University of Groningen. Left ventricular diastolic function and cardiac disease Muntinga, Harm Jans

Transcription

1 University of Groningen Left ventricular diastolic function and cardiac disease Muntinga, Harm Jans IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2000 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Muntinga, H. J. (2000). Left ventricular diastolic function and cardiac disease: a radionuclide angiography study Groningen: s.n. Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date:

2 Chapter 3 Normal values and reproducibility of left ventricular filling parameters by radionuclide angiography H.J. Muntinga 1, F. van den Berg 1, H.R. Knol 2, M.G. Niemeyer 1, P.K. Blanksma 3, H. Louwes 4 and E.E. van der Wall 5 1 Department of Cardiology, Martini Hospital, Groningen, 2 Northern Centre for Health Care Research, University of Groningen, Groningen, 3 Department of Cardiology, University Hospital Groningen, Groningen, 4 Department of Nuclear Medicine, Martini Hospital, Groningen, 5 Department of Cardiology, University Hospital Leiden, Leiden, The Netherlands. Summary Background: In physiologic situations age, heart rate (HR) and left ventricular ejection fraction (EF) may influence left ventricular filling rate. In this study, we determined normal values for radionuclide angiography (RNA) derived diastolic filling parameters, the correlations with age, HR and EF and their reproducibility. Methods: The study was performed in 20 patients, years old (mean 57), with normal findings at coronary angiography and left ventriculography. The first RNA was performed at rest (RNA1). Then, five minutes bicycle ergometry was performed and the patients were allowed five minutes rest before RNA was repeated (RNA2). From the left ventricular time activity curve we determined peak filling rate (PFR), time to peak filling rate (TPFR) and atrial contribution (AC) to ventricular filling. Results: Values for PFR1 were 2.2 ± 0.6 EDV/s (PFR2 2.4 ± 0.7 EDV/sec, r=0.82), for TPFR1 198 ± 22 ms (TPFR2 203 ± 24 ms, r=0.45) and for AC1 31 ± 11 % (AC2 31 ± 10 %, r = 0.72). The correlation s of PFR and TPFR with age were statistically significant (respectively r=-0.68 and r=0.48, P<0.05). PFR was also influenced by HR and EF (respectively r=0.51 and r=0.50, P<0.05). TPFR however was not influenced by HR and EF, whereas AC was positively correlated with HR (r=0.79, P<0.01). Conclusions: Radionuclide angiography is a reliable and reproducible method to assess parameters of diastolic left ventricular filling in individual patients. It may therefore be used to serially follow diastolic function. When used for interindividual comparison the dependency of RNA derived left ventricular filling parameters on age, HR and EF should however be considered. International Journal of Cardiac Imaging 1997;13:

3 Reproducibility of normal diastolic function parameters 31 F or the quantitative interpretation of diagnostic test results used to evaluate heart function it is necessary to define limits of normal. Healthy volunteers, 1,2,3,4,5 patients with a low probability of coronary artery disease, 1,5,6,7 and patients with normal coronary angiograms 2 are commonly used to provide standards in assessing the accuracy of diagnostic tests in cardiology. In the context of tests used to evaluate diastolic heart function, the definition of normal is often based on findings in patients with a normal exercise electrocardiogram and a normal Doppler echocardiogram. 1,8 Alternatively, clinically normal subjects with normal findings at coronary arteriography and left ventriculography may also be used to define the normal range. 2 Until now several studies reported on radionuclide angiographic assessment of normal diastolic function. 1,2,3,5,6,16 Unfortunately, only few studies address the reproducibility of the normal values for the patient group used. 2 In the present study a description is made of normal values of diastolic function parameters, together with the reproducibility of these parameters for patients with normal findings in coronary arteriography. The parameters were assessed using a computer program which automatically defines filling parameters derived by radionuclide angiography. METHODS Patients. The present study was performed in 20 subjects, 8 male persons, mean age 57 ± 11 years, with normal findings at coronary arteriography and left ventriculography. Coronary angiography was performed as a diagnostic test because of precordial pain of unknown origin. All results of the cardiac angiography were reviewed by one person. Patients with significant coronary abnormalities were excluded, as were patients with detectable valvar abnormalities. Nine patients were successfully treated for hypertension. Left ventricular hypertrophy was excluded with electrocardiography and with echocardiography. Three patients had paroxysmal atrial fibrillation. In thirteen patients all medication if any was used was withdrawn more than 48 hours before evaluation. The other seven patients remained on oral therapy, which included oral nitrates (1 patient), diuretics (1), calcium channel blockers (1), beta-blocking agents (5) and angiotensin-converting enzyme inhibitors (1). Radionuclide angiography. Each patient's red blood cells were labelled with technetium 99m, after intravenous administration of pyrophosphate. The total dosage was MBq. Left ventricular function was evaluated by radionuclide angiography using a gamma camera (Siemens Orbiter) with an all-purpose parallel-hole collimator interfaced with a Pinnacle computer (Medasys Inc., Ann Arbor). Rest supine multigated images were obtained in left anterior oblique view with a caudal tilt, so that the left and right ventricles were entirely separated. Only a 5% cycle-length-window with forward

4 32 Chapter 3 TABLE 3.1. Radionuclide angiographic values of systolic and diastolic parameters in the study population. Mean ± SD Range Age (years) 57 ± HR (beats/min) 68 ± EF (%EDV) 62 ± PFR (EDV/s) 2.2 ± PFR (SV/s) 3.5 ± TES (ms) 382 ± TPFR (ms, from TES) 198 ± TPFR' (ms, from t=0) 579 ± AC A(%SV) B 31 ± AC (%EDV) 19 ± FIGURE HR = heart 3.1. rate; Time-activity EF = ejection curve fraction; of the PFR left ventricle = peak filling (Figure rate; 3.1A) EDV and = end-diastolic its first derivative volume; (Figure SV = 3.1B). stroke Ejection volume; TES fraction = time is the to end ratio of of systole; stroke TPFR volume = time (SV) to and peak end-diastolic filling rate. volume (EDV). Atrial contribution (AC) is expressed as a fraction of SV or EDV. In Figure 3.1B PFR is the rapid instantaneous filling rate during early rapid filling expressed as EDV per second and as SV per second. TPFR is measured from the beginning of relaxation (TPFR) and from the beginning of contraction (TPFR'). gating was accepted. 9 Acquisition was completed after 150,000 counts per frame of 20 ms. Temporal smoothing was performed by 5 Fourier harmonics. 10,11 Measurements. From the systolic part of the left ventricular time-activity curve (TAC) we measured the time to end of systole (TES) and the ejection fraction (EF). From the diastolic part of the curve we took measurements only when early diastolic filling could be separated from late diastolic filling by a diastasis period. The diastolic parameters included peak filling rate (PFR) and time to peak filling rate (TPFR). The PFR was normalised to end-diastolic volume (EDV/sec) and to stroke volume (SV/sec). The TPFR was measured from the beginning of contraction (TPFR') and from the beginning of relaxation (Figure 3.1). Further diastolic filling was divided into rapid filling and atrial contribution (AC) as percentage of stroke volume and end-diastolic volume. Reproducibility data. Reproducibility was assessed by performing 2 different radionuclide angiograms. They were separated by a supine bicycle ergometry of 5 minutes and 5 minutes of rest, so that heart rate in both studies was not automatically the same. After a new adjustment of the collimator the 2nd data collection was started. Statistical analysis. Measured data are presented as mean ± 1 SD. Linear regression analysis was used to test the relation between the radionuclide angiographic variables on the two occasions. This was also used to test the relation between age, HR and EF and the radionuclide angiographic data. Statistical analysis for paired samples was performed by a Student's t-test.

5 Reproducibility of normal diastolic function parameters 33 RESULTS Normal values. The mean values, standard deviations and ranges of systolic and diastolic parameters measured by the first radionuclide angiography together with age and heart rate are listed in Table 3.1. A description of the values of diastolic filling parameters with radionuclide angiography with comparable techniques in normal individuals is made in Table 3.2. Most authors had used different selection criteria for their study population, including a low probability for coronary artery disease, normal volunteers and patients with normal findings in coronary angiography. Age was another difference between these studies. Lee et al., Miller et al. and Iskandrian et al. had selected patients of all age categories. 2,6,16 Iskandrian et al. had divided their population according to age. 16 Arora et al. had selected two groups of different ages. 1 All authors except Iskandrian et al. had used 5 harmonics for temporal smoothing. Taking these differences into account, the values of PFR (normalised to both EDV and SV), TPFR (measured from the beginning of filling and from the beginning of emptying) and AC (%SV) were approximately the same as in our study. Reproducibility. Two patients were not evaluable for the second RNA study, and they were excluded from this part of the study. The reproducibility data are presented in Table 3.3. Compared to the first RNA no changes of the means of the variables were present in the second RNA. The average difference between the individual values of the two RNA studies ranged from 3 % of the initial value in TPFR' to 20 % of the initial value TABLE 3.2. Values of diastolic filling parameters assessed with radionuclide angiography with comparable techniques in normal individuals. Lee Arora Miller Iskandrian Group I Group I I Group I Group I I Ref. no Screening low. vol. vol. vol. & low. low. No. of subjects NCAG Age Harmonics PFR (EDV/s) 3.0 ± ± ± ± ± ±0.6 Range PFR (SV/s) 4.5 ± Range TPFR (ms) ± ± ±46 Range TPFR' (ms) ± ± Range AC (%SV) - 16 ± 4 31 ± Range ref.no. = reference number; low. = low probability for coronary artery disease; vol. = normal volunteers; NCAG = normal coronary angiogram; PFR = peak filling rate; EDV = end diastolic volume; SV = stroke volume; TPFR = time to peak filling rate measured from the beginning of relaxation; TPFR' = time to peak filling rate measured from the beginning of contraction; AC = atrial contribution to diastolic filling.

6 34 Chapter 3 TABLE 3.4. Correlations between age, heart rate and ejection fraction and diastolic parameters of the in AC (%EDV). The values of the parameters of the first and the second RNA correlated TAC. well. Only the correlation PFR of TPFR1 to TPFR2 TES was TPFR relatively TPFR' low (r=0.45). AC AC Correlations of Age, (EDV/s) HR and (SV/s) EF to (ms) parameters (ms) of the (ms) TAC. (%SV) The results (%EDV) of the regression Age analysis of * Age, HR * and 0.17 EF + in 0.48 relation + to 0.45 the + mentioned diastolic HR (beats/min) * * 0.79 * 0.82 * parameters EF (%EDV) of the 0.50 TAC + are listed in Table The negative relation 0.28 of age 0.50 with + left ventricular P<0.01; + P<0.05; filling All corelations was reflected not listed in are a not statistically significant (P>0.05). significant correlation with PFR expressed HR = heartrate as (beats/min); EDV/sec EF and = ejection SV/sec fraction (P<0.01, of the Figure left ventricle; 3.2A) EDV and = both end-diastolic TPFR volume; and TPFR' = peak filling rate; SV = stroke volume; TES = time to end of systole; TPFR = time to peak filling rate P<0.05, measured Figure from the 3.2B). beginning Age of relaxation; was however TPFR' = not time correlated to peak filling to rate AC measured (r=0.15 from and the r=-0.08 beginning for AC of contraction; expressed AC as = atrial %SV contribution. and as %EDV respectively). The positive influence of heart rate on diastolic filling was expressed in the relation with PFR normalised to EDV (P<0.05), TPFR' (P<0.01) and AC expressed as fractions of EDV and of SV (P<0.01). PFR normalised to EDV, TPFR' and AC (%EDV) were also correlated to EF. DISCUSSION Normal values. In the radionuclide angiographic evaluation of diastolic left ventricular function, the matter of interest is the left ventricular filling pattern. In the absence of mitral valve abnormalities, this filling pattern is the result of the diastolic atrial ventricular TABLE 3.3. Reproducibility data of 18 patients who underwent two sequential RNA's. Average RNA 1 (mean ± SD) RNA 2 (mean ± SD) difference between the studies (mean ± SD) Maximum Difference Between the studies HR (beats/min) 69 ± ± 9 4 ± EF (%EDV) 63 ± 8 62 ± 8 5 ± PFR (EDV/s) 2.22 ± ± ± PFR (SV/s) 3.53 ± ± ± TES (ms) 380 ± ± ± TPFR (ms) 198 ± ± ± TPFR' (ms) 579 ± ± ± AC (%SV) 31 ± ± 10 5 ± AC (%EDV) 20 ± 9 19 ± 7 4 ± HR = heartrate (beats/min); EF = ejection fraction of the left ventricle; EDV = end-diastolic volume; PFR = peak filling rate; SV = stroke volume; TES = time to end of systole; TPFR = time to peak filling rate measured from the beginning of relaxation; TPFR' = time to peak filling rate measured from the beginning of contraction; AC = atrial contribution; SD = standard deviation; r = correlation coefficient relating the first and second study. r A B FIGURE 3.2A The relation between peak filling rate (PFR) and age in the 20 patients with normal coronary angiograms. EDV/s = end-diastolic volume per second. B The relation of time to peak filling rate (TPFR) to age in the 20 patients with normal coronary angiograms.

7 Reproducibility of normal diastolic function parameters 35 pressure gradient, 12 which in addition is influenced by many factors including left ventricular relaxation and compliance. 13,14,15 Some filling parameters therefore show a clear relation with physiological (e.g. age and exercise) and pathological (e.g. ischaemia and hypertension) conditions that influence these factors. 11 It is therefore of great importance to describe the normal values and the reliability of the test in individual cases. The comparison of the normal test results with the findings of other investigators, must however be made with caution, for also a comparison of the tests has to be made. Previously, other investigators have described the values of PFR, TPFR and AC in healthy volunteers and patients with a low probability of coronary artery disease. 1,2,3,5,16 The measured values of PFR (normalised to end diastolic volume and to stroke volume), TPFR (both from beginning of contraction and from beginning of relaxation) and AC in these studies roughly coincided with our mean values. Also, the ranges were approximately the same. Relationship with age. With the normal ageing process, the heart is subjected to a number of anatomic changes which include a decreased amount of myocytes, fibrous tissue proliferation, loss of elastic fibres and increased left ventricular wall mass. 8 With these findings, a decreased left ventricular compliance and a decreased relaxation rate are likely to be present with increasing age. The described relation between age and PFR and TPFR was found earlier by other authors. 1,2,3,5,6,16 The increasing AC with rising age which was earlier described, is thought to be a compensation of decreased early diastolic filling. 1 This is however not supported by the present study. Relationship with ejection fraction. In the present study the measured filling parameters were also related to the ejection fraction. As was described before, PFR normalised to end-diastolic volume, but not PFR normalised to stroke volume, is influenced by EF. 11 This relation expresses the influence of systolic performance on ventricular relaxation. PFR normalised to stroke volume and AC expressed as a percentage of stroke volume are however relatively independent of EF. Relationship with heart rate. Heart rate is also an independent variable influencing the timing and maximum rate of early rapid diastolic filling. 3,6,11 Its positive effect on PFR is believed to be resulting from the disappearance of the diastasis period in tachycardia. In this situation PFR is highly heart rate dependent. 11 In every TAC in our study however a diastasis period was present. Despite this, the relation between HR and PFR was still statistically significant, as was reported before. 2 An increased left ventricular relaxation in higher heart rates may be the explanation for this. Also negative correlations between HR and TPFR' (measured from the beginning of contraction) and TES were measured. They indicate a decreased duration of the contraction-relaxation cycle in higher heart rates mainly due to a reduction of the duration of ejection.

8 36 Chapter 3 Reproducibility. On PFR and TPFR excellent reproducibility data were reported before by Miller et al. 2 In this study the second data collection was not, as in our study, preceded by a new adjustment of the collimator. Also, the patients had been lying quietly between the measurements while in our study they had a short period of exercise. The release of catecholamines during exercise has a potential effect on diastolic function. In view of the redressment of heart rate in the following period of rest, this effect was probably not long-lasting in our study. Although these factors may have been of influence on the results, the present study shows good and consequent individual reproducibility of the values of the measured variables. Limitations. As mentioned, the methodological limitations of radionuclide angiography should not be underestimated. Not only technical considerations, as temporal smoothing and resolution are important in the assessment of diastolic function, also methodological considerations as normalisation, influence the outcome and the reliability of the measurements. In addition, in the present study only 20 patients were investigated, which decreases the power of describing normal values and reference values for other investigators. Conclusions. We demonstrated the values of filling parameters with radionuclide angiography in patients with normal findings in coronary angiography. A good individual reproducibility of the values was found. RNA may therefore be used to serially follow diastolic function. Unfortunately, the diagnostic use of radionuclide derived left ventricular filling parameters remains limited because of the wide ranges and the relatively high standard deviations of the normal values. Also, the dependency on age, heart rate and ejection fraction make the interpretation of the diagnostic test results difficult. References 1. Arora RR, Machac J, Goldman ME, Butler RN, Gorlin R Horowitz SF. Atrial kinetics and left ventricular diastolic filling in the healthy elderly. J Am Coll Cardiol 1987;9: Miller TR, Grossman SJ, Schectman KB, Biello DR, Ludbrook PA, Ehsani AA. Left ventricular diastolic filling and its association with age. Am J Cardiol 1986;58: Bonow RO, Vitale DF, Bacharach SL, Maron BJ, Green MV. Effects of aging on asynchronous left ventricular regional function and global ventricular filling in normal human subjects. J Am Coll Cardiol 1988;11: Kuo LC, Quinones MA, Rokey R, Sartori M, Abinader EG, Zoghbi WA. Quantification of atrial contribution to left ventricular filling by pulsed doppler echocardiography and the effect of age in normal and diseased hearts. Am J Cardiol 1987;59: Bowman LK, Lee FA, Jaffe CC, Mattera J, Wackers FJTh, Zaret BL. Peak filling rate normalized to mitral stroke volume: a new doppler echocardiographic filling index validated by radionuclide angiographic techniques. J Am Coll Cardiol 1988;12:

9 Reproducibility of normal diastolic function parameters Lee KJ, Southee AE, Bautovich GJ, Freedman B, McLaughlin AF, Rossleigh MA, Hutton BF,Morris JG. Normalised radionuclide measures of left ventricular diastolic function. Eur J Nucl Med 1989;15: Miyatake K, Okamoto M, Kinoshita N, Owa M, Nakasone I, Sakakibara H, Nimura Y.Augmentation of atrial contribution to left ventricular inflow with aging as assessed by intracardiac doppler flowmetry. Am J Cardiol 1984;53: Nixon JV, Burns CA. Cardiac effects of aging and diastolic dysfunction in the elderly. In: Gaasch WH, LeWinter MM, ed. Left ventricular diastolic dysfunction and heart failure. Philadelphia: Lea & Febiger, 1994; Juni JE, Chen CC. Effects of gating modes on the analysis of left ventricular function in the presence of heart rate variation. J Nucl Med 1988;29: Bacharach SL, Green MV, Vitale D, White G, Douglas MA, Bonow RO, Larson SM.Optimum fourier filtering of cardiac data: a minimum-error method: concise communication. J Nucl Med 1983;24: Udelson JE, Bonow RO. Radionuclide angiographic evaluation of left ventricular diastolic function. In: Gaasch WH, LeWinter MM, ed. Left ventricular diastolic dysfunction and heart failure. Philadelphia: Lea & Febiger, 1994; Ishida Y, Meisner JS, Tsujioka K, Gallo JI, Yoran C, Frater RWM, Yellin EL. Left ventricular filling dynamics: influence of left ventricular relaxation and left atrial pressure. Circulation 1986;74: Bonow RO, Udelson JE. Left ventricular diastolic dysfunction as a cause of congestive heart failure. Mechanisms and management. Ann Intern Med 1992;117: Grossman W. Diastolic dysfunction in congestive heart failure. N Engl J Med 1991;325: Shintani H, Glantz A. The left ventricular diastolic pressure-volume relation, relaxation, and filling. In: GaaschWH, LeWinter MM, eds. Left ventricular diastolic dysfunction and heart failure. Philadelphia: Lea & Febiger, 1994; Iskandrian AS, Hakki AH. Age-related changes in left ventricular diastolic performance. Am Heart J 1986;112:75-78.

10 Editorial comment on chapter 3 When is a visually or mathematically diagnosed scan abnormality clinically important? International Journal of Cardiac Imaging 1997;13:173 M. Pillay Dr. Daniel den Hoed Clinic, University Hospital Rotterdam This question is frequently addressed as more and more sensitive and sophisticated diagnostic tools become available. The first step in setting up a diagnostic test is to evaluate its reliability in terms of sensitivity, specificity and reproducibility under a number of conditions. In a diagnostic imaging department, this is usually achieved by repeat imaging and analyses. In this respect the authors have satisfied the basic requirement necessary to introduce the test. The next question, of course, is the test being utilised on a regular basis? If so, is there an expectation to supplement the results of this paper with additional data (preferably a combination of diseased and normal subjects) so that validity of the repeated measurements will be tested to determine the specificity, sensitivity and accuracy of the method. Without this additional data the reproducibility figures derived from repeated measurements in small studies can surely only be of importance as a quality control procedure. It is also well recognised that the normality in parameters derived by radionuclide cardiac angiography are geographically and population biased so that multicentre trials to establish normal values are essential. The pooling of data using meta-analyses has limitations and inherent deficiencies so that choice of analysis should be avoided where possible. Presenting p-values alone can lead to their being given more merit than they deserve. There is a tendency to equate statistical significance with medical importance or biological relevance. Small differences of no real interest can be statistically significant with large sample sizes, whereas clinically important effects may be statistically non-significant only because the number of subsets studied was small. This phenomenon is clearly evident in this study where a number of p-values are shown to indicate significant differences and where correlation coefficients demonstrate large spread of individual data. Including data on diseased subjects will test the reliability of the hypotheses. The estimation and use of confidence intervals would haven been more appropriate. In larger studies on impaired left ventricular function and diastolic filling Bonow et al. and Pace et al. have concluded that the differences between the diastolic filling characteristics of patients with coronary artery disease without prior infarction and those of normal volunteers could not be explained on the basis of any differences in age, in heart rate or in left ventricular end-diastolic size. This finding is not supported by this

11 Editorial comment on chapter 3 40 study and may be a result of the geographic, population sample and methodological bias. The evaluation of reproducibility is surely of much importance provided that studies are designed to go beyond laboratory based quality control and include data to test hypotheses on a broader base, i.e. sensitivity, specificity and accuracy. References 1. Bonow RO et al. Impaired left ventricular diastolic filling in patients with coronary artery disease: assessment with radionuclide angiography. Circulation 1981;64: Pace L et al. Diagnosis of coronary artery disease by radionuclide angiography: effect of combining indices of left ventricular function. J Nucl Med 1989;30: