 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 105 No. 6 June 2000, pp. 1260-1270
,
,,
From * Virginia Neurological Institute, University of Virginia,
Charlottesville, Virginia; Department of Pediatric Cardiology,
Indiana University Medical Center, Indianapolis, Indiana; § Department
of Pediatric Cardiology, Geisinger Medical Center, Danville,
Pennsylvania; Division of Pediatric Cardiology, University of
Florida Health Sciences Center, Gainesville, Florida; ¶ Valley
Children's Hospital, Fresno, California; # Department of Biostatistics,
University of North Carolina, Chapel Hill, North Carolina; ** Mayo
Clinic, Rochester, Minnesota; Department of Pediatrics,
Fairview-University Medical School, University of Minnesota,
Minneapolis, Minnesota; and §§ Metabolic Research and Analysis, Inc,
Fresno, California.
 |
ABSTRACT |
|---|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Objectives. A multicenter retrospective study was conducted to investigate the possible metabolic causes of pediatric cardiomyopathy and evaluate the outcome of patients treated with L-carnitine.
Methods. Seventy-six patients diagnosed with cardiomyopathy were treated with L-carnitine in addition to conventional cardiac treatment, and 145 patients were treated with conventional treatment only. There were 101 males and 120 females between 1 day and 18 years old. Cardiomyopathy diagnoses included dilated (148 patients), hypertrophic (42 patients), restrictive (16 patients), mixed diagnosis (11 patients), and 4 with an unknown type. Of 76 L-carnitine-treated patients, 29 (38%) had evidence to suggest a disorder of metabolism, and of 145 control patients, 15 (10%) were suspected to have a disorder of metabolism. These metabolic disorders were thought to be the cause for the cardiomyopathy of the patients. The duration of L-carnitine treatment ranged from 2 weeks to >1 year. Information was collected on length of survival (time-to-event), clinical outcome, echocardiogram parameters, and clinical assessments. Data were collected at intervals from baseline to study endpoint, death, transplant, or last known follow-up visit.
Results. L-Carnitine-treated patients were younger than control patients and had poorer clinical functioning at baseline, yet they demonstrated lower mortality and a level of clinical functioning and clinical severity comparable to control patients on conventional therapy by the end of the study. An analysis of the interaction between clinical outcome and concomitant medications unexpectedly revealed that the population of patients treated with angiotensin-converting enzyme (ACE) inhibitors (40% of patients) had significantly poorer survival (although their greater likelihood for poor survival may possibly have made them more likely to receive ACE inhibitors).
Conclusion. Results suggest that L-carnitine provides clinical benefit in treating pediatric cardiomyopathy. There is a need for further exploration of potential explanatory factors for the higher mortality observed in the population of patients treated with ACE inhibitors. Key words: pediatric, cardiology, cardiomyopathy, L-carnitine, metabolism, genetics.
Cardiomyopathy is a serious disease with a poor prognosis
and high mortality. Estimates of incidence vary widely but range between 2 to 8 cases per 100 0001 to 17.2 per
100 0002 among all age groups. In the first year of life,
the incidence has been estimated at 1 in 10 000 live
births3 and ~5000 cases are reported annually for the
pediatric population. Survival rates in children remain dismal; despite
treatment, a review of reports in the literature4-13 finds a median mortality of ~37% with even poorer survival noted for
older children.
Conventional approaches to treatment, especially in older children and
for the more intractable forms of cardiomyopathy, are often only
palliative. The development of effective treatments has been hampered
by a poor understanding of the underlying cellular and molecular
biology of the disease. Recent advances, however, have begun to yield
some important clues into the nature of myocardial disease, including
contributions from genetic and metabolic studies. With these insights
have come implications for treatment based on a better understanding of
the pathogenesis of heart failure.
The objective of this study was to investigate the treatment of certain
cardiomyopathy-associated metabolic disorders using levocarnitine
(L-carnitine), the fatty acid shuttle of the cell. An
evolving literature describes the integral role that
L-carnitine plays in myocardial metabolism. Cardiomyopathy
is an associated symptom in metabolic disorders, where the
intramitochondrial accumulation of toxic organic acid intermediates
leads to the depletion of L-carnitine stores, such as in
fatty acid oxidation defects and mitochondrial disorders.
L-Carnitine conjugates with organic acid intermediates and
serves to remove them from the mitochondria with excretion in the
urine. Therefore, cardiomyopathy-associated metabolic disorders have
seemed to be reasonable candidates for treatment with
L-carnitine.
Therefore, we have had great interest in evaluating clinical data from
patients treated for pediatric cardiomyopathy, who have received
conventional therapy in comparison to patients who were treated for a
possible underlying metabolic cause. A group of pediatric cardiologists
and metabolic specialists were identified who had used
L-carnitine over a 10-year period to treat several forms of
cardiomyopathy. The present study is a retrospective evaluation of that
collective experience.
A brief review of the cause and pathogenesis of cardiomyopathies and a
survey of recent advances in the molecular biology and genetics of
these disorders are instructive. Here, we review the highlights of
these developments as a basis for discussing the role that
L-carnitine plays in myocardial energy metabolism and its
therapeutic potential.
Classification
The definition of cardiomyopathy has undergone many
changes.14 In the strictest sense, cardiomyopathy refers
to primary myocardial disease of unknown origin, so called idiopathic cardiomyopathy. In the broader sense, the term includes specific heart
muscle diseases resulting from known disorders or causes (eg,
hypertensive and metabolic). An expanding classification scheme
continues to organize by cause and morphology. The classification system originally proposed by Goodwin et al15 and
subsequently adopted and modified by the World Health
Organization/International Society and Federation of
Cardiologists16 reflects a distinctly empirical approach;
the division between primary and secondary
cardiomyopathiesa was created by
recognizing cause as essentially morphologic (ie, unknown cause;
I-III) or physiologic (ie, known cause; IV-XI) in nature (Table
1). Dilated, hypertrophic, and
restrictive cardiomyopathies (the 3 primary cardiomyopathies) have been
classified morphologically. The remaining cardiomyopathies are divided
according to putative cause.
TABLE 1
OVERVIEW OF THE CARDIOMYOPATHIES
Current Classification of Cardiomyopathies
General Pathophysiologic and Morphologic Features of Myocardial Disease
In general, myocardial disease is characterized by at least some histomorphologic derangement and diminished compliance of the ventricular myocardium, usually the left ventricle, although diffuse, multichamber involvement is not infrequent. Histopathologic studies often find some form of fibrosis, diffuse loss of myocytes, or myocyte and myofibrillar disarray. Nonspecific findings often complicate diagnosis. The 3 forms of primary disease differ mostly in terms of type and location of morphologic remodeling and resulting hemodynamic dysfunction.
The distinguishing feature of dilated cardiomyopathy, accounting for >90% of all cases, is a notably enlarged ventricle (left ventricle) or ventricles, disproportionately thinner septal and free wall thickness, and diminished myocardial contractility. In the diffuse form, all chambers are involved, whereas in the nondiffuse form, 1 or more chambers remain uninvolved.
Hypertrophic cardiomyopathy is characterized by a notably exaggerated ventricular mass (ie, often 3 times greater than normal for body size) with a typically disproportionate interventricular septum. The ventricular chamber is stiff and noncompliant, and the volume of the ventricular cavities (characteristically the left ventricle) is grossly diminished. Atria, however, may be dilated.
Restrictive cardiomyopathy, the least prevalent form, is characterized by diffuse fibrosis and rigidity of the endocardium and subendocardium resulting in severely decreased atrial and ventricular compliance. The ventricular chamber may be reduced in volume and compensatory hypertrophy may be present.
Multidetermined Nature of Myocardial Disease
Currently, the diagnosis of cardiomyopathy as idiopathic or as a specific heart muscle disease depends almost entirely on whether an explicit cause is found. If the cause remains indeterminate after clinical evaluation and testing, a diagnosis of primary myocardial disease is made. Such empirical diagnoses have primarily been responsible for the current dual classification system. It is evident from the literature, however, that an evolution in understanding of myocardial disease is emerging.
Confounds to Diagnosis
In the past, a diagnosis of specific heart muscle disease implied a single cause. This idea has generally been discounted.14,17 A clinical finding that leads to a diagnosis of specific heart muscle disease will frequently stop the search for additional contributory factors.17 A diagnosis of toxic cardiomyopathy, for example, will stop the investigation from finding a gene defect, viral cause, or underlying congenital abnormality. Diagnosis is made even more difficult because there is often no direct relationship between pathogenesis and myocardial condition.18 There are >100 specific disease conditions identified17 producing cardiomyopathy, each sharing clinical and morphologic features, often indistinguishable.
Mechanisms of Myocardial Dysfunction
With idiopathic disease, it is important to consider the complex interrelationships among acquired, heritable, and environmental factors. Lethality and symptom severity vary according to the impact of a gene defect, intensity of an environmental stress, and severity of an acquired disease. A variety of disorders are known to manifest only when several factors interact in concert. Organic acidemias, fatty acid oxidation defects, and urea cycle defects are examples of heritable, occasionally benign disease states in which symptoms may manifest only during catabolic crisis, as during acute infection. Diet can play a role, where the quantity or mix of amino acids, carbohydrate, or lipid consumption can overload a partially defective metabolic pathway.
The myocardium is similarly susceptible to a spectrum of inherited,
acquired, and environmental factors. Mutations in the genes coding for
-myosin heavy chain,19 troponin T, and
-tropomysin are associated with familial hypertrophic
cardiomyopathy.20 Familial restrictive cardiomyopathy have
been identified.21,22 Such instances of heritable forms of
cardiomyopathy likely effect the contractile apparatus directly; in
-myosin heavy chain mutation, a myosin head malfunction is thought
to underlie myocardial dysfunction.
Heritable disorders of metabolism with myocardial involvement include glycogen storage disease type 2 (Pompe's disease), glutaric aciduria type II, the fatty acyl-CoA dehydrogenase deficiencies, carnitine membrane transport defects, and carnitine acyl translocase and transferase defects. Mitochondrial DNA mutations have been identified in some patients with idiopathic dilated cardiomyopathy.18 These disorders result in impaired myocardial energy metabolism. Many nonmetabolic genetic disorders also have associated cardiomyopathy, including Noonan's syndrome, muscular dystrophies, and some chromosomal anomalies.23
Viral myocarditis is believed to precede some cases of dilated cardiomyopathy.24,25 Although controversial, there is some evidence that enteroviruses causing myocarditis may eventually lead to dilated cardiomyopathy.26 Much of the evidence is indirect and suggests that dilated cardiomyopathy develops secondary to the acute enteroviral syndrome, the enterovirus, therefore, having an indirect role. Recent advances in molecular biology, however, have uncovered the presence of enteroviral genomes in the myocardium of patients with idiopathic dilated cardiomyopathy.27,28 Other studies, however, have found enteroviral DNA in comparable proportions of control subjects29 or have not found enteroviral DNA in dilated cardiomyopathy patients at all.30 Although there is no clinical or prognostic importance in these findings, it would seem unlikely if enteroviral presence in the myocardium proved benign.
There is considerable interest in the idea that autoimmune disorders may play a significant role in dilated cardiomyopathy, particularly in idiopathic disease. There is suggestion that a chronic autoimmune response may develop in the wake of infectious myocarditis and that dilated cardiomyopathy results from the long-term autoimmune state.24 There is also evidence that acute myocarditis may be an autoimmune disease31 because cytolytic cardiac autoantibodies have been detected32 in inflammatory disease. These autoantibodies act specifically on sarcolemma and myolemma.
Between 30% and 40% of patients with dilated cardiomyopathy have cardiac autoantibodies, and for the remainder of patients, it is believed that autoimmunity does not play a role. Although putative causes are often ascribed to cardiomyopathy, it is unlikely that any single pathogenic culprit is singly responsible.
Energy Production in the Heart
Myocardial contractility depends primarily on lipid
metabolism
the oxidation of fatty acidsin the mitochondria to
provide energy (adenosine 5'-triphosphate [ATP]); less important
sources of fuel include glucose and lactate. Fatty acid oxidation is a complex process involving ~20 steps and 18 enzymes. The major steps
of mitochondrial energy production include uptake of fatty acids, the
carnitine cycle, -oxidation, Krebs cycle, and oxidative phosphorylation to ATP.34 Impairments to the carnitine cycle and -oxidation will be considered here, with an emphasis on
defects that lead to L-carnitine deficiency.
Other defects in mitochondrial functioning, ie, in the Krebs cycle, oxidation/phosphorylation coupling, and electron transport chain, although essential to energy production, are beyond the scope of this review and excellent reviews are available elsewhere.35 Also, because the myocardium depends only marginally on glycogenolysis and glycolysis, disorders of these pathways will not be considered, nor will overall diagnostic considerations or supportive measures because these are likewise presented elsewhere in definitive texts.35
Significance of L-Carnitine Deficiency
The significance of L-carnitine lies in its primary
role of shuttling fatty acids across the mitochondrial membrane
delivering them for -oxidation and the production of energy (ATP).
Clinical manifestations secondary to carnitine deficiency syndromes
have been well-described36-38 and almost always impact
muscle tissue. Deficiency of L-carnitine, depending on severity, results in accumulation of lipid in muscle, muscle myopathy, and weakness and can involve the myocardium.
L-Carnitine is a natural substance obtained from diet and endogenous synthesis from muscle protein degradation with the final step of synthesis being hepatic. Muscle carnitine exists in a free form and an esterified form (bound to organic acids removed from mitochondria). Carnitine deficiency can be primary attributable to a recessively inherited defect in muscle transport of carnitine or secondary attributable to decreased availability of free carnitine with many causes. These secondary deficiencies can be attributable to decreased dietary intake (as with TPN), decreased absorption (as with cystic fibrosis), increased loss (as with dialysis or renal Fanconi's syndrome, and increased use), and excretion of esterified (acyl) carnitine (as with organic acidurias).
Plasma carnitine concentrations may be observed in a range from profoundly low levels to normal levels depending on the severity of the primary disorder and dietary intake. In general, clinical manifestations related strictly to the deficiency state are significantly correlated with plasma and/or tissue L-carnitine levels. Although a primary or secondary carnitine deficiency syndrome may be the only concomitant entity accompanying cardiomyopathy, it may also be counted among the many factors that are potentially contributory.
The Carnitine Cycle
In order for ATP to be ultimately produced from free fatty acids,
they must be transported from the cytosol across the mitochondrial membranes and into the mitochondrial matrix, where they are
metabolized. To accomplish this, they are first activated to form fatty
acyl CoA in the cytosol by the enzyme acyl CoA synthetase. Since the resulting fatty acyl CoA cannot traverse mitochondrial membranes alone,
the enzyme carnitine acyl transferase I, located on the outer surface
of the inner mitochondrial membrane, must conjugate fatty acyl CoA and
L-carnitine to form a new complex, fatty acyl carnitine.
This compound may then be transported across the inner mitochondrial
membrane by the integral enzyme, carnitine acyl translocase. On arrival
into the mitochondrial matrix, acyl carnitine transferase II transfers
the fatty acyl carnitine to CoA. This fatty acyl CoA complex
subsequently undergoes -oxidation to produce acetyl CoA.
Acetyl CoA is metabolized by Krebs cycle to provide H+ for the electron transport chain which, in turn, produces ATP. In the process, free carnitine is recycled into the mitochondrial matrix and is transported to the cytosol by carnitine acyl translocase. Alternatively, carnitine may conjugate with acyl CoA (by acyl carnitine transferase II) and may also be transported out by translocase. The significance of this latter mechanism is that it allows for the removal of acyl carnitine derivatives (eg, toxic organic acids that accumulate during pathologic catabolism) from the cell for excretion.
The carnitine shuttle, therefore, is integral to normal intramitochondrial events, ATP production, and regulation of the acyl CoA:free CoA ratio; availability of free CoA is crucial for the production of ATP and an imbalance in the acyl CoA:free CoA ratio signals metabolic disturbance. An increased acyl CoA:CoA ratio inhibits pyruvate dehydrogenase activity, an enzyme normally allowing pyruvate to enter Krebs cycle. Administration of L-carnitine can correct this imbalance and restore normal functioning of the dehydrogenase enzyme. L-Carnitine also reduces long-chain acyl CoA in heart tissue. This lowers the inhibition of adenine-nucleotide translocase and facilitates the formation and transportation of ATP from long-chain acyl CoA to the cytosol.39
Defects of the Carnitine Cycleb
Four defects of the carnitine cycle, not all of which create a carnitine deficiency, are known.34 These include: 1) carnitine transport defect, 2) carnitine palymitoyltransferase I deficiency, 3) carnitine/acylcarnitine translocase deficiency, and 4) carnitine palymitoyltransferase II deficiency. Carnitine transport defect creates a primary systemic carnitine deficiency, resulting in markedly low plasma and tissue carnitine levels, and often presents with cardiomyopathy and skeletal muscle weakness among other symptoms. In documented cases, treatment with L-carnitine supplementation has led to improvement in skeletal muscle and ventricular function.
In carnitine palmitoyltransferase II deficiency, normal to elevated carnitine levels are typical, accompanied by normal muscle and cardiac function.
In contrast, carnitine/acylcarnitine translocase deficiency can present with severe muscle weakness, mild hypertrophic cardiomyopathy, and low total and free carnitine levels. Of the 2 forms of carnitine palmitoyltransferase II deficiency, the less common but more severe form presents with low plasma and tissue carnitine levels, cardiomegaly, and cardiomyopathy. Because of the rarity of the disorder, there are no data on effects of treatment with L-carnitine.
Defects of the -Oxidation Cycle
The fatty acid oxidation cycle occurs in 4 steps and involves 13 enzymes. The fatty acyl-CoA ester provided at the end of the carnitine
cycle is metabolized and shortened by 2 carbons per cycle during
-oxidation, releasing an acetyl-CoA moiety to the Krebs cycle with
each iteration. Of the potential number of enzyme
deficiencies or impairments, there are 5 identified enzyme deficiency states and there is evidence that 4 can create a secondary L-carnitine deficiency; of these, 3 have been associated
with cardiomyopathy. Long-chain acyl-CoA dehydrogenase
(LCAD)c deficiency, long-chain
L-3-hydroxyacyl-CoA dehydrogenase (LCHAD) deficiency, and medium-chain acyl-CoA dehydrogenase deficiency create
low to low-normal plasma and tissue carnitine levels and often an
associated cardiomyopathy.
The logic underlying treatment with L-carnitine in this group of disorders lies in the fact that enzyme deficiencies create an accumulation of the normally metabolized corresponding substrate; L-carnitine provides a route of removal of these often toxic organic intermediates and this accounts for the low tissue and plasma-free L-carnitine levels observed. Supplementation restores these levels, at least in plasma, and can allow normalization of mitochondrial energy production and improvement in myocardial functioning.40 Restoration of normal tissue levels is often difficult attributable either to transport defects or continuous urinary loss. Fortunately, even partial restoration of tissue losses can have significant clinical benefit.
Clinical Basis for L-Carnitine Supplementation
There is considerable precedent in the literature for using L-carnitine to treat various forms of cardiomyopathy. Virtually all reports to date involve small groups of patients (<20) with diverse causes and varying methods of assessing outcome. Not all reports demonstrate an improved outcome; this is primarily dependent on the pathogenic mechanism responsible for L-carnitine deficiency, progression of the disease at treatment initiation, and nature of any accompanying disorders. For example, treatment of disorders of L-carnitine transport (ie, L-carnitine transport defects), which typically result in low plasma levels and deficient tissue levels, rarely restore tissue levels to normal. Nonetheless, there are numerous reports of amelioration or dramatic resolution of cardiomyopathy and associated signs and symptoms in L-carnitine transport defect patients.42-44 Although the normal route of L-carnitine transport is impaired, passive diffusion is thought to allow sufficient quantities of L-carnitine to reach tissue.45
L-Carnitine supplementation has been recommended in secondary deficiency syndromes resulting from the fatty acid oxidation disorders medium-chain acyl-CoA dehydrogenase,45 LCAD3,46,47 and LCHAD.46 It has also been shown to be effective in hypertrophic cardiomyopathy associated with mitochondrial myopathy, demonstrating clinical and echocardiographic improvement.48
There are a host of other disorders that lead to secondary carnitine deficiency and occasionally patients develop a cardiomyopathy. These include organic acidurias, such as methylmalonic aciduria and propionic aciduria. Certainly, any organic aciduria leading to carnitine deficiency could increase the risk for cardiomyopathy.
Conventional Treatment
Despite advances in understanding myocardial disease, treatment
has remained relatively unchanged in the past decade in terms of novel
therapies. Besides efforts directed to treat primary causal factors
(eg, toxic and infectious), conventional therapy consists primarily of
pre- and afterload reducing agents, inotropic agents, and -blockers
and Ca-blockers. Treatment is primarily directed at functional
improvement, and there is little evidence that long-term outcome is
changed. Here, we present an analysis of our collective clinical
experience in treating these disorders with both conventional therapy
and L-carnitine therapy.
| |
METHODS |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Study Design and Objectives
This investigation had a retrospective, multicenter protocol for cardiomyopathy in pediatric-aged patients and included 232 cases. The objective of this investigation was to compare the outcome of cardiomyopathy patients treated with L-carnitine plus conventional therapies with control patients treated with only conventional therapies. Outcome was assessed primarily in terms of overall outcome for mortality or transplantation, time to outcome, and changes in the patient's clinical severity, clinical functioning, and echocardiographic parameters obtained at time of diagnosis (baseline) and at end of study.
Investigators and Patients
Eight investigators at 7 centers reviewed all cases within their service for patients who had been diagnosed with cardiomyopathy and who met inclusion and exclusion criteria (Table 2). A total of 87 L-carnitine-supplemented patients and 145 control patients were identified who met inclusion and exclusion criteria; however, a minimum of 2 weeks of treatment with L-carnitine was required for treated cases to be included for analysis. Of the 232 total patients identified, 11 cases (5%) did not meet this requirement. Four patients died from cardiomyopathy and/or other cause before the end of 2 weeks of treatment; in 3 patients, it is unknown why treatment was stopped at <2 weeks (2 of these 3 patients were alive by end of study and 1 eventually went to transplant); 1 patient went to transplant before the end of 2 weeks of treatment; 1 patient received <2 weeks of treatment and went to transplant just after 2 weeks of starting treatment; 1 patient received <2 weeks of treatment and died from an unknown cause the month after starting treatment. All summaries and statistical analyses were performed on the 221 remaining patients.
|
The distribution of patients by center and treatment group is presented in Table 3. There was an approximately equal proportion of males and females in the study (Table 4), with slightly more females enrolled in both treatment groups. Patients in the control group were significantly older than were patients in the L-carnitine-treated group (6.5 years vs 2.1 years old, respectively; P < .001). The majority of patients were diagnosed with dilated cardiomyopathy, followed by the hypertrophic form. The remainder of patients had restrictive or mixed forms (Table 5).
|
|
|
The distribution of patients with a suspected metabolic cause for their cardiomyopathy diagnoses are shown in Tables 6 and 7; the most frequent metabolic abnormalities identified were glutaric acidemia type II, dicarboxylic aciduria, and long chain acyl CoA dehydrogenase deficiency. Certainly, these categories are not always diagnostic of a metabolic disorder, such as dicarboxylic aciduria. We placed any patient into this category who had an abnormality of organic acids, carnitine level, tissue biopsy, enzyme study, or family history suggestive of a metabolic disorder.
|
|
L-Carnitine Treatment
Patients were treated with intravenous L-carnitine and/or oral L-carnitine. Intravenous L-carnitine was generally given to inpatients (eg, for treatment of acute catabolic episodes at the time of hospitalization) and oral L-carnitine was provided to outpatients.
The mean dose of carnitine administered was 96 mg/kg and ranged from 14 mg/kg/day to 455 mg/kg/day. Table 8 summarizes the distribution of patients by dose duration.
|
Data Collection
Hospital and clinic charts were reviewed and data were collected on patient outcome, clinical severity, clinical functioning, echocardiographic parameters, concomitant medications, concomitant illnesses, L-carnitine treatment dose and duration, and patient demographics.
Data were captured by clinical visit using a sampling schedule, recording at or as close to the following intervals as possible: baseline (day of diagnosis of cardiomyopathy), 7 days, 3 months, 6 months, 9 months, 1 year, every 6 months thereafter up to 3 years, and yearly thereafter.
Data were collected on all patients until the study endpoint of July 31, 1994. At that point, patient outcome was coded as alive, died, transplanted, or lost to follow-up. For patients who underwent transplantation, data were not collected after the date of transplantation. Patients who died were further categorized as having died from cardiomyopathy, other disorder (unspecified), cardiomyopathy and a comorbid disorder (unspecified), or an unknown cause.
Clinical Severity and Clinical Functioning Scales
Three scales were developed before the investigation to assess clinical severity and clinical functioning. A set of signs and symptoms, considered by investigators to reflect key indicators of the patient's clinical status and acuity resulting from their cardiomyopathy, was developed by investigators. A total of 37 signs and symptoms were categorized into respiratory, cardiovascular, and systemic groups. The occurrence of each sign or symptom was recorded as documented in the record.
Each sign or symptom was weighted according to it's relative clinical severity. Cough, for example, was weighted as a 1, whereas pulmonary edema was weighted as a 3. The composite (summed score) of recorded signs and symptoms was deemed a proxy of clinical severity. Possible scores ranged between 0 and 71 points inclusive.
Two similar scales were developed to measure functional level, depending on patient age. The first was developed for school-aged children and the second for nonschool-aged children. The latter, however, was to be scored for all patients. The scales consisted of 5 points: 1 = normal activity (for age); 2 = mildly decreased activity; 3 = moderately decreased activity; 4 = no unnecessary activity allowed; and 5 = patient hospitalized. Scores were assigned based on the patient chart.
Statistical Methods
All statistical analyses were performed with BMDP statistical
analysis software (SPSS, Chicago, IL). Distributions of demographic and
baseline variables were summarized with descriptive statistics for each
treatment group. The mean ages of patients in the 2 treatment groups
were compared with Student's t test. Comparisons between the distributions of clinical outcomes for the treatment groups at
study termination were made with 
2
statistics.
Proportional hazards regression models were applied to assess times to clinical outcomes for the treatment groups and subgroups of clinical interest. For these analyses, both orthotopic transplantations and deaths were included as treatment failure outcomes. Patients who had died from an unknown cause (an exact cause was unavailable from the chart of 9 patients) or died from a known cause that was unrelated to cardiomyopathy (1 patient) were managed as censored in all time-to-event analyses and were excluded from clinical outcome summary tables. These patients were included in all other summary tables, however. Patients whose last follow-up visit occurred before end of study (July 31, 1994) were managed as censored in time-to-event analyses and were included in all summaries.
The proportional hazards model additionally enabled assessments of covariates, particularly those such as clinical severity with noteworthy imbalance for the treatment groups at baseline. The Wald statistic was used to make comparisons for parameters in the model for the treatment groups and for subgroups according to the covariates.
Incidence Density
For each treatment group and for subgroups within them according to whether the patient was treated with angiotensin-converting enzyme (ACE) inhibitors, time-to-clinical-outcome was further described with means and standard deviations for the duration of follow-up for the patients with and without treatment failure as well as the corresponding numbers of such patients. These quantities were then transformed to incidence densities by dividing the proportion of patients with treatment failure in a group by the mean duration of follow-up for all patients in that group. The incidence density is interpretable as the number of treatment failures per person-day of follow-up. The ratio of the incidence density for the control group to that for the L-carnitine-treated group represents the extent to which treatment failures occur sooner over time for the control group. Such incidence density ratios were determined for all patients and for the subgroups according to whether patients were treated with ACE inhibitors.
Analysis of Variance (ANOVA)
Repeated-measures ANOVAs were used to compare treatment groups at baseline and at end of study (posttreatment) for echocardiographic, clinical severity, and clinical functioning variables. Only patients with both baseline and end-of-study scores available were included for analysis of each variable.
| |
RESULTS |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Clinical Outcome
The distributions of clinical outcome are summarized by treatment group in Table 9. These distributions were significantly different (P = .010) in an overall sense. The pattern for such differences corresponded to tendencies for L-carnitine-treated patients to have lower mortality from cardiomyopathy as the primary diagnosis than do control patients (6.8% vs 17.9%), and less transplantation among L-carnitine-treated patients than control patients (9.6% vs 15.0%). L-Carnitine patients had higher mortality from cardiomyopathy with a comorbid diagnosis (9.6% vs 2.1%).
|
An analysis of the association between clinical outcome and concomitant medications unexpectedly revealed that the population of patients who received ACE inhibitors had significantly poorer survival. Although data were captured on whether ACE inhibitors were used, complete and reliable information was not available for dose and duration and dose relationship analyses were not possible.
Overall, 85 of the 213 total patients (40%) in Table 9 were treated with ACE inhibitors. Among the 85 ACE inhibitor-treated patients, 42 (49%) died from cardiomyopathy or underwent transplantation, and among 128 patients not treated with an ACE inhibitor, only 26 (20%) died from cardiomyopathy or underwent transplantation (P < .001; Table 10). The difference in clinical outcome between these 2 groups is mostly accounted for by more patients undergoing transplantation in the ACE inhibitor-treated group (25.9%) in comparison to the untreated group (4.7%), so ACE inhibitor treatment might possibly have been a consequence of their deterioration before transplantation rather than a possible cause of such deterioration.
|
Within the L-carnitine-treated group, 36 of the 73 patients (49%) received ACE inhibitors, and within the control group, 49 of the 140 patients (35%) received ACE inhibitors (Table 11). When distributions of clinical outcomes for L-carnitine treatment was examined in combination with and without ACE inhibitor treatment, overall differences were significant under both conditions (P = .047). Table 11 shows the comparison between treatment groups for each ACE inhibitor subgroup separately. The comparisons produced similar results to the overall outcome analysis in Table 9, although the bigger differences for mortality and transplantation were found in the ACE inhibitor-treated group, whereas the lowest proportions for mortality from cardiomyopathy and transplantation were found in the L-carnitine-treated, ACE inhibitor-untreated group.
|
Survival (Time-to-Event)
The comparison of time to death or transplantation between groups had exploratory modification to account for use of ACE inhibitor treatment. Better survival for ACE inhibitor-untreated versus ACE inhibitor-treated patients was clearly significant among all patients (P < .001; Fig 1).
|
Two further analyses were performed: 1) patients who were treated with L-carnitine and no ACE inhibitors versus all control patients, and 2) control patients who received ACE inhibitors versus control patients who did not receive ACE inhibitors. Significant improvement in survival was observed for L-carnitine-treated patients who did not receive ACE inhibitors versus control patients (Fig 2; P = .046), and improved survival was noted for control patients who did not receive ACE inhibitors compared with those who did (Fig 3; P = .0001).
|
|
An exploratory proportional hazards regression model was fit to compare survival for patients who were younger than and older than 1 year, 9 months of age; this was the age at which the maximum separation in survival time (regardless of treatment) was suggested. Patients younger than 1 year, 9 months of age at the time of diagnosis demonstrated better survival with descriptive P = .065, than those older than 1 year, 9 months of age (Fig 4).
|
Incidence Density
The incidence density (ie, the number of deaths or transplantations per person-day of follow-up) was used for further description of times to clinical outcome for the treatment groups in relation to ACE inhibitor use (Table 12). The incidence density was 1.35 times greater in the control group than in the L-carnitine-treated group for both ACE inhibitor-treated and -untreated patients. This ratio thereby describes the greater extent of deaths or transplantations per person-day of follow-up time for the control group. Also, the incidence density in the ACE inhibitor-treated group was ~4.4 times higher than the non-ACE inhibitor-treated group in both the L-carnitine-treated and control groups, indicating a substantially greater extent of events per person-day of follow-up among ACE inhibitor-treated patients.
|
Echocardiographic Data
Mean ejection fraction values are shown by treatment group at baseline and at end of study in Fig 5. Repeated-measures ANOVA of the ejection fraction values did not indicate any noteworthy differences between the treatment groups with respect to ejection fraction, but there was a significant time effect (P = .001); the ejection fraction values were similarly higher at the end of study than baseline for both groups.
|
Clinical Severity, Clinical Functioning, and Patient Status at Baseline
Mean clinical functioning scores and clinical severity scores by treatment group and time are shown in Figs 6 and 7, respectively (higher severity and functioning scores indicate a clinically worse condition). The repeated-measures ANOVA models indicated significant treatment and time effects for both clinical functioning and clinical severity (P < .001 for both effects). The model for clinical functioning also produced a highly significant interaction between treatment and time, indicating that the change between baseline and end of study differed for the treatment groups. This interaction comes from the L-carnitine patients improving to have nearly the same clinical functioning and severity as control patients at the end of the study, although they were substantially worse at baseline. In this regard, the percent hospitalized at baseline was twice as large for the L-carnitine group as the control group (Table 13).
|
|
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Although a better prognosis might be attributed to patients 3 times younger at onset than their control counterparts, L-carnitine-treated patients were also manifestly clinically worse as demonstrated by significantly poorer clinical severity scores, clinical functioning scores, and ejection fraction values. Moreover, 2.5 times as many patients were hospitalized at baseline in the L-carnitine-treated group compared with the control group (Table 13).
Level of clinical functioning, age of patient at time of treatment, and use of ACE inhibitors become important covariates in determining efficacy. L-Carnitine-treated subjects, compared with control subjects, likely represent patients suffering from sufficient morbidity, progression of disease, and advanced risk of death to prompt clinicians to seek this alternative treatment. In contrast, patients treated conventionally may have been more clinically stable and thought to have a better prognosis to forestall consideration of alternative intervention, although a higher proportion of these patients eventually died or received transplant.
The unfavorable experience of patients receiving ACE inhibitor treatment was unexpected and required subgroup analysis of the data. The implications of this observation, in view of the efficacy reports in the literature and the observational nature of this investigation, are unclear and must be approached tentatively. Nonetheless, the unfavorable survival of patients on ACE inhibitors (Fig 1) deserves further attention. The follow-up period in this investigation is longer than that of many reports and may suggest uncertainty for improvement in long-term mortality from ACE inhibitor use, although there might possibly be short- to intermediate-term improvements in myocardial function. It is difficult to precisely quantify the relationship between ACE inhibitors, survival, and other parameters recorded in this investigation because the dose, frequency, and precise duration of concomitant medications were often not available from clinical charts.
In the L-carnitine-treated group, the poorer survival of ACE inhibitor-treated patients was a suggestive trend. This finding is perhaps analogous to the protective effect against hepatotoxicity suggested for L-carnitine in patients treated with valproic acid. In this series of patients, this exploratory finding may be the most supportive to the benefits from L-carnitine treatment. Evaluating the benefit from L-carnitine for patients not treated with ACE inhibitors was more challenging because of reduced sample size, effects of other possible concomitant medications, and significant imbalances of baseline clinical severity and age relative to the corresponding control patients.
Among the clinical parameters, each showed improvement by the end of study; ejection fraction values improved for both treated and control groups, slightly more for treated patients. Likewise for clinical severity and clinical functioning, treated and control patients ended the study within ~1 scale point of each other. This is a clinically significant difference in improvement for L-carnitine-treated patients because they started with substantially poorer measures at baseline.
| |
CONCLUSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
In summary, our study showed a significant improvement in clinical severity and functioning of L-carnitine-treated patients versus control patients. Attributable to the retrospective nature of this study, several demographic and clinical parameters between treatment groups were imbalanced, making further conclusions difficult. ACE inhibitors were associated with unfavorable long-term outcome, but these exploratory results need further investigation because the study did not originally intend to evaluate the outcome of the pediatric use of ACE inhibitors. Metabolic causes of cardiomyopathy were, in general, not identified or considered, and appropriate testing was not ordered. Only 34 of 221 patients had any evidence to suggest a metabolic cause. We suspect the observed frequency to be low attributable to the lack of consideration or investigation for these disorders. Further prospective studies to investigate metabolic causes and L-carnitine therapy of pediatric cardiomyopathy would be helpful.
| |
FOOTNOTES |
|---|
a The term secondary cardiomyopathy, in fact, has been deemed inappropriate and has been supplanted by specific heart muscle disease (of known cause).
b Although diagnosed rarely in the general population, heritable defects creating primary carnitine deficiency states are briefly reviewed to make inclusive factors that may be unrecognized in contributing to idiopathic disease.
c Many of the previously diagnosed cases of LCAD have now been reclassified as LCHAD.
Received for publication Oct 8, 1998; accepted Aug 17, 1999.
Reprint requests to (E.H.) SAS Institute Inc, SAS Campus Dr, S6050, Cary, NC 27513. E-mail: ed.helton{at}sas.com
| |
ABBREVIATIONS |
|---|
ATP, adenosine 5'-triphosphate; LCAD, long-chain acyl-CoA dehydrogenase; LCHAD, long-chain L-3-hydroxyacyl-CoA dehydrogenase; ACE, angiotensin-converting enzyme; ANOVA, analysis of variance.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
