After the 24-h washout period, seven of the nine patients underwent cardiac catheterization for hemodynamic evaluation of the immediate response to oral levodopa administration. In the remaining two patients, vascular access was not available in one and consent for catheterization was refused in the other patient. There was an 88% increase in the cardiac index from 1.7 ± 0.3 to 3.2 ± 0.7 ml/min per m2(p < 0.01) (Fig. 1), which occurred at 100 ± 14.8 min after levodopa administration. The stroke volume index increased from 16.1 ± 3.2 to 31.2 ± 7.0 ml/m2per beat (p < 0.01) at peak effect (Fig. 2). These changes were accompanied by an increase in oxygen consumption from 138.6 ± 24.4 to 188.3 ± 30.8 ml/m2(p < 0.01) (Fig. 3). Systemic vascular resistance decreased from 36.8 ± 8.0 to 21.9 ± 5.5 indexed Wood’s units (p < 0.01) (Fig. 4), whereas pulmonary vascular resistance decreased from 5.5 ± 1.0 to 3.1 ± 0.6 indexed Wood’s units (p = 0.07). There was no significant change in central venous pressure (15.7 ± 3.4 vs. 15.2 ± 2.2 mm Hg), systemic ventricular end-diastolic pressure (15.7 ± 2.8 vs. 15 ± 3.8 mm Hg), heart rate (105.2 ± 3.7 vs. 101.2 ± 4.8 beats/min) or mean arterial pressure (69 ± 8.1 vs. 78.2 ± 5.1 mm Hg).
The serum dopamine level increased from 108.5 ± 59.2 pg/ml to 1,375.8 ± 567.9 pg/ml (p = 0.03) at 90 ± 16 min after levodopa administration. There was no significant change in the serum epinephrine (37.7 ± 36.5 vs. 1.1 ± 1.1 pg/ml) or norepinephrine (1.8 ± 1.0 vs. 10.6 ± 7.7 pg/ml) level. Fig. 5and Fig. 6, derived from the catecholamine and catheterization data of Patient 2, are representative of the correlation between serum levodopa and dopamine levels and cardiac index. As seen in Fig. 5, levodopa levels were accompanied by lower but parallel levels of dopamine (derived from levodopa). The increases in serum dopamine correlate well temporally with increases in the cardiac index outlined in Fig. 6.
2.5 One-Month Follow-Up.
Patients returned at 1 month for follow-up of their maintenance doses of levodopa, pyridoxine and metoclopramide. Catecholamine levels (dopamine: 258.8 ± 89.6 pg/ml; epinephrine: 3.2 ± 1.4 pg/ml; norepinephrine: 2.4 ± 0.8 pg/ml), exercise tolerance (9.5 ± 1.8 min) and clinical examinations were unchanged at follow-up compared with those measured at hospital discharge. No gastrointestinal, cardiovascular or other symptoms were identified by inquiry, and metoclopramide was discontinued. Patients have continued on levodopa and pyridoxine, with incremental increases in the total dosage commensurate with increasing weight to maintain the same dose per kilogram schedule, for a median of 19.5 months (range 2 to 38). One of the nine patients (Patient 3) required intermittent hospitalization for intravenous inotropic and diuretic therapy for worsening heart failure. Two patients (Patients 2 and 7) died 12 months and 3 months after initiation of levodopa administration with symptoms of intractable congestive heart failure. The remaining patients have not required further inpatient treatment.
Our study is based on the previous work by Rajfer et al. ([1, 2]) and DeMarco et al. (), who reported on the administration of levodopa to adult patients with severe congestive heart failure. The basic protocol, including dose escalation, was patterned after these studies. In contrast to the adult studies, however, we administered pyridoxine and metoclopramide prophylactically to diminish clinical side effects. In 10 adult patients, Rajfer et al. () demonstrated a 33.3% increase in the cardiac index (1.8 ± 0.1 to 2.4 ± 0.2 liters/min per m2) and a 28.6% increase in the stroke volume index (21 ± 2 to 27 ± 2 ml/beat per m2) (both p < 0.01) approximately 1 h after levodopa administration. There was no change in mean arterial pressure, heart rate, left ventricular filling pressure or central venous pressure. Systemic vascular resistance decreased 20.5%, from 23.8 ± 1.4 to 18.9 ± 1.5 Wood’s units (p < 0.01). There were significant increases in serum levodopa and dopamine levels (p < 0.01). Continued improvement in the cardiac index was reported in five patients followed up for 6.8 ± 1.7 months. In 17 adults, DeMarco et al. () confirmed the findings of Rajfer et al., but also reported a significant increase in norepinephrine levels at peak levodopa effect. Our clinical results are similar. At cardiac catheterization, we identified an 88% increase in the cardiac index, a 94% increase in the stroke volume index and a 40% decrease in systemic vascular resistance after levodopa administration. No changes were detected in ventricular filling pressure, mean arterial pressure or heart rate. Serum dopamine levels increased without significant changes in epinephrine or norepinephrine levels. Because elevations of epinephrine and norepinephrine have been suggested as promoting beta-receptor downregulation in chronic congestive heart failure ([6–8]), these findings suggest that levodopa may be advantageous for the oral treatment of congestive heart failure in children.
We surmise from our data that the dopamine levels achieved in this study may be responsible for stimulation of the dopamine receptor, type 1 (DA1), as well as for low grade beta1receptor effects (). This is supported by the observation that fluid intake/output ratios, and therefore renal perfusion, appear to improve after initiation of levodopa. Echocardiographic follow-up data of load-dependent ejection variables (e.g., shortening fraction) demonstrated less striking improvement in ventricular function compared with the changes documented at cardiac catheterization. However, serial determinations of acetylene rebreathing cardiac output, exercise testing and, anecdotally, load-independent echocardiographic variables (e.g., velocity of circumferential fiber shortening [corrected]/wall stress) did yield better comparative, noninvasive data. We believe that stroke volume and cardiac output may increase in a dilated ventricle without detection of significant dimension changes by echocardiography because of decreases in left ventricular end-systolic wall stress, afterload reduction and improvements in oxygen transport secondary to DA1and beta1receptor stimulation by dopamine ([4, 9]). In addition, data on the effects of levodopa and dopamine on contractility in the right ventricle are limited ().
Our study is limited by the small patient group involved. This limits the statistical power to detect significant clinical side effects. The data are promising, however, and suggest that levodopa may provide beneficial oral therapy for children with severe, refractory congestive heart failure. Further study is necessary to determine the long-term clinical utility of oral levodopa in the pediatric population. Randomized, blinded studies of the differential effects of levodopa versus placebo in this patient population are planned.
We thank Jo Riddell, Margie DeHo and Julie Schuster for editorial assistance.
☆ This study was supported by U.S. Public Health Service Grant M01 RR 08084 from the General Clinical Research Centers Program, National Center for Research Resources, National Institutes of Health, Bethesda, Maryland and was presented in part at the 68th Annual Scientific Sessions of the American Heart Association, Anaheim, California, November 1995.
- dopaminergic receptor, type 1
- electrocardiogram, electrocardiographic
- Received October 7, 1996.
- Revision received March 7, 1997.
- Accepted March 12, 1997.
- The American College of Cardiology
- Rajfer SI,
- Rossen JD,
- Nemanich JW,
- Douglas FL,
- David F,
- Osinski J
- Rajfer SI,
- Corow KM,
- Lang RM,
- Neumann A,
- Carroll JS
- Dreyer WJ,
- Fisher DJ