Echocardiographic Abnormalities in Patients with Cystic Fibrosis* StephenS. Hirschfeld, M.D.; David G. Fleming, M.D.; Carl Doershuk, M.D.; and Jerome LiebfiUJn, M.D.
An echographic study was undertaken to evaluate left (LV) and right ventricnlar (RV) function in 30 patients with cystic fibrosis. Echographic recording of the pnlmonary and aortic valve echogram permitted measurement of the phases of right and left ventricnlar systole. The ratio of the LV preejection period/LV ejection time (LPEP/LVET) and shortening of the LV internal dimension %SID was employed to reftect LV function, whDe RV preejection period/RV ejection time (RPEP/RVEI') has exceUent correlation with pnlmonary artery diastoHc pressure. RPEP/RVET and two other echographlc measurements, right ventricnlar 'Wall (RVW) and internal dimension (RVD) were compared with pnlmonary function tests and clinical scores. RPEP/RVET correlated weD
with pen:ent vital capaclty(%VC), r = -0.73, pen:ent residual volume (%RVol) r = +0.72, and clinical score, r = -0.77. MnltiUnear repesslon of RPEP/RVET, RVD, and RVW improved correlation for % VC (r = -0.80), %RVoi, r = +0.82, and clinical score, r = -0.84. Patients in overt right heart failure exhibited elevated RPEP/RVET (mean= OAS) when compared to patients not in right heart failure '(mean = .33). Marked diminution of LV function was present in two patients. A variety of cardiovascnlar abnormaHties were demonstrated echographically and were valnable in asse q the degree of cardiae involvement in patients with cystic
The majority of patients with cystic fibrosis (CF) die as a consequence of cor pulmonale, the cardiac complication of chronic pulmonary disease. 14 Progressive loss of pulmonary function and the subsequent hypoxemia lead to pulmonary hypertension and right ventricular hypertrophy ( RVH) or cor pulmonale.4.11 Right heart failure ( RHF) occurs later in the natural history of chronic cor pulmonale and most patients with cystic fibrosis expire within one to two years of the development of right heart failure. Early detection of right ventricular dysfunction and pulmonary hypertension would be of utmost importance as improvement in functional pulmonary status with more intensive therapy may improve hypoxemia, alleviate cor pulmonale, and prolong patient life.• To date, detection of the presence of cor pulmonale has been difficult in patients with cystic fibrosis because recognition of signs of right ventricular enlargement and right heart failure may be delayed by the manifestations of chronic emphysema and pulmonary disease. 3 Routine laboratory tests such as the standard electrocardiogram and chest roentgenogram often do not indicate the onset and progression of cor pul-
monale until it is well advanced. 14 Orthogonal electrocardiography (vectorcardiography), which has more accurately reflected RVH in cystic fibrosis, does not permit quanti.B.cation of cardiac or pulmonary involvement, and may reflect chronic hemodynamic problems rather than acute changes.8 The echocardiogram has recently "been shown to reflect thickening of the right ventricular free wall ( RVW) and dilation of the right ventricular internal dimension (RVD) early in the course of cystic fibrosis.'·' These echographic measurements have been shown to correlate well with pulmonary function tests and clinical status. As the underlying hemodynamic abnormality in cor pulmonale is pulmonary hypertension, the echocardiogram may be a particularly useful tool in cystic fibrosis. We have demonstrated that the pulmonary valve echogram offers a reliable means of timing the events of right ventricular systole. 10 Two right ventricular systolic time intervals ( RVSTI) have been measured: 1) right ventricular ejectiOJ1 time ( RVET); and 2) right ventricular pre-ejection period (RPEP). The ratio RPEP/RVET provided a reliable assessment of the diastolic pressure in the pulmonary vascular bed in children with congenital heart disease, and sequential changes of RPEP/ RVET reflected hemodynamic alterations occurring within the pulmonary circulation.11.12 Left ventricular systolic time intervals (LVSTI) have been measured from the aortic val~e echogram10 and
°From Case Westem Reserve University School of Mediciue, Department of Pediatrica, Rainbow Babies and Chilthenl Hospital, Cleveland. Manuscript received Mav 25; revision &901'Pted September 11. &prim teqfl6#8: Dt. Hirschfeld. BalnbotD &: Childrtm~ Hos7'ittrl, 2101 Adelbert &ad, Cletieltmtl 44106
CHEST, 75: 3, MARCH, 1979
fibrosis.
ECHOCARDIOGRAPHIC ABNORMAUTIES IN CYsnC ABROSIS 351
have been employed to assess left ventricular function.18 The shortening of the left ventricular internal dimension ( JSID) has also been valuable in assessing LV dysfunction. 14 This echographic study was undertaken to measure RPEP/RVET, LPEP/LVET, JSID, RVW, and RVD in patients with cystic fibrosis. The sensitivity of each measurement in assessing cardiac and pulmonary status was evaluated PATIENTS AND METHoDS
Thirty patients with cystic fibrosis were enrolled in the study (Table 1 ) . There were 19 females and 11 males, age 0.3 to 35 years. Most ( 24/30) patients had been admitted to the hospital for intensive pulmonary toilet. The mean age of all patients was 14.5 years with a standard error of 1.8 years. The patients generally had severe cystic fibrosis as reflected by a mean clinical score of 52± SE of 3.0. The severity of illness was also reflected by the deaths of nine patients during the study and clinical evidence for right heart failure in 18/30 patients. Pulmonary functional status was evaluated employing a 9 liter Collins spirometeru and a total body plethysmograph.te The vital capacity as a percentage of normal (we) and residual volume as a percentage of normal ( IRVol) were Table 1--SumlJIGlY of Data (30 Patienla) Age ofPt
9.5 7 13 13.5 10 20.5 23 26
19 30 35 35
18 10 14.5 24 14.5 19 23 19 7 10 20 26
.3 4.5 6 4 4 3.5
RPEP/RVET
VC%
.50
30
.33 .46 .29 .46 .38 .47
68
.54
50
.38 .40 .48 .58 .41 .28 .50
.41 .36 .25 .39 .24 .41 .33 .55 .53 .36 .33 .26 .24 .30 .40
41
64 27
51 55
53 40 40 47 50 91 27
43 53 115 45 62 43 72
32 41
RVol%
Score
RHF
370
39 72
+ 0 + 0 + + +
250
307 155 285 298 346 333 361 330 203 555 233 100 190 313 330 98 219 90 340 247 388 365
48
71 26 58
38
40
66
45 34 40 48
81 37 35 47 72 48 86 34
45 34 38
+
0
+ 0 + + 0 +
calculated. These measurements were obtained in 24 of 30 patients over six years of age. The clinical score was calculated according to the method of Shwachman and Kulczycki.U The presence of right heart failure was defined by evidence on physical examination of systemic venous congestion (hepatomegaly, edema, elevated jugular venous pressure) or tricuspid insufficiency. The measurement of RVSTI and LVSTI was accomplished from a simultaneous, rapid speed ( 100 mm/sec), recording of the electrocardiogram and pulmonary or aortic valve echograms in a manner previously described.to RVET was measured from opening to closure of the pulmonary valve echo, while RPEP was measured from the onset of ventricu-
~--
-
-
- ----;:-------====--- - ---- ...
-
~
----
~
-
.
0 0 0 0 0
+ + + + 0 + 0 0 0
+
RPEP - right pre-ejection period; RVET - right ventricular ejection time; VC% - vital capacity; RVol% "" residual volume; RHF - right heart failure.
352 HIRSCHFELD ET AL
FIGURE I. The measurement of right ventricular systolic time intervals was demonstrated from the posterior pulmonary valve cusp echogram. The right pre-ejection period ( RPEP) was measured from the onset of the QRS to the rapid descent of the pulmonary cusp. The right ventricular ejection time ( RVET) was measured from pulmonary cusp opening to closure, the point at which the leaflet reapproaches its closed, thickened baseline. There was marked phasic variation of the time intervals. PA: pulmonary artery.
FIGURE 2. Measurement of left ventricular systolic time intervals was accomplished from the aortic valve (Ao) echogram. LPEP was left pre-ejection period and L VET left ventricular ejection time.
CHEST, 75: 3, MARCH, 1979
lar depolarization, indicated by the ECG, to pulmonary valve opening ( Fig 1). LPEP/LVET was measured in a similar manner from the aortic valve echogram (Fig 2) . RVW and RVD in millimeters were measured at end-diastole and expressed as an absolute measurement without indexing for body surface area ( Fig 3) . Measurement of RPEP/RVET and LPEP/LVET should be measured during quiet respiration, if possible. Five separate measurements were made and averaged for the final PEP /VET ratio. Diastolic and systolic LV dimensions were measured and the %SID was computed (Fig 3 ).14 ,19 Values were compared to published normal values.2o Echocardiography, pulmonary function study, and clinical assessment were performed within a 24-48 hour period. Statistical analysis was performed by linear and multiple linear regression analysis. A correlation coefficient ( r value) was obtained. An r value greater than 0.7 indicated excellent correlation, between 0.5-0.7 good, and below 0.5 was poor. The standard error of the estimate was computed, as was the 95 percent confidence limits for the mean %VC, %RVol, and clinical score. The differences between patient groups was evaluated by Student t-test. Group characteristics were expressed as the mean± one standard error ( ±SE). RESULTS
Echocardiographic Findings RPEPIRVET: Mean± SE for the group was 0.38 ± 0.02 with a range of 0.24-0.55. There was excellent correlation of RPEPI RVET with $\'C ( r = -0.73), ~RVol (r = + 0.72) and clinical score (r = -0.77) (Table 1). RPEPI RVET increased with patient age ( r = + 0.53). The mean ratio was 0.48 ± .02 for •
•
.
'
.
' . . . ,v
~-
..... ,
.. --
.' '
~. ~·-
I
-'
-
---
_..,
....~
.
} CHEST WALL
.: _. _It :---~~
·· -
RVW RVD
patients in right heart failure compared to 0.33 + .02 in those without RHF ( P < 0.01 ) . Only 2/16 patients with RHF had RPEP/RVET less than 0.40, while no patient without RHF had an RPEPIRVET over0.40. RVW: Mean + SE RVW was 3.6 ± 0.23 mm, with a range from 2-5 mm. Correlation of RVW was good with $\'C (r = -0.56), $RVol (r = + 0.65) and clinical score ( r = -0.56). The RVW was related to RPEPIRVET (r = 0.62) . RVD: Mean RVD ± SE was 23 ± 2 mm with a range of 12-41 mm. RVD 'correlated well with $\'C (r = -0.51), but poorly with ~RVol and clinical score. There was no relationship between RVD/m2 and RPEPIRVET, but there was good correlation between RVW and RVD ( r = + 0.58) . LPEPILVET: Mean+ SE LPEPILVET for the group was 0.36 ± .02, only slightly increased from normal. The range was 0.20-0.52. Twelve patients had LPEP I LVET ;::: 0.40, which is the upper limit of normal and suggestive of left ventricular dysfunction. Seven had a LPEPI LVET greater than 0.45. The ratio was not significantly different ( P > .10) in patients in right heart failure ( 0.38 + .02) and those not demonstrating right heart failure ( 0.35 + .03). There was poor correlation between RPEP I RVETandLPEPILVET (r = + 0.48). LVD and ~SID: The mean LVD was 39 ± 10 mm (range 23-70 mm) and was not significantly different from normal. There was no difference in the group with right heart failure and normal patients. In two patients, $SID was depressed below the normal range ( 25-35 percent); $SID was 13 and 10 percent respectively in these patients. The ventricular septum moved abnormally in five patients with right heart failure and dilated right ventricles, so that Table Z-Sipi/ieanl Repeuion Equalioru
-SEPTUM
R
LV
} LVW
3. The ultrasonic beam transverses a position through the body of the right ventricle ( RV) and left ventricle (LV}. All measurements were made at the onset of the QRS complex. Right ventricular wall ( RVW) was measured from the undersurface of the chest wall to RV endocardium. RV dimension was measured from the RV endocardium to the right surface of the interventricular septum. The LV was dilated to 7.0 em and the calculated shortening of the LV internal dimension was markedly reduced and indicated the presence of myopathy. FIGURE
CHEST, 75: 3, MARCH, 1979
VC%
-0.73 -0.51 =72.6-0.94 (RVD) =78-8 (RVW) -0.56 = 112-115 (RPEP /RVET)0.78 (RVD)+1.6 (RVW) -0.80
r120-173 (RPEP/RVET)
±12 ±14 ±14
±5.6 ±6.6 ±6.6
t888
±10
±4.7
(RPEP/RVET)-87 RVol% =62+RVW ""-138+781(RPEP/RVET) +2.2(RVD) +11 (RVW) Score
SEE 95%CL
I= 100-123(RPEP/RVET) =78.8-8.3 (RVW) = 113-135(RPEP/RVET)0.35(RVD) +0.29(RVW)
0.72 0.65
±66 ±31 ±68 ±32
0.85 -0.77 -0.56
±62 ±29 ±10 ±4.7 ±14 ±6.6
-0.82
±9
±4.2
R = correlation coefficient; SEE - standard error of estimate; 95%CL = 95% confidence limits for the mean; RPEP - right pre-ejection period; RVD .. right ventricular dimension; RVET = right ventricular ejection time; RVol% = percent residual volume; VC% - percent vital capacity.
ECHOCARDIOQRAPHIC ABNORMALITIES IN CYSTIC FIBROSIS 353
ISID could not be computed in many patients. Multiple Ccmelations: When three parameters, RVD, RVW, and RPEP/RVET, were correlated with we, IRVol, and clinical score, the coefficient of correlation was improved in predicting we (r -.80, mvol (r = 0.82) and clinical score (r = - 0.84) (Table2). Sequential and Catheterization: Ten patients developed signs of right heart failure during the course of the study and permitted evaluation of intervention with digoxin. The effects of digoxin on RPEP/RVET, LPEP/ LVET, and ISID were insignffi.cant in all ten patients. In patients who had evidence of right heart failure, the echograms demonstrated marked inspiratory, expiratory variation of both RPEP/RVET and LPEP/LVET. With expiration, there was prolongation of LPEP or RPEP and associated shortening of the LVET or RVET, resulting in prolongation of each ventricular ratio. Four patients with severe right heart failure underwent cardiac catheterization. Each patient had severe pulmonary hypertension with a mean pulmonary arterial pressure above 50 mm Hg. The RPEPI RVET was elevated in all patients and ranged from 0.40-0.56. There was respiratory variation in the RPEP/RVET as described above, and this corresponded to a 15-30 mm Hg decrease in the systolic and diastolic pulmonary arterial pressure. Three patients had an elevated pulmonary arterial wedge pressure, suggesting left ventricular dysfunction and in each patient the LPEP I LVET was increased (Table3).
=
+
DISCUSSION
Echographic recording of the aortic and pulmonic valves has facilitated measurement of RV and LVSTI. A systolic time interval is a phase of electromechanical systole, the duration of which is governed by four basic factors: 1) preload, indicated by ventricular end diastolic volume; 2) afterload or resistance to ventricular emptying, reflected by arterial diastolic pressure; 3) contractile state of the myocardium; and 4) rate and sequence of intraventricular electrical conduction. 18 The use of the ratios LPEP /LVET and RPEP/
Patient No.
s
PA Pressure D Mean
11 13 24 30
80 85 55 60
40
50 40 40
60 70 50 50
BP
CI
LPAW
110/70 100/70 100/80 105/60
4.3 5.5 3.2 4.1
22 16 28
8
BP - blood pressure; CI - cardiac index; D - diastolic; LPAW • left pulmonary artery wedge; PA - pulmonary artery; S • systOlic.
354 HIRSCHFELD ET AL
RVET eliminates the influence of heart rate and age, so that patients of different ages can be compared and the same subject may be evaluated sequentially. 11·18 An elevated RPEP/RVET has been predictive in indicating elevated diastolic pressure ( afterload) in the pulmonary circulation, but may also be prolonged if RV contractility is diminished.11·12 An increased RPEP/RVET in a patient with cystic fibrosis may, therefore, be a result of either increased pulmonary artery pressure, as was the case in patients who underwent cardiac catheterization, or contractile abnormalities. The latter may be a consequence of intermittent pulmonary hypertension, hypoxia, or acidosis. In either case, RPEP/RVET should provide an excellent guide to the severity of pulmonary involvement and provide an index to the contractile state of the right ventricular myocardium. In our study, RPEP/RVET showed excellent correlation with other indicators of the severity of cystic fibrosis, namely We, IRVol, and ShwachmanKulczycki score. The ratio also provided excellent separation of those patients in right heart failure from those demonstrating no signs of right heart failure, for when RPEP/RVET was 0.40 or greater, the patient was in potential or overt right heart failure. Sequential evaluation was particularly important in that deterioration in right ventricular function was predicted in three patients, who developed signs of right heart failure shortly after an echogram demonstrated an elevated RPEP/RVET. The presence of marked phasic respiratory changes was present in all patients in RHF. This appears to reflect the signffi.cant variations in pulmonary arterial pulse amplitude, pulsus paradoxus, that has been suggested in patients with increased intrathoracic pressure secondary to asthma. 18 Impairment of right ventricular filling may also occur in patients with severe cystic fibrosis. The impaired right ventricular filling results in diminished right and left ventricular stroke volume or preload when the intrathoracic pressure increases during expiration. The phasic respiratory changes in preload (ventricular filling) produce the respiratory variation in RPEP/RVET and LPEP/LVET seen in patients with the most severe cystic fibrosis and right heart failure. Because of the variation in ratios, at least five separate measurements should be averaged for the final ratio. In patients with respiratory variation, either measurement was markedly elevated and provided an index of the severity of the disease. Digitalization did not produce reduction in LPEP/LVET or RPEP/RVET, although clinical improvement was evident in most patients. Twelve patients had increased LPEP/LEVT ( > 0.40) which indicated that while left ventricular dysfunc-
CHEST, 75: 3, MARCH, 1979
tion may be present in a few patients, it was not a major determinant in influencing pulmonary function tests or the presence of right heart failure. Marked depression of left ventricular function suggested by decreased %SID and/ or very elevated LPEPI LEVT demonstrated that in an occasional patient the left ventricle may be markedly myopathic. The increased pulmonary wedge pressure measured at catheterization further attests to the importance of left ventricular dysfunction in patients with cystic fibrosis. Other investigators have demonstrated good correlation between RVW/m2, RVD/m2 and clinical severity of cystic fibrosis. 7•8 It is also doubtful that RVW can be measured with the accuracy that has been reported in the other echographic studies of cystic fibrosis. The beam width of the standard ultrasonic transducers permits measurement of structures to the nearest millimeter at best and not tenths of millimeters as reported in other studies. Our study did not index RVW or RVD for body surface area as there is no evidence that either measurement has a constant relationship to surface area in young subjects. RVW showed good correlation with RPEP /RVET suggesting the thickening of the RVW was occurring concomitantly with the elevated pulmonary vascular resistance or RV dysfunction. When RVW and RVD are combined with RPEP/RVET, correlations further improved between echocardiography and clinical severity of cystic fibrosis. The clinical score, ~c. and mvol could be estimated by the regression formula using the three echographic parameters. Problems with measurement of RPEP/RVET have been in recording the pulmonary valve echogram. Air trapping, which is very common in cystic fibrosis, may interfere with recording the pulmonary valve echogram. In this study, we were able to visualize the pulmonary valve in approximately twothirds of the patients. Pulmonary valve recording was not hindered by advanced pulmonary disease or increased age. The patient group in which the pulmonary valve could not be recorded had the same mean age and clinical score as the reported patient population. In summary, RPEP/RVET, alone or combined with RVD and RVW, is an excellent index of the severity of the cardiovascular complications of cystic fibrosis. Patients in RHF or imminent RHF may be separated from those without RHF. Left ventricular function may be abnormal, and in an occasional patient the left ventricle may be myopathic. Further evaluation employing correlation of cardiac catheterization and echograpbic findings will be necessary in a large group of patients before the ability of
CHEST, 75: 3, MARCH, 1979
this technique to detect early development of cor pulmonale and response to therapy can be completely evaluated.
1 Liebman J, Lucas R, Moss A, et al: Cor pulmonale and related cardiovascu1ar eHects of cystic fibrosis. In Cystic Fibrosis, Projections into the Future. Mangos JA, Talamo RC, eds. Miami, Symposia Specialists, 1976, pp 41-79 2 Wood RE, Boat TF, Doershuk CF: State of the art: Cystic fibrosis. Am Rev Respir Dis 113:833-878,1976 3 Siassi B, Moss AJ, Dooley RR: Clinical recognition of cor pulmonale in cystic fibrosis. J Pediatrics 78:794-805, 1971 4 Goldring RM, Fishman AP, Turino JM, et al: Pulmonary hypertension and cor pulmonale in cystic fibrosis of the pancreas. J Pediatrics 65:501-524, 1964 5 Thomas AJ: Chronic pulmonary heart disease. Br Heart J 34:653-657, 1972 6 Liebman J, Doershuk CF, Rapp C, et al: The vectorcardiogram in cystic fibrosis: Diagnostic significance and correlation with pulmonary function tests. Circulation
35:552-569, 1967
7 Rosenthal, A, Tucker CR, Williams RG, et al: Echocardiographic assessment of cor pulmonale in cystic fibrosis. Pediat Clin North Am 23:327-343, 1976 8 Ryssing E:Assessment of cor pulmonale in cystic fibrosis by echocardiography. Acta Pediatr Scand 66:753-756, 1977 9 Gewitz M, Eshaghpour E, Holsclaw DS, et al: Echocardiography in cystic fibrosis. Am J Diseases Child 131 :275-280, 1977 10 Hirschfeld S, Meyer R, Schwartz D, et al: Measurement of right and left ventricular systolic time intervals by echocardiography. Circulation 51:304-309, 1975 11 Hirschfeld S, Meyer R, Schwartz DC, et al: The echocardiographic assessment of pulmonary artery pressure and pulmonary vascular resistance. Circulation 52:641-650, 1975 12 Riggs T, Hirschfeld S, Borkat G, et al: Right ventricular systolic time intervals in the assessment of the pulmonary vascular bed. Circulation 57:939-947, 1978 13 Weissler AM, Lewis RP, Leighton RF: The systolic time intervals as a measure of left ventricular performance in man. In Progress in Cardiology, (Vol!) (Yu PN, Goodwin JF eds). Philadelphia, Lea and Febiger, 1972, pp 155183 14 Fortuin NJ, Hood WP, Craige E: Evaluation of left ventricular function by echocardiography. Circulation 4626-35, 1972 15 Helliesen PJ, Cook CD, Friedlander L, et al: Studies of respiratory physiology in children. Pediatrics 22:80-93, 1958 16 Dubois AB, Botelho SY, Bedell GN, et al: A rapid plethysmographic method for measuring thoracic gas volume. J Clin Investigation 35:322-335, 1956 17 Shwachman H, ICulczycki LL: Long-term study of one hundred five patients with cystic fibrosis. Am J Dis Child 96:6-15, 1958 18 Galant SP, Croney CE, Shaw ICC: The value of pulsus paradoxus in assessing the child with status asthamaticus. Pediatrics 61:46-51,1978 19 Meyer RA: Pediatric Echocardiography. Philadelphia, Lea and Febiger, 1977, pp 257-265 20 Gutgesell HP, Paquet M, DuH DR, et al: Evaluation of left ventricular size and function by echocardiography. Circulation 56:457-462, 1977
ECHOCARDIOGRAPHIC ABNORMAliTIES IN CYme FIBROSIS 355
An echographic study was undertaken to evaluate left (LV) and right ventricnlar (RV) function in 30 patients with cystic fibrosis. Echographic recording of the pnlmonary and aortic valve echogram permitted measurement of the phases of right and left ventricnlar systole. The ratio of the LV preejection period/LV ejection time (LPEP/LVET) and shortening of the LV internal dimension %SID was employed to reftect LV function, whDe RV preejection period/RV ejection time (RPEP/RVEI') has exceUent correlation with pnlmonary artery diastoHc pressure. RPEP/RVET and two other echographlc measurements, right ventricnlar 'Wall (RVW) and internal dimension (RVD) were compared with pnlmonary function tests and clinical scores. RPEP/RVET correlated weD
with pen:ent vital capaclty(%VC), r = -0.73, pen:ent residual volume (%RVol) r = +0.72, and clinical score, r = -0.77. MnltiUnear repesslon of RPEP/RVET, RVD, and RVW improved correlation for % VC (r = -0.80), %RVoi, r = +0.82, and clinical score, r = -0.84. Patients in overt right heart failure exhibited elevated RPEP/RVET (mean= OAS) when compared to patients not in right heart failure '(mean = .33). Marked diminution of LV function was present in two patients. A variety of cardiovascnlar abnormaHties were demonstrated echographically and were valnable in asse q the degree of cardiae involvement in patients with cystic
The majority of patients with cystic fibrosis (CF) die as a consequence of cor pulmonale, the cardiac complication of chronic pulmonary disease. 14 Progressive loss of pulmonary function and the subsequent hypoxemia lead to pulmonary hypertension and right ventricular hypertrophy ( RVH) or cor pulmonale.4.11 Right heart failure ( RHF) occurs later in the natural history of chronic cor pulmonale and most patients with cystic fibrosis expire within one to two years of the development of right heart failure. Early detection of right ventricular dysfunction and pulmonary hypertension would be of utmost importance as improvement in functional pulmonary status with more intensive therapy may improve hypoxemia, alleviate cor pulmonale, and prolong patient life.• To date, detection of the presence of cor pulmonale has been difficult in patients with cystic fibrosis because recognition of signs of right ventricular enlargement and right heart failure may be delayed by the manifestations of chronic emphysema and pulmonary disease. 3 Routine laboratory tests such as the standard electrocardiogram and chest roentgenogram often do not indicate the onset and progression of cor pul-
monale until it is well advanced. 14 Orthogonal electrocardiography (vectorcardiography), which has more accurately reflected RVH in cystic fibrosis, does not permit quanti.B.cation of cardiac or pulmonary involvement, and may reflect chronic hemodynamic problems rather than acute changes.8 The echocardiogram has recently "been shown to reflect thickening of the right ventricular free wall ( RVW) and dilation of the right ventricular internal dimension (RVD) early in the course of cystic fibrosis.'·' These echographic measurements have been shown to correlate well with pulmonary function tests and clinical status. As the underlying hemodynamic abnormality in cor pulmonale is pulmonary hypertension, the echocardiogram may be a particularly useful tool in cystic fibrosis. We have demonstrated that the pulmonary valve echogram offers a reliable means of timing the events of right ventricular systole. 10 Two right ventricular systolic time intervals ( RVSTI) have been measured: 1) right ventricular ejectiOJ1 time ( RVET); and 2) right ventricular pre-ejection period (RPEP). The ratio RPEP/RVET provided a reliable assessment of the diastolic pressure in the pulmonary vascular bed in children with congenital heart disease, and sequential changes of RPEP/ RVET reflected hemodynamic alterations occurring within the pulmonary circulation.11.12 Left ventricular systolic time intervals (LVSTI) have been measured from the aortic val~e echogram10 and
°From Case Westem Reserve University School of Mediciue, Department of Pediatrica, Rainbow Babies and Chilthenl Hospital, Cleveland. Manuscript received Mav 25; revision &901'Pted September 11. &prim teqfl6#8: Dt. Hirschfeld. BalnbotD &: Childrtm~ Hos7'ittrl, 2101 Adelbert &ad, Cletieltmtl 44106
CHEST, 75: 3, MARCH, 1979
fibrosis.
ECHOCARDIOGRAPHIC ABNORMAUTIES IN CYsnC ABROSIS 351
have been employed to assess left ventricular function.18 The shortening of the left ventricular internal dimension ( JSID) has also been valuable in assessing LV dysfunction. 14 This echographic study was undertaken to measure RPEP/RVET, LPEP/LVET, JSID, RVW, and RVD in patients with cystic fibrosis. The sensitivity of each measurement in assessing cardiac and pulmonary status was evaluated PATIENTS AND METHoDS
Thirty patients with cystic fibrosis were enrolled in the study (Table 1 ) . There were 19 females and 11 males, age 0.3 to 35 years. Most ( 24/30) patients had been admitted to the hospital for intensive pulmonary toilet. The mean age of all patients was 14.5 years with a standard error of 1.8 years. The patients generally had severe cystic fibrosis as reflected by a mean clinical score of 52± SE of 3.0. The severity of illness was also reflected by the deaths of nine patients during the study and clinical evidence for right heart failure in 18/30 patients. Pulmonary functional status was evaluated employing a 9 liter Collins spirometeru and a total body plethysmograph.te The vital capacity as a percentage of normal (we) and residual volume as a percentage of normal ( IRVol) were Table 1--SumlJIGlY of Data (30 Patienla) Age ofPt
9.5 7 13 13.5 10 20.5 23 26
19 30 35 35
18 10 14.5 24 14.5 19 23 19 7 10 20 26
.3 4.5 6 4 4 3.5
RPEP/RVET
VC%
.50
30
.33 .46 .29 .46 .38 .47
68
.54
50
.38 .40 .48 .58 .41 .28 .50
.41 .36 .25 .39 .24 .41 .33 .55 .53 .36 .33 .26 .24 .30 .40
41
64 27
51 55
53 40 40 47 50 91 27
43 53 115 45 62 43 72
32 41
RVol%
Score
RHF
370
39 72
+ 0 + 0 + + +
250
307 155 285 298 346 333 361 330 203 555 233 100 190 313 330 98 219 90 340 247 388 365
48
71 26 58
38
40
66
45 34 40 48
81 37 35 47 72 48 86 34
45 34 38
+
0
+ 0 + + 0 +
calculated. These measurements were obtained in 24 of 30 patients over six years of age. The clinical score was calculated according to the method of Shwachman and Kulczycki.U The presence of right heart failure was defined by evidence on physical examination of systemic venous congestion (hepatomegaly, edema, elevated jugular venous pressure) or tricuspid insufficiency. The measurement of RVSTI and LVSTI was accomplished from a simultaneous, rapid speed ( 100 mm/sec), recording of the electrocardiogram and pulmonary or aortic valve echograms in a manner previously described.to RVET was measured from opening to closure of the pulmonary valve echo, while RPEP was measured from the onset of ventricu-
~--
-
-
- ----;:-------====--- - ---- ...
-
~
----
~
-
.
0 0 0 0 0
+ + + + 0 + 0 0 0
+
RPEP - right pre-ejection period; RVET - right ventricular ejection time; VC% - vital capacity; RVol% "" residual volume; RHF - right heart failure.
352 HIRSCHFELD ET AL
FIGURE I. The measurement of right ventricular systolic time intervals was demonstrated from the posterior pulmonary valve cusp echogram. The right pre-ejection period ( RPEP) was measured from the onset of the QRS to the rapid descent of the pulmonary cusp. The right ventricular ejection time ( RVET) was measured from pulmonary cusp opening to closure, the point at which the leaflet reapproaches its closed, thickened baseline. There was marked phasic variation of the time intervals. PA: pulmonary artery.
FIGURE 2. Measurement of left ventricular systolic time intervals was accomplished from the aortic valve (Ao) echogram. LPEP was left pre-ejection period and L VET left ventricular ejection time.
CHEST, 75: 3, MARCH, 1979
lar depolarization, indicated by the ECG, to pulmonary valve opening ( Fig 1). LPEP/LVET was measured in a similar manner from the aortic valve echogram (Fig 2) . RVW and RVD in millimeters were measured at end-diastole and expressed as an absolute measurement without indexing for body surface area ( Fig 3) . Measurement of RPEP/RVET and LPEP/LVET should be measured during quiet respiration, if possible. Five separate measurements were made and averaged for the final PEP /VET ratio. Diastolic and systolic LV dimensions were measured and the %SID was computed (Fig 3 ).14 ,19 Values were compared to published normal values.2o Echocardiography, pulmonary function study, and clinical assessment were performed within a 24-48 hour period. Statistical analysis was performed by linear and multiple linear regression analysis. A correlation coefficient ( r value) was obtained. An r value greater than 0.7 indicated excellent correlation, between 0.5-0.7 good, and below 0.5 was poor. The standard error of the estimate was computed, as was the 95 percent confidence limits for the mean %VC, %RVol, and clinical score. The differences between patient groups was evaluated by Student t-test. Group characteristics were expressed as the mean± one standard error ( ±SE). RESULTS
Echocardiographic Findings RPEPIRVET: Mean± SE for the group was 0.38 ± 0.02 with a range of 0.24-0.55. There was excellent correlation of RPEPI RVET with $\'C ( r = -0.73), ~RVol (r = + 0.72) and clinical score (r = -0.77) (Table 1). RPEPI RVET increased with patient age ( r = + 0.53). The mean ratio was 0.48 ± .02 for •
•
.
'
.
' . . . ,v
~-
..... ,
.. --
.' '
~. ~·-
I
-'
-
---
_..,
....~
.
} CHEST WALL
.: _. _It :---~~
·· -
RVW RVD
patients in right heart failure compared to 0.33 + .02 in those without RHF ( P < 0.01 ) . Only 2/16 patients with RHF had RPEP/RVET less than 0.40, while no patient without RHF had an RPEPIRVET over0.40. RVW: Mean + SE RVW was 3.6 ± 0.23 mm, with a range from 2-5 mm. Correlation of RVW was good with $\'C (r = -0.56), $RVol (r = + 0.65) and clinical score ( r = -0.56). The RVW was related to RPEPIRVET (r = 0.62) . RVD: Mean RVD ± SE was 23 ± 2 mm with a range of 12-41 mm. RVD 'correlated well with $\'C (r = -0.51), but poorly with ~RVol and clinical score. There was no relationship between RVD/m2 and RPEPIRVET, but there was good correlation between RVW and RVD ( r = + 0.58) . LPEPILVET: Mean+ SE LPEPILVET for the group was 0.36 ± .02, only slightly increased from normal. The range was 0.20-0.52. Twelve patients had LPEP I LVET ;::: 0.40, which is the upper limit of normal and suggestive of left ventricular dysfunction. Seven had a LPEPI LVET greater than 0.45. The ratio was not significantly different ( P > .10) in patients in right heart failure ( 0.38 + .02) and those not demonstrating right heart failure ( 0.35 + .03). There was poor correlation between RPEP I RVETandLPEPILVET (r = + 0.48). LVD and ~SID: The mean LVD was 39 ± 10 mm (range 23-70 mm) and was not significantly different from normal. There was no difference in the group with right heart failure and normal patients. In two patients, $SID was depressed below the normal range ( 25-35 percent); $SID was 13 and 10 percent respectively in these patients. The ventricular septum moved abnormally in five patients with right heart failure and dilated right ventricles, so that Table Z-Sipi/ieanl Repeuion Equalioru
-SEPTUM
R
LV
} LVW
3. The ultrasonic beam transverses a position through the body of the right ventricle ( RV) and left ventricle (LV}. All measurements were made at the onset of the QRS complex. Right ventricular wall ( RVW) was measured from the undersurface of the chest wall to RV endocardium. RV dimension was measured from the RV endocardium to the right surface of the interventricular septum. The LV was dilated to 7.0 em and the calculated shortening of the LV internal dimension was markedly reduced and indicated the presence of myopathy. FIGURE
CHEST, 75: 3, MARCH, 1979
VC%
-0.73 -0.51 =72.6-0.94 (RVD) =78-8 (RVW) -0.56 = 112-115 (RPEP /RVET)0.78 (RVD)+1.6 (RVW) -0.80
r120-173 (RPEP/RVET)
±12 ±14 ±14
±5.6 ±6.6 ±6.6
t888
±10
±4.7
(RPEP/RVET)-87 RVol% =62+RVW ""-138+781(RPEP/RVET) +2.2(RVD) +11 (RVW) Score
SEE 95%CL
I= 100-123(RPEP/RVET) =78.8-8.3 (RVW) = 113-135(RPEP/RVET)0.35(RVD) +0.29(RVW)
0.72 0.65
±66 ±31 ±68 ±32
0.85 -0.77 -0.56
±62 ±29 ±10 ±4.7 ±14 ±6.6
-0.82
±9
±4.2
R = correlation coefficient; SEE - standard error of estimate; 95%CL = 95% confidence limits for the mean; RPEP - right pre-ejection period; RVD .. right ventricular dimension; RVET = right ventricular ejection time; RVol% = percent residual volume; VC% - percent vital capacity.
ECHOCARDIOQRAPHIC ABNORMALITIES IN CYSTIC FIBROSIS 353
ISID could not be computed in many patients. Multiple Ccmelations: When three parameters, RVD, RVW, and RPEP/RVET, were correlated with we, IRVol, and clinical score, the coefficient of correlation was improved in predicting we (r -.80, mvol (r = 0.82) and clinical score (r = - 0.84) (Table2). Sequential and Catheterization: Ten patients developed signs of right heart failure during the course of the study and permitted evaluation of intervention with digoxin. The effects of digoxin on RPEP/RVET, LPEP/ LVET, and ISID were insignffi.cant in all ten patients. In patients who had evidence of right heart failure, the echograms demonstrated marked inspiratory, expiratory variation of both RPEP/RVET and LPEP/LVET. With expiration, there was prolongation of LPEP or RPEP and associated shortening of the LVET or RVET, resulting in prolongation of each ventricular ratio. Four patients with severe right heart failure underwent cardiac catheterization. Each patient had severe pulmonary hypertension with a mean pulmonary arterial pressure above 50 mm Hg. The RPEPI RVET was elevated in all patients and ranged from 0.40-0.56. There was respiratory variation in the RPEP/RVET as described above, and this corresponded to a 15-30 mm Hg decrease in the systolic and diastolic pulmonary arterial pressure. Three patients had an elevated pulmonary arterial wedge pressure, suggesting left ventricular dysfunction and in each patient the LPEP I LVET was increased (Table3).
=
+
DISCUSSION
Echographic recording of the aortic and pulmonic valves has facilitated measurement of RV and LVSTI. A systolic time interval is a phase of electromechanical systole, the duration of which is governed by four basic factors: 1) preload, indicated by ventricular end diastolic volume; 2) afterload or resistance to ventricular emptying, reflected by arterial diastolic pressure; 3) contractile state of the myocardium; and 4) rate and sequence of intraventricular electrical conduction. 18 The use of the ratios LPEP /LVET and RPEP/
Patient No.
s
PA Pressure D Mean
11 13 24 30
80 85 55 60
40
50 40 40
60 70 50 50
BP
CI
LPAW
110/70 100/70 100/80 105/60
4.3 5.5 3.2 4.1
22 16 28
8
BP - blood pressure; CI - cardiac index; D - diastolic; LPAW • left pulmonary artery wedge; PA - pulmonary artery; S • systOlic.
354 HIRSCHFELD ET AL
RVET eliminates the influence of heart rate and age, so that patients of different ages can be compared and the same subject may be evaluated sequentially. 11·18 An elevated RPEP/RVET has been predictive in indicating elevated diastolic pressure ( afterload) in the pulmonary circulation, but may also be prolonged if RV contractility is diminished.11·12 An increased RPEP/RVET in a patient with cystic fibrosis may, therefore, be a result of either increased pulmonary artery pressure, as was the case in patients who underwent cardiac catheterization, or contractile abnormalities. The latter may be a consequence of intermittent pulmonary hypertension, hypoxia, or acidosis. In either case, RPEP/RVET should provide an excellent guide to the severity of pulmonary involvement and provide an index to the contractile state of the right ventricular myocardium. In our study, RPEP/RVET showed excellent correlation with other indicators of the severity of cystic fibrosis, namely We, IRVol, and ShwachmanKulczycki score. The ratio also provided excellent separation of those patients in right heart failure from those demonstrating no signs of right heart failure, for when RPEP/RVET was 0.40 or greater, the patient was in potential or overt right heart failure. Sequential evaluation was particularly important in that deterioration in right ventricular function was predicted in three patients, who developed signs of right heart failure shortly after an echogram demonstrated an elevated RPEP/RVET. The presence of marked phasic respiratory changes was present in all patients in RHF. This appears to reflect the signffi.cant variations in pulmonary arterial pulse amplitude, pulsus paradoxus, that has been suggested in patients with increased intrathoracic pressure secondary to asthma. 18 Impairment of right ventricular filling may also occur in patients with severe cystic fibrosis. The impaired right ventricular filling results in diminished right and left ventricular stroke volume or preload when the intrathoracic pressure increases during expiration. The phasic respiratory changes in preload (ventricular filling) produce the respiratory variation in RPEP/RVET and LPEP/LVET seen in patients with the most severe cystic fibrosis and right heart failure. Because of the variation in ratios, at least five separate measurements should be averaged for the final ratio. In patients with respiratory variation, either measurement was markedly elevated and provided an index of the severity of the disease. Digitalization did not produce reduction in LPEP/LVET or RPEP/RVET, although clinical improvement was evident in most patients. Twelve patients had increased LPEP/LEVT ( > 0.40) which indicated that while left ventricular dysfunc-
CHEST, 75: 3, MARCH, 1979
tion may be present in a few patients, it was not a major determinant in influencing pulmonary function tests or the presence of right heart failure. Marked depression of left ventricular function suggested by decreased %SID and/ or very elevated LPEPI LEVT demonstrated that in an occasional patient the left ventricle may be markedly myopathic. The increased pulmonary wedge pressure measured at catheterization further attests to the importance of left ventricular dysfunction in patients with cystic fibrosis. Other investigators have demonstrated good correlation between RVW/m2, RVD/m2 and clinical severity of cystic fibrosis. 7•8 It is also doubtful that RVW can be measured with the accuracy that has been reported in the other echographic studies of cystic fibrosis. The beam width of the standard ultrasonic transducers permits measurement of structures to the nearest millimeter at best and not tenths of millimeters as reported in other studies. Our study did not index RVW or RVD for body surface area as there is no evidence that either measurement has a constant relationship to surface area in young subjects. RVW showed good correlation with RPEP /RVET suggesting the thickening of the RVW was occurring concomitantly with the elevated pulmonary vascular resistance or RV dysfunction. When RVW and RVD are combined with RPEP/RVET, correlations further improved between echocardiography and clinical severity of cystic fibrosis. The clinical score, ~c. and mvol could be estimated by the regression formula using the three echographic parameters. Problems with measurement of RPEP/RVET have been in recording the pulmonary valve echogram. Air trapping, which is very common in cystic fibrosis, may interfere with recording the pulmonary valve echogram. In this study, we were able to visualize the pulmonary valve in approximately twothirds of the patients. Pulmonary valve recording was not hindered by advanced pulmonary disease or increased age. The patient group in which the pulmonary valve could not be recorded had the same mean age and clinical score as the reported patient population. In summary, RPEP/RVET, alone or combined with RVD and RVW, is an excellent index of the severity of the cardiovascular complications of cystic fibrosis. Patients in RHF or imminent RHF may be separated from those without RHF. Left ventricular function may be abnormal, and in an occasional patient the left ventricle may be myopathic. Further evaluation employing correlation of cardiac catheterization and echograpbic findings will be necessary in a large group of patients before the ability of
CHEST, 75: 3, MARCH, 1979
this technique to detect early development of cor pulmonale and response to therapy can be completely evaluated.
1 Liebman J, Lucas R, Moss A, et al: Cor pulmonale and related cardiovascu1ar eHects of cystic fibrosis. In Cystic Fibrosis, Projections into the Future. Mangos JA, Talamo RC, eds. Miami, Symposia Specialists, 1976, pp 41-79 2 Wood RE, Boat TF, Doershuk CF: State of the art: Cystic fibrosis. Am Rev Respir Dis 113:833-878,1976 3 Siassi B, Moss AJ, Dooley RR: Clinical recognition of cor pulmonale in cystic fibrosis. J Pediatrics 78:794-805, 1971 4 Goldring RM, Fishman AP, Turino JM, et al: Pulmonary hypertension and cor pulmonale in cystic fibrosis of the pancreas. J Pediatrics 65:501-524, 1964 5 Thomas AJ: Chronic pulmonary heart disease. Br Heart J 34:653-657, 1972 6 Liebman J, Doershuk CF, Rapp C, et al: The vectorcardiogram in cystic fibrosis: Diagnostic significance and correlation with pulmonary function tests. Circulation
35:552-569, 1967
7 Rosenthal, A, Tucker CR, Williams RG, et al: Echocardiographic assessment of cor pulmonale in cystic fibrosis. Pediat Clin North Am 23:327-343, 1976 8 Ryssing E:Assessment of cor pulmonale in cystic fibrosis by echocardiography. Acta Pediatr Scand 66:753-756, 1977 9 Gewitz M, Eshaghpour E, Holsclaw DS, et al: Echocardiography in cystic fibrosis. Am J Diseases Child 131 :275-280, 1977 10 Hirschfeld S, Meyer R, Schwartz D, et al: Measurement of right and left ventricular systolic time intervals by echocardiography. Circulation 51:304-309, 1975 11 Hirschfeld S, Meyer R, Schwartz DC, et al: The echocardiographic assessment of pulmonary artery pressure and pulmonary vascular resistance. Circulation 52:641-650, 1975 12 Riggs T, Hirschfeld S, Borkat G, et al: Right ventricular systolic time intervals in the assessment of the pulmonary vascular bed. Circulation 57:939-947, 1978 13 Weissler AM, Lewis RP, Leighton RF: The systolic time intervals as a measure of left ventricular performance in man. In Progress in Cardiology, (Vol!) (Yu PN, Goodwin JF eds). Philadelphia, Lea and Febiger, 1972, pp 155183 14 Fortuin NJ, Hood WP, Craige E: Evaluation of left ventricular function by echocardiography. Circulation 4626-35, 1972 15 Helliesen PJ, Cook CD, Friedlander L, et al: Studies of respiratory physiology in children. Pediatrics 22:80-93, 1958 16 Dubois AB, Botelho SY, Bedell GN, et al: A rapid plethysmographic method for measuring thoracic gas volume. J Clin Investigation 35:322-335, 1956 17 Shwachman H, ICulczycki LL: Long-term study of one hundred five patients with cystic fibrosis. Am J Dis Child 96:6-15, 1958 18 Galant SP, Croney CE, Shaw ICC: The value of pulsus paradoxus in assessing the child with status asthamaticus. Pediatrics 61:46-51,1978 19 Meyer RA: Pediatric Echocardiography. Philadelphia, Lea and Febiger, 1977, pp 257-265 20 Gutgesell HP, Paquet M, DuH DR, et al: Evaluation of left ventricular size and function by echocardiography. Circulation 56:457-462, 1977
ECHOCARDIOGRAPHIC ABNORMAliTIES IN CYme FIBROSIS 355
