Late Recovery of Ventricular Function in Children with Idiopathic Dilated
Cardiomyopathy

	October 1999 - The prognosis for children with IDC (idiopathic
dilated cardiomyopathy) varies. Children who do not show improvement in LV
(left ventricular) function have a high one year mortality rate.
Improvement in LV fractional shortening (LVFS) to more than 15% is associated
with better survival. However, complete recovery of LV function has not been
studied.
	The clinical features and echocardiograms of 63 children with IDC
were reviewed. Sixteen patients (group 1) were identified who showed
progressive improvement in LVFS, ultimately recovering to normal. They were
compared to 47 patients (group 2) in whom LVFS remained low. Group 1 LVFS at
first exam was 14%, z = -11, and improved to normal range (34%, z = -1).
Group 2 initial LVFS was 14%, z = -9, and did not change significantly
(16%, z = -7).
	The LV was dilated at first exam in all patients (z = 7). Recovery in
group 1 was associated with a decrease in LV size to within normal range
(z = 1.3), while LV size in group 2 patients remained increased (z = 6). The
average follow-up time at which LV function was noted to be normal was 4.5
years. The total length of follow-up was 6.5 years.
	Conclusions: Complete recovery of LV function is possible in children
with IDC. Recovery may occur within the first year after initial examination
in some patients, but longer periods are needed in most patients for whom LV
function ultimately returned to normal.

Long version

	Mortality rates for children with DCM are highest during the first 6
months after initial exam and decline thereafter. One year survival has been
reported to be 70%, decreasing to 65% by 5 years and to nearly 50% beyond 10
years.
	Echocardiogram measures of ventricular dysfunction at time of initial
evaluation have little prognostic value, but persistent depression of left
ventricular fractional shortening (LVFS) - less than 15% for 1-3 months - is
associated with significantly poorer survival. The one year survival for such
patients was 46%, and 5 year survival was 29%, compared to higher than 90%
survival for children in whom early LV improvement is seen.
	Consequently, early heart transplant is recommended for children with
extended low LV function. Most patients can be stabilized with drug therapy
and show LV function improvement. However, the potential for recovering normal
LV function has not been studied. This study identifies a subset of children
with DCM who showed complete recovery of LV function.

Patient Selection

	The cardiomyopathy database was reviewed to identify 129 patients
with DCM. Children with myocarditis or known structural, metabolic, toxic,
collagen-vascular, and neuromuscular disorders were excluded, as were patients
specifically referred for heart transplant (to avoid extreme cases).
	Onset was defined as the date LV dysfunction was first documented.
Laboratory studies included EKGs, chest x-rays, echocardiograms, carnitine
blood level, heart cath, and heart biopsy. Plasma amino acids and urine
organic acids were measured in selected infants. Echos were reviewed and the
following measurements obtained:

1) minor-axis LV end-diastolic and end-systolic dimensions (LVDd, LVDs)
2) LV fractional shortening (LVFS) - calculated in the fashion LVFS = (LVDd
- LVDs)/LVDd.

	63 patients were identified in whom at least 2 follow-up echos were
available for review. Group 1 was 16 patients in whom LV function improved to
normal or near-normal. They were compared with 47 children (group 2) in whom
LVFS remained depressed.

Statistical Analysis

	Average age was 2 years at first examination for group 1, ranging
from one day old to 10 years. Patients in group 2 averaged 4.5 years of age.
Eleven of 16 group 1 patients (69%) were less than 2 years old when first
examined; nineteen of 47 patients (40%) in group 2 were more than 2 years old
at first exam.
	LVFS at first exam in group 1 was 13.6%, z = -10.8, and was similar
to group 2 (13.6%, z = -8.9). LVFS improved to normal range in group 1
patients (33.4%, z = -0.9) at an average follow-up interval of 4.5 years.
	However, LVFS did not change significantly in group 2. Length of
follow-up ranged from 1-16 years, with an average of 6.5 years. Ventricular
function recovered to normal within the first follow-up year in 4 patients,
during the second year in one patient, and between 3 and 14 years in 11
patients. (JON: I assume 31 children did not improve at all)

Recovery of LVFS

	Change in average LVFS during the first year after initial exam: LVFS
increased from 14% to 24% in group 1 and from 13% to 19% in group 2. Group 1
average LVFS continued to slowly increase, reaching normal range (higher than
29%) roughly 4-5 years after first exam. Average LVFS for group 2 patients
generally remained lower than 20%.
	The LV was significantly dilated (LVDd z > 2) at first exam in all
patients. For the purpose of comparison, LVDd is expressed as a z score
because of the wide variation in patient age and size. Initial LVDd z = 6.9
in group one and 6.4 in group 2. Recovery of ventricular function in group 1
was associated with a proportionate decrease in chamber size to within normal
range (z = 1.3). In contrast, group 2 continued to show LV dilatation
(z = 6.2).

Meds and heart class

	All patients in both groups took digoxin, diuretics (Lasix), and an
ACE inhibitor (captopril, enalapril or quinapril) according to the wishes of
their cardiologist. None were treated with beta-blockers or antiarrhythmic
drugs. Ten group 1 patients required intravenous inotropic support on their
first hospital admission. Two children required multiple hospitalizations for
CHF during the first year after onset and were treated with intravenous
inotropes and diuretics. They were classified as functional heart class 4 and
were evaluated for heart transplant but were not listed because of high
pulmonary artery pressures and unstable home environments. Both patients
showed recovery of LVFS after 2.5 and 5 years, respectively.
	At the most recent follow-up visit, all group 1 patients were class
one. All heart medications had been discontinued in 2 patients. Eleven
patients continued to take diuretics, although at a reduced dose. Fourteen
patients continued to take ACE inhibitors, 13 of whom were also taking
digoxin.
	Fifteen patients (32%) in group 2 died or had heart transplant one
month to 2 years after first exam. All surviving patients continue to take
digoxin, diuretics, and ACE inhibitors. All are in heart class 1 or 2.

Discussion

	The outcomes of IDC patients vary. Spontaneous improvement has been
seen in about 30% of adults. In pediatric patients from our institution, one
year mortality is high, especially for children with persistent depression of
LVFS (less than 15%). Currently, there is a limited ability to identify risk
factors at first exam that conclusively separate patients with poor prognosis
from those who will show improvement. However, more severely depressed LVEF,
lack of improvement in ventricular function, and more spherical ventricle
shape are associated with poorer prognosis.
	It is noteworthy that late recovery of LV function was seen in a few
patients who did not have increased EF until more than one year after first
exam. Others have found no relation between symptom duration and improvement.
Younger age at onset may be associated with better survival. Our patients who
showed recovery of LV function tended to be younger than those in whom FS did
not improve. However, younger age alone does not predict recovery. Neither
initial LVFS or LV size helped identify which patients would ultimately
recover LV function.
	Many causes are currently lumped together as idiopathic
cardiomyopathy. In recent years, disorders of heart energy metabolism,
including systemic carnitine deficiency, fatty acid beta-oxidation,
respiratory chain enzyme deficiencies, and mitochondrial DNA mutations have
emerged as potentially treatable metabolic causes of DCM but may remain
undiagnosed and called idiopathic. A comprehensive clinical approach to
evaluating children with cardiomyopathy is in order.
	It has recently become evident that "idiopathic" DCM may also result
from acute viral myocarditis. Heart biopsy has become the "gold standard" for
diagnosing myocarditis. Although the presence of a lymphocytic infiltrate and
adjacent heart cell necrosis may confirm a myocarditis diagnosis, absence of
inflammation does not prohibit the diagnosis because of the rapid clearance
of inflammatory cells. None of our patients who had a heart biopsy had any
features of myocarditis. However, this cannot be certain of much of anything.
	There are some limitations to our study. It was a review with patients
managed by their individual cardiologist. The interval between echos was not
standardized, and variable drug therapy was noted. The patient population may
may have been more severe than average so the proportion of DCM patients who
have complete recovery of LV function may be higher than we suggest.
	Fifteen of 47 patients in group 2 died or had heart transplant within
2 years of first exam. NOTE THAT THE 32 SURVIVING GROUP 2 PATIENTS CONTINUED
TO SHOW AN AVERAGE LVFS IN THE RANGE OF 18-20% RATHER THAN DETERIORATION.

Conclusions:

	Long-term prognosis for children with recovered DCM is uncertain.
Whether they are at risk for later deterioration of LV function is unknown.
The need for continued drug therapy is also unknown. In the absence of
congestion, there is little need for diuretics. Similarly, it should be
possible to discontinue digoxin after a suitable period of normal LV function.
	The ability of ACE inhibitors to modify ventricular remodeling and to
limit or reverse progression of LV enlargement has led to their use as first
line drugs in patients with LV dysfunction regardless of symptoms. Their
beneficial effects on ventricular chamber size, hypertrophy, and heart
fibrosis may not only halt deterioration but in some people may restore normal
ventricular shape, size and function. However, when ACE inhibitors can be
discontinued remains unclear.
	The role of beta-blockers in children with DCM remains to be seen.
Recent trials with metoprolol and Coreg have shown improved blood flow at rest
and exercise, improved exercise ability, increased EF, and lower mortality.
Different reasons have been proposed to explain this, including reduced
metabolic demand, protection from nervous system overstimulation, and
increased heart beta receptor density.
	Recently, Shaddy reported improvement in functional class and LVFS in
15 children with DCM and persistent symptoms, who were treated with
metoprolol. These children had symptoms despite conventional therapy with
digoxin, diuretics, and ACE inhibitors, until adding the beta-blocker.
	It seems that some patients with IDC who take regular triple drug
therapy may experience recovery of ventricular function. The addition of
newer generation beta-blockers like Coreg may offer recovery to more patients.
A multi-institution, randomized, placebo-controlled trial is needed to
address this.

Alan B. Lewis, MD
Division of Cardiology
Childrens Hospital Los Angeles, Department of Pediatrics
University of Southern California, Los Angeles
Am Heart J 138(2):334-338, 1999