Abstract
Patients with severe cystic fibrosis can develop cor pulmonale, but little is known about the function of the right ventricle (RV) early in the disease. We hypothesized that such patients might have subclinical RV dysfunction, detectable by tissue Doppler echocardiography, and related to the severity of lung disease. We studied 21 clinically stable patients (Group 1), five patients with severe lung disease (Group 2), and 23 age-matched healthy subjects. Patients had impaired RV systolic function. The mean (SD) systolic velocities of the RV free wall were 8.9 (1.7) cm/s in Group 1, 7.7 (1.0) in Group 2, and 10.8 (1.9) in healthy subjects (p < 0.001). The velocities of the tricuspid annulus were less in patients (p < 0.0001). Patients had a greater isovolumic relaxation time (p < 0.001), indicating RV diastolic dysfunction. RV wall thickness was greater in patients (0.4 [0.1] versus 0.3 [0.1] cm/m2, p < 0.01). RV systolic function was related to C-reactive protein (r = − 0.66, p < 0.001) and FEV1 (r = 0.62, p = 0.003) and diastolic function to interleukin-6 (r = 0.64, p < 0.005). Patients with cystic fibrosis have subclinical RV dysfunction, which correlates with the severity of lung disease. Tissue Doppler echocardiography provides a quantifiable indicator useful for detection and monitoring of disease progression.
In patients with cystic fibrosis (CF), the clinical assessment of right ventricular function is unreliable (1), and usually dysfunction is diagnosed only when it produces the clinical picture of cor pulmonale. In 1951 it was estimated that chronic cor pulmonale was present in 70% of infants and children dying of CF (1), but since then the natural history of CF has changed dramatically. Chronic pulmonary infection and persistent systemic inflammation may exert a significant influence on cardiac function, but this has not been studied in detail. There have been very few recent studies of right ventricular function in patients with CF, although it has now been established in experimental studies that inflammatory mediators can depress myocardial function (2). Recurrent pulmonary infection may exert a significant influence on cardiac function, but this has not been studied in detail.
Conventional imaging techniques have serious limitations when applied to the right ventricle. Radionuclide ventriculography involves exposure to radiation, is relatively expensive, has restricted availability, and is not easily repeatable. Echocardiography is a bedside, innocuous, and widely available technique, but the irregular shape of the right ventricle, its retrosternal location, and the frequent coexistence of lung hyperinflation, make standard “gray-scale” imaging suboptimal for routine assessment of right ventricular function.
Tissue Doppler Echocardiography (TDE) allows measurement of regional myocardial velocities and time intervals in systole and diastole (3). Accurate and reproducible tracings can be obtained even when the gray-scale image is suboptimal (4). As it provides quantitative measures of regional function, this new modality may be more sensitive for detecting subclinical abnormalities.
We tested the hypotheses that patients with CF have subclinical impairment of right ventricular function during clinical stability, and that TDE may detect this impairment at a stage when conventional echocardiographic indices are still normal. We expected the use of TDE to be superior to that of conventional gray-scale imaging. We further considered that differences might be related to the severity of lung disease or the frequency of infections, and so we included a further group of patients with severe disease in order to test possible causal relationships by correlating the tissue Doppler indices of right ventricular function to abnormalities of respiratory function and evidence of systemic inflammation.
We studied 26 patients with proven CF (sweat sodium and chloride > 70 mmol/L and relevant genotype) and 23 age- and sex-matched healthy subjects (Table 1). Twenty-one patients (Group 1) were clinically stable and ambulatory at the time of the echocardiographic study, and were recruited from outpatients attending the regional CF center over a period of 18 mo. Inclusion criteria were: age older than 16 yr, absence of signs and symptoms of right ventricular failure, absence of an exacerbation of the respiratory symptoms (no change in symptoms or reduction of FEV1 of > 10% of the usual value in the month previous to the study), absence of oral corticosteroid treatment or metabolic disease and willingness to undergo an echocardiographic study. Five patients (Group 2) were in-patients who had had chronic hypoxia for at least 2 mo; two were awaiting transplantation, and one died from respiratory failure 2 wk later. All patients in Group 2 were receiving nebulized short acting beta-2 agonist therapy, long-term oxygen treatment, and intravenous antibiotics. All patients in Group 1 were receiving regular treatment with short acting inhaled beta-2 agonists.
| Healthy Subjects (n = 23) | Group 1–Stable CF (n = 21) | Group 2–Severe CF (n = 5) | ||||
|---|---|---|---|---|---|---|
| Male, % | 57 | 57 | 60 | |||
| Age, yr | 22.1 (2.9) | 23.2 (4.3) | 23.7 (1.5) | |||
| BSA, m2 | 1.87 (0.07) | 1.76 (0.11) | 1.43 (0.19) | |||
| Female, % | 43 | 43 | 40 | |||
| Age, yr | 23.1 (3.7) | 26.0 (3.2) | 21.5 (3.5) | |||
| BSA, m2 | 1.75 (0.09) | 1.52 (0.18) | 1.25 (0.07) |
Healthy subjects were recruited from among medical students and staff at our institution. Inclusion criteria were no history of chronic respiratory disease, no history of smoking, no chronic medications, and willingness to undergo an echocardiographic study. Healthy subjects were recruited over a period similar to that of the patients.
The protocol was approved by the Local Research Ethics Committee, and all subjects gave informed consent.
All patients were assessed clinically by the same chest physician (A-AI) at the time of their recruitment into the study. Clinical records were reviewed, and for each patient the Shwachmann score (5) and the number of episodes of exacerbations of the respiratory symptoms treated with antibiotics during the 18 mo prior to recruitment into this study were documented. An exacerbation of respiratory symptoms was defined as increased sputum production with increased purulence, with or without systemic symptoms (loss of weight and appetite, tiredness) and a reduction of FEV1 of more than 10% of the patient's usual value.
Body composition in patients was determined by Dual Energy X-Ray Absorptiometry (DXA, Hologic 2000) (6). The fat-free mass (FFM) corrected for height (FFM:height2, kg/m2) was determined (6).
All subjects were examined in a semisupine, left lateral position, by the same observer (AAI). EKG, respiratory movements, and the phonocardiogram were recorded continuously. Patients in Group 1, and the normal healthy subjects, were studied with an Acuson XP 10 scanner interfaced with a 3.5/4.5 MHz transducer. The five patients in Group 2 were imaged in a different hospital with a Vingmed System V scanner, without a phonocardiogram, and isovolumic relaxation times were not measured. Gray-scale images and conventional blood-pool Doppler data were acquired in standard planes from parasternal and apical windows. M-mode tracings of the right and left ventricle were obtained in the parasternal long-axis view with the cursor placed at the tip of the mitral valve leaflets.
Pulmonary arterial flow was recorded with pulsed-wave Doppler, placing the sample volume centrally between the leaflets of the pulmonary valve in a short-axis view at the base of the heart. The mitral and tricuspid Doppler signals were recorded in the apical four-chamber view, with the Doppler sample volume placed at the tip of the mitral or tricuspid valve. All valves were scanned with color Doppler, and any regurgitant jet of more than trace severity was documented.
TDE was performed in apical four-chamber planes with the pulsed-wave Doppler sample volume (length = 4 mm) placed successively at four locations (Figure 1): lateral mitral annulus, medial mitral annulus, lateral tricuspid annulus, and right ventricular free wall midway between the apex of the right ventricle and the tricuspid annulus. When the right ventricular free wall was not adequately imaged from the apical four-chamber view, the echocardiographic transducer was tilted medially, and the TDE recording was performed using this slightly off-axis view. At least 10 cardiac cycles were recorded from each site on a strip-chart recorder at a speed of 100 mm/s. The whole examination was also recorded on S-VHS videotape.
Fig. 1. Apical four-chamber view used for the recording of TDE velocities of the mitral and tricuspid annuli. The placement of the sample volume is indicated by the white arrows. Abbreviations: LV = left ventricle; RV = right ventricle.
[More] [Minimize]The spectral tissue Doppler traces were digitized with a personal computer, using custom-made software (7). The gray-scale images and the conventional Doppler traces were measured off-line, on calibrated frozen video frames, using the software of a Hewlett-Packard Sonos 1500 scanner (Hewlett-Packard, Cupertino, CA).
From the EKG and phonocardiogram, ventricular systole was defined as the onset of the R-wave of the EKG to the first high-frequency component of the second heart sound, and diastole was measured from the second heart sound to the onset of the R-wave or the first deflection of the QRS complex.
Isovolumic relaxation times were measured from the second heart sound to the onset of the Doppler diastolic signal, on both inflow Doppler and tissue Doppler traces (Figure 2). Peak myocardial velocities and the corresponding velocity-time integrals were traced with an electronic caliper. Velocity-time integrals were normalized by dividing by the RR interval.
Fig. 2. Representative TDE trace recorded from the tricuspid annulus in a patient in Group 1. The isovolumic relaxation time is indicated by the arrows. Abbreviations: Phono = phonocardiographic trace; S1 = first heart sound; S2 = second heart sound; Resp = respiratory trace; sv = peak systolic velocity; ivrt = isovolumic relaxation time; E = early diastolic velocity; A = late diastolic velocity.
[More] [Minimize]The thickness of the right ventricular anterior wall (RVAW), the ventricular septum, and the left ventricular posterior wall were measured at end-diastole. The dimensions of the left ventricle were measured at end-diastole and systole, all from M-mode traces and using the recommendations of the American Society of Echocardiography (8).
Left ventricular ejection fraction and diastolic mass were estimated from M-mode tracings, using the Teichholz formula (9) and the cubed formula (10), respectively. Right ventricular ejection fraction was calculated using the single plane area-length method (11), from end-systolic and end-diastolic frames of an apical four-chamber view of the right ventricle. Measurements obtained during four end-expiratory cardiac cycles were averaged for each parameter.
All patients had spirometry (Vitalograph Ltd, Buckingham, UK) within 1 wk of the echocardiographic study, and the FEV1 and FVC were documented (best result of three successive attempts). Arterialized ear lobe capillary blood was sampled for measurement of PaO2 and PaCO2 , and capillary blood oxygen saturation was measured with a finger pulse oximeter.
Circulating levels of C-reactive protein (CRP), Interleukin 6 (IL-6), tumor necrosis factor alpha (TNFα), and of the soluble receptors I and II of TNFα were measured in venous blood by ELISA (12) at the time of the echocardiographic study in all patients. For patients in Group 1, CRP levels measured at the end of the most recent course of antibiotic treatment for an exacerbation of the respiratory symptoms, when the systemic inflammation reaches its nadir, were used for calculations (12).
The data were analyzed with SPSS version 6.1 software (SPSS, Chicago, IL). The type of distribution fitting each data set was assessed by obtaining histograms and normality plots of the data and with the Kolmogorov-Smirnov test. The groups were compared using the unpaired samples t test, for data with a normal distribution and equal variance, or the Mann-Whitney U test, for data where assumptions of normality and equal variance of the data were not met. Patients were then matched by age to healthy subjects and the groups were compared using Wilcoxon's matched-pairs signed-ranks test. Correlations between variables were assessed by obtaining scatter graphs of paired variables, and by calculating Pearsons and Spearman's Rank correlation coefficients. Multiple stepwise regression was used to assess the effect of lung function and inflammation on TDE parameters. In all cases the significance level was set at p < 0.05. Intraobserver variability for the TDE was assessed by reanalyzing five randomly selected traces several months after they had been recorded. The coefficient of variation was calculated for each of the measured parameters as the ratio of the standard deviation of each specific measurement to its mean, expressed as a percentage.
The feasibility of TDE was better than for all the other methods, particularly for the study of the right ventricle where systolic velocities could be recorded in all patients and healthy subjects. In contrast, right ventricular ejection fraction could be estimated by gray-scale imaging in only 64% of the subjects.
Myocardial velocities. The systolic velocities of the right ventricular free wall and of the tricuspid annulus were less in patients in Group 1 than in healthy subjects (Table 2 and Figure 3). The peak systolic and the diastolic velocities at the mitral annulus were not different between the three groups, but systolic and diastolic velocity-time integrals were less in patients in Group 1 than in healthy subjects. All absolute systolic velocities and velocity-time integrals were even lower in patients in Group 2 compared with those in Group 1 and with healthy subjects.
| Right Ventricular Wall | Tricuspid Annulus | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Healthy Subjects | Group 1 | Group 2 | Healthy Subjects | Group 1 | Group 2 | |||||||
| Peak systolic velocity, cm/s | 10.8 (1.9) | 8.9 (1.7)† | 7.7 (1.0) | 13.1 (2.2) | 10.7 (1.8)‡ | 9.5 (0.7) | ||||||
| E velocity, cm/s | 12.1 (3.3) | 10.6 (2.4) | 12.2 (5.4) | 14.7 (3.4) | 12.3 (2.7)§ | 13.4 (4.2) | ||||||
| A velocity, cm/s | 7.9 (1.8) | 8.4 (2.7) | 6.9 (2.6) | 9.7 (2.5) | 9.3 (3.0) | 8.7 (2.2) | ||||||
| Systolic velocity-time integral, cm‖ | 2.2 (0.4) | 1.8 (0.5)† | 1.3 (0.6) | 2.7 (0.5) | 2.1 (0.4)‡ | 1.7 (0.6) | ||||||
| Diastolic velocity-time integral, cm‖ | 3.1 (0.6) | 2.4 (0.6)‡ | 1.8 (0.9) | 3.8 (0.8) | 2.9 (0.7)‡ | 1.9 (0.8) | ||||||
| Isovolumic relaxation time (milliseconds)‖ | 35.3 (16.0) | 57.7 (22.1)† | N/A | 31.3 (21.7) | 49.8 (19.3)† | N/A | ||||||
Fig. 3. TDE traces recorded at the lateral tricuspid annulus from representative subjects in each study group. Top panel: normal control subject; middle panel: mild CF; bottom panel: severe CF. Note the different velocity scale of the recordings. The arrows denote the peak systolic velocity.
[More] [Minimize]The early diastolic velocity of the tricuspid valve annulus was lower in patients from Group 1 than in healthy subjects. All the other diastolic velocities and velocity time integrals were not different between the three groups. There was no correlation between left ventricular mass (either in absolute value or normalized to body surface area or to FFMI) and any of the tissue Doppler indices.
Regional isovolumic relaxation times. The isovolumic relaxation times at the right ventricular and lateral tricuspid annular locations were longer in patients (Table 2). The duration of isovolumic relaxation times of the left ventricle (LV), measured at the mitral annular sites, was not different between patients from Group 1 and healthy subjects.
Reproducibility. The coefficients of intraobserver variation for the tissue Doppler measurements ranged from 2.1% (for the systolic velocity-time integral of the medial mitral annulus) to 15.8% (for the isovolumic relaxation time of the tricuspid valve annulus), with a mean (SD) of 5.4% (3.2%).
Comparison between patients and healthy subjects. The right ventricular free wall was thicker in patients in Group 1 than in healthy subjects, whether expressed as absolute values (0.6 ± 0.2 cm versus 0.5 ± 0.1 cm, p < 0.05) or normalized to the body surface area (BSA) (0.4 ± 0.1 versus 0.3 ± 0.1 cm/m2, p < 0.005). The absolute values of the dimensions of the cardiac chambers, and of the thickness of the left ventricular walls, and left ventricular mass, were not different between patients in Group 1 and healthy subjects, either in absolute values or after normalization to BSA (Table 3). Right ventricular ejection fraction was not different between patients and healthy subjects. Fractional shortening and ejection fraction of the left ventricle showed a trend towards higher values in healthy subjects than in patients.
| Parameter | Healthy Subjects | Group 1 | Group 2 | p Value† | ||||
|---|---|---|---|---|---|---|---|---|
| RVAW | 0.3 (0.1) | 0.4 (0.1) | 0.4 (0.2) | 0.002 | ||||
| RVEDD | 0.9 (0.3) | 1.0 (0.2) | 1.9 (0.1) | 0.2 | ||||
| IVSD | 0.5 (0.1) | 0.5 (0.1) | 0.6 (0.1) | 0.8 | ||||
| LVEDD | 2.8 (0.3) | 2.9 (0.4) | 3.4 (1.3) | 0.2 | ||||
| LVESD | 1.7 (0.2) | 1.8 (0.2) | 2.1 (0.8) | 0.01 | ||||
| LVPW | 0.5 (0.1) | 0.6 (0.1) | 0.6 (0.1) | 0.03 | ||||
| LVFS | 39.1 (7.0) | 35.3 (5.8) | 36.4 (1.6) | 0.05 | ||||
| LVEF | 68.8 (9.0) | 64.3 (7.9) | 68.3 (6.1) | 0.08 | ||||
| RVEF | 41.8 (14.6) | 44.3 (10.3) | 51.5 (11.8) | 0.62 |
The acceleration time of the pulmonic outflow was shorter in patients in Group 1 (129.0 ± 22.8 ms) than in healthy subjects (153.0 ± 18.6 ms, p = 0.0001), and reduced further in patients in Group 2 (81.0 ± 19.2). The values were within the healthy reference range (i.e. > 100 ms) both for patients in Group 1 and for the healthy subjects. The normalized mitral and tricuspid inflow velocity-time integrals were not different.
The mean (SD) FEV1 of patients was 60.7 (24.4) percent predicted in Group 1 and 15.2 (3.8) percent predicted for those in Group 2. The Shwachmann score was greater in patients from Group 1 than in those from Group 2, in accordance with the lesser severity of the disease in these patients (Table 4). The FFM corrected for height was less in the patients with severe lung disease than in those with stable disease (13.8 [1.7] and 16.5 [.9] kg/m2, respectively, p < 0.05). Nine patients in Group 1 and three in Group 2 had an absolute total FFM less than the lower 5th percentile for the healthy age- and sex-matched UK population (6). However, no difference in any of the TDE indices was found between the patients with a low and those with a normal FFM.
| Parameter (units) | Group 1 | Group 2 | ||
|---|---|---|---|---|
| Shwachmann score | 82.1 (12.2) | 42.0 (2.6)† | ||
| PaO2 , mm Hg | 77.4 (7.5) | 55.2 (6.1)† | ||
| PaCO2 , mm Hg | 36.4 (2.4) | 40.7 (7.7) | ||
| SaO2 , % | 95.4 (1.7) | 88.0 (5.7)† | ||
| FEV1, % pred | 60.7 (23.2) | 15.3 (3.9)‡ | ||
| FVC, % pred | 75.5 (20.1) | 31.3 (19.3)‡ | ||
| IL-6, pg/ml | 4.1 (3.5) | 15.3 (6.7)† | ||
| TNFα, pg/ml | 3.2 (1.5) | 3.3 (1.4) | ||
| TNFα soluble receptor I, pg/ml | 1,134.1 (369.0) | 1,183.2 (307.4) | ||
| TNFα soluble receptor II, pg/ml | 3,014.9 (953.8) | 3,025.5 (1,457.3) | ||
| CRP at echo study, mg/L | 89.7 (109.9) | 395.2 (186.1)† | ||
| CRP at end of exacerbation, mg/L | 64.0 (70.6) | N/A |
Relationships between the various parameters were assessed in the 21 clinically stable patients. The tissue Doppler echocardiographic indices of systolic function correlated with clinical, spirometric, and inflammatory parameters (Table 5).
| Peak Systolic Velocity– Right Ventricle | Normalized IVRT– Tricuspid Valve Annulus | |||
|---|---|---|---|---|
| Number of exacerbations | r = − 0.46 | r = 0.77 | ||
| p = 0.02 | p < 0.001 | |||
| FEV1 | r = 0.62 | r = − 0.76 | ||
| p = 0.003 | p < 0.001 | |||
| CRP at end of exacerbation | r = − 0.66 | r = 8.0 | ||
| p < 0.001 | p > 0.5 |
The Shwachmann score correlated inversely with IL-6 (r = −0.58, p < 0.05), and directly with the acceleration time at the pulmonary valve (r = 0.74, p < 0.01). In the clinically stable patients, there was an inverse correlation between the number of episodes of exacerbation of the respiratory symptoms and the midright ventricular peak systolic velocity (Figure 4A) and velocity-time integral. The right ventricular isovolumic relaxation time at midwall was positively correlated to the number of episodes of exacerbation (Figure 4B) and the peak systolic velocity of the right ventricle also correlated to the Shwachmann score (r = 0.46, p < 0.05).
Fig. 4. Scatter plots of correlations between TDE and clinical, spirometric and inflammatory parameters. Circles = healthy subjects; triangles = mild CF; squares = severe CF. (A) Correlation between the peak systolic velocity of the right ventricular free wall (PSVRV) (cm/s) and the number of exacerbations of infection in the 18 mo prior to recruitment in the study. (B) Correlation between the isovolumic relaxation time of the right ventricle (RVIVRT) free wall (ms) and the number of exacerbations of infection in the 18 mo prior to recruitment in the study. (C ) Correlation between the peak systolic velocity of the right ventricular free wall (PSVRV) (cm/s) and the FEV1.
In the clinically stable patients, both the FEV1 and FVC correlated positively with the peak systolic velocity of the right ventricular free wall (Figure 4C), and with the diastolic velocity-time integrals of the right ventricular free wall and of the tricuspid valve annulus.
Inflammatory mediators and CRP were all significantly greater in both groups of patients when compared with control values (Table 4). Circulating IL-6 concentration was greater in patients with severe disease than in the other group (p < 0.01). The CRP measured at the end of the last antibiotic treatment in the clinically stable patients correlated negatively to the peak systolic velocity and to the systolic velocity-time integral of the right ventricle, and positively to the isovolumic relaxation time of both the right ventricle and the tricuspid valve annulus. The peak early diastolic velocity of the tricuspid annulus correlated with the level of IL-6 (r = 0.64, p < 0.005).
A multiple stepwise regression confirmed the effect of FEV1 (F = 11.9, p < 0.05) and of the number of exacerbations (F = 17.0, p < 0.01) on the peak systolic velocity of the right ventricle. Addition of CRP to the analysis did not add to the relationship to pulmonary function (F = 2.6, p = 0.12). The following variables were significant in relation to the isovolumetric relaxation time of the tricuspid annulus: FEV1 (F = 5.2, p < 0.05), number of exacerbations (F = 6.3, p < 0.05); CRP level (F = 1.5, p = 0.22) was not significant. Even when FEV1 was entered on the first step of the regression, IL-6 still correlated to the peak early diastolic velocity of the tricuspid annulus (F = 16.0, p < 0.001). FEV1 was not related to the diastolic velocity of the tricuspid annulus.
We detected subclinical dysfunction of the right ventricle in young adults with cystic fibrosis and chronic bacterial colonization of the respiratory system, when they were clinically stable and free from an exacerbation of the respiratory symptoms. This dysfunction affected both systolic contractility, as demonstrated by reduced peak velocity of right ventricular longitudinal shortening, and diastolic filling, as indicated by prolongation of the isovolumic relaxation time.
M-mode echocardiographic studies in the 1970s and early 1980s demonstrated increased thickness of the right ventricular free wall and dilatation of the right ventricle in CF (13-15). Right and left ventricular systolic time intervals, as well as right ventricular wall thickness and internal dimension, were related to vital capacity, residual volume, and clinical scores of the severity of disease (16, 17), though these studies provided only limited morphologic and functional information.
Panidis and colleagues (17) studied 17 patients, using two-dimensional and Doppler echocardiography, and found no differences between them and healthy subjects. As in our study, no patient had tricuspid regurgitation measurable by continuous wave Doppler. Hoffmann and colleagues (18) found similar right ventricular dimensions in three groups of patients with increasing severity of cor pulmonale, and lower fractional shortening of the right ventricle in patients with more severe pulmonary hypertension.
Radionuclide angiography correlates well to the ejection fraction measured by echocardiography in patients with chronic obstructive pulmonary disease (19). Burghuber and colleagues (20) found right ventricular ejection fraction to be lower in patients with pulmonary hypertension (45 ± 2% versus 59 ± 7%), and inversely correlated to pulmonary arterial pressure. Right ventricular contractility, however, was greater in patients with pulmonary hypertension, which they suggested might be due to “preserved or even increased contractility in the face of an increased afterload.” Piepsz and colleagues (21) reported a moderate decrease of the right ventricular ejection fraction in children with CF, although in some patients with terminal lung disease the ejection fraction was still normal.
Thus, conventional methods for the assessment of right ventricular structure and function yield contradictory results in patients with CF because of the suboptimal results of gray-scale imaging.
The clinical application of TDE was described by McDicken and colleagues (22). Whereas conventional Doppler echocardiography is used to record the blood flow within the heart and great vessels, which occurs at relatively high velocities and gives signals of rather low amplitude, TDE uses similar principles to record shifts in ultrasonic frequencies from moving tissue rather than from blood. By using different filters, these signals, which are of low velocity and high amplitude, can be recorded preferentially. Precise measurement of regional myocardial velocities is achieved by placing an electronic “sample volume” in the region of interest. The technique has been extensively validated in vitro and in vivo (23-25) and is currently under active investigation for an expanding number of clinical applications (26, 27). Its theoretical basis means that it is a powerful technique for quantifying regional myocardial function. The greater feasibility of TDE of the right ventricle in both our patient and our control groups reflects the high signal-to-noise ratio of this method.
In one earlier study, subclinical right ventricular systolic dysfunction (detected with exercise echocardiography and radionuclide ventriculography) was identified in approximately 20% of outpatients with CF (28). Our study has demonstrated that patients with CF, who are stable and free of clinically apparent right ventricular failure, nonetheless have lower systolic velocities in the right ventricular free wall and at the lateral tricuspid annulus when compared with age- and sex-matched healthy subjects. Together with the longer isovolumic relaxation times at the same locations, this suggests impaired systolic and diastolic right ventricular function at this subclinical stage.
Supporting the validity of our findings is the earlier report on TDE of the right ventricle in healthy subjects (29), with values very similar to our own findings in the healthy subjects. The myocardial velocities are even lower in patients with respiratory failure and end-stage CF lung disease. This provides further support to our hypotheses that the severity of lung disease relates to the degree of right ventricular impairment detectable with TDE.
A recent consensus conference (30) identified the need to “improve the accuracy and validity of [...] measures of biological activity and clinical efficacy in CF.” The present study provides insight into the pathophysiology of right ventricular dysfunction and suggests that TDE may be useful for monitoring right ventricular function in patients with chronic lung disease. It might be possible to identify at a subclinical stage patients with an increased risk of developing right ventricular failure, and this may have important implications for treatment, which need to be tested in further studies.
Right ventricular dysfunction in our patients may be due to pulmonary hypertension, or to the direct effect of inflammatory mediators on the right ventricle, acting in a cumulative manner during repeated episodes of pulmonary infections.
Evidence for a role of pulmonary hypertension. The patients had greater pulmonary arterial pressures, as shown by shorter pulmonary acceleration times, than did healthy subjects, although patients in Group 1 had pulmonary acceleration times still in the normal range (> 100 ms). A possible hypothesis may be that patients in Group 1 experience bouts of pulmonary hypertension during exacerbations of respiratory symptoms, which may exert a cumulative effect leading to right ventricular hypertrophy. Support for this interpretation comes from studies in CF and chronic obstructive pulmonary disease (31). Moreover, it was demonstrated recently that subclinical pulmonary hypertension develops in a substantial proportion of patients with cystic fibrosis and stable lung disease, and that it correlates with the frequency and severity of episodes of arterial oxygen desaturation and with FEV1 (32). A catabolic response with depletion of FFM is common in adults with CF (6), though this is not simple malnutrition. The potential subclinical pulmonary hypertension may explain the preservation of cardiac muscle despite depletion of FFM.
Patients in Group 1 had a thicker right ventricular free wall than did healthy subjects. Right ventricular hypertrophy begins very early in the course of CF (15). It is known that even in its early stages, left ventricular hypertrophy is associated with diastolic dysfunction, which may be subclinical, but can be detected with TDE (33). Our findings suggest that in the right ventricle, mild hypertrophy is associated with impairment of both systolic and diastolic function.
Evidence for a role of inflammation. Patients with CF have chronic and sustained inflammation, which increases at the time of an exacerbation of respiratory symptoms. CRP levels reach a nadir at the end of a successful course of antibiotics, but then increase gradually, even before worsening of respiratory symptoms (6). There is great interindividual variability in the magnitude of this inflammatory response: some patients have normal levels of circulating inflammatory mediators at the end of the treatment, whereas others have markedly elevated values (6, 12).
Although this is only circumstantial evidence, we found a clear relationship linking the level of the systemic inflammatory response, as indicated by the CRP at the end of the exacerbation, and the systolic velocity and velocity-time integral of the right ventricular free wall. There is also a positive correlation of the CRP with the right ventricular and tricuspid annular isovolumic relaxation time. Thus, this index of inflammation relates in the expected physiologic manner to indices of regional systolic and diastolic function. Stepwise regression demonstrated that a low FEV1 was related to decreased systolic function of the right ventricle and to impaired relaxation at the tricuspid annulus. CRP appears to be a surrogate for the severity of impairment of lung function. These findings suggest a primary effect on the systolic and diastolic function of the right ventricular myocardium, though inflammatory mediators may have independent effects. However, this was not unequivocally demonstrated by this study, and the relationship to IL-6 may reflect the link of the inflammatory mediators to lung injury.
CF is associated with increased circulating levels of inflammatory mediators (IL-6, IL-1, TNFα), which are generated as a result of severe pulmonary inflammation. Some inflammatory mediators (IL-1β, TNFα) are potent depressants of cardiac contractility (2). Long-term expression of TNFα within the heart may produce cardiac decompensation (34). It is likely that chronic or repeated bouts of pulmonary infection and inflammation have a cumulative effect on cardiac function. If inflammation were the cause of right ventricular abnormalities, one would expect the left ventricle to be impaired also. The finding of reduced systolic and diastolic velocity-time integrals of the mitral annulus, and of a trend for the fractional shortening of the left ventricle to be lower in patients, provide further circumstantial evidence for an inflammatory-mediated mechanism.
The assessment of ventricular function by TDE was not corroborated by an independent method, such as radionuclide ventriculography. However, the ability of TDE to detect regional changes in myocardial velocities has been extensively validated. In experimental studies myocardial velocities have been found to correlate very closely with local shortening assessed by implanted microsonic crystals and with invasively measured indices of regional contractility (35). Detailed mapping of velocities may provide more insight into the physiology of ventricular function than a global index such as the ejection fraction.
In conclusion, patients with cystic fibrosis have subclinical right ventricular systolic and diastolic dysfunction as shown by Tissue Doppler Echocardiography, even when they are clinically stable and free from signs of heart failure. The severity of the impairment parallels the severity of the lung disease and of the inflammatory process. Tissue Doppler Echocardiography has a better feasibility than conventional gray-scale imaging, and may prove useful for monitoring right ventricular function in patients with chronic lung disease or other diseases associated with right ventricular involvement, as well as for testing new treatments for cystic fibrosis.
The writers thank Lisette S. Nixon B.Sc., who performed the assays for the measurement of the inflammatory mediators.
Supported by the Cystic Fibrosis Trust UK.
| 1. | Royce SWCor pulmonale in infancy and early childhood: report on 34 patients, with special reference to the occurrence of pulmonary heart disease in cystic fibrosis of the pancreas. Pediatrics81951255 |
| 2. | Kelly RA, Smith TWCytokines and cardiac contractile function. Circulation951997778781 |
| 3. | Palka P, Lange A, Fleming AD, Fenn LN, Bouki KP, Shaw TRD, Fox KAA, McDicken WN, Sutherland GRAge-related transmural peak mean velocities and peak velocity gradients by Doppler myocardial imaging in normal subjects. Eur Heart J171996940950 |
| 4. | Sutherland GR, Lange A, Palka P, Grubb N, Fleming A, McDicken WNDoes Doppler myocardial imaging give new insights or simply old information revisited (editorial). Heart761996197199 |
| 5. | Shwachmann H, Kulczycki LLong term study of 105 patients with cystic fibrosis: studies made over a five to fourteen year period. Am J. Dis Child9619586 |
| 6. | Ionescu AA, Nixon LS, Evans WD, Stone MD, Lewis-Jenkins V, Chatham K, Shale DJBone density, body composition and inflammatory status in cystic fibrosis. Am J Respir Crit Care Med1622000789794 |
| 7. | Mannaerts HF, Balk AH, Simoons ML, Tijssen J, van der Borden SG, Zondervan P, Sutherland GR, Roelandt JR. Changes in left ventricular function and wall thickness in heart transplant recipients and their relation to acute rejection: an assessment by digitised M mode echocardiography. Br Heart J 1992;.68: 356–364. |
| 8. | Sahn DJ, DeMaria A, Kisslo J, Weyman ARecommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation58197810721083 |
| 9. | Teicholz LE, Cohen MV, Sonnenblick BM, Gorlin RStudy of left ventricular geometry and function by B-scan ultrasonography in patients with and without asynergy. N Engl J Med2911964120128 |
| 10. | Feigenbaum H, Popp RL, Wolfe SBUltrasound measurements of the left ventricle: correlations with angiocardiography. Arch Intern Med11291972461475 |
| 11. | Kaul S, Tei C, Hopkins JM, Shah PMAssessment of right ventricular function using two-dimensional echocardiography. Am Heart J1071984526531 |
| 12. | Nixon LS, Yung B, Bell SC, Elborn S, Shale DJCirculating immunoreactive Interleukin-6 in cystic fibrosis. Am J Respir Crit Care Med157199817641769 |
| 13. | Gewitz M, Eshaghpour E, Holsclaw DS, Miller HA, Kawai NEchocardiography in cystic fibrosis. Am J Dis Child1311977275280 |
| 14. | Allen HD, Taussig LM, Gaines JA, Sahn DJ, Goldberg SJEchocardiographic profiles of the long-term cardiac changes in cystic fibrosis. Chest751979428433 |
| 15. | Hirschfeld SS, Fleming DG, Doershuk C, Liebman JEchocardiographic abnormalities in patients with cystic fibrosis. Chest751979351355 |
| 16. | Lester LA, Egge AC, Hubbard VS, Camerini-Otero CS, Fink RJEchocardiography in cystic fibrosis: a proposed scoring system. J Pediatr971980742748 |
| 17. | Panidis IP, Ren JF, Holsclaw DS, Kotler MN, Mintz GS, Ross JCardiac function in patients with cystic fibrosis: evaluation by two-dimensional and Doppler echocardiography. J Am Coll Cardiol61985701706 |
| 18. | Hoffmann H, Vogt L, Volkner EQuantitative two-dimensional echocardiography: study results in patients with cystic fibrosis. Kinderarztl Prax581990523530 |
| 19. | Starling MR, Crawford MH, Sorensen SG, O'Rourke RAA new two-dimensional echocardiographic technique for evaluating right ventricular size and performance in patients with obstructive lung disease. Circulation661982612620 |
| 20. | Burghuber OC, Salzer-Muhar U, Gotz M. Right ventricular contractility is preserved in patients with cystic fibrosis and pulmonary artery hypertension. Scand J Gastroenterol 1988;143(Suppl):93–98. |
| 21. | Piepsz A, Ham HR, Millet E, Dab IDetermination of right ventricular ejection fraction in children with cystic fibrosis. Pediatr Pulmonol319872428 |
| 22. | McDicken WN, Sutherland GR, Moran CM, Gordon LColour Doppler velocity imaging of the myocardium. Ultrasound Med Biol181992651654 |
| 23. | Miyatake K, Yamagishi M, Tanaka N, Uematsu M, Yamazaki N, Mine Y, Sano A, Hirama MA new method for the evaluation of left ventricular wall motion by color-coded tissue Doppler imaging: in vitro and in vivo studies. J Am Coll Cardiol251995717724 |
| 24. | Bach DS, Armstrong WF, Donovan CL, Muller DWMQuantitative Doppler tissue imaging for assessment of regional myocardial velocities during transient ischemia and reperfusion.Am Heart J1321996721725 |
| 25. | Derumeaux G, Ovize M, Loufoua J, Andre-Fouet X, Minaire Y, Cribier A, Letac BDoppler tissue imaging quantitates regional wall motion during myocardial ischemia and reperfusion. Circulation97199819701977 |
| 26. | Nagueh SF, Middleton KJ, Kopelen HA, Zoghbi WA, Quinones MADoppler tissue imaging: a non-invasive technique for evaluation of left ventricular relaxation and estimation of filling pressures.J Am Coll Cardiol30199715271533 |
| 27. | Garcia MJ, Rodriguez L, Ares M, Griffin BP, Thomas JD, Klein ALDifferentiation of constrictive pericarditis from restrictive cardiomyopathy: assessment of left ventricular diastolic velocities in longitudinal axis by Doppler tissue imaging. J Am Coll Cardiol271996108114 |
| 28. | Moskowitz WB, Gewitz MH, Heyman S, Ruddy RM, Scanlin TFCardiac involvement in cystic fibrosis: early non-invasive detection and vasodilator therapy. Pediatr Pharmacol (New York)51985139148 |
| 29. | Kukulski T, Hubbert L, Arnold M, Wranne B, Hatle L, Sutherland GRSNormal regional right ventricular function and its changes with age: a Doppler myocardial imaging study. J Am Soc Echocardiogr132000194204 |
| 30. | Ramsey BW, Boat TFOutcome measures for clinical trials in cystic fibrosis. J Pediatr1241994177192 |
| 31. | MacNee WPathophysiology of cor pulmonale in chronic obstructive pulmonary disease. Part One. Am J Respir Crit Care Med1501994833852 |
| 32. | Fraser KL, Tullis DE, Sasson Z, Hyland RH, Thornley KS, Hanly PJPulmonary hypertension and cardiac function in adult cystic fibrosis: role of hypoxemia. Chest115199913211328 |
| 33. | Pearson AC, Labovitz AJ, Mrosek D, Williams GA, Kennedy HLAssessment of diastolic function in normal and hypertrophied hearts: comparison of Doppler echocardiography and M-mode echocardiography. Am Heart J113198714171425 |
| 34. | Mann DL. The effect of tumor necrosis factor-alpha on cardiac structure and function: a tale of two cytokines. J Card Fail 1996;2(4 Suppl): S165–172. |
| 35. | Gorcsan J, Strum DP, Mandarino WA, Gulati UK, Pinsky MRQuantitative assessment of alterations in LV contractility with color-coded tissue Doppler echo: comparison with sonomicromanometry and pressure volume relations. Circulation95199724232433 |
