- Email: [email protected]
Outcome of pregnancy in patients with congenitally corrected transposition of the great arteries
Outcome of Pregnancy in Patients With Congenitally Corrected Transposition of the Great Arteries Judith Therrien,
MD,
Isobel Barnes,
BSC, MSC,
and Jane Somerville,
MD
To assess maternal and fetal outcome of pregnancy in patients with congenitally corrected transposition of the great arteries, we reviewed 19 patients (18 retrospectively) who had 45 pregnancies. Their ages ranged from 18 to 40 years (mean 27) at the time of pregnancy. Thirty-six percent of the pregnancies were undertaken while patients were cyanosed, 7% in patients with unpaced complete heart block, and 16% were undertaken after surgical repair of the associated anomalies. Change in functional class and maternal cardiovascular complications during pregnancy were analyzed as well as number of live births, miscarriages, elective termination of pregnancies, timing of delivery, and incidence of
cardiac defects in the live offspring. Five patients (26% of patients) developed cardiovascular complications during pregnancy, namely congestive heart failure (3 patients), worsening cyanosis (1 patient), and cerebrovascular accident (1 patient). No maternal deaths occurred. There were 27 live births (60%), 12 miscarriages (27%), and 6 elective terminations of pregnancy (13%). Cyanosis was a significant risk factor for miscarriage. One live offspring had congenital heart disease. Close follow-up of these patients during pregnancy, by a team of experienced physicians, is recommended. 䊚1999 by Excerpta Medica, Inc. (Am J Cardiol 1999;84:820 – 824)
ongenitally corrected transposition of the great arteries (CCTGA) is a combination of atriovenC tricular and ventriculoarterial discordance. The ana-
CCTGA on the database. Twenty had conceived between 1954 and 1998, with a total of 46 pregnancies. Twelve patients were attending the outpatient service at the time of their pregnancy; 8 patients were referred after pregnancies had been completed. One patient was followed prospectively. One patient was excluded because of concomitant Eisenmenger syndrome. Medical records were reviewed in all patients and additional investigations such as echocardiograms were reviewed when available. The cardiac diagnosis of atrioventricular and ventriculoarterial discordance and associated anomalies was established by cardiac catheterization, echocardiography, or at surgery. The patient population was divided into 2 groups according to their basic cardiac anatomy: group 1, CCTGA with no or minor associated anomalies (small atrial septal defect, mild pulmonary stenosis, or mild isolated structural left atrioventricular valve anomaly); group 2, CCTGA with ventricular septal defect and pulmonary outflow tract obstruction (valvar, subvalvar, or pulmonary atresia). Maternal outcome: Recorded cardiovascular complications and maternal mortality occurring during or after pregnancy (within 1 year) were noted. Change in patients’ functional class during and after (up to 1 year) pregnancy was also analyzed as an outcome variable. Fetal outcome: The number of live births with timing of delivery, miscarriages, and elective termination were studied, as well as the incidence of cardiac defects in live offspring. Miscarriages were defined as spontaneous abortion at or before 20 weeks of gestation and prematurity as delivery at or before 37 weeks of gestation. Analysis of data: The effect of associated cardiac lesions, maternal age, baseline New York Heart Association (NYHA) class before pregnancy, cyanosis
1
tomic left ventricle becomes the pulmonary ventricle, whereas the anatomic right ventricle, guarded by a tricuspid valve, is the systemic ventricle. Usually, there are associated defects such as ventricular septal defect, pulmonary stenosis (at valvar or subvalvar level), or anatomically abnormal left-sided (tricuspid) atrioventricular valve with varying degrees of regurgitation. Complete heart block may be congenital or acquired. Symptoms depend on the severity of the associated lesions and the function of the systemic right ventricle.2 There are ongoing concerns about the long-term ability of the morphologic right ventricle and tricuspid valve to support systemic pressure3– 6 and their capacity to handle the volume load of pregnancy.3,7,8 The paucity of information about the outcome of pregnancy in adult patients with CCTGA9 led us to study the outcome of pregnancy and its cardiovascular effects in women with CCTGA.
METHODS
Selection of patients: The database of the Grown Up Congenital Heart Unit, a designated quaternary center for adolescents and adults with congenital heart disease, was searched retrospectively for adult female patients aged ⱖ16 years with confirmed diagnosis of CCTGA, with or without associated lesions, who had become pregnant. There were 39 female patients with From the Royal Brompton Hospital, National Heart and Lung Institute and Imperial College School of Medicine, London, United Kingdom. Manuscript received January 14, 1999; revised manuscript received and accepted May 12, 1999. Address for reprints: Judith Therrien, MD, The Toronto Hospital, 200 Elizabeth Street, 12 EN Room 213, Toronto, Ontario, Canada. E Mail: [email protected].
820
©1999 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 84 October 1, 1999
0002-9149/99/$–see front matter PII S0002-9149(99)00444-0
(defined as clinical central cyanosis or an oxygen saturation of ⱕ85%), surgery (palliative or reparative), and complete heart block on fetal and maternal outcome were studied. All analyses were performed using a random-effects logistic regression procedure using the statistical package EGRET (SERC, Seattle, Washington), version 1.00.03. A p value ⬍0.05 was considered statistically significant.
RESULTS
Patient characteristics: GROUP 1: Six patients had 13
pregnancies at a mean age of 27 years. Five patients (with 10 pregnancies) were in NYHA class I before pregnancy and 1 patient was in class I before her first 2 pregnancies and in class II before her third. All patients were acyanotic and 1 patient (with 3 pregnancies) had a bioprosthetic left atrioventricular valve replacement for left atrioventricular valve regurgitation before her third pregnancy. Two patients (with 4 pregnancies) had complete heart block of which 1 (with 2 pregnancies) was treated with a dual-chamber pacemaker before pregnancy. No patient was taking medication before pregnancy. GROUP 2: Thirteen patients had 32 pregnancies at a mean age of 27 years. Eight patients (with 16 pregnancies) were in NYHA class I before pregnancy and 5 (with 16 pregnancies) were in class II before pregnancy. Four patients (with 13 pregnancies) were cyanotic before each pregnancy and 1 (with 4 pregnancies) was cyanotic before her first 3 pregnancies. Two patients (4 pregnancies) who had had cyanosis underwent previous palliative surgery (modified right and left Blalock Taussig shunt in 1 patient and Waterston shunt in 1 patient). Eight patients (with 16 pregnancies) were acyanotic, 5 (with 6 pregnancies) having had intracardiac repair before pregnancy (ventricular septal defect patch closure and left ventricular to pulmonary artery conduit in 2 patients, ventricular septal defect patch closure, pulmonary valvotomy, and left atrioventricular valve replacement in 2 patients, and Fontan procedure in 1 patient). Five patients (with 6 pregnancies) had associated complete heart block; 4 (with 5 pregnancies) were treated with endovenous pacemaker before pregnancy. Two patients were taking diuretics and 3 were taking angiotensin-converting enzyme inhibitors before pregnancy (Table I). Maternal outcome: Cardiovascular complications occurred in 5 patients (26% of patients) (group 1, 1 patient; group 2, 4 patients) with 5 completed pregnancies (11% of the pregnancies) at the later stage of pregnancy. Details of the patients with cardiovascular complications during pregnancy are summarized in Table II. Congestive heart failure was the cause of deterioration in 3 patients from worsening systemic right ventricular dysfunction (patients 1 and 2), worsening left atrioventricular valve regurgitation (patient 2), and atrial fibrillation from a stenotic left atrioventricular prosthetis (patient 3). Worsening cyanosis occurred in 1 patient (no. 4) and cerebrovascular accident in 1 patient (no. 5). Deterioration in NYHA functional class occurred in the first 4 patients (21%) (Figure 1). No cardiovascular complications occurred
TABLE I Patient Characteristics Before Pregnancy Group 1 No or Minor Associated Defects (n ⫽ 13 pregnancies)
Group 2 VSD and POTO (n ⫽ 32 pregnancies)
27 (19–40)
27 (18–39)
12 (92%) 1 (8%) 0
16 (50%) 16 (50%) 16 (50%)
Mean age (yr) at pregnancy (range) NYHA I II Cyanosis Surgery Palliative Repair Complete heart block With pacemaker Without pacemaker
0 1 (8%)
4 (12%) 6 (18%)
2 (16%)
5 (15%)
2 (16%)
1 (3%)
POTO ⫽ pulmonary outflow tract obstruction; VSD ⫽ ventricular septal defect.
with spontaneous miscarriage or elective termination of pregnancy. There were no maternal deaths. Univariate analysis showed that older maternal age (p ⫽ 0.025) and the presence of untreated complete heart block (p ⫽ 0.01) were significant risk factors for worsening functional class during pregnancy. However, on multivariate analysis, neither retained their significance (Table III). Fetal outcome: There were 27 live births (60%) (group 1, 10; group 2, 17). Twenty-two (49%) were born at term and 5 (11%) prematurely at 31 to 35 weeks of gestation (group 2, 5). Twelve spontaneous miscarriages (27%) occurred at 10 to 20 weeks of gestation (group 2, 12), and 6 elective termination of pregnancies (13%) were performed at the request of the patient. One live offspring (4%) (group 2) had congenital heart disease— hypoplastic left heart syndrome (Table IV). Potential risk factors for miscarriage were examined with results shown in Table V. Pregnancies that ended in elective termination were not included in the analysis and, owing to sparse data, the statistical significance of associate defects and complete heart block could not be determined. NYHA class II (p ⬍0.001) and cyanosis before pregnancy (p ⬍0.001) were significant risk factors for miscarriage. Multivariate analysis showed that once cyanosis was taken into account, no other risk factor added significantly to the prediction of miscarriage (Table V).
DISCUSSION
Maternal outcome: In this study, 74% of patients had an uneventful pregnancy. Cardiovascular complications occurred in 5 patients (26% of the patients) with 5 pregnancies (11% of the pregnancies). Complications intrinsic to CCTGA, namely the development of congestive heart failure and systemic tricuspid valve regurgitation occurred in 2 patients (10%). Dur-
CONGENITAL HEART DISEASE/CONGENITALLY CORRECTED TRANSPOSITION OF GREAT ARTERIES
821
— — 0 I
32
29
4 (2nd)
5 (4th)
VSD/PS (2)
0
⫹ II
26 3 (1st)
CHB ⫽ complete heart block; CHF ⫽ congestive heart failure; CVA ⫽ cerebrovascular accident; LAVV ⫽ left atrioventricular valve; LAVVR ⫽ left atrioventricular valve regurgitation; LV ⫽ left ventricular; PA ⫽ pulmonary artery; pm ⫽ pacemaker; PS ⫽ pulmonary stenosis; PV ⫽ pulmonary valve; RV ⫽ right ventricular; VSD ⫽ ventricular septal defect; — ⫽ no information available.
I — —
II —
29th wk gestation 12th wk gestation — —
0 II
39 2 (4th)
VSD/PS Ebstein LAVV (2) VSD/PS (2)
0 II
VSD closure LV to PA conduit VSD closure PV valvotomy LAVVR 0
⫹ (0) 0
0 0
⫹ ⫹⫹
Worsening cyanosis CVA
—
II 0 0 CHF
III ⫹⫹ ⫹⫹⫹ CHF
I ⫹⫹⫹ ⫹⫹ ⫹⫹
⫹ (0) ⫹ (⫹) 0 0 0 I 40 1 (2nd)
Mother # (pregnancy #)
Isolated CCTGA (1) VSD/PS (2)
Cyanosis Diagnosis (group)
Surgery Before Pregnancy
CHB (pm)
⫹⫹
48 hours after delivery 29th wk gestation 36th wk gestation
CHF
NYHA at 1 Year Postpartum RV Dysfunction at Time of Deterioration (0–3⫹) LAVVR at Time of Deterioration (0–3⫹) Cause of Maternal Deterioration Timing of Maternal Deterioration RV Dysfunction Before Pregnancy (0–3⫹) LAVVR Before Pregnancy (0–3⫹) NYHA Before Pregnancy Age at Pregnancy (yrs)
TABLE II Maternal Complications
822 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 84
FIGURE 1. Change in NYHA functional class before (<3 months), during, and after (<1 year) pregnancy. N ⴝ number of pregnancies; I ⴝ NYHA class 1; II ⴝ NYHA class II; III ⴝ NYHA class III; IV ⴝ NYHA class IV.
TABLE III Risk Factors for Maternal Deterioration During Pregnancy
Associated defects Group 1 Group 2 Mean maternal age (yr) (range) Functional class I II Cyanosis Surgery Complete heart block without pacemaker
Same Functional Class During Pregnancy (n ⫽ 41)
Worsening Functional Class During Pregnancy (n ⫽ 4)
12 (29%) 29 (71%) 26 (18–38)
1 (25%) 3 (75%) 34 (26–40)*
27 15 15 9 1
(66%) (34%) (37%) (22%) (2%)
1 3 1 2 2
(25%) (75%) (25%) (50%) (50%)†
*p ⫽ 0.025; †p ⫽ 0.01.
TABLE IV Fetal Outcome Group 1—None or Trivial Associated Defects (n ⫽ 13 pregnancies) Term delivery Premature delivery Miscarriage Termination Cardiac defect in live offspring
10 (77%) 0 0 3 (23%) 0
Group 2—VSD and POTO (n ⫽ 32 pregnancies) 12 5 12 3 1
(37.5%) (16%) (37.5%) (9%) (3%)
Abbreviations as in Table I.
ing pregnancy, blood volume increases by 50%, reaching its peak around the 32nd week. At delivery, there is a transient increase in cardiac output, which persists for days to weeks. Blood volume returns to pregravidad level only 4 to 6 weeks postpartum.10 Congestive heart failure from systemic right ventricular dysfunction and/or worsening left atrioventricular valve regurgitation occurred in the later stage of pregnancy, at the time when blood volume reached its peak. Isolated cases of worsening left atrioventricular valve regurgitation and systemic right ventricular failOCTOBER 1, 1999
TABLE V Risk Factors for Miscarriage
Associated defects Group 1 Group 2 Mean maternal age (yr) (range) Functional class I II Cyanosis Surgery Complete heart block without pacemaker
Live Births (n ⫽ 27)
Miscarriages (n ⫽ 12)
10 (37%) 17 (59%) 27 (20–40)
0 (0%) 12 (100%) 28 (21–36)
23 4 3 4 3
3 9 10 5 0
(85%) (15%) (11%) (15%) (11%)
(25%)* (75%) (83%)† (42%)
*p ⫽ 0.001; †p ⬍0.001.
ure during late pregnancy in 2 patients with CCTGA have been reported previously.7,8 A preliminary report of 17 pregnant patients with CCTGA9 found late maternal deterioration from worsening left atrioventricular valve regurgitation in 1 patient (6%). Similarly, systemic right ventricular failure during pregnancy is reported in about 13% of patients with complete transposition after the Mustard procedure.11–16 Our study is the first to report on maternal complications of CCTGA patients and confirms a similar incidence of congestive heart failure from systemic right ventricular failure and/or worsening left atrioventricular valve regurgitation during pregnancy. The reason why pregnancy with its volume load precipitates systemic right ventricular and tricuspid valve dysfunction in some patients with CCTGA, and not others, remains to be established. On univariate analysis, older age seemed to predispose to functional class deterioration during pregnancy. This is of interest given the natural history of progressive right ventricular dysfunction and left atrioventricular valve regurgitation seen in patients with CCTGA over time.3– 6 Likewise, univariate analysis suggested that the presence of untreated complete heart block, an associated lesion with a high preva-lence in CCTGA, could predispose to symptomatic deterioration. The increase in cardiac output during pregnancy initially results from an increase in stroke volume, but later on, relies mainly on an increase in heart rate to sustain itself.10 Patients with complete heart block may have to rely exclusively on an increase in stroke volume to maintain an adequate cardiac output throughout the pregnancy, putting an additional burden on their systemic right ventricle. However, our small sample size did not permit us to reach any definitive conclusion, and these hypotheses remain to be verified by a larger study. Other complications related to associated lesions, namely ventricular septal defect and pulmonary stenosis, were present in 2 patients. Worsening cyanosis from an increased right to left shunting during pregnancy can be expected in such patients. This is because systemic vascular resistance decreases during pregnancy proportionally more than pulmonary vas-
cular resistance,10 resulting in an increased right to left shunting through the ventricular septal defect. Similarly, neurologic events can occur in these patients from microcytic erythrocytosis if cyanosed,17 or from paradoxical emboli when a deep venous thrombosis is present.18 Avoidance of prolonged bed rest, prophylactic elastic stockings, and subcutaneous heparin can be used to prevent deep venous thrombosis in such patients. Fetal outcome: In previous studies, the presence of cyanosis has been established as a risk factor for premature delivery, low birth weight, and spontaneous abortion.10,19 The incidence of spontaneous abortion increases proportionally with maternal hypoxemia. An oxygen saturation of ⱕ85% results in fetal hypoxemia threatening its survival and may also result in cardiac malformation leading to fetal death.20 Our findings are in accord with previous studies and emphasize the negative impact of cyanosis on fetal outcome. The incidence of congenital heart disease in live offspring of our study was low— 4%. Maternal transmission of congenital heart disease varies between 2% and 18%10,21 in published reports depending on the type of maternal cardiac anomaly. The precise transmission rate for women with CCTGA is not known. Thus, in this group, our finding falls in the lower end of the generally accepted range of transmission and may have been underestimated given that the prevalence of congenital heart disease in the terminated pregnancies was unknown. Study limitations: Because of the retrospective nature of the study (except for 1 patient), it was not possible to assess serially the hemodynamic effect of pregnancy on systemic right ventricular function and its left atrioventricular valve. This would be better addressed by a prospective study with serial echocardiographic studies. Also, because of the small sample size of our study due to the rare condition of this disease, risk factors for maternal complications could not be elucidated. Acknowledgment: We gratefully acknowledge the expert statistical help provided by Jessica Harris and Paul Cullinan.
1. Allwork SP, Bentall UH, Becker AE, Cameron H, Gerlis LM, Wilkonson JL,
Anderson RH. Congenitally corrected transposition of the great arteries: morphologic study of 32 cases. Am J Cardiol 1976;38:911–920. 2. Lundstrom U, Bull C, Wyse RKH, Somerville J.The natural and unatural history of congenitally corrected transposition. Am J Cardiol 1990;65:1222–1229. 3. Connelly MS, Liu PP, Williams WG, Webb GD, Robertson P, McLaughlin PR. Congenitally corrected transposition of the great arteries in the adults: functional status and complications. J Am Coll Cardiol 1996;27:1238 –1243. 4. Presbitero P, Somerville J, Rabajoli P, Stone S, Conte MR. Corrected transposition of the great arteries without associated defects in the adult patients: clinical profile and follow up. Br Heart J 1995;74:57–59. 5. Sano T, Riesenfeld T, Karl TR, Wilkinson JL. Intermediate term outcome after intracardiac repair of associated cardiac defects in patients with atrioventricular and ventriculoarterial discordance (abstr). Circulation 1995;92(suppl II):II-272–II-278. 6. Graham TP, Parrish MD, Boucek RJ, Boerth RC, Breitweser JA, Thompson S, Robertson RM, Morgan JR, Friesinger GC. Assessment of ventricular size and function in congenitally corrected transposition of the great arteries. Am J Cardiol 1983;51:244 –251. 7. Haiderer O, Bahler R, Carson P, Kennedy JH. Congenital corrected transposition of the great vessels. J Thorac Cardiovasc Surg 1971;6:933–937.
CONGENITAL HEART DISEASE/CONGENITALLY CORRECTED TRANSPOSITION OF GREAT ARTERIES
823
8. Lam D, Melvin D, Cheitlin MD, Pooper RW, Szarnicki RJ. Rapid development
of biventricular heart failure in corrected transposition of the great arteries during pregnancy. Am Heart J 1988;116:1111–1115. 9. Conolly HM, Warnes CA. Outcome of pregnancy in women with congenitally corrected transposition of the great arteries (abstr). Circulation 1995;92(suppl I):I–517. 10. Perloff JK. Pregnancy in congenital heart disease. In: Zorab R, ed. Congenital Heart disease in Adults. Philadelphia: WB Saunders, 1991:124 –140. 11. Rousseil MP, Irion O, Beguin F, Jacques O, Adamec R, Lerch R, Friedli B, Rifat K.. Successful term pregnancy after Mustard operation for transposition of the great arteries. Eur J Obstet Gynecol Reprod Biol 1995;59 :111–113. 12. Dellinger EH, Hadi HA. Maternal transposition of the great arteries in pregnancy. A case report. J Repro Med 1994;39:324 –326. 13. Clarkson PM, Wilson NJ, Neutze JM, North RA, Calder AL, Barratt-Boyes BG. Outcome of pregnancy after the Mustard operation for transposition of the great arteries with intact ventricular septum. J Am Coll Cardiol 1994;24:190 –193. 14. Lao TT, Sermer M, Colman JM. Pregnancy following surgical correction for transposition of the great arteries. Obstet Gynecol 1994;83:665– 668.
824 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 84
15. Megerian G, Bell JG, Hunta JC, Bottalico JN, Weiner S. Pregnancy outcome following Mustard procedure for transposition of the great arteries: a report of five cases and review of the literature. Obstet Gynecol 1994;83: 512–516. 16. Lynch-Salomon DI, Maze SS, Combs CA. Pregnancy after Mustard repair for transposition of the great arteries. Obstet Gynecol 1993;82:676 – 679. 17. Ammash N, Warnes CA. Cerebrovascular events in adult patients with cyanotic congenital heart disease. J Am Coll Cardiol 1996;28:768 –772. 18. Goldhaber SZ. Venous thrombosis: prevention, treatment, and relationship to paradoxical embolization. Cardiol Clin 1994;12:505–516. 19. Presbitero P, Somerville J, Stone S, Aruta E, Spiegelhalter D, Rabajoli F. Pregnancy in cyanotic congenital heart disease: outcome of mother and fetus. Circulation 1994;89:2673–2677 20. Somerville J. Congenital heart disease in adolescents and adults. In: Weatherall DJ, Ledingham JGG, Warrell DA, eds. Oxford Textbook of Medicine. London: Oxford Medical Publications, 1996:2398 –2431. 21. Bitsch M, Johansen C, Wennerold A, Osler M. Eisenmenger’s syndrome and pregnancy. Eur J Obstet Gynecol Reprod Bio 1988;28:69.
OCTOBER 1, 1999
MD,
Isobel Barnes,
BSC, MSC,
and Jane Somerville,
MD
To assess maternal and fetal outcome of pregnancy in patients with congenitally corrected transposition of the great arteries, we reviewed 19 patients (18 retrospectively) who had 45 pregnancies. Their ages ranged from 18 to 40 years (mean 27) at the time of pregnancy. Thirty-six percent of the pregnancies were undertaken while patients were cyanosed, 7% in patients with unpaced complete heart block, and 16% were undertaken after surgical repair of the associated anomalies. Change in functional class and maternal cardiovascular complications during pregnancy were analyzed as well as number of live births, miscarriages, elective termination of pregnancies, timing of delivery, and incidence of
cardiac defects in the live offspring. Five patients (26% of patients) developed cardiovascular complications during pregnancy, namely congestive heart failure (3 patients), worsening cyanosis (1 patient), and cerebrovascular accident (1 patient). No maternal deaths occurred. There were 27 live births (60%), 12 miscarriages (27%), and 6 elective terminations of pregnancy (13%). Cyanosis was a significant risk factor for miscarriage. One live offspring had congenital heart disease. Close follow-up of these patients during pregnancy, by a team of experienced physicians, is recommended. 䊚1999 by Excerpta Medica, Inc. (Am J Cardiol 1999;84:820 – 824)
ongenitally corrected transposition of the great arteries (CCTGA) is a combination of atriovenC tricular and ventriculoarterial discordance. The ana-
CCTGA on the database. Twenty had conceived between 1954 and 1998, with a total of 46 pregnancies. Twelve patients were attending the outpatient service at the time of their pregnancy; 8 patients were referred after pregnancies had been completed. One patient was followed prospectively. One patient was excluded because of concomitant Eisenmenger syndrome. Medical records were reviewed in all patients and additional investigations such as echocardiograms were reviewed when available. The cardiac diagnosis of atrioventricular and ventriculoarterial discordance and associated anomalies was established by cardiac catheterization, echocardiography, or at surgery. The patient population was divided into 2 groups according to their basic cardiac anatomy: group 1, CCTGA with no or minor associated anomalies (small atrial septal defect, mild pulmonary stenosis, or mild isolated structural left atrioventricular valve anomaly); group 2, CCTGA with ventricular septal defect and pulmonary outflow tract obstruction (valvar, subvalvar, or pulmonary atresia). Maternal outcome: Recorded cardiovascular complications and maternal mortality occurring during or after pregnancy (within 1 year) were noted. Change in patients’ functional class during and after (up to 1 year) pregnancy was also analyzed as an outcome variable. Fetal outcome: The number of live births with timing of delivery, miscarriages, and elective termination were studied, as well as the incidence of cardiac defects in live offspring. Miscarriages were defined as spontaneous abortion at or before 20 weeks of gestation and prematurity as delivery at or before 37 weeks of gestation. Analysis of data: The effect of associated cardiac lesions, maternal age, baseline New York Heart Association (NYHA) class before pregnancy, cyanosis
1
tomic left ventricle becomes the pulmonary ventricle, whereas the anatomic right ventricle, guarded by a tricuspid valve, is the systemic ventricle. Usually, there are associated defects such as ventricular septal defect, pulmonary stenosis (at valvar or subvalvar level), or anatomically abnormal left-sided (tricuspid) atrioventricular valve with varying degrees of regurgitation. Complete heart block may be congenital or acquired. Symptoms depend on the severity of the associated lesions and the function of the systemic right ventricle.2 There are ongoing concerns about the long-term ability of the morphologic right ventricle and tricuspid valve to support systemic pressure3– 6 and their capacity to handle the volume load of pregnancy.3,7,8 The paucity of information about the outcome of pregnancy in adult patients with CCTGA9 led us to study the outcome of pregnancy and its cardiovascular effects in women with CCTGA.
METHODS
Selection of patients: The database of the Grown Up Congenital Heart Unit, a designated quaternary center for adolescents and adults with congenital heart disease, was searched retrospectively for adult female patients aged ⱖ16 years with confirmed diagnosis of CCTGA, with or without associated lesions, who had become pregnant. There were 39 female patients with From the Royal Brompton Hospital, National Heart and Lung Institute and Imperial College School of Medicine, London, United Kingdom. Manuscript received January 14, 1999; revised manuscript received and accepted May 12, 1999. Address for reprints: Judith Therrien, MD, The Toronto Hospital, 200 Elizabeth Street, 12 EN Room 213, Toronto, Ontario, Canada. E Mail: [email protected].
820
©1999 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 84 October 1, 1999
0002-9149/99/$–see front matter PII S0002-9149(99)00444-0
(defined as clinical central cyanosis or an oxygen saturation of ⱕ85%), surgery (palliative or reparative), and complete heart block on fetal and maternal outcome were studied. All analyses were performed using a random-effects logistic regression procedure using the statistical package EGRET (SERC, Seattle, Washington), version 1.00.03. A p value ⬍0.05 was considered statistically significant.
RESULTS
Patient characteristics: GROUP 1: Six patients had 13
pregnancies at a mean age of 27 years. Five patients (with 10 pregnancies) were in NYHA class I before pregnancy and 1 patient was in class I before her first 2 pregnancies and in class II before her third. All patients were acyanotic and 1 patient (with 3 pregnancies) had a bioprosthetic left atrioventricular valve replacement for left atrioventricular valve regurgitation before her third pregnancy. Two patients (with 4 pregnancies) had complete heart block of which 1 (with 2 pregnancies) was treated with a dual-chamber pacemaker before pregnancy. No patient was taking medication before pregnancy. GROUP 2: Thirteen patients had 32 pregnancies at a mean age of 27 years. Eight patients (with 16 pregnancies) were in NYHA class I before pregnancy and 5 (with 16 pregnancies) were in class II before pregnancy. Four patients (with 13 pregnancies) were cyanotic before each pregnancy and 1 (with 4 pregnancies) was cyanotic before her first 3 pregnancies. Two patients (4 pregnancies) who had had cyanosis underwent previous palliative surgery (modified right and left Blalock Taussig shunt in 1 patient and Waterston shunt in 1 patient). Eight patients (with 16 pregnancies) were acyanotic, 5 (with 6 pregnancies) having had intracardiac repair before pregnancy (ventricular septal defect patch closure and left ventricular to pulmonary artery conduit in 2 patients, ventricular septal defect patch closure, pulmonary valvotomy, and left atrioventricular valve replacement in 2 patients, and Fontan procedure in 1 patient). Five patients (with 6 pregnancies) had associated complete heart block; 4 (with 5 pregnancies) were treated with endovenous pacemaker before pregnancy. Two patients were taking diuretics and 3 were taking angiotensin-converting enzyme inhibitors before pregnancy (Table I). Maternal outcome: Cardiovascular complications occurred in 5 patients (26% of patients) (group 1, 1 patient; group 2, 4 patients) with 5 completed pregnancies (11% of the pregnancies) at the later stage of pregnancy. Details of the patients with cardiovascular complications during pregnancy are summarized in Table II. Congestive heart failure was the cause of deterioration in 3 patients from worsening systemic right ventricular dysfunction (patients 1 and 2), worsening left atrioventricular valve regurgitation (patient 2), and atrial fibrillation from a stenotic left atrioventricular prosthetis (patient 3). Worsening cyanosis occurred in 1 patient (no. 4) and cerebrovascular accident in 1 patient (no. 5). Deterioration in NYHA functional class occurred in the first 4 patients (21%) (Figure 1). No cardiovascular complications occurred
TABLE I Patient Characteristics Before Pregnancy Group 1 No or Minor Associated Defects (n ⫽ 13 pregnancies)
Group 2 VSD and POTO (n ⫽ 32 pregnancies)
27 (19–40)
27 (18–39)
12 (92%) 1 (8%) 0
16 (50%) 16 (50%) 16 (50%)
Mean age (yr) at pregnancy (range) NYHA I II Cyanosis Surgery Palliative Repair Complete heart block With pacemaker Without pacemaker
0 1 (8%)
4 (12%) 6 (18%)
2 (16%)
5 (15%)
2 (16%)
1 (3%)
POTO ⫽ pulmonary outflow tract obstruction; VSD ⫽ ventricular septal defect.
with spontaneous miscarriage or elective termination of pregnancy. There were no maternal deaths. Univariate analysis showed that older maternal age (p ⫽ 0.025) and the presence of untreated complete heart block (p ⫽ 0.01) were significant risk factors for worsening functional class during pregnancy. However, on multivariate analysis, neither retained their significance (Table III). Fetal outcome: There were 27 live births (60%) (group 1, 10; group 2, 17). Twenty-two (49%) were born at term and 5 (11%) prematurely at 31 to 35 weeks of gestation (group 2, 5). Twelve spontaneous miscarriages (27%) occurred at 10 to 20 weeks of gestation (group 2, 12), and 6 elective termination of pregnancies (13%) were performed at the request of the patient. One live offspring (4%) (group 2) had congenital heart disease— hypoplastic left heart syndrome (Table IV). Potential risk factors for miscarriage were examined with results shown in Table V. Pregnancies that ended in elective termination were not included in the analysis and, owing to sparse data, the statistical significance of associate defects and complete heart block could not be determined. NYHA class II (p ⬍0.001) and cyanosis before pregnancy (p ⬍0.001) were significant risk factors for miscarriage. Multivariate analysis showed that once cyanosis was taken into account, no other risk factor added significantly to the prediction of miscarriage (Table V).
DISCUSSION
Maternal outcome: In this study, 74% of patients had an uneventful pregnancy. Cardiovascular complications occurred in 5 patients (26% of the patients) with 5 pregnancies (11% of the pregnancies). Complications intrinsic to CCTGA, namely the development of congestive heart failure and systemic tricuspid valve regurgitation occurred in 2 patients (10%). Dur-
CONGENITAL HEART DISEASE/CONGENITALLY CORRECTED TRANSPOSITION OF GREAT ARTERIES
821
— — 0 I
32
29
4 (2nd)
5 (4th)
VSD/PS (2)
0
⫹ II
26 3 (1st)
CHB ⫽ complete heart block; CHF ⫽ congestive heart failure; CVA ⫽ cerebrovascular accident; LAVV ⫽ left atrioventricular valve; LAVVR ⫽ left atrioventricular valve regurgitation; LV ⫽ left ventricular; PA ⫽ pulmonary artery; pm ⫽ pacemaker; PS ⫽ pulmonary stenosis; PV ⫽ pulmonary valve; RV ⫽ right ventricular; VSD ⫽ ventricular septal defect; — ⫽ no information available.
I — —
II —
29th wk gestation 12th wk gestation — —
0 II
39 2 (4th)
VSD/PS Ebstein LAVV (2) VSD/PS (2)
0 II
VSD closure LV to PA conduit VSD closure PV valvotomy LAVVR 0
⫹ (0) 0
0 0
⫹ ⫹⫹
Worsening cyanosis CVA
—
II 0 0 CHF
III ⫹⫹ ⫹⫹⫹ CHF
I ⫹⫹⫹ ⫹⫹ ⫹⫹
⫹ (0) ⫹ (⫹) 0 0 0 I 40 1 (2nd)
Mother # (pregnancy #)
Isolated CCTGA (1) VSD/PS (2)
Cyanosis Diagnosis (group)
Surgery Before Pregnancy
CHB (pm)
⫹⫹
48 hours after delivery 29th wk gestation 36th wk gestation
CHF
NYHA at 1 Year Postpartum RV Dysfunction at Time of Deterioration (0–3⫹) LAVVR at Time of Deterioration (0–3⫹) Cause of Maternal Deterioration Timing of Maternal Deterioration RV Dysfunction Before Pregnancy (0–3⫹) LAVVR Before Pregnancy (0–3⫹) NYHA Before Pregnancy Age at Pregnancy (yrs)
TABLE II Maternal Complications
822 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 84
FIGURE 1. Change in NYHA functional class before (<3 months), during, and after (<1 year) pregnancy. N ⴝ number of pregnancies; I ⴝ NYHA class 1; II ⴝ NYHA class II; III ⴝ NYHA class III; IV ⴝ NYHA class IV.
TABLE III Risk Factors for Maternal Deterioration During Pregnancy
Associated defects Group 1 Group 2 Mean maternal age (yr) (range) Functional class I II Cyanosis Surgery Complete heart block without pacemaker
Same Functional Class During Pregnancy (n ⫽ 41)
Worsening Functional Class During Pregnancy (n ⫽ 4)
12 (29%) 29 (71%) 26 (18–38)
1 (25%) 3 (75%) 34 (26–40)*
27 15 15 9 1
(66%) (34%) (37%) (22%) (2%)
1 3 1 2 2
(25%) (75%) (25%) (50%) (50%)†
*p ⫽ 0.025; †p ⫽ 0.01.
TABLE IV Fetal Outcome Group 1—None or Trivial Associated Defects (n ⫽ 13 pregnancies) Term delivery Premature delivery Miscarriage Termination Cardiac defect in live offspring
10 (77%) 0 0 3 (23%) 0
Group 2—VSD and POTO (n ⫽ 32 pregnancies) 12 5 12 3 1
(37.5%) (16%) (37.5%) (9%) (3%)
Abbreviations as in Table I.
ing pregnancy, blood volume increases by 50%, reaching its peak around the 32nd week. At delivery, there is a transient increase in cardiac output, which persists for days to weeks. Blood volume returns to pregravidad level only 4 to 6 weeks postpartum.10 Congestive heart failure from systemic right ventricular dysfunction and/or worsening left atrioventricular valve regurgitation occurred in the later stage of pregnancy, at the time when blood volume reached its peak. Isolated cases of worsening left atrioventricular valve regurgitation and systemic right ventricular failOCTOBER 1, 1999
TABLE V Risk Factors for Miscarriage
Associated defects Group 1 Group 2 Mean maternal age (yr) (range) Functional class I II Cyanosis Surgery Complete heart block without pacemaker
Live Births (n ⫽ 27)
Miscarriages (n ⫽ 12)
10 (37%) 17 (59%) 27 (20–40)
0 (0%) 12 (100%) 28 (21–36)
23 4 3 4 3
3 9 10 5 0
(85%) (15%) (11%) (15%) (11%)
(25%)* (75%) (83%)† (42%)
*p ⫽ 0.001; †p ⬍0.001.
ure during late pregnancy in 2 patients with CCTGA have been reported previously.7,8 A preliminary report of 17 pregnant patients with CCTGA9 found late maternal deterioration from worsening left atrioventricular valve regurgitation in 1 patient (6%). Similarly, systemic right ventricular failure during pregnancy is reported in about 13% of patients with complete transposition after the Mustard procedure.11–16 Our study is the first to report on maternal complications of CCTGA patients and confirms a similar incidence of congestive heart failure from systemic right ventricular failure and/or worsening left atrioventricular valve regurgitation during pregnancy. The reason why pregnancy with its volume load precipitates systemic right ventricular and tricuspid valve dysfunction in some patients with CCTGA, and not others, remains to be established. On univariate analysis, older age seemed to predispose to functional class deterioration during pregnancy. This is of interest given the natural history of progressive right ventricular dysfunction and left atrioventricular valve regurgitation seen in patients with CCTGA over time.3– 6 Likewise, univariate analysis suggested that the presence of untreated complete heart block, an associated lesion with a high preva-lence in CCTGA, could predispose to symptomatic deterioration. The increase in cardiac output during pregnancy initially results from an increase in stroke volume, but later on, relies mainly on an increase in heart rate to sustain itself.10 Patients with complete heart block may have to rely exclusively on an increase in stroke volume to maintain an adequate cardiac output throughout the pregnancy, putting an additional burden on their systemic right ventricle. However, our small sample size did not permit us to reach any definitive conclusion, and these hypotheses remain to be verified by a larger study. Other complications related to associated lesions, namely ventricular septal defect and pulmonary stenosis, were present in 2 patients. Worsening cyanosis from an increased right to left shunting during pregnancy can be expected in such patients. This is because systemic vascular resistance decreases during pregnancy proportionally more than pulmonary vas-
cular resistance,10 resulting in an increased right to left shunting through the ventricular septal defect. Similarly, neurologic events can occur in these patients from microcytic erythrocytosis if cyanosed,17 or from paradoxical emboli when a deep venous thrombosis is present.18 Avoidance of prolonged bed rest, prophylactic elastic stockings, and subcutaneous heparin can be used to prevent deep venous thrombosis in such patients. Fetal outcome: In previous studies, the presence of cyanosis has been established as a risk factor for premature delivery, low birth weight, and spontaneous abortion.10,19 The incidence of spontaneous abortion increases proportionally with maternal hypoxemia. An oxygen saturation of ⱕ85% results in fetal hypoxemia threatening its survival and may also result in cardiac malformation leading to fetal death.20 Our findings are in accord with previous studies and emphasize the negative impact of cyanosis on fetal outcome. The incidence of congenital heart disease in live offspring of our study was low— 4%. Maternal transmission of congenital heart disease varies between 2% and 18%10,21 in published reports depending on the type of maternal cardiac anomaly. The precise transmission rate for women with CCTGA is not known. Thus, in this group, our finding falls in the lower end of the generally accepted range of transmission and may have been underestimated given that the prevalence of congenital heart disease in the terminated pregnancies was unknown. Study limitations: Because of the retrospective nature of the study (except for 1 patient), it was not possible to assess serially the hemodynamic effect of pregnancy on systemic right ventricular function and its left atrioventricular valve. This would be better addressed by a prospective study with serial echocardiographic studies. Also, because of the small sample size of our study due to the rare condition of this disease, risk factors for maternal complications could not be elucidated. Acknowledgment: We gratefully acknowledge the expert statistical help provided by Jessica Harris and Paul Cullinan.
1. Allwork SP, Bentall UH, Becker AE, Cameron H, Gerlis LM, Wilkonson JL,
Anderson RH. Congenitally corrected transposition of the great arteries: morphologic study of 32 cases. Am J Cardiol 1976;38:911–920. 2. Lundstrom U, Bull C, Wyse RKH, Somerville J.The natural and unatural history of congenitally corrected transposition. Am J Cardiol 1990;65:1222–1229. 3. Connelly MS, Liu PP, Williams WG, Webb GD, Robertson P, McLaughlin PR. Congenitally corrected transposition of the great arteries in the adults: functional status and complications. J Am Coll Cardiol 1996;27:1238 –1243. 4. Presbitero P, Somerville J, Rabajoli P, Stone S, Conte MR. Corrected transposition of the great arteries without associated defects in the adult patients: clinical profile and follow up. Br Heart J 1995;74:57–59. 5. Sano T, Riesenfeld T, Karl TR, Wilkinson JL. Intermediate term outcome after intracardiac repair of associated cardiac defects in patients with atrioventricular and ventriculoarterial discordance (abstr). Circulation 1995;92(suppl II):II-272–II-278. 6. Graham TP, Parrish MD, Boucek RJ, Boerth RC, Breitweser JA, Thompson S, Robertson RM, Morgan JR, Friesinger GC. Assessment of ventricular size and function in congenitally corrected transposition of the great arteries. Am J Cardiol 1983;51:244 –251. 7. Haiderer O, Bahler R, Carson P, Kennedy JH. Congenital corrected transposition of the great vessels. J Thorac Cardiovasc Surg 1971;6:933–937.
CONGENITAL HEART DISEASE/CONGENITALLY CORRECTED TRANSPOSITION OF GREAT ARTERIES
823
8. Lam D, Melvin D, Cheitlin MD, Pooper RW, Szarnicki RJ. Rapid development
of biventricular heart failure in corrected transposition of the great arteries during pregnancy. Am Heart J 1988;116:1111–1115. 9. Conolly HM, Warnes CA. Outcome of pregnancy in women with congenitally corrected transposition of the great arteries (abstr). Circulation 1995;92(suppl I):I–517. 10. Perloff JK. Pregnancy in congenital heart disease. In: Zorab R, ed. Congenital Heart disease in Adults. Philadelphia: WB Saunders, 1991:124 –140. 11. Rousseil MP, Irion O, Beguin F, Jacques O, Adamec R, Lerch R, Friedli B, Rifat K.. Successful term pregnancy after Mustard operation for transposition of the great arteries. Eur J Obstet Gynecol Reprod Biol 1995;59 :111–113. 12. Dellinger EH, Hadi HA. Maternal transposition of the great arteries in pregnancy. A case report. J Repro Med 1994;39:324 –326. 13. Clarkson PM, Wilson NJ, Neutze JM, North RA, Calder AL, Barratt-Boyes BG. Outcome of pregnancy after the Mustard operation for transposition of the great arteries with intact ventricular septum. J Am Coll Cardiol 1994;24:190 –193. 14. Lao TT, Sermer M, Colman JM. Pregnancy following surgical correction for transposition of the great arteries. Obstet Gynecol 1994;83:665– 668.
824 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 84
15. Megerian G, Bell JG, Hunta JC, Bottalico JN, Weiner S. Pregnancy outcome following Mustard procedure for transposition of the great arteries: a report of five cases and review of the literature. Obstet Gynecol 1994;83: 512–516. 16. Lynch-Salomon DI, Maze SS, Combs CA. Pregnancy after Mustard repair for transposition of the great arteries. Obstet Gynecol 1993;82:676 – 679. 17. Ammash N, Warnes CA. Cerebrovascular events in adult patients with cyanotic congenital heart disease. J Am Coll Cardiol 1996;28:768 –772. 18. Goldhaber SZ. Venous thrombosis: prevention, treatment, and relationship to paradoxical embolization. Cardiol Clin 1994;12:505–516. 19. Presbitero P, Somerville J, Stone S, Aruta E, Spiegelhalter D, Rabajoli F. Pregnancy in cyanotic congenital heart disease: outcome of mother and fetus. Circulation 1994;89:2673–2677 20. Somerville J. Congenital heart disease in adolescents and adults. In: Weatherall DJ, Ledingham JGG, Warrell DA, eds. Oxford Textbook of Medicine. London: Oxford Medical Publications, 1996:2398 –2431. 21. Bitsch M, Johansen C, Wennerold A, Osler M. Eisenmenger’s syndrome and pregnancy. Eur J Obstet Gynecol Reprod Bio 1988;28:69.
OCTOBER 1, 1999