- Email: [email protected]
Thyrotoxicosis and heart failure that complicate pregnancy
American Journal of Obstetrics and Gynecology (2004) 190, 211e7
www.elsevier.com/locate/ajog
Thyrotoxicosis and heart failure that complicate pregnancy Jeanne S. Sheffield, MD,* F. Gary Cunningham, MD Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center at Dallas, Dallas, Tex Received June 17, 2003; revised July 21, 2003; accepted July 25, 2003
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Pregnancy Graves’ disease Heart failure
Objective: When untreated, Graves’ thyrotoxicosis has profound cardiovascular effects, although it rarely causes heart failure in otherwise healthy patients. Preliminary observations suggest that pregnant women are the exception. To further elucidate this association, we studied both immediate and long-term outcomes in women who had thyrotoxicosis and heart failure during pregnancy. Study design: We reviewed clinical outcomes of pregnant women with Graves’ disease and heart failure at our institution from 1974 through 2001. Women with other underlying heart disease were excluded. A standardized antithyroid regimen and serial echocardiography and/or chest radiography were performed. Results: The 13 women with thyrotoxicosis and heart failure were either noncompliant with antithyroid therapy or had no medical care during pregnancy. Six women had heart failure before fetal viability; decompensation was precipitated by hemorrhage, sepsis, or both. The other 7 women were in the last trimester when heart failure developed; in 4 women, the heart failure was precipitated by severe preeclampsia-eclampsia and in 2 women was precipitated by sepsis. Overall, 11 of 13 women had an underlying obstetric event. In follow-up of 11 women from 2 to 25 years, resolution of cardiomyopathy was confirmed after successful treatment of thyrotoxicosis. Conclusion: Normal pregnancy mimics and amplifies some of the hyperdynamic cardiovascular changes that are caused by thyrotoxicosis. When they occur simultaneously, there is usually a compensated high-output state. In some women, however, common pregnancy complications that include hemorrhage with associated anemia, sepsis, and severe preeclampsia-eclampsia will precipitate heart failure. The immediate treatment of heart failure and the correction of precipitating pregnancy factors usually results in good outcome. Long-term follow-up confirmed that thyrotoxic cardiac dysfunction is reversible with successful antithyroid therapy. Ó 2004 Elsevier Inc. All rights reserved.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
* Reprint requests: Jeanne S. Sheffield, MD, Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390-9032. E-mail: Jeanne.Sheffi[email protected] 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/S0002-9378(03)00944-X
There are a number of profound effects of excessive thyroid hormones on cardiovascular function.1 Contractility is directly increased by tri-iodothyronineeresponsive cardiac genes that augment cardiac protein synthesis. Contractility is increased indirectly by diminished systemic vascular resistance with the resultant
212 increased heart rate and cardiac output. Without intervention, these two mechanisms result in increased left ventricular mass and a high-output state. In this case, any further cardiac demands (such as from exercise) cause significant but reversible left ventricular dysfunction.2-5 Thus, this basal high-output state, although compensated, may result in heart failure when there is increased cardiac burden. Despite this, nonpregnant patients with thyrotoxicosis seldom develop overt heart failure.1,6 By contrast, almost 10% of women with untreated thyrotoxicosis have heart failure during pregnancy.7,8 The propensity for pregnant women to have heart failure has been explained by the combined effects of thyroxine-induced cardiovascular changes along with the very similar pregnancy-induced hyperdynamic changes.9,10 Even then, heart failure does not usually develop unless precipitated by factors that further magnify cardiac workload.9 These include severe preeclampsiaeclampsia, sepsis, hemorrhage, and anemia.11,12 Our purposes now are to extend our previous observations of pregnant women with hyperthyroidism and with cardiac failure9 and to study its long-term consequences.
Material and methods The pregnant women who are described in this report had uncontrolled thyrotoxicosis that was further complicated by congestive heart failure. They were admitted to the obstetrics service at Parkland Hospital from 1974 through 2001. The diagnosis of Graves’ disease was made clinically with laboratory confirmation with the standard criteria.9,10,13 All of the women had a diffuse toxic goiter and other variable findings that included peripheral tremor, resting tachycardia, weight loss, palpitations, weakness, hyperreflexia, nervousness, or irritability and Graves’ ophthalmopathy and dermopathy. Over the study period, laboratory tests for thyroid disease continually evolved. Accordingly, total serum thyroxine and tri-iodothyronine, T3-resin uptake ratio, and free thyroxine index were measured until they were replaced by contemporaneous assays for serum thyroid-stimulating hormone and free thyroxine. Heart failure was diagnosed with the use of the Framingham criteria and confirmed by chest radiography that showed cardiomegaly and pulmonary edema.14 Echocardiography was not available widely during the early years of the study. When available, however, cardiac structure and function were assessed with echocardiograph, or, in some earlier cases, by radionuclide scanning ventriculography. Women were excluded if they had underlying cardiac abnormalities. At our institution, clinical treatment of pregnant women with thyrotoxicosis and heart failure is similar to that for thyroid storm.15 Briefly, they are admitted to the obstetrics intensive care unit, and furosemide is given intravenously for pulmonary edema. Treatment
Sheffield and Cunningham of thyrotoxicosis included a 1000-mg oral loading dose of propylthiouracil followed in 1 hour by 500 to 1000 mg of potassium iodide, which was given either intravenously or orally. Modified doses of both drugs were then continued at 8-hour intervals. They also were given three doses of intravenous hydrocortisone or methylprednisolone. Early in the study, some women were given digoxin, and later b-blockade was used for extreme tachycardia. Vigorous treatment was given for obstetrics complications that precipitate or worsen heart failure. Severe preeclampsia or eclampsia was treated with magnesium sulfate to prevent or stop eclamptic seizures and with intravenous hydralazine to control severe hypertension and afterload failure.16 Severe anemia was corrected with red blood cell transfusions. Infection was treated aggressively with intravenous antimicrobial therapy, and uterine curettage was performed for women with a septic abortion. All of these women were followed by at least one of the authors to ensure homogeneity of treatment. Data retrieval was begun at the sentinel admission, and long-term follow-up was provided by record review. Data retrieval and analysis were approved by the Institutional Review Board of the University of Texas Southwestern Medical Center.
Results During the 28-year study period, slightly over 300,000 women were delivered at our institution. The recognized incidence of overt thyrotoxicosis averaged approximately 1 in 1500 pregnancies, as determined for three epochs: 1:2000 for approximately 120,000 women who were delivered from 1974 through 19859; 1:1500 for approximately 125,000 women who were delivered from 1991 through 1999; and 1:1000 for 11,000 women who underwent serum thyrotropin screening from 2000 through 2001. From all of these, we identified 150 women who were overtly thyrotoxic; in 13 women (9%), pregnancy was further complicated by heart failure. At diagnosis, in 10 women the total thyroxine level ranged from 19 to O40 mg/dL. The other 3 women had free thyroxine values that ranged from 4.2 to 6.1 ng/dL. Clinical information for the 13 women with heart failure is listed in Table I. Six of these 13 women had heart failure before the time of expected fetal viability. In 3 of these women, hemorrhage and sepsis with incomplete abortion precipitated heart failure. In 2 other women, septicemia from pyelonephritis triggered heart failure. Four of the 6 women had anemia with a hematocrit level of !25%. The remaining woman had no precipitating factors that were identified. The other seven women who are mentioned in Table I were in the third trimester when heart failure developed. Four of these seven women had attended prenatal
213
Sheffield and Cunningham Table I
Clinical and laboratory data from 13 Pregnant Women with thyrotoxicosis and heart failure
Associated Demographic precipitating event 1. 32 G2P1 None except pregnancy; (18 wk) mild anemia (hematocrit = 25%)
2. 26 G3P2 Molar pregnancy; (20 wk) hemorrhage; anemia (hematocrit 22%)
3. 20 G2P1 Incomplete abortion; (20 wk) hemorrhage; anemia (hematocrit 23%) 4. 27 G3P2 Septic abortion (T 40(C); (20 wk) hemorrhage; anemia (hematocrit 19%) 5. 21 G3P1 Pyelonephritis; sepsis (22 wk) (T 39.4(C); anemia (hematocrit 16%) 6. 20 G2P1 Pyelonephritis; sepsis (23 wk) (T 40(C); anemia (hematocrit 25%) 7. 22 G2P0 Twins: preterm PROM; (29 wk) chorioamnionitis; sepsis (T 39(C) 8. 28 G2P1 Severe preeclampsia (32 wk) (BP, 170/120 mm Hg) 9. 22 G4P3 Severe preeclampsia (34 wk) (BP, 220/140 mm Hg)
10. 34 G4P3 None except pregnancy; (35 wk) mild anemia (hematocrit = 28%) 11. 21 G1P0 Chorioamnionitis; (35 wk) sepsis (T 39(C); anemia (hematocrit 28%)
12. 30 G4P3 Severe preeclampsia; (36 wk) (BP, 200/150 mm Hg); anemia (hematocrit 20%) 13. 17 G1P0 Eclampsia (37 wk) (BP, 160/95 mm Hg)
Cardiac evaluation Initial
Follow Up
Clinical course
Long-term follow up
Four years, clinically E: Dilated right E: Four follow- Reversal of pulmonary euthyroid with normal hypertension and up studies atrium and cardiac function. ventricular (see Fig 2) ventricle; hypertrophy by 4 wk pulmonary after delivery, hypertension with PTU treatment (RVSP, 60 mm begun at 16 wk Hg) Chest radiograph clear with On T3 replacement therapy E: Dilated right E: 5 dd normal no residual cardiomegaly atrium and after 131I 2 y, clinically 8 d after delivery ventricle; 2+ euthyroid, normal mitral and third pregnancy tricuspid regurgitation None None Chest radiography returned Twenty-five years, chest to normal after delivery radiographs and clinical examinations normal. Chest radiograph returned Three years, euthyroid RS: Ejection RS: 25 dd to normal at 18 d taking PTU; normal fraction 0.35 ejection cardiac examination fraction 0.53 None None Ongoing PTU therapy; chest None radiograph returned to normal at 14 d None None Chest radiograph normal Twenty years, euthyroid at 2 mo after 131I at 10 y; normal cardiac examination Mildly depressed Echocardiogram normal Two years, normal chest left ventricular at 5 mo radiograph and cardiac dysfunction evaluation None None Chest radiograph returned None to normal at 4 d Chest radiograph returned Five years, normal chest None RS: 11 dd to normal at 2 d radiograph and cardiac ejection evaluation fraction, 0.59 None None Chest radiograph returned Three years, normal chest radiograph and cardiac to normal at 4 d and evaluation normal at 1 mo Echocardiogram normal Five years, took PTU RS: 6 dd E: Moderately at 3 mo 2 years, now euthyroid ejection depressed left with normal chest fraction, ventricular radiograph and cardiac 0.51 with function with evaluation mild globally abnormal depressed anterior wall function motion None None Chest radiograph returned Ten years, euthyroid after to normal at 2 wk PTU for 5 years; normal chest, radiograph and cardiac evaluation Two years, 131I at Chest radiography and E: 35 dd E: Severely normal cardiac depressed 2 mo, euthyroid on T3 evaluationdnormal ventricular replacement, normal at 5 mo function cardiac function
E, Echocardiography; RS, radionuclide scan, PTU, propylthiouracil; PROM, premature rupture of membranes; BP, blood pressure.
214
Sheffield and Cunningham
Figure 1 Sequential chest radiographs and echocardiography findings from patient 1. A, Pregnancy at 18 weeks of gestation with cardiomegaly and pulmonary edema. B, One week later shows clearing of pulmonary edema and persistent cardiomegaly. C, Four weeks after delivery shows normal heart and lungs.
clinic but were noncompliant with antithyroid medication; the other three women had no prenatal care and were diagnosed at admission. Severe preeclampsiaeclampsia precipitated heart failure in four of the seven women, two of the women had sepsis, and in one woman no other factors were identified. One woman
with severe preeclampsia also had a hematocrit level of 20%. Pulmonary edema and heart failure promptly responded to aggressive diuretic therapy and treatment of the underlying obstetric complication. This was confirmed by clinical examination and by radiographic
Sheffield and Cunningham
215
Figure 2 Serial right ventricular systolic pressures and echocardiography results from patient 1. LV, Left ventricle; RV, right ventricle; RA, right atrium.
improvement that demonstrated clearing of infiltrates and resolution of cardiomegaly. In seven women, cardiac function was assessed with the use of echocardiography, radionuclide ventriculography, or both. These were done within 1 to 2 days of admission in five women and were abnormal in all of them. Cardiac function was usually normal by the time of the follow-up study. However, an accurate interval for resolution cannot be ascertained because the interval to the second study was 5 days to 5 months. In at least two women, cardiac dysfunction persisted after the clinical failure had resolved. The first of these women still had depressed global function at 6 days that had normalized by 3 months. The other women had heart failure at 18 weeks. There was cardiomegaly and pulmonary edema on the chest radiograph (Figure 1, A). Echocardiography showed right atrial and ventricular dilation and moderately severe pulmonary hypertension with a right ventricular systolic pressure of 60 mm Hg (Figure 2). Pulmonary edema and left ventricular dysfunction improved with furosemide treatment (Figure 1, B). Treatment of thyrotoxicosis was begun, and by 35 weeks, at the time of spontaneous delivery, cardiomegaly had resolved, and pulmonary hypertension had improved markedly, with a systolic pressure of 40 mm Hg. One month after delivery, the chest radiography results were normal (Figure 1, C), and right ventricular pressures were within normal range. She was treated twice with radioiodine at 1 and 2 years and became hypothyroid, so replacement treatment was given.
Comment The incidence of Graves’ thyrotoxicosis was approximately 1 in 1700 deliveries during the 28-year period under study. A review of the clinical course of our women
with hyperthyroid allows us to make a number of clinical observations. First, we found that 9% of pregnant women with Graves’ disease had heart failure. All of these 13 women were untreated and thus were overtly thyrotoxic because they had no medical care during their pregnancy or they were noncompliant with antithyroid therapy. The second and equally important finding was that 11 of the 13 women also had an underlying obstetric complication that precipitated heart failure. Third, congestive heart failure responded rapidly to aggressive diuresis and treatment of the coexistent obstetric complications. Finally, with appropriate treatment of Graves’ disease, cardiomyopathy ultimately resolved, with an excellent long-term prognosis. The high incidence of heart failure in these women confirms previous reports that hyperthyroid-induced heart failure during pregnancy is much more common than in nonpregnant patients.1,17,18 This disparity can be explained best by the additive hemodynamic effects from normal pregnancy-induced cardiovascular changes on those of thyrotoxicosis (Table II).19-21 The net result of thyrotoxicosis and pregnancy is substantively increased cardiac output from diminished systemic vascular resistance and hypervolemia of pregnancy.9,10,20 Easterling et al21 have confirmed these additive effects on cardiac function (Table II). In most pregnant women, the high-output state is compensated, and ventricular function remains normal. Further decompensation is precipitated usually by the additional hemodynamic burdens that are imposed by any of a number of common pregnancy complications. Indeed, all but 2 of the 13 women had R1 obstetric complication that are known to further increase cardiac work and precipitate heart failure.9,11 In the first half of pregnancy, heart failure was precipitated in 5 of 6 women by either septicemia or hemorrhage and anemia, or both. Of the 7 women between 29 weeks of gestation and term,
216
Sheffield and Cunningham
Table II Indices of normal cardiovascular function and changes from thyrotoxicosis, normal pregnancy, and pregnancy complicated by thyrotoxicosis
Normal nonpregnant Hyperthyroidism1 Normal pregnancy19 Pregnant hyperthyroidism21 Before treatment After treatment
Systemic vascular resistance (dyne/sec/cme5)
Heart rate (beats/min)
Cardiac output (L/min)
Blood volume (% normal)
1500e1700 700e1200 945e1475
72e84 88e130 73e93
4.0e6.0 O7.0 5.2e7.2
100 105 14820
609 698
89 84
11.2 9.5
d d
heart failure was precipitated in 4 women by severe preeclampsia-eclampsia and in 2 women by sepsis. The rapid clinical improvement in these pregnant women is similar to that described for nonpregnant patients.22 In addition to treatment for heart failure and thyrotoxicosis, aggressive treatment is also imperative for coexistent and precipitating obstetric conditions. Aspects of this treatment include afterload reduction with severe preeclampsia-eclampsia, blood replacement for hemorrhage and anemia, and vigorous treatment for infection and sepsis. Although initial treatment results in rapid symptomatic and clinical improvement, underlying ventricular dysfunction persists until the treatment of thyrotoxicosis.1,6 In one study of nonpregnant individuals, Umpierrez et al22 showed that ventricular function normalizes over 6 to 12 months with antithyroid therapy. This is consistent with the report of Easterling et al21 who treated pregnant women with thyrotoxicosis. These latter investigators showed significant and substantial improvement, but not complete resolution, of the high-output state by a mean of 8 weeks after treatment was begun. In our study, we could not ascertain the duration that was necessary for the normalization of ventricular function because serial follow-up echocardiography was not done. We obtained a 2- to 25-year (mean, 7.4 years) followup in 11 of 13 women who are reported here. All the women eventually became euthyroid and had normal cardiac findings, as ascertained by chest radiography, clinical findings, or both. In the extreme case, one patient with pulmonary hypertension had gradual resolution over 5 months with antithyroid treatment. Concentric ventricular hypertrophy also resolved over this time, and her condition was entirely normal 4 years later. This response is similar to those in other case reports of the resolution of pulmonary hypertension in nonpregnant individuals with treatment.23-26 The clinical implications of these findings, taken in toto with previous observations,7-9 are straightforward. Pregnant women with untreated thyrotoxicosis should do well if they are seen early and adhere to an antithyroid regimen. For those women who are noncompliant or who do not seek early prenatal care, uncontrolled
thyrotoxicosis causes heart failure in approximately 10%. In most of those women with cardiac decompensation, there is usually a coexistent obstetric complication that precipitates the heart failure. These include sepsis and hemorrhage; in later gestation, severe preeclampsia-eclampsia is another common precipitating finding. Women with cardiac decompensation responds readily to treatment for pulmonary edema and the precipitating cause(s). Finally, even when there is profoundly abnormal ventricular dysfunction that is caused by long-standing thyrotoxic cardiomyopathy, the long-term prognosis is good with the treatment of thyrotoxicosis.
References 1. Klein I, Ojamaa K. Thyroid hormone and the cardiovascular system. N Engl J Med 2001;344:501-9. 2. Forfar JC, Muir AL, Sawers SA, Toft AD. Abnormal left ventricular function in hyperthyroidism: evidence for a possible reversible cardiomyopathy. N Engl J Med 1982;307:1165-70. 3. Kahaly GJ, Wagner S, Nieswandt J, Mohr-Kahaly S, Ryan TJ. Stress echocardiography in hyperthyroidism. J Clin Endocrinol Metab 1999;84:2308-13. 4. Piatnek-Leunissen D, Olson RE. Cardiac failure in the dog as a consequence of exogenous hyperthyroidism. Circ Res 1967;20: 242-52. 5. Veille JC. Maternal and fetal cardiovascular response to exercise during pregnancy. Semin Perinatol 1996;4:250-62. 6. Klein I, Ojamaa K. Thyrotoxicosis and the heart. Endocrinol Metab Clin North Am 1998;27:51-62. 7. Kriplani A, Buckshee K, Bhargava VL, Takkar D, Ammini AC. Maternal and perinatal outcome in thyrotoxicosis complicating pregnancy. Eur J Obstet Gynecol Reprod Biol 1994;54:159-63. 8. Millar LK, Wing DA, Leung AS, Koonings PP, Montoro MN, Mestman JH. Low birth weight and preeclampsia in pregnancies complicated by hyperthyroidism. Obstet Gynecol 1994;8:946-9. 9. Davis LE, Lucas MJ, Hankins GDV, Roark ML, Cunningham FG. Thyrotoxicosis complicating pregnancy. Am J Obstet Gynecol 1989;160:63-70. 10. Mestman JH. Hyperthyroidism in pregnancy. Endocrinol Metab Clin North Am 1998;27:127-49. 11. Cunningham FG, Pritchard JA, Hankins GDV, Anderson PL, Lucas MJ, Armstrong KF. Peripartum heart failure: Idiopathic cardiomyopathy or compounding cardiovascular events? Obstet Gynecol 1986;67:157-68.
Sheffield and Cunningham 12. Witlin AG, Mabie WC, Sibai BM. Peripartum cardiomyopathy: a longitudinal echocardiographic study. Am J Obstet Gynecol 1997;177:1129-32. 13. Burrow GN. The management of thyrotoxicosis in pregnancy. N Engl J Med 1985;313:562-5. 14. McKee PA, Castelli WP, McNamara PM, Kannel WB. The natural history of congestive heart failure: the Framingham Study. N Engl J Med 1971;285:1441-6. 15. Cunningham FG, Gant NF, Leveno KJ, Gilstrap LC, Hauth JC, Wenstrom KA. Endocrinological disorders. In: Cunningham F, Gant NF, Leveno KJ, Gilstrap LC III, Hauth JC, Wenstrom KD, editors. Williams obstetrics. 21st ed. New York: McGraw-Hill; 2001. p. 1342-43. 16. Lucas MJ, Leveno KJ, Cunningham FG. A comparison of magnesium sulfate with phenytoin for the prevention of eclampsia. N Engl J Med 1995;333:201-5. 17. Sandler G, Wilson GM. The nature and prognosis of heart disease in thyrotoxicosis: a review of 150 patients treated with 131I. Q J Med 1959;28:347-70. 18. Likoff WB, Levine SA. Thyrotoxicosis as the sole cause of heart failure. Am J Med Sci 1943;206:425-9.
217 19. Clark SL, Cotton DB, Lee W, Bishop C, Hill T, Southwick J, et al. Central hemodynamic assessment of normal term pregnancy. Am J Obstet Gynecol 1989;161:1439-42. 20. Pritchard JA. Changes in the blood volume during pregnancy and delivery. Anesthesiology 1965;26:393-9. 21. Easterling TR, Schmucker BC, Carlson KL, Millard SP, Benedetti TJ. Maternal hemodynamics in pregnancies complicated by hyperthyroidism. Obstet Gynecol 1991;78:348-52. 22. Umpierrez G, Challapalli S, Patterson C. Congestive heart failure due to reversible cardiomyopathy in patients with hyperthyroidism. Am J Med Sci 1995;310:99-102. 23. Nakchbandi IA, Wirth JA, Inzucchi SE. Pulmonary hypertension caused by Graves’ thyrotoxicosis. Chest 1999;116:1483-5. 24. Thurnheer R, Jenni R, Russi EW, Greminger P, Speich R. Hyperthyroidism and pulmonary hypertension. J Intern Med 1997; 242:185-8. 25. Yanai-Landau H, Amital H, Bar-Dayan Y, Levy Y, Gur H, Lin HC, et al. Autoimmune aspects of primary pulmonary hypertension. Pathobiology 1995;63:71-5. 26. Okura H, Takatsu Y. High-output heart failure as a cause of pulmonary hypertension. Intern Med 1994;33:363-5.
www.elsevier.com/locate/ajog
Thyrotoxicosis and heart failure that complicate pregnancy Jeanne S. Sheffield, MD,* F. Gary Cunningham, MD Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center at Dallas, Dallas, Tex Received June 17, 2003; revised July 21, 2003; accepted July 25, 2003
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Pregnancy Graves’ disease Heart failure
Objective: When untreated, Graves’ thyrotoxicosis has profound cardiovascular effects, although it rarely causes heart failure in otherwise healthy patients. Preliminary observations suggest that pregnant women are the exception. To further elucidate this association, we studied both immediate and long-term outcomes in women who had thyrotoxicosis and heart failure during pregnancy. Study design: We reviewed clinical outcomes of pregnant women with Graves’ disease and heart failure at our institution from 1974 through 2001. Women with other underlying heart disease were excluded. A standardized antithyroid regimen and serial echocardiography and/or chest radiography were performed. Results: The 13 women with thyrotoxicosis and heart failure were either noncompliant with antithyroid therapy or had no medical care during pregnancy. Six women had heart failure before fetal viability; decompensation was precipitated by hemorrhage, sepsis, or both. The other 7 women were in the last trimester when heart failure developed; in 4 women, the heart failure was precipitated by severe preeclampsia-eclampsia and in 2 women was precipitated by sepsis. Overall, 11 of 13 women had an underlying obstetric event. In follow-up of 11 women from 2 to 25 years, resolution of cardiomyopathy was confirmed after successful treatment of thyrotoxicosis. Conclusion: Normal pregnancy mimics and amplifies some of the hyperdynamic cardiovascular changes that are caused by thyrotoxicosis. When they occur simultaneously, there is usually a compensated high-output state. In some women, however, common pregnancy complications that include hemorrhage with associated anemia, sepsis, and severe preeclampsia-eclampsia will precipitate heart failure. The immediate treatment of heart failure and the correction of precipitating pregnancy factors usually results in good outcome. Long-term follow-up confirmed that thyrotoxic cardiac dysfunction is reversible with successful antithyroid therapy. Ó 2004 Elsevier Inc. All rights reserved.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
* Reprint requests: Jeanne S. Sheffield, MD, Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390-9032. E-mail: Jeanne.Sheffi[email protected] 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/S0002-9378(03)00944-X
There are a number of profound effects of excessive thyroid hormones on cardiovascular function.1 Contractility is directly increased by tri-iodothyronineeresponsive cardiac genes that augment cardiac protein synthesis. Contractility is increased indirectly by diminished systemic vascular resistance with the resultant
212 increased heart rate and cardiac output. Without intervention, these two mechanisms result in increased left ventricular mass and a high-output state. In this case, any further cardiac demands (such as from exercise) cause significant but reversible left ventricular dysfunction.2-5 Thus, this basal high-output state, although compensated, may result in heart failure when there is increased cardiac burden. Despite this, nonpregnant patients with thyrotoxicosis seldom develop overt heart failure.1,6 By contrast, almost 10% of women with untreated thyrotoxicosis have heart failure during pregnancy.7,8 The propensity for pregnant women to have heart failure has been explained by the combined effects of thyroxine-induced cardiovascular changes along with the very similar pregnancy-induced hyperdynamic changes.9,10 Even then, heart failure does not usually develop unless precipitated by factors that further magnify cardiac workload.9 These include severe preeclampsiaeclampsia, sepsis, hemorrhage, and anemia.11,12 Our purposes now are to extend our previous observations of pregnant women with hyperthyroidism and with cardiac failure9 and to study its long-term consequences.
Material and methods The pregnant women who are described in this report had uncontrolled thyrotoxicosis that was further complicated by congestive heart failure. They were admitted to the obstetrics service at Parkland Hospital from 1974 through 2001. The diagnosis of Graves’ disease was made clinically with laboratory confirmation with the standard criteria.9,10,13 All of the women had a diffuse toxic goiter and other variable findings that included peripheral tremor, resting tachycardia, weight loss, palpitations, weakness, hyperreflexia, nervousness, or irritability and Graves’ ophthalmopathy and dermopathy. Over the study period, laboratory tests for thyroid disease continually evolved. Accordingly, total serum thyroxine and tri-iodothyronine, T3-resin uptake ratio, and free thyroxine index were measured until they were replaced by contemporaneous assays for serum thyroid-stimulating hormone and free thyroxine. Heart failure was diagnosed with the use of the Framingham criteria and confirmed by chest radiography that showed cardiomegaly and pulmonary edema.14 Echocardiography was not available widely during the early years of the study. When available, however, cardiac structure and function were assessed with echocardiograph, or, in some earlier cases, by radionuclide scanning ventriculography. Women were excluded if they had underlying cardiac abnormalities. At our institution, clinical treatment of pregnant women with thyrotoxicosis and heart failure is similar to that for thyroid storm.15 Briefly, they are admitted to the obstetrics intensive care unit, and furosemide is given intravenously for pulmonary edema. Treatment
Sheffield and Cunningham of thyrotoxicosis included a 1000-mg oral loading dose of propylthiouracil followed in 1 hour by 500 to 1000 mg of potassium iodide, which was given either intravenously or orally. Modified doses of both drugs were then continued at 8-hour intervals. They also were given three doses of intravenous hydrocortisone or methylprednisolone. Early in the study, some women were given digoxin, and later b-blockade was used for extreme tachycardia. Vigorous treatment was given for obstetrics complications that precipitate or worsen heart failure. Severe preeclampsia or eclampsia was treated with magnesium sulfate to prevent or stop eclamptic seizures and with intravenous hydralazine to control severe hypertension and afterload failure.16 Severe anemia was corrected with red blood cell transfusions. Infection was treated aggressively with intravenous antimicrobial therapy, and uterine curettage was performed for women with a septic abortion. All of these women were followed by at least one of the authors to ensure homogeneity of treatment. Data retrieval was begun at the sentinel admission, and long-term follow-up was provided by record review. Data retrieval and analysis were approved by the Institutional Review Board of the University of Texas Southwestern Medical Center.
Results During the 28-year study period, slightly over 300,000 women were delivered at our institution. The recognized incidence of overt thyrotoxicosis averaged approximately 1 in 1500 pregnancies, as determined for three epochs: 1:2000 for approximately 120,000 women who were delivered from 1974 through 19859; 1:1500 for approximately 125,000 women who were delivered from 1991 through 1999; and 1:1000 for 11,000 women who underwent serum thyrotropin screening from 2000 through 2001. From all of these, we identified 150 women who were overtly thyrotoxic; in 13 women (9%), pregnancy was further complicated by heart failure. At diagnosis, in 10 women the total thyroxine level ranged from 19 to O40 mg/dL. The other 3 women had free thyroxine values that ranged from 4.2 to 6.1 ng/dL. Clinical information for the 13 women with heart failure is listed in Table I. Six of these 13 women had heart failure before the time of expected fetal viability. In 3 of these women, hemorrhage and sepsis with incomplete abortion precipitated heart failure. In 2 other women, septicemia from pyelonephritis triggered heart failure. Four of the 6 women had anemia with a hematocrit level of !25%. The remaining woman had no precipitating factors that were identified. The other seven women who are mentioned in Table I were in the third trimester when heart failure developed. Four of these seven women had attended prenatal
213
Sheffield and Cunningham Table I
Clinical and laboratory data from 13 Pregnant Women with thyrotoxicosis and heart failure
Associated Demographic precipitating event 1. 32 G2P1 None except pregnancy; (18 wk) mild anemia (hematocrit = 25%)
2. 26 G3P2 Molar pregnancy; (20 wk) hemorrhage; anemia (hematocrit 22%)
3. 20 G2P1 Incomplete abortion; (20 wk) hemorrhage; anemia (hematocrit 23%) 4. 27 G3P2 Septic abortion (T 40(C); (20 wk) hemorrhage; anemia (hematocrit 19%) 5. 21 G3P1 Pyelonephritis; sepsis (22 wk) (T 39.4(C); anemia (hematocrit 16%) 6. 20 G2P1 Pyelonephritis; sepsis (23 wk) (T 40(C); anemia (hematocrit 25%) 7. 22 G2P0 Twins: preterm PROM; (29 wk) chorioamnionitis; sepsis (T 39(C) 8. 28 G2P1 Severe preeclampsia (32 wk) (BP, 170/120 mm Hg) 9. 22 G4P3 Severe preeclampsia (34 wk) (BP, 220/140 mm Hg)
10. 34 G4P3 None except pregnancy; (35 wk) mild anemia (hematocrit = 28%) 11. 21 G1P0 Chorioamnionitis; (35 wk) sepsis (T 39(C); anemia (hematocrit 28%)
12. 30 G4P3 Severe preeclampsia; (36 wk) (BP, 200/150 mm Hg); anemia (hematocrit 20%) 13. 17 G1P0 Eclampsia (37 wk) (BP, 160/95 mm Hg)
Cardiac evaluation Initial
Follow Up
Clinical course
Long-term follow up
Four years, clinically E: Dilated right E: Four follow- Reversal of pulmonary euthyroid with normal hypertension and up studies atrium and cardiac function. ventricular (see Fig 2) ventricle; hypertrophy by 4 wk pulmonary after delivery, hypertension with PTU treatment (RVSP, 60 mm begun at 16 wk Hg) Chest radiograph clear with On T3 replacement therapy E: Dilated right E: 5 dd normal no residual cardiomegaly atrium and after 131I 2 y, clinically 8 d after delivery ventricle; 2+ euthyroid, normal mitral and third pregnancy tricuspid regurgitation None None Chest radiography returned Twenty-five years, chest to normal after delivery radiographs and clinical examinations normal. Chest radiograph returned Three years, euthyroid RS: Ejection RS: 25 dd to normal at 18 d taking PTU; normal fraction 0.35 ejection cardiac examination fraction 0.53 None None Ongoing PTU therapy; chest None radiograph returned to normal at 14 d None None Chest radiograph normal Twenty years, euthyroid at 2 mo after 131I at 10 y; normal cardiac examination Mildly depressed Echocardiogram normal Two years, normal chest left ventricular at 5 mo radiograph and cardiac dysfunction evaluation None None Chest radiograph returned None to normal at 4 d Chest radiograph returned Five years, normal chest None RS: 11 dd to normal at 2 d radiograph and cardiac ejection evaluation fraction, 0.59 None None Chest radiograph returned Three years, normal chest radiograph and cardiac to normal at 4 d and evaluation normal at 1 mo Echocardiogram normal Five years, took PTU RS: 6 dd E: Moderately at 3 mo 2 years, now euthyroid ejection depressed left with normal chest fraction, ventricular radiograph and cardiac 0.51 with function with evaluation mild globally abnormal depressed anterior wall function motion None None Chest radiograph returned Ten years, euthyroid after to normal at 2 wk PTU for 5 years; normal chest, radiograph and cardiac evaluation Two years, 131I at Chest radiography and E: 35 dd E: Severely normal cardiac depressed 2 mo, euthyroid on T3 evaluationdnormal ventricular replacement, normal at 5 mo function cardiac function
E, Echocardiography; RS, radionuclide scan, PTU, propylthiouracil; PROM, premature rupture of membranes; BP, blood pressure.
214
Sheffield and Cunningham
Figure 1 Sequential chest radiographs and echocardiography findings from patient 1. A, Pregnancy at 18 weeks of gestation with cardiomegaly and pulmonary edema. B, One week later shows clearing of pulmonary edema and persistent cardiomegaly. C, Four weeks after delivery shows normal heart and lungs.
clinic but were noncompliant with antithyroid medication; the other three women had no prenatal care and were diagnosed at admission. Severe preeclampsiaeclampsia precipitated heart failure in four of the seven women, two of the women had sepsis, and in one woman no other factors were identified. One woman
with severe preeclampsia also had a hematocrit level of 20%. Pulmonary edema and heart failure promptly responded to aggressive diuretic therapy and treatment of the underlying obstetric complication. This was confirmed by clinical examination and by radiographic
Sheffield and Cunningham
215
Figure 2 Serial right ventricular systolic pressures and echocardiography results from patient 1. LV, Left ventricle; RV, right ventricle; RA, right atrium.
improvement that demonstrated clearing of infiltrates and resolution of cardiomegaly. In seven women, cardiac function was assessed with the use of echocardiography, radionuclide ventriculography, or both. These were done within 1 to 2 days of admission in five women and were abnormal in all of them. Cardiac function was usually normal by the time of the follow-up study. However, an accurate interval for resolution cannot be ascertained because the interval to the second study was 5 days to 5 months. In at least two women, cardiac dysfunction persisted after the clinical failure had resolved. The first of these women still had depressed global function at 6 days that had normalized by 3 months. The other women had heart failure at 18 weeks. There was cardiomegaly and pulmonary edema on the chest radiograph (Figure 1, A). Echocardiography showed right atrial and ventricular dilation and moderately severe pulmonary hypertension with a right ventricular systolic pressure of 60 mm Hg (Figure 2). Pulmonary edema and left ventricular dysfunction improved with furosemide treatment (Figure 1, B). Treatment of thyrotoxicosis was begun, and by 35 weeks, at the time of spontaneous delivery, cardiomegaly had resolved, and pulmonary hypertension had improved markedly, with a systolic pressure of 40 mm Hg. One month after delivery, the chest radiography results were normal (Figure 1, C), and right ventricular pressures were within normal range. She was treated twice with radioiodine at 1 and 2 years and became hypothyroid, so replacement treatment was given.
Comment The incidence of Graves’ thyrotoxicosis was approximately 1 in 1700 deliveries during the 28-year period under study. A review of the clinical course of our women
with hyperthyroid allows us to make a number of clinical observations. First, we found that 9% of pregnant women with Graves’ disease had heart failure. All of these 13 women were untreated and thus were overtly thyrotoxic because they had no medical care during their pregnancy or they were noncompliant with antithyroid therapy. The second and equally important finding was that 11 of the 13 women also had an underlying obstetric complication that precipitated heart failure. Third, congestive heart failure responded rapidly to aggressive diuresis and treatment of the coexistent obstetric complications. Finally, with appropriate treatment of Graves’ disease, cardiomyopathy ultimately resolved, with an excellent long-term prognosis. The high incidence of heart failure in these women confirms previous reports that hyperthyroid-induced heart failure during pregnancy is much more common than in nonpregnant patients.1,17,18 This disparity can be explained best by the additive hemodynamic effects from normal pregnancy-induced cardiovascular changes on those of thyrotoxicosis (Table II).19-21 The net result of thyrotoxicosis and pregnancy is substantively increased cardiac output from diminished systemic vascular resistance and hypervolemia of pregnancy.9,10,20 Easterling et al21 have confirmed these additive effects on cardiac function (Table II). In most pregnant women, the high-output state is compensated, and ventricular function remains normal. Further decompensation is precipitated usually by the additional hemodynamic burdens that are imposed by any of a number of common pregnancy complications. Indeed, all but 2 of the 13 women had R1 obstetric complication that are known to further increase cardiac work and precipitate heart failure.9,11 In the first half of pregnancy, heart failure was precipitated in 5 of 6 women by either septicemia or hemorrhage and anemia, or both. Of the 7 women between 29 weeks of gestation and term,
216
Sheffield and Cunningham
Table II Indices of normal cardiovascular function and changes from thyrotoxicosis, normal pregnancy, and pregnancy complicated by thyrotoxicosis
Normal nonpregnant Hyperthyroidism1 Normal pregnancy19 Pregnant hyperthyroidism21 Before treatment After treatment
Systemic vascular resistance (dyne/sec/cme5)
Heart rate (beats/min)
Cardiac output (L/min)
Blood volume (% normal)
1500e1700 700e1200 945e1475
72e84 88e130 73e93
4.0e6.0 O7.0 5.2e7.2
100 105 14820
609 698
89 84
11.2 9.5
d d
heart failure was precipitated in 4 women by severe preeclampsia-eclampsia and in 2 women by sepsis. The rapid clinical improvement in these pregnant women is similar to that described for nonpregnant patients.22 In addition to treatment for heart failure and thyrotoxicosis, aggressive treatment is also imperative for coexistent and precipitating obstetric conditions. Aspects of this treatment include afterload reduction with severe preeclampsia-eclampsia, blood replacement for hemorrhage and anemia, and vigorous treatment for infection and sepsis. Although initial treatment results in rapid symptomatic and clinical improvement, underlying ventricular dysfunction persists until the treatment of thyrotoxicosis.1,6 In one study of nonpregnant individuals, Umpierrez et al22 showed that ventricular function normalizes over 6 to 12 months with antithyroid therapy. This is consistent with the report of Easterling et al21 who treated pregnant women with thyrotoxicosis. These latter investigators showed significant and substantial improvement, but not complete resolution, of the high-output state by a mean of 8 weeks after treatment was begun. In our study, we could not ascertain the duration that was necessary for the normalization of ventricular function because serial follow-up echocardiography was not done. We obtained a 2- to 25-year (mean, 7.4 years) followup in 11 of 13 women who are reported here. All the women eventually became euthyroid and had normal cardiac findings, as ascertained by chest radiography, clinical findings, or both. In the extreme case, one patient with pulmonary hypertension had gradual resolution over 5 months with antithyroid treatment. Concentric ventricular hypertrophy also resolved over this time, and her condition was entirely normal 4 years later. This response is similar to those in other case reports of the resolution of pulmonary hypertension in nonpregnant individuals with treatment.23-26 The clinical implications of these findings, taken in toto with previous observations,7-9 are straightforward. Pregnant women with untreated thyrotoxicosis should do well if they are seen early and adhere to an antithyroid regimen. For those women who are noncompliant or who do not seek early prenatal care, uncontrolled
thyrotoxicosis causes heart failure in approximately 10%. In most of those women with cardiac decompensation, there is usually a coexistent obstetric complication that precipitates the heart failure. These include sepsis and hemorrhage; in later gestation, severe preeclampsia-eclampsia is another common precipitating finding. Women with cardiac decompensation responds readily to treatment for pulmonary edema and the precipitating cause(s). Finally, even when there is profoundly abnormal ventricular dysfunction that is caused by long-standing thyrotoxic cardiomyopathy, the long-term prognosis is good with the treatment of thyrotoxicosis.
References 1. Klein I, Ojamaa K. Thyroid hormone and the cardiovascular system. N Engl J Med 2001;344:501-9. 2. Forfar JC, Muir AL, Sawers SA, Toft AD. Abnormal left ventricular function in hyperthyroidism: evidence for a possible reversible cardiomyopathy. N Engl J Med 1982;307:1165-70. 3. Kahaly GJ, Wagner S, Nieswandt J, Mohr-Kahaly S, Ryan TJ. Stress echocardiography in hyperthyroidism. J Clin Endocrinol Metab 1999;84:2308-13. 4. Piatnek-Leunissen D, Olson RE. Cardiac failure in the dog as a consequence of exogenous hyperthyroidism. Circ Res 1967;20: 242-52. 5. Veille JC. Maternal and fetal cardiovascular response to exercise during pregnancy. Semin Perinatol 1996;4:250-62. 6. Klein I, Ojamaa K. Thyrotoxicosis and the heart. Endocrinol Metab Clin North Am 1998;27:51-62. 7. Kriplani A, Buckshee K, Bhargava VL, Takkar D, Ammini AC. Maternal and perinatal outcome in thyrotoxicosis complicating pregnancy. Eur J Obstet Gynecol Reprod Biol 1994;54:159-63. 8. Millar LK, Wing DA, Leung AS, Koonings PP, Montoro MN, Mestman JH. Low birth weight and preeclampsia in pregnancies complicated by hyperthyroidism. Obstet Gynecol 1994;8:946-9. 9. Davis LE, Lucas MJ, Hankins GDV, Roark ML, Cunningham FG. Thyrotoxicosis complicating pregnancy. Am J Obstet Gynecol 1989;160:63-70. 10. Mestman JH. Hyperthyroidism in pregnancy. Endocrinol Metab Clin North Am 1998;27:127-49. 11. Cunningham FG, Pritchard JA, Hankins GDV, Anderson PL, Lucas MJ, Armstrong KF. Peripartum heart failure: Idiopathic cardiomyopathy or compounding cardiovascular events? Obstet Gynecol 1986;67:157-68.
Sheffield and Cunningham 12. Witlin AG, Mabie WC, Sibai BM. Peripartum cardiomyopathy: a longitudinal echocardiographic study. Am J Obstet Gynecol 1997;177:1129-32. 13. Burrow GN. The management of thyrotoxicosis in pregnancy. N Engl J Med 1985;313:562-5. 14. McKee PA, Castelli WP, McNamara PM, Kannel WB. The natural history of congestive heart failure: the Framingham Study. N Engl J Med 1971;285:1441-6. 15. Cunningham FG, Gant NF, Leveno KJ, Gilstrap LC, Hauth JC, Wenstrom KA. Endocrinological disorders. In: Cunningham F, Gant NF, Leveno KJ, Gilstrap LC III, Hauth JC, Wenstrom KD, editors. Williams obstetrics. 21st ed. New York: McGraw-Hill; 2001. p. 1342-43. 16. Lucas MJ, Leveno KJ, Cunningham FG. A comparison of magnesium sulfate with phenytoin for the prevention of eclampsia. N Engl J Med 1995;333:201-5. 17. Sandler G, Wilson GM. The nature and prognosis of heart disease in thyrotoxicosis: a review of 150 patients treated with 131I. Q J Med 1959;28:347-70. 18. Likoff WB, Levine SA. Thyrotoxicosis as the sole cause of heart failure. Am J Med Sci 1943;206:425-9.
217 19. Clark SL, Cotton DB, Lee W, Bishop C, Hill T, Southwick J, et al. Central hemodynamic assessment of normal term pregnancy. Am J Obstet Gynecol 1989;161:1439-42. 20. Pritchard JA. Changes in the blood volume during pregnancy and delivery. Anesthesiology 1965;26:393-9. 21. Easterling TR, Schmucker BC, Carlson KL, Millard SP, Benedetti TJ. Maternal hemodynamics in pregnancies complicated by hyperthyroidism. Obstet Gynecol 1991;78:348-52. 22. Umpierrez G, Challapalli S, Patterson C. Congestive heart failure due to reversible cardiomyopathy in patients with hyperthyroidism. Am J Med Sci 1995;310:99-102. 23. Nakchbandi IA, Wirth JA, Inzucchi SE. Pulmonary hypertension caused by Graves’ thyrotoxicosis. Chest 1999;116:1483-5. 24. Thurnheer R, Jenni R, Russi EW, Greminger P, Speich R. Hyperthyroidism and pulmonary hypertension. J Intern Med 1997; 242:185-8. 25. Yanai-Landau H, Amital H, Bar-Dayan Y, Levy Y, Gur H, Lin HC, et al. Autoimmune aspects of primary pulmonary hypertension. Pathobiology 1995;63:71-5. 26. Okura H, Takatsu Y. High-output heart failure as a cause of pulmonary hypertension. Intern Med 1994;33:363-5.