Natriuretic peptide and echocardiography after operation of atrial septal defect

International Journal of Cardiology 89 (2003) 45–52 www.elsevier.com / locate / ijcard

Natriuretic peptide and echocardiography after operation of atrial septal defect a, b a d c K.W.E. Groundstroem *, T.E. Iivainen , J.T. Lahtela , T.J. Talvensaari , T.A. Paakkala , a b A.I. Pasternack , A.J. Uusitalo a

Department of Internal Medicine, Tampere University Hospital, FIN-33250 Tampere, Finland b Department of Clinical Physiology, Tampere University Hospital, Tampere, Finland c Department of Diagnostic Radiology, Tampere University Hospital, Tampere, Finland d ¨ ¨ Central Hospital, Hameenlinna , Finland Department of Internal Medicine, Kanta-Hame Received 26 March 2002; received in revised form 18 July 2002; accepted 20 July 2002

Abstract Patients benefit from surgical seclusion of atrial septal defect but have excessive cardiovascular morbidity after the operation. We evaluated haemodynamics and looked for abnormalities of cardiac structures and function late after surgical seclusion of the defect. Serum N-terminal natriuretic peptide measurement and transthoracic and transoesophageal echocardiography were performed in 61 patients aged 43615 years (mean6standard deviation) 2165 years after surgery. The findings were compared with 67 control subjects. The patients had higher serum N-terminal atrial natriuretic peptide concentration than the control subjects (0.4060.32 vs. 0.2460.12 nmol / l, P50.0001). Peptide levels correlated with current age (P50.0001) and age at operation (P50.0014), but not with age in the control subjects. In the patients, echocardiography measurements of cardiac dimensions correlated with hormone levels (atrial natriuretic peptide concentration with left atrial end-systolic diameter (P50.042), left ventricular end-diastolic (P50.021) and end-systolic diameter (P50.042). There were only 10 patients (16%) without any abnormality in echocardiography. Their peptide concentration was 0.2560.18 nmol / l (P5not significant compared to the control subjects). The association between increasing N-terminal atrial peptide levels and operation age together with echocardiography findings support the clinical consensus of treating atrial septal defect patients in their childhood and adolescence.  2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Adult; Atrial natriuretic peptide; Atrial septal defect; Echocardiography; Postoperative; Surgery

1. Introduction Atrial septal defect is one of the most common congenital heart diseases [1]. It is often not recognized until adulthood [2]. Although patients benefit from surgical seclusion of the defect, they still have cardiovascular morbidity after the operation [3,4]. *Corresponding author. Tel.: 1358-3-247-4140; fax: 1358-3-2474157. E-mail address: [email protected] (K.W.E. Groundstroem).

Patients operated in childhood have a better survival rate and less complications than patients operated in adulthood [3–6]. Atrial natriuretic peptide and its biologically active C-terminal fraction, are synthesized, stored and secreted in equimolar amounts by the myocytes of the cardiac atria due to atrial wall stretch, but also by the ventricles [7,8]. The peptides are cleared by the natriuretic peptide-C receptor and are degraded by the ectoenzyme neutral endopeptidase 24.11, both of which are widely expressed in the kidneys, lungs, and

0167-5273 / 02 / $ – see front matter  2002 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0167-5273(02)00427-8

46

K.W.E. Groundstroem et al. / International Journal of Cardiology 89 (2003) 45–52

the vascular wall [8,9]. The C-terminal fraction has a diuretic, natriuretic and vasodilative effect and it inhibits the renin–angiotensin–aldosterone system [9,10]. Excessive amounts of the peptides are released into the circulation in congestive heart failure [7–9], congenital heart diseases [10–13], and valvular heart diseases [14]. The N-terminal fraction can be measured more reliably because of its better stability and longer half-life [7,8]. Although the correlation between atrial peptide levels and prognosis in adult atrial septal defect patients is not well established they do have prognostic value in heart failure patients [15]. We measured serum N-terminal atrial natriuretic peptide in 61 atrial septal defect patients late after operation and in 67 age, sex and body mass index matched control subjects. The peptide levels were compared and contrasted with electrocardiography, chest radiogram and findings in transthoracic and transoesophageal echocardiography.

2. Patients and control subjects

2.1. Atrial septal defect patients Two groups of operated atrial septal defect patients were included in the study. The first group consisted of 93 patients (57 females (f) / 36 males (m)) located in a computer search of our hospital records of

cardiac operations between 1 January 1966 and 31 December 1978. Patients with ostium primum defects were excluded. Six patients had died (4 f / 2 m) and one male patient with ventricular septal defect and pulmonary valve stenosis was excluded from the study. Of the remaining 86 atrial septal defect patients (53 f / 33 m), 51 lived in our area and were contacted by mail. Two female and four male patients aged 3667 years refused to participate in the study. Afterwards, three female patients and one male patient who had been operated on simultaneously for another congenital heart defect and two female patients with chronic atrial fibrillation and one female with paradoxal septal motion in transthoracic echocardiography, were excluded from the study, leaving 38 patients (23 f / 15 m). The second group consisted of 24 patients, who in 1993 and 1994 visited our cardiac outpatient clinic. One female patient with paradoxal septal motion in echocardiographic study was excluded from the study leaving 23 patients (18 f / 5 m). The clinical characteristics of the 61 patients (41 f / 20 m) are shown in Table 1. Fifty-seven patients had atrial septal defect of the secundum type and three of the sinus venosus type, two of whom had an anomalous pulmonary venous connection. One patient had both secundum and sinus venosus type atrial septal defect with an anomalous pulmonary venous connection. The method of atrial septal defect repair was direct suture in 56 patients and patch repair in three patients (data missing for

Table 1 Clinical characteristics of 67 control subjects and 61 patients operated for atrial septal defect Parameters

Age (years) BMI (kg / m 2 ) BSA (m 2 ) BPs (mmHg) BPd (mmHg) HR (beats / min) Age at operation (years) Follow-up time (years) Q p / Q s ratio PAPs (mmHg) PAPm (mmHg)

Control subjects (n567, 44 f / 23 m)

Atrial septal defect patients (n561, 41 f / 20 m)

P value

Mean

S.D.

Range

Mean

S.D.

Range

42 24 1.80 123 74 60

12 3 0.17 10 9 8

22–68 19–32 1.49–2.15 104–138 46–88 38–83

43 25 1.79 131 76 69 20 21 2.6 29 18

15 4 0.19 16 12 12 13 5 0.7 6 3

17–71 17–35 1.43–2.22 102–170 40–100 45–101 4–48 5–35 1.5–5.0 17–49 10–30

0.75 0.15 0.73 0.01 0.13 0.001

BMI, body mass index; BPd, diastolic blood pressure; BPs, systolic blood pressure; BSA, body surface area; f, female; HR, heart rate; m, male; PAPs, preoperative systolic pulmonary artery pressure; PAPm, preoperative mean pulmonary artery pressure; Q p / Q s ratio, preoperative size of oxymetric shunt; S.D., standard deviation.

K.W.E. Groundstroem et al. / International Journal of Cardiology 89 (2003) 45–52

two patients). There were no patients with renal failure in our study, none had known chronic atrial fibrillation. Sixteen of the 61 patients used medication; digitalis (4), b-blockers (4), Ca 21 -blockers (1), chinidine sulfate (1), nitrates (2), diuretics (2), angiotensin-converting enzyme inhibitors (3), insulin or oral antidiabetic medication (2), thyroxin (2), carbamazepine (1), pilocarpin and timolol for glaucoma (1), warfarin or acetosalisylic acid (2) and antidepressant alpratsolam (2).

2.2. Control subjects The 67 control subjects were recruited among 98 employees in companies from the Tampere region. After history and clinical examination, electrocardiogram and transthoracic echocardiography, 28 subjects were excluded. In 14, the reason was history of cardiovascular disease or sphygmomanometric blood pressure $140 / 90 [16], in three abnormal electrocardiogram, in two non-cardiovascular disorders, in one regular medication and in eight poor quality echocardiogram. During matching of control subjects and patients for sex, age and body mass index another three subjects were excluded, leaving 67 healthy individuals. Their clinical characteristics are shown in Table 1. The Ethics Committee of Tampere University Hospital approved the study protocol that complied with the Declaration of Helsinki.

3. Methods Medical history was obtained from interviews and patient records. Arterial blood pressure and heart rate were measured in recumbent position at echocardiography after the patient had rested for at least 5 min. Systolic blood pressure $140 and / or diastolic blood pressure $90 mmHg in repeated measurements were defined as hypertension [16]. Standard 12 lead electrocardiogram was obtained on the day of clinical examination.

3.1. Chest roentgenogram Posteroanterior and lateral chest radiograph was performed in the upright position. The target-film distance was 200 cm. Cardiac volume was deter-

47

mined by routinely used ellipsoid technique and correlated with body surface area [17,18]. Cardiac volume was considered enlarged when the volume was above the limits used locally (.450 cc / m 2 in females and .500 cc / m 2 in males) [19]. Features of possible congestive failure were noted.

3.2. Echocardiography Transthoracic echocardiography was performed as previously described using an Acuson computed sonography 128 / XP10 system (Mountain View, CA, USA) [20]. M-mode dimensions were normalized for body surface area. Transoesophageal echocardiography was undertaken in standard fashion [20] with a 5.6-MHz 64channel Acuson biplane transducer connected to the same echocardiographic system. The presence of an interatrial communication was defined as an anatomic defect and / or transseptal blood flow visualized with colour Doppler and two 10-ml intravenous injections of plasma expander (Hemaccel  ), one at baseline and another during the Valsalva maneuver. Valvular regurgitation was evaluated using transoesophageal scanning only. The severity of pathologic mitral regurgitation was based on the onset, number, duration, length, width, and color of the regurgitant flow [20]. A small central monocoloured regurgitation jet was considered physiologic or trivial. A multicoloured holosystolic jet less than 30 mm long and 10 mm wide indicated mild regurgitation. Jets exceeding these dimensions but not reaching into the pulmonary veins were considered as moderate regurgitation and jets extending into the pulmonary veins markers of severe regurgitation. Aortic and pulmonary valve regurgitation was evaluated from the ratio between the immediate subvalvular ventricular outflow tract diameter and colour regurgitation jet width. Regurgitating jets occupying ,25, 26–50, and .50% were considered markers of mild, moderate and severe regurgitation, respectively [20]. Tricuspid regurgitation jets ,10 mm diameter and ,30 mm length were was classified as physiologic or trivial [20]. In the plane with the larger colour Doppler regurgitant jet, jet areas ,1 / 3, 1 / 3–2 / 3, and .2 / 3 of the right atrial area were considered as markers of mild, moderate and severe regurgitation, respectively [20].

48

K.W.E. Groundstroem et al. / International Journal of Cardiology 89 (2003) 45–52

3.3. N-terminal atrial natriuretic peptide measurement Blood samples for atrial natriuretic peptide measurement were drawn from an antecubital vein in sitting position before noon. After centrifugation the samples were stored at 270 8C. Peptide concentrations were measured with commercial IRMA method using two monoclonal antibodies (Medix, Kauniainen, Finland). The interassay coefficients of variation at the levels of 0.30 and 0.70 nmol / l were 6.5 and 5.1%, respectively [21].

4. Statistics The results are expressed as mean6standard deviation (S.D.), and also as median and interquartile range (IQR) when the distribution was skewed. The Kolmogorov–Smirnov test was used to assess the normality of data distribution. In the case of normal distribution (P$0.05) differences between means were tested using the Student’s t-test, otherwise the Mann–Whitney U-test was used. The multiple linear stepwise regression analysis was used to evaluate the associations between independent variables. The control subjects’ mean62 S.D. was used as the reference range for atrial and ventricular dimensions. P-values ,0.05 were considered significant. The statistical

analyses were done using SPSS statistical software (V5.0.1, Chicago, IL, USA).

5. Results

5.1. Clinical characteristics of patients and control subjects The patients had higher systolic blood pressure and heart rate than the control subjects (Table 1). The clinical characteristics of the patients from the hospital records did not differ from those obtained from the outpatient clinic except that the latter had higher diastolic blood pressure (78613 and 70613 mmHg, respectively, P50.03). Cardiac volumes in chest radiogram did not differ.

5.2. N-terminal atrial natriuretic peptide The patients had higher N-terminal atrial natriuretic peptide concentration than the controls: mean (S.D.) 0.4060.32 and 0.2460.12 nmol / l; medians (IQR): 0.30 (0.21–0.47) and 0.23 nmol / l (0.18– 0.29), respectively, P50.001. Peptide concentrations in patient quartiles are shown in Fig. 1. In the patients, peptide levels correlated with current age (P50.0001) and age at operation (P5

Fig. 1. Quartiles of N-terminal natriuretic peptide concentrations in 61 patients operated for atrium secundum defect (s) and 67 control subjects (1). Solid lines indicate median values.

K.W.E. Groundstroem et al. / International Journal of Cardiology 89 (2003) 45–52

49

Table 2 Echocardiography findings in 61 patients operated for atrial septal defect

Any abnormality in echocardiography Transoesophageal echocardiography Residual interatrial communication Tricuspid regurgitation Mild Moderate Severe Mitral regurgitation Mild Moderate Severe Aortic regurgitation Mild Moderate Pulmonary regurgitation mild Fig. 2. Patient age at operation in 61 atrium secundum defect patients divided into postoperative quartiles of N-terminal natriuretic peptide concentration. Bars indicate standard deviation.

0.0014), but not with age in the controls (P5not significant, NS) (Fig. 2). Serum N-terminal atrial natriuretic peptide levels were similar in the patients found in the hospital records and from the outpatient clinic (0.3560.24 and 0.4960.42 nmol / l, NS). Both patient groups had higher serum peptide level than the control subjects (P50.03 and P50.001, respectively). In the patients the systolic blood pressure correlated with hormone concentrations (P50.04), but the diastolic did not (NS). In the control group there was no correlation between blood pressure and peptide concentrations (NS). The peptide concentration did not correlate with body mass index in the patients or controls (NS for both).

5.3. Echocardiography In all 51 / 61 patients (84%) had one or more echocardiography abnormalities (Table 2). The patients had greater right ventricular end-diastolic diameter, left atrial end-systolic diameter, and left ventricular end-systolic diameter and lower left ventricular fractional shortening than the control subjects (1564 and 1262 mm / m 2 , P50.001; 2163 and 1962 mm / m 2 , P50.001; 1964 and 1762 mm / m 2 , P50.008; 3567 and 3964%, P50.001, respectively). Left ventricular end-diastolic diameter did not differ (2964 and 2863 mm / m 2 , NS). Trans-

Transthoracic echocardiography RV visually large and / or hypokinetic Hypokinetic interventricular septal motion LAESD/ BSA .21 mm / m 2 in males or .23 mm / m 2 in females RVEDD/ BSA .16 mm / m 2 LVEDD/ BSA .33 mm / m 2 LVESD/ BSA .21 mm / m 2 LVFS ,31% IVSD/ BSA .6 mm / m 2 LVPWD/ BSA .6 mm / m 2

n

%

51

84

17 a

28

25 6 1

40 10 2

16 6 1

26 10 2

3 3 18

5 5 30

30 21

49 34

16 15 5 11 18 2 1

26 25 8 18 30 3 2

BSA, body surface area; IVSD, interventricular septal end-diastolic thickness; LAESD,left atrial end-systolic diameter; LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; LVFS, fractional shortening of left ventricle; LVPWD, left ventricular posterior wall end-diastolic thickness; RVEDD, right ventricular enddiastolic diameter; RV, right ventricle. a In nine patients there was passage of contrast bubbles across the interatrial septum but no visible defect.

oesophageal echocardiography findings are presented in Table 2.

5.4. Echocardiography and N-terminal atrial natriuretic peptide concentration In the patients, N-terminal atrial natriuretic peptide concentrations correlated with left atrial end-systolic diameter (P50.042), left ventricular end-diastolic diameter (P50.021) and left ventricular end-systolic diameter (P50.042), but not with left ventricular fractional shortening or right ventricular end-diastolic diameter (NS for both). There were 17 patients with a residual interatrial communication (eight with a visible defect and nine with contrast bubble passage across the interatrial

50

K.W.E. Groundstroem et al. / International Journal of Cardiology 89 (2003) 45–52

septum only). Ten of them had peptide levels in the third and fourth quartiles. However, three had concomitant mitral regurgitation, one mitral and more than trivial tricuspid regurgitation, two more than trivial tricuspid regurgitation and two moderate aortic incompetence. The 10 patients without lesions in echocardiography and control subjects had similar serum N-terminal atrial natriuretic peptide level (0.2560.18 and 0.2460.12 nmol / l, NS i.e. lower than in patients with any abnormality in echocardiography (0.4360.34 nmol / l, P50.03).

5.5. Chest roentgenogram None of the patients had blood congestion in the lungs in the plain chest roentgenogram. The cardiac volume was above normal in 11 patients (18%). Six of these patients had N-terminal atrial natriuretic peptide concentrations in the highest quartile, two in the third, two in the second and one in the first. In echocardiography nine of these patients had larger cardiac chamber(s) than the controls: left atrial endsystolic diameter (4), right ventricular end-diastolic diameter (6), left ventricular end-diastolic diameter (2). The chest radiogram was judged as normal in eight patients with serum peptide concentration in the highest quartile. Seven of these eight patients had one or more abnormality in echocardiography, moderate aortic regurgitation (1), residual interatrial communication (1), prolapse of the mitral valve (1), large and / or hypokinetic right ventricle (5), increased left atrial end-systolic diameter (2), enlarged left ventricular end-systolic diameter (1), low left ventricular fractional shortening (3) or hypokinetic interventricular septum (2).

5.6. Electrocardiogram There were 23 patients (38%) with supraventricular or ventricular arrhythmias, conduction defects or ventricular hypertrophy. Eight were in the highest N-terminal atrial natriuretic peptide concentration quartile and the other quartiles contained five patients each.

6. Discussion

6.1. Impact of operation age on postoperative Nterminal atrial natriuretic peptide secretion After surgical seclusion of atrial septal defect the patients had higher N-terminal atrial natriuretic peptide concentration than the control subjects. Patient age at operation correlated significantly with serum peptide levels. We have previously shown that patients operated for atrial septal defect have higher N-terminal atrial natriuretic peptide concentrations than a matched control population [22] and that patients operated at older age have more cardiac lesions [20]. In their study that included postoperative atrial septal defect patients Bolger et al. [13] found that atrial peptide levels are elevated after routine surgical closure. Many changes caused by shunting of blood may be irreversible [23] and continue to drive neurohormonal activation. Our finding that N-terminal atrial natriuretic peptide concentrations are higher in patients with lesions in echocardiography support this hypothesis.

6.2. Cardiac dimensions and N-terminal natriuretic peptide concentration Bolger et al. [13] found that visually assessed global ventricular function correlated with atrial peptide concentration in patients with congenital heart disease patients and chronic heart failure. None of our patients had symptoms of heart failure and left ventricular fractional shortening did not correlate with hormone levels. However, left ventricular enddiastolic and end-systolic diameters did correlate with hormone concentrations in these hearts with postoperative segmental alterations in cardiac contraction [24]. Right ventricular end-diastolic diameter did not correlate with N-terminal atrial natriuretic peptide concentrations. This may seem surprising because a compromised right ventricle is one of the most prominent residual defects after atrial septal defect closure [6,20]. However, atrial peptides are secreted from both the right and the left heart. Also, after cardiopulmonary bypass the heart is mechanically restricted to move and function in the same manner as before the operation [24], which is more likely to

K.W.E. Groundstroem et al. / International Journal of Cardiology 89 (2003) 45–52

51

influence right heart size and function. In our study, left atrial end-systolic diameter correlated with serum N-terminal atrial natriuretic peptide concentration in the patients [13].

N-terminal atrial peptide levels and operation age together with echocardiography findings support the clinical consensus of treating atrial septal defect patients in their childhood and adolescence.

6.3. Postoperative chest radiogram and electrocardiography in atrial septal defect patients

Acknowledgements

Finding enlarged cardiac volume in plain chest roentgenogram in 18% of our patients is in agreement with previous reports [4,5]. Most of these patients had N-terminal atrial natriuretic peptide concentrations in the highest quartiles and also echocardiography abnormalities indicating that further examination of such patients is warranted. Among the patients with normal radiographic cardiac volume, higher peptide concentrations were likely to imply echocardiography abnormalities. In atrial septal defect patients, the high prevalence of postoperative electrocardiography abnormalities impair their value as markers of cardiac dysfunction [5]. In our patients, electrocardiography abnormalities were seen in 38% of patients and their distribution in the peptide concentration quartiles was quite even.

6.4. Study limitations We do not have data on preoperative N-terminal natriuretic peptide levels in our patients. The significance of hormone levels in atrial septal defect patients is uncertain because the long term peptide secretion pattern after surgical closure of is not known. In short term studies of children under 15 years with different types of congenital heart defects, the plasma concentrations of atrial peptide may diminish [25–27]or remain elevated [27] 24–48 h after an operation for congenital heart disease. Because there were only 10 patients without lesions in echocardiography, our study does not give reference values for N-terminal atrial peptide concentrations after surgical seclusion of atrial septal defect to be used as in patients with heart failure [15,28,29] or valvular heart disease [30].

6.5. Conclusion In conclusion, the association between increasing

This study was supported by the Maud Kuistila Memorial Foundation, Turku, Finland, the Medical Research Fund of Tampere University Hospital, Tampere, Finland and the Aarne Koskelo Foundation, Helsinki, Finland. The authors are indebted to Miss Elina Soimasuo RN for technical assistance.

References [1] Samanek M, Slavik Z, Zborilova B et al. Prevalence, treatment, and outcome of heart disease in live-born children: A prospective analysis of 91,823 live-born children. Pediatr Cardiol 1989;10:205– 11. [2] Seldon WA, Rubinstein C, Fraser AA. The incidence of atrial septal defect in adults. Br Heart J 1962;24:557–60. [3] Murphy JG, Gersh BJ, McGoon MD et al. Long-term outcome after surgical repair of isolated atrial septal defect. Follow-up at 27 years. New Engl J Med 1990;323:1645–50. [4] Siltanen P. Atrial septal defect of secundum type in adults. Acta Med Scand 1968;497(suppl):1–151. [5] Huysmans HA, Vrakking M, van Boven WJP. Late follow-up after surgical correction of atrial septal defect of the secundum type. Z Kardiol 1989;78(suppl 7):43–5. [6] Meijboom F, Hess J, Szatmari A, Utens EMWJ, McGhie J, Deckers JW. Long-term follow-up (9 to 20 years) after surgical closure of atrial septal defect at a young age. Am J Cardiol 1993;72:1431–4. [7] Chen HH, Burnett JC. The natriuretic peptides in heart failure: Diagnostic and therapeutic potentials. Proc Assoc Am Phys 1999;111:406–16. [8] Bonow RO. New insights into the cardiac natriuretic peptides. Circulation 1996;93:1946–50. [9] Nicholls MG. Minisymposium: The natriuretic peptide hormones. The natriuretic peptides in heart failure. J Intern Med 1994;235:515– 26. [10] Uchiyama M, Satokata I, Sakai K. Plasma atrial natriuretic peptide in congenital heart disease. Acta Paediatr 1987;76:669–70. [11] Kikuchi K, Nishioka K, Ueda T et al. Relationship between plasma atrial natriuretic polypeptide concentration and hemodynamic measurements in children with congenital heart diseases. J Pediatr 1987;111:335–42. ¨ H, Thaulow E, Stokke O et al. Serum N-terminal [12] Holmstrom proatrial natriuretic factor in children with congenital heart disease. Eur Heart J 1996;17:1737–46. [13] Bolger AP, Sharma R, Li W et al. Neurohormonal activation and the chronic heart failure syndrome in adults with congenital heart disease. Circulation 2002;106:92–9.

52

K.W.E. Groundstroem et al. / International Journal of Cardiology 89 (2003) 45–52

[14] Lang RE, Dietz R, Merkel A, Unger T, Ruskoaho H, Ganten D. Plasma atrial natriuretic peptide values in cardiac disease. J Hypertens 1986;4(suppl 2):119–23. [15] Gottlieb SS, Kukin ML, Ahern D et al. Prognostic importance of atrial natriuretic peptide in patients with chronic heart failure. J Am Coll Cardiol 1989;13:1534–9. [16] Guidelines Sub-Committee. Guidelines for the management of mild hypertension: memorandum from a World Health Organization / International Society of Hypertension meeting. J Hypertens 1993;11:905. [17] Jonsell S. Method for determination of heart size by teleroentgenography (Heart Volume Index). Acta Radiol 1939;20:325–40. [18] Keats TE, Enge JP. Cardiac measurement by the cardiac volume method. Radiology 1965;85:850–5. [19] Amundsen P. The diagnostic value of conventional radiological examination of the heart in adults. Acta Radiol 1959;1(suppl):181. [20] Groundstroem KWE, Iivainen TE, Talvensaari T et al. Late postoperative follow-up of ostium secundum defect. Eur Heart J 1999;20:904–9. ¨ [21] Tikkanen I, Helenius T, Heinanen T et al. N-terminal atrial natriuretic peptide (ANP-N) in heart failure (abstract). Scand J Clin Lab Invest 1995;55(suppl 223):645. [22] Iivainen TE, Groundstroem KWE, Lahtela JT, Talvensaari TJ, Pasternack AI, Uusitalo AJ. Serum N-terminal atrial natriuretic peptide in adult patients late after surgical repair of atrial septal defect. Eur J Heart Fail 2000;2:161–5. [23] Espino-Vela J, Alvarado-Toro A. Natural history of atrial septal defect. Cardiovasc Clin 1971;2:103–25.

[24] Wranne B, Pinto FJ, Siegel LC, Miller DC, Schnittger I. Abnormal postoperative interventricular motion: New intraoperative transesophageal echocardiographic evidence supports a novel hypothesis. Am Heart J 1993;126:161–7. [25] Ationu A, Singer DRJ, Smith A, Elliot M, Burch M, Carter ND. Studies of cardiopulmonary bypass in children: implications for regulation of brain natriuretic peptide. Cardiovasc Res 1993;27:1538–41. [26] Kindelan AA, Navero JLP, De La Rosa II et al. Relationship between hemodynamic changes and blood hormone concentrations after cardiac surgery in children with congenital heart disease. Crit Care Med 1994;22:1754–61. [27] Agnoletti G, Scotti C, Panzali AF et al. Plasma levels of atrial natriuretic factor (ANF) and urinary excretion of ANF, arginine vasopressin and catecholamines in children with congenital heart disease: effect of cardiac surgery. Eur J Cardiothorac Surg 1993;7:533–9. [28] Fyhrquist F, Tikkanen I. Atrial natriuretic peptide in congestive heart failure. Am J Cardiol 1988;62(suppl 2):20–4. [29] Hara H, Ogihara T, Shima J et al. Plasma Atrial natriuretic peptide level as an index for the severity of congestive heart failure. Clin Cardiol 1987;10:437–42. [30] Iivanainen AM, Tikkanen I, Tilvis R, Heikkila¨ J, Kupari M. Associations between atrial natriuretic peptides, echocardiographic findings and mortality in an elderly population sample. J Intern Med 1997;241:261–8.