|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Heart Failure and Cardiac Transplantation Center, Cardiology Division, (Drs. Koelling, Dec, and Semigran), and the Pulmonary Unit (Dr. Ginns), Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA.
Correspondence to: Marc J. Semigran, MD, MGH Heart Failure and Cardiac Transplantation Center, Massachusetts General Hospital, Bigelow 645, 55 Fruit St, Boston, MA 02114; e-mail: semigran.marc{at}mgh.harvard.edu
| Abstract |
|---|
|
|
|---|
Background: Although myocardial fibrosis has been described in autopsy specimens of patients with cystic fibrosis, the possibility that myocardial dysfunction may occur during life in adult patients with cystic fibrosis has not been explored.
Methods: To assess the possibility of cardiac dysfunction occurring in cystic fibrosis, we studied 40 patients with advanced cystic fibrosis with first-pass radionuclide ventriculography and compared them to 9 patients with advanced bronchiectasis and 18 normal control subjects.
Results: Indexes of right ventricular systolic function were similarly impaired in patients with cystic fibrosis and patients with bronchiectasis. Left ventricular ejection fraction of patients with cystic fibrosis, patients with bronchiectasis, and normal control subjects did not differ. Fractional left ventricular filling at 50% of diastole, an index of diastolic function, was significantly lower in patients with cystic fibrosis (54 ± 13%, mean ± SD) in comparison to patients with bronchiectasis (66 ± 4%, p = 0.009) or normal control subjects (69 ± 14, p = 0.0002). The contribution of atrial systole to total diastolic left ventricular filling was greater in patients with cystic fibrosis (38 ± 18%) than in patients with bronchiectasis (21 ± 4%, p = 0.01) or normal control subjects (25 ± 12%, p = 0.01).
Conclusions: Patients with advanced cystic fibrosis demonstrate impaired left ventricular distensibility when compared to normal control subjects and patients with bronchiectasis. Patients with cystic fibrosis may be at risk of heart failure due to right ventricular dysfunction or left ventricular diastolic dysfunction.
Key Words: cystic fibrosis diastolic dysfunction radionuclide ventriculography
| Introduction |
|---|
|
|
|---|
As the longevity of patients with cystic fibrosis is extended with the use of newer antibiotics, lung transplantation, and the possibility of successful gene therapy, a process of progressive fibrosis of myocardium may have functional significance and could lead to the development of heart failure in these patients. In order to determine if patients with cystic fibrosis, without symptomatic heart failure, would exhibit impairment of myocardial function, we compared indexes of systolic and diastolic function in these patients with a group of patients with bronchiectasis who did not have abnormal epithelial chloride conductance, and with a control population.
| Materials and Methods |
|---|
|
|
|---|
Cardiopulmonary Exercise Testing
Cardiopulmonary exercise testing was performed by previously reported methods.9
In brief, upright bicycle ergometry (pedal ergometer; Warren E. Collins; Braintree, MA) was used with a 12.5 W/min or 25 W/min incremental ramp protocol based on predicted exercise capacity, and was performed following an overnight fast with all patients breathing air. Breath-by-breath expired gas analysis was performed by a metabolic cart (Beckman OM1; Beckman Instruments; Fullerton, CA) interfaced to the ergometer. Peak oxygen uptake was defined as the highest oxygen consumption (
O2) measured during the last minute of symptom-limited exercise. The anaerobic threshold was determined by the V-slope method.10
Exercise testing was delayed if patients required a course of antibiotics until their pulmonary infection had improved.
First-Pass Radionuclide Ventriculography
Rest and exercise first-pass radionuclide ventriculography was performed at the time of bicycle ergometry. Patients were seated upright on the cycle ergometer facing a multicrystal nuclear camera (System 77; Baird Corporation; Bedford, MA). Images were acquired at 0.025-s intervals after central venous bolus administration of autologous erythrocytes labeled with 10 mCi 99mTc at rest and with 15 mCi at peak exercise. Left and right ventricular regions of interest were chosen based on maximum intensity of radioactive counts as depicted by color and identification of valvular planes. Time-activity curves were generated for each region of interest by imaging software (SIM-400; Scinticor; Milwaukee, WI). Only cycles with > 70% of the maximal end-diastolic activity in the end-diastolic frame were included in the analysis. Lung background counts were subtracted from the time-activity curves representing each cardiac cycle. The background curves were then summed to create a single representative cardiac cycle. Peak activity was considered to be end-diastole and minimum activity was considered to be end-systole.
The ejection fraction of each ventricle was calculated as (end-diastolic counts - end-systolic counts)/end-diastolic counts. The left ventricular end-diastolic volume (LVEDV) was determined using a previously described geometric area-length method.11 Atrial filling contribution, time to peak filling rate, and filling at 50% of diastole were determined from the time-activity curves.12 The period of atrial contraction was taken to begin 40 ms prior to the beginning of the PR interval of a 12-lead ECG (thereby compensating for electromechanical delay) and terminate at end-diastole. Peak filling rate and filling at 50% of diastole were determined from the diastolic portion of the time-activity curve.
Statistical Analysis
Rates and proportions were compared using
2 tests of general association or Fisher exact tests where appropriate. The unpaired Student t test with assumption of equal variance was used to compare the means of normally distributed continuous variables. Correlations of demographic data, physical examination findings, pulmonary function data, with radionuclide ventriculography measurements were performed using the Spearman rank- correlation method for nonparametric data. Data are presented as mean ± SD. For all analyses, a two-tailed p value < 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
O2, and
O2 at the anaerobic threshold. Each of the patients with cystic fibrosis and patients with bronchiectasis reached a pulmonary limit to exercise. Peak workload achieved by patients with cystic fibrosis was less than observed in patients with bronchiectasis, but no difference was seen with respect to peak
O2 or
O2 at the anaerobic threshold. The resting heart rate of the patients with cystic fibrosis was higher than either the patients with bronchiectasis (87 ± 23 beats/min) or the normal control subjects (82 ± 13 beats/min), although the heart rate at peak exercise did not differ. The mean systemic BP at rest was similar in all three groups; however, the systemic BP achieved at peak exercise was lower in the patients with cystic fibrosis than the patients with bronchiectasis. The PaO2 of the patients with cystic fibrosis and patients with bronchiectasis was similar at rest and at peak exercise.
|
|
|
| Discussion |
|---|
|
|
|---|
Ventricular Dysfunction in Patients With Cystic Fibrosis
Although right ventricular dysfunction in patients with cystic fibrosis has been described by previous investigators,3
4
5
6
14
left ventricular function has not been studied extensively. Tomlin et al15
described chronic cor pulmonale as a complication of fibrocystic disease of the pancreas in 1952. Allen et al,16
using echocardiography, identified increased right ventricular wall thickness and cavity size in 37 children with cystic fibrosis, observing that these abnormalities occurred in patients with mild as well as severe disease. Ryland and Reid17
identified right ventricular hypertrophy at postmortem examination in 83% of 36 patients with cystic fibrosis > 3 years old.
Our study expands on these observations, as we observed that abnormalities in right ventricular systolic function are present at rest and persist to peak exercise in patients with cystic fibrosis. Although 25% of the patients had RVEFs < 0.30, it is important to note that no patients with cystic fibrosis in the population had more than mild (1+/4) pedal edema, and only 8% had evidence of jugular venous distention. Only 2 of the 40 patients with cystic fibrosis required therapy with diuretic medications. None of the patients in this pretransplant population was recommended to have combined heart-lung transplantation. One potential cause of right ventricular systolic dysfunction may be an increase in afterload related to destruction of pulmonary parenchyma. In support of this, we found a correlation between FEV1 and RVEF. We did not find a correlation between RVEF and pulmonary arterial systolic pressure estimated from the velocity of the tricuspid regurgitant jet; however, pulmonary arterial systolic pressure may not be an accurate measure of afterload in the presence of right ventricular dilation such as was observed in the patient population.
The etiology of the observed left ventricular diastolic dysfunction in patients with cystic fibrosis is potentially multifold. Right ventricular dilation may decrease left ventricular distensibility through the interventricular septum.18 The echocardiographic observation of septal flattening in some of the patients with cystic fibrosis supports this mechanism; however we did not observe an association of the presence of septal flattening with impaired indexes of ventricular filling. Other determinants of ventricular distensibility, such as tachycardia and increased levels of neurohormonal mediators, may have had a greater impact on ventricular filling and obscured the effect of the right ventricular pressure and volume overload on left ventricular function.
Studies of left ventricular function in patients with cystic fibrosis have given clues that the disease may be associated with a pathologic process that directly effects the myocardium. Several case reports have described children with cystic fibrosis with signs and/or symptoms of left ventricular dysfunction who were later found to have myocardial fibrosis at autopsy.1 2
Allen et al16 showed that patients with cystic fibrosis and severe pulmonary disease had thickened septa, posterior walls, and aortas compared with normal subjects by two-dimensional echocardiography. Our study confirms that the interventricular septa of patients with cystic fibrosis are thickened as compared to the patients with bronchiectasis. Benson et al19 studied 31 patients with cystic fibrosis with exercise radionuclide ventriculography and found that only 2 patients had LVEF < 55% at rest, but 29% had impaired cardiac performance with exercise. More recently, Herout et al20 described skeletal and cardiac muscle pathologic changes in 26 children with cardiomyopathy and cystic fibrosis, 24 of whom subsequently died. Ambrosi et al21 studied 67 infants and children with cystic fibrosis and found that 17 patients had LVEFs < 0.45, and 35% of these had perfusion defects by scintigraphy compared to 8% of the patients with LVEFs > 0.45. They surmised that these abnormalities were due to myocardial fibrosis.
A more likely clinical manifestation of myocardial fibrosis would be the development of impaired left ventricular diastolic relaxation. Johnson et al22 studied 25 patients with cystic fibrosis using Doppler echocardiography, examining left ventricular filling patterns, and found that the proportion of atrial filling was higher in these patients compared to control subjects, and correlated with worsening pulmonary disease. Although the Doppler signal from atrial filling is inversely associated with left ventricular relaxation, transmitral flow velocities have been shown to be closely related to both atrial filling pressure and left ventricular diastolic pressure.23 For this reason, we chose to assess the patients with cystic fibrosis with radionuclide ventriculography to assess left ventricular volumetric changes during diastole.
Causes of myocardial fibrosis in patients with cystic fibrosis are unknown. Zimmerman et al24 reported on pathology findings in two patients dying of cystic fibrosis-associated cardiomyopathy in whom myocardial edema was found and lymph stasis was seen, suggesting that cystic fibrosis may be complicated by a disorder of cardiac lymphatic drainage. Others have hypothesized that the fibrosis seen on pathology samples occurs because of malnourishment and/or hypovitaminosis.2 Levels of neurohormonal mediators such as aldosterone and angiotensin II are increased in patients with cystic fibrosis, and these signaling molecules have previously been shown to cause myocyte hypertrophy and interstitial fibrosis.25 26 Certainly, chronic hypoxia can lead to myocardial fibrosis,27 and this may have occurred in the patients with cystic fibrosis we studied, as indicated by the need for supplemental oxygen in many of them. Finally, the fundamental abnormality of cystic fibrosis lies in chloride transport in epithelial cells. Whether this abnormality alters either myocyte calcium handling or myocardial collagen production is unknown.
The implications of the finding of diastolic dysfunction in this population are twofold. First, advances in medical therapy for patients with cystic fibrosis may lead to a population of patients in whom dyspnea may have both pulmonary and cardiac contributions. Secondly, patients who have successful lung transplantation and long-term survival from improved allograft preservation may eventually acquire symptomatic left heart failure from progressive left ventricular noncompliance. The study of ventricular function in cystic fibrosis patients after transplantation is warranted.
Limitations
This study is a review of cardiovascular findings in a population of patients under evaluation for lung transplantation. Because of the advanced degree of the lung disease in these patients, the results of this study cannot necessarily be extrapolated to the general population of patients with cystic fibrosis without further study. Further study of a broader age range of patients may allow demonstration of an association between diastolic parameters and age in our study.
The indexes of diastolic function used in this study were derived from changes in chamber volume over time measured by radionuclide ventriculography. These indexes have been shown previously to be dependent on loading conditions, ejection fraction, and heart rate.28 29 30 31 We found no significant differences among the groups with respect to BP or LVEF at rest. Patients with cystic fibrosis did have higher resting heart rates, on average, than control subjects or patients with bronchiectasis. Previous investigators have shown that higher heart rate increases the peak ventricular filling rate, increases the early diastolic filling contribution, and decreases the atrial filling contribution; however, despite the tendency of a higher heart rate to improve diastolic relaxation, the patients with cystic fibrosis were observed to have impaired diastolic filling parameters. Future assessment of diastolic function independent of loading conditions and heart rate may require the patients and controls to undergo micromanometer left-heart and right-heart catheterization simultaneous with radionuclide ventriculography and right atrial pacing.
We chose to compare indexes of myocardial systolic and diastolic function in patients with cystic fibrosis with a group of patients with bronchiectasis who did not have abnormal epithelial chloride conductance but in whom the pulmonary pathology is similar to that of cystic fibrosis. The patients with cystic fibrosis did have a greater degree of impairment of pulmonary function than the patients with bronchiectasis, and we cannot completely exclude the possibility that this contributed to the left ventricular diastolic dysfunction observed in the patients with cystic fibrosis; however, the differences in pulmonary function did not cause resting right and left ventricular loading conditions to differ, given the limitation imposed by the sample size to determine a difference. This supports our conclusion that our findings are characteristic of cystic fibrosis and not the greater degree of impairment of pulmonary function.
| Summary |
|---|
|
|
|---|
| Footnotes |
|---|
O2 = oxygen consumption Received for publication October 19, 2001. Accepted for publication November 8, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. H. Alzeer, A. F. Al-Mobeirek, H. A. K. Al-Otair, U. A. F. Elzamzamy, I. A. Joherjy, and A. S. Shaffi Right and Left Ventricular Function and Pulmonary Artery Pressure in Patients With Bronchiectasis Chest, February 1, 2008; 133(2): 468 - 473. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Lindqvist, K. Caidahl, G. Neuman-Andersen, C. Ozolins, S. Rantapaa-Dahlqvist, A. Waldenstrom, and E. Kazzam Disturbed Right Ventricular Diastolic Function in Patients With Systemic Sclerosis: A Doppler Tissue Imaging Study Chest, August 1, 2005; 128(2): 755 - 763. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |