Author + information
- Received August 21, 2001
- Revision received December 12, 2001
- Accepted December 18, 2001
- Published online March 20, 2002.
- Harald M Gabriel, MD*,
- Maria Heger, MD*,
- Petra Innerhofer, MD*,
- Manfred Zehetgruber, MD*,
- Gerald Mundigler, MD*,
- Maria Wimmer, MD†,
- Gerald Maurer, MD, FACC* and
- Helmut Baumgartner, MD, FACC*,* ()
- ↵*Reprint requests and correspondence:
Dr. Helmut Baumgartner, Department of Cardiology, University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria.
Objectives The purpose of the study was to assess the long-term outcome of patients with small ventricular septal defects (VSDs) considered not to require surgical closure during childhood.
Background Although patients with small VSDs have generally been considered not to require surgery, more recent data suggest that a significant percentage of these patients develop serious problems during adult life.
Methods A total of 229 consecutive patients (115 females) with a VSD considered too small to require surgery during childhood as defined by normal pulmonary artery pressure, less than 50% shunt, pulmonary vascular resistance ≤200 dynes·s·cm−5, no VSD-related aortic regurgitation (AR), and no symptoms and who had no additional hemodynamically relevant heart defect were followed in an adult congenital heart disease program. Physical examination, electrocardiography, and echocardiography were performed in all patients in one- to three-year intervals; exercise tests and Holter monitoring were performed in 140 and 127 patients, respectively.
Results Follow-up was completed in 222 patients (97%). Mean age at last visit was 30 ± 10 years. Spontaneous VSD closure was observed in 14 patients (6%). No patients died, four patients (1.8%) had an episode of endocarditis, of whom two required aortic valve replacement, and one additional patient (0.4%) had surgical closure for hemodynamic reasons. For 118 patients who entered the study between 1993 and 1996 and were prospectively followed for 7.4 ± 1.2 years, event-free survival with end points defined as death, endocarditis or heart surgery was 99.1 ± 0.8% at three years, 96.5 ± 1.7% at six years and 95.5 ± 1.9% at eight years. At last visit, 94.6% of all patients studied were symptom free. Left ventricular (LV) size by echocardiography was normal in 198 (89%) patients, borderline in 23 patients and definitely enlarged in only one patient. None had systolic LV dysfunction, and pulmonary artery pressure (PAP) was normal in all patients. Mean exercise capacity was 92 ± 21% of expected, and 87% of patients had no arrhythmias on Holter monitoring, with the remainder showing benign rhythm disorders.
Conclusions Outcome in well-selected patients with a small VSD is good. Surgical closure does not appear to be required during childhood as long as left-to-right shunt is <50% and signs of LV volume overload are absent, when PAP is not elevated, and no VSD-related AR or symptoms are present.
It had been well accepted for many years that patients with a small ventricular septal defect (VSD) as defined by a left-to-right shunt of <50%, normal pulmonary artery pressure (PAP), and absence of symptoms do not require surgical repair (1–5). Such criteria for operation were based on the observation that VSDs frequently close spontaneously (1,4–6), that patients with such small defects are unlikely to develop pulmonary hypertension (3,4,7)and that the clinical outcome was assumed to be good (1,3,4,7). However, more recent studies that followed patients with “small” VSDs into adulthood reported a worrying rate of complications such as endocarditis (8,9), congestive heart failure (8), development of significant aortic regurgitation (AR) (8,9), occurrence of arrhythmias and even sudden death (2,8,9). These observations raised the question of whether criteria for surgical repair of small VSDs should be more liberal than previously thought, particularly as the surgical risk in these patients is now considered to be very low (10). However, presently available data are limited by small patient numbers (8)and, probably, selection bias (8,9). In addition, patients in these previous studies were not followed on a regular basis in a specialized congenital cardiac center. Therefore, we report the outcome in a large unselected group of adolescents and adults with VSDs considered too small to require surgery during childhood and who were regularly followed in our adult congenital heart disease program.
The patient population consisted of 229 consecutive patients with small VSDs (female/male 115/114). All patients fulfilled the following inclusion criteria: 1) Surgical closure had not been performed during childhood for the following reasons: normal PAP, Qp/Qs <2 (<50% shunt), pulmonary vascular resistance ≤200 dynes·s·cm−5, no VSD-related AR and no symptoms; and 2) no additional hemodynamically relevant heart defect was present. The patient population represents the total number of patients followed in the congenital heart disease program of the Department of Cardiology, University of Vienna; patients fulfilled these criteria at their first visit to the Cardiology Department. The adult congenital heart disease unit was established in 1993, and it presently follows approximately 6,500 patients. Most patients had been transferred directly from the Department of Pediatric Cardiology; this normally happens at an age between 14 and 18 years. Patients with episodes of endocarditis, arrhythmias or other cardiac problems prior to their first visit were not excluded from the study.
At first visit, the study protocol included patient history, physical examination, 12-lead electrocardiogram (ECG) and transthoracic echocardiography. These examinations were repeated in one- to three-year intervals. Since 1997, all patients were asked at their regular visit to have at least one bicycle exercise test and 24-h Holter monitoring during follow-up. These tests were eventually performed in 140 and 127 patients, respectively.
All echocardiograms were performed and interpreted at one laboratory. Standard transthoracic M-mode, two-dimensional and Doppler echocardiography were performed with commercially available ultrasound systems. The VSDs were classified by their location into perimembranous, inlet, outlet supracristal, outlet infracristal, and trabecular. A left ventricular end-diastolic diameter (LVEDD) >57 mm was considered abnormal. For a patient size below average, the LVEDD index (LVEDDi) was also taken into account (≤30 mm/m2). A fractional shortening of <28% was considered abnormal. The peak pressure difference between right and left ventricle (LV) was calculated from the continuous-wave Doppler-measured maximum VSD jet velocity using the simplified Bernoulli equation. Systolic right ventricular-to-right atrial pressure difference was calculated from the continuous-wave Doppler-measured peak tricuspid regurgitant velocity with the simplified Bernoulli equation. When pulmonary stenosis was excluded, systolic PAP was obtained by adding 5 mm Hg for the right atrial pressure as long as the gradient did not indicate pulmonary hypertension and tricuspid regurgitation was not hemodynamically relevant. A systolic PAP ≤32 mm Hg was considered normal. If no adequate Doppler signal of the tricuspid regurgitant flow could be obtained, PAP was calculated by subtracting the left ventricular to right ventricular systolic gradient (LV-RV gradient) from the systemic arterial pressure measured with cuff, provided that no aortic stenosis was present.
Aortic stenosis was classified mild or moderate as long as the aortic valve area was >1.0 cm2. Patients with an aortic valve area <1 cm2were excluded. Aortic, mitral and tricuspid regurgitation were semiquantified using color Doppler and pulsed-wave Doppler (trace, mild, moderate and severe).
24-h Holter monitoring and exercise test
Two-channel recording was used for 24-h Holter monitoring. Exercise capacity was evaluated with a bicycle exercise test using the standard protocol of the Austrian Society of Cardiology and was indicated in percent of expected capacity (11).
Ever since the establishment of the adult congenital heart disease program in 1993, patients were prospectively followed. Patients were seen in one- to three-year intervals. Besides careful patient history and physical examination, ECGs and echocardiograms were repeated at each visit. In 1993, all patients with small VSDs registered in the database of the Department of Cardiology were invited to a follow-up visit. If patients did not return to the appointed visit at this time or during the later prospective follow-up, they were contacted. If patients could not be reached, their relatives and their physician were contacted to find out about patient outcome.
For the assessment of outcome, the following events were defined: 1) major events: death, culture-positive endocarditis, surgical VSD closure; and 2) minor events: symptoms of congestive heart failure requiring medical therapy, arrhythmias requiring medical therapy, hospitalization for cardiac disease related to the VSD.
Descriptive data are given as means ± SD. For the descriptive presentation, all patients seen between 1993 and 2001 were included, and the available follow-up data before establishment of the special program were also considered (retrospectivedata).
For all patients who were recruited during the first four years of the adult congenital heart disease program (i.e., 1993 to 1996) and who were prospectivelyfollowed until 2001, event-free survival was analyzed by the Kaplan-Meier method (12).
Follow-up was completed in 222 patients (97%). Mean age at the last visit was 30 ± 10 years. The age distribution of the study population is shown in Figure 1. Mean follow-up of the entire study group (including retrospective data) was 7.4 ± 7.3 years. The longest follow-up was 29 years. Thirty-one percent of the patients were under surveillance for more than 10 years, and 9% of patients were followed for more than 20 years.
Defect locations were as follows: perimembranous, 194 (84.8%); trabecular, 30 (13.1%); outlet infracristal, 4 (1.7%); inlet, 1 (0.4%). Forty-two patients (18%) presented at follow-up with an aneurysm of the membranous septum. Spontaneous closure of the VSD during follow-up was observed in 14 patients (6%) at an age of 21.5 ± 5.6 years (range, 17 to 31 years).
Mortality and morbidity
No patients died during follow-up.
Four patients (1.8%) had an episode of endocarditis at age 24, 27, 34 and 55 years, respectively. All of them had a perimembranous defect. One patient with bicuspid aortic valve developed aortic valve endocarditis and required valve replacement. A Ross procedure was successfully performed. A second patient suffered endocarditis of a tricuspid aortic valve and received a homograft. The other two patients had mitral valve endocarditis and were treated medically. One of these patients had mitral valve prolapse.
Surgical closure of the VSD
In the two patients with endocarditis who required aortic valve replacement, the VSD was closed at the same time. Another patient required surgical repair of a ruptured sinus of Valsalva aneurysm when he was 27 years old and the VSD was closed. Only one patient underwent surgery for the VSD itself during follow-up. During childhood, her defect was considered too small to require surgery. When she was first seen in our outpatient clinic at age 35 she was still asymptomatic, but echocardiography suggested LV volume overload (LVEDDi 39 mm/cm2). By echocardiography, a maximal defect diameter of approximately 5 mm was measured. Heart catheterization revealed a Qp/Qs of 2:1. The PAP was normal (28/10/17 mm Hg). Based on these data, elective surgical closure was eventually performed. Surgery was successful in all patients, and no complications occurred.
Event-free survival during prospective follow-up
The 118 patients who were seen during the first four years of the congenital heart disease program (i.e., 1993 to 1996) were prospectively followed until 2001 (7.4 ± 1.2 years). Their age at entry was 20.8 ± 8.1 years. Follow-up was 96% complete. Event-free survival with end points defined as death, endocarditis or heart surgery was 99.1 ± 0.8% at three years, 96.5 ± 1.7% at six years and 95.5 ± 1.9% at eight years (Fig. 2).
Requirement of medical treatment or hospitalization
In addition to those patients who had endocarditis or surgery, no patient was admitted to a hospital for cardiac problems related to the VSD during follow-up. No patient required treatment for congestive heart failure. Two patients were on antiarrhythmic drugs (sotalol and verapamil) for supraventricular arrhythmias. They became symptomatic at 24 and 45 years, respectively. Seven patients required medical treatment for arterial hypertension (angiotensin-converting enzyme inhibitors, beta-blockers and/or diuretics); two of them were also on lipid-lowering drugs. Hypertension was detected at a mean age of 40.0 ± 11.6 years.
Patient presentation at last follow-up
At their last visit, 210 patients (94.6%) were free of symptoms; 8 patients (3.6%) complained of mild exertional shortness of breath and 4 patients (1.8%) reported atypical chest pain.
In one patient, the ECG showed permanent regular ventricular pacing. Pacemaker implantation in this patient had been performed before study entry. Of the remaining 221 patients, 215 presented with normal heart axis, while 6 patients had left axis deviation.
Incomplete right bundle branch block was present in 23 patients, complete right bundle branch block was present in 2 patients, and complete left bundle branch block in 1 patient. No first-degree or higher-degree heart block was found in any patient. Of three patients with signs of LV hypertrophy, two had arterial hypertension.
Considering both LVEDD and LVEDDi, the LV was found to be definitely enlarged in only one patient. This patient also developed symptoms (limited exercise capacity) and is currently being more closely evaluated for eventual surgical repair. The LV size was borderline in 23 patients but without significant change during follow-up. The remaining 198 patients (89%) had normal LV size. All patients had normal LV function. The PAP could not be estimated in five patients for technical reasons. In 217 patients with available data, systolic PAP was normal. Additional echocardiographic findings were AR in 12 patients (trace in 7, mild in 5), bicuspid aortic valve in 3 patients, moderate aortic stenosis in 1 patient, aortic valve replacement (autograft, homograft) in 2 patients, mild mitral regurgitation in 46 patients, mild tricuspid regurgitation (more than physiologic) in 6 patients, mild pulmonary stenosis in 1 patient and double-chambered right ventricle (gradient <50 mm Hg) in 5 patients.
Exercise testing and Holter monitoring
In the 140 patients (63%) who underwent bicycle exercise testing, mean exercise capacity was 92 ± 21% of expected (Fig. 3). More than 75% of the patients had an exercise capacity >80% of expected, and only 14 patients (10%) had an exercise capacity below 70%. Thirty-six patients had an exercise capacity greater than 100% (100% to 166%).
Twenty-four hour Holter monitoring could be performed in 127 patients (57.2%). Of these, 111 patients (87%) had no pathologic findings. Frequent premature atrial beats (>1%) could be detected in four patients (3%) and supraventricular tachycardia in two patients, with atrioventricular-node–re-entrant tachycardia in one patient. Ventricular arrhythmias such as bigeminy, couplets or triplets were found in nine patients (7%).
In this large group of 222 patients with small VSDs the outcome was excellent: The vast majority of patients remained free of any symptoms; AR rarely developed, being present in only 5% at last follow-up and trivial or mild in all instances; endocarditis was infrequent with 1.8%; arrhythmias were found in only 13% and were benign in all patients. This outcome is remarkably better than that recently reported by two other groups of investigators (8,9). However, in both previous studies there is a high likelihood of a selection bias. The population studied by Otterstad et al. (8)comprises patients who were referred for evaluation of possible cardiac surgery from all over Norway. Thus, it is likely that the spectrum of disease severity differed from that in our population. The fact that half the patients had mild symptoms such as exercise dyspnea or chest discomfort whereas ours were all asymptomatic during childhood appears to support this assumption. In the study reported by Neumayer et al. (9), patients were at least partially referred to this special unit because of complications such as endocarditis or because of complaints. In addition, they included patients who had neither heart catheterization nor echocardiography. In contrast, most patients in our study were transferred from the Department of Pediatric Cardiology within the same institution. The policy is that all patients are sent to the adult unit regardless of how small the defect is. In addition, patients in whom surgery is withheld are carefully selected: They must have small shunts (Qp/Qs definitely <2.0), normal PAP, no signs of LV volume overload, no VSD-related AR or aortic cusp prolapse, and no symptoms. Finally, all patients with small VSDs are followed in the adult congenital cardiac disease program on a regular basis. It has to be emphasized that the results of the present study may only apply to such carefully selected low-risk patients.
Although we had no mortality in our study, 11 of 69 patients died during follow-up in the report by Otterstad et al. (8). Four of these deaths were suspected to be VSD related (one bacterial aortic valve endocarditis; three heart failure). In the series reported by Neumayer et al. (9), three deaths occurred in 188 patients. Two patients died suddenly and one died from gram-negative septicemia.
LV volume overload and pulmonary vascular disease
It has previously been reported (3,4,7)that patients with small VSDs are unlikely to develop pulmonary hypertension, and shunt flow is unlikely to increase with advancing age. For the latter, shunt flow has even been observed to decrease. However, although PAP can be measured with satisfying accuracy even by echocardiography, shunt flow and LV volume load are more difficult to assess and may be underestimated in some patients, particularly when a single evaluation is performed in the specialized center. One might suspect that at least in a part of those patients who present with LV dilation and/or congestive heart failure at late follow-up, volume overload was originally underestimated. Indeed, 5% of patients in the series by Otterstad et al. (8)were eventually found to have significant left-to-right shunt.
In the Second Natural History Study (NHS–2) (2), patients with normal pulmonary vascular resistance but significant shunt flow had a significantly worse outcome compared to patients with small shunts. Shunt-flow calculations regardless of the method used are prone to significant error, and the LV should, therefore, be carefully evaluated for signs of volume overload. Patients with borderline findings should be carefully followed. Based on currently available data (2,8,9)we believe that patients with signs of volume overload should undergo surgery even in the absence of pulmonary hypertension or symptoms.
Development of AR or presence of AR at late follow-up has been reported within the wide range of 2% to 20% in previous studies (2,7–9,13). This variety is not surprising, AR can be directly related to the VSD. This is mainly the case in outlet defects (supracristal/infracristal) but can also occur in perimembranous defects (8,14,15). Patients with obvious prolapse of an aortic cusp into the VSD have the highest risk of developing progressive AR (14). Thus, there is consensus that outlet VSDs together with aortic valve prolapse and all VSDs with directly related AR that is more than trivial, particularly when progression of AR is observed, should be corrected even if the defect is small (16,17). In studies that excluded patients with AR at presentation such as the NHS–2 (2), development of AR at follow-up was rarely observed (2%). However, 8 of these 12 patients had developed AR after VSD closure, emphasizing that AR in patients with VSD is not always directly related to the VSD itself but rather a consequence of associated lesions such as bicuspid valve, subaortic stenosis, and abnormalities of the ascending aorta. It is not surprising that, in these cases, development or progression of AR cannot be prevented by VSD closure (2,15).
The occurrence of endocarditis has been described as the major risk in small VSDs (1,4,13). Again, the incidence of this complication varies widely in previous reports, ranging from 1% to 15% (2,3,8,9,13). Variations reflect the difficulty of estimating the true incidence of this complication (3). Shah et al. (18)estimated the risk of a 15-year-old with a VSD developing bacterial endocarditis by age 70 to be 11.5%. Corone et al. (7)followed 790 patients for 25 years and reported endocarditis in 3.7% (2.4 per 1,000 patient-years). The highest percentage of 15% (5.7 per 1,000 patient-years) was reported by Otterstad et al. (13)for a heterogeneous group of 109 patients with a high percentage of AR.
In contrast, bacterial endocarditis occurred rarely in patients with small VSDs in the NHS–2 (1.3 per 1,000 patient-years) (2). Similarly, Keith et al. (3)reported endocarditis to occur infrequently in early adult life. Based on the NHS–2 data, Gersony et al. (19)reported that the incidence of endocarditis was significantly less after VSD closure but not eliminated, whereas Kidd and associates (2)reported from the same study an endocarditis rate of 1.4 versus 1.6 per 1,000 patient-years for nonoperated and postoperative follow-up, respectively. The fact that they excluded patients with additional AR may explain the discrepant results, and it highlights the fact that endocarditis is in part related to additional valvular disease in this patient population. In our series, endocarditis infrequently occurred (1.8% or 2.4 per 1,000 patient-years), and one patient with aortic valve endocarditis had a bicuspid valve; another patient with mitral valve endocarditis had mitral valve prolapse with regurgitation. Residual VSD and valvular disease after surgery may explain why patients remain at risk of endocarditis even after VSD closure. Thus, although endocarditis is definitely a matter of concern in VSD patients, the risk of endocarditis alone may not justify the need to close all small VSDs (20).
Spontaneous defect closure
The high chance of spontaneous closure is one of the major reasons why small VSDs are followed conservatively (5,20). Reported closure rates vary with size and location of VSD, age at presentation and patient population. Alpert et al. (21)reported that small VSDs have a >50% chance of spontaneous closure by five years of age and a greater than 80% chance by adolescence. Spontaneous closure becomes less likely during adolescence and adult life. In 74 patients over 21 years of age (including 38 patients with small defects) admitted to the NHS–1 (22), only one closure was observed. In NHS–2 (2), spontaneous closure was reported in 15% of patients. In the study by Neumayer et al. (9), spontaneous closure occurred in 10% of patients, similar to our finding of 6%.
A higher-than-normal prevalence of arrhythmias in patients with VSD has previously been reported. In NHS–2 (2), 14% of medically managed patients had multiform premature ventricular contractions; 8% had couplets; ventricular tachycardias were seen in 3%, and even sudden death was reported. However, it is not specified how many of these patients had small VSDs. Neumayer et al. (9)reported 9% of patients in their series had supraventricular and ventricular arrhythmias. These numbers are similar to ours. The tendency to develop arrhythmias cannot be expected to be positively affected by surgery. In NHS–2 (2), the prevalence of arrhythmias was even significantly higher in surgically treated patients. Houyel et al. (23)also reported arrhythmias to be frequent after surgical VSD closure. Thus, the occurrence of arrhythmias is a possible complication in patients with VSD, but cannot be avoided by surgical repair and is even more likely after surgery.
Quality of life and socioeconomic considerations
In the present study, only 5% of patients reported mild symptoms, whereas 95% were asymptomatic. Similarly, in NHS–2 (2), 94% of patients with small VSDs were in New York Heart Association functional class I. Backer et al. in their study (10)argue that a definite socioeconomic stigma is associated with living with an uncorrected cardiac defect. Patients may not be allowed or may hesitate themselves to participate freely in physical education and sports and may have difficulty obtaining health and life insurance. In view of the low risk of repair in the era of modern cardiac surgery, the investigators use this as one of the arguments to consider surgery even in patients with small VSDs (10). However, the fact of having had cardiac surgery and the fact of having a median sternotomy scar may cause psychological problems in itself and operation does not replace the need for communication with the insurance company or the physical education teacher (20). In addition, patients with a small VSD should have no restrictions placed on physical activities, and there is no evidence that competitive sports are more dangerous with a small VSD than with an intact septum nor do data confirm the safety of competitive sports after surgery. Thus, possible “socioeconomic stigma” cannot be accepted as an argument for rigorous VSD closure.
Although most patients in this study were sent to the adult unit by the Department of Pediatric Cardiology where they had been followed regardless of how small the defect was, some selection bias cannot be excluded. However, if anything, this bias would likely be weighted toward sicker or more symptomatic patients. If such a referral bias were indeed present, it would only mean that the outcome of the overall population of patients with small VSDs would be even better than reported.
Another limitation of the present study is that only 62% of patients agreed to have exercise testing, and only 57% agreed to have Holter monitoring. This may have affected the results on exercise capacity and the occurrence of arrhythmias. However, patients who are not totally asymptomatic or do not feel entirely healthy are in general more likely to agree to further testing. Thus, if there is any bias, this should have worsened the results rather than improved them.
A major limitation of the present study is that many of our patients are still young adults, and problems may increase with advancing age. However, although average age at last follow-up was significantly higher in the series by Otterstad et al. (8), it was very similar in the study by Neumayer et al. (9), the second previous report with markedly worse patient outcome compared to the one in our population. Nevertheless, we will continue to follow these patients carefully to learn about outcome at advanced age.
The results of the present study confirm that it is safe to withhold surgery in patients with an isolated small VSD as long as patients are carefully selected and carefully followed in a specialized congenital heart disease program. Surgery does not appear to be required as long as left-to-right shunt is definitely <50% and signs of LV volume overload are absent, when pulmonary pressure is not elevated, and no VSD-related AR and/or symptoms are present.
- aortic regurgitation
- left ventricle/ventricular
- left ventricular end-diastolic diameter
- left ventricular end-diastolic diameter index
- LV-RV gradient
- left ventricular to right ventricular systolic gradient
- pulmonary artery pressure
- shunt ratio (pulmonary to systemic flow)
- ventricular septal defect
- Received August 21, 2001.
- Revision received December 12, 2001.
- Accepted December 18, 2001.
- American College of Cardiology Foundation
- Bloomfield D.K.
- Keith J.D.,
- Rose V.,
- Collins G.,
- Kidd B.S.
- Campbell M.
- Dickinson D.F.,
- Arnold R.,
- Wilkinson J.L.
- Alpert B.S.,
- Cook D.H.,
- Varghese P.J.,
- Rowe R.D.
- Corone P.,
- Doyon F.,
- Gaudeau S.,
- et al.
- Neumayer U.,
- Stone S.,
- Somerville J.
- Boehm H.,
- Buerklen R.,
- Dienstl F.,
- et al.
- Otterstad J.E.,
- Nitter-Hauge S.,
- Myhre E.
- Tatsuno K.,
- Konno S.,
- Ando M.,
- Sakakibara S.
- Somerville J.,
- Brandao A.,
- Ross D.N.
- Shah P.,
- Singh W.S.,
- Rose V.,
- Keith J.D.