Correspondence Cardiovascular Division, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115, USA

Email kbaughman@partners.org

The case

A 49-year-old woman was admitted to hospital 8 days after giving birth to twins. No complications had occurred during her first pregnancy 3 years previously. During her second pregnancy, the patient gained 20.9 kg in weight. Echocardiography was performed on two occasions because of the twin pregnancy and the older age of the patient, revealing normal diastolic and systolic diameters, well-preserved ventricular function and a normal estimated right ventricular systolic pressure. In the last month of gestation, the patient began to notice edema in the lower extremities. Two weeks before delivery, she also developed symptoms of dyspnea and orthopnea that worsened over time, and eventually had to sleep almost upright in a chair. She went on to have a normal delivery at full term and was discharged 2 days later.

The patient's lower-extremity edema failed to improve after delivery, and her shortness of breath worsened. She was diagnosed by her family physician as having heart failure and was referred to a cardiologist. Echocardiography showed an ejection fraction of 35%. The patient was referred to hospital for further management. Compared with baseline measurements, repeat echocardiograms revealed increased diastolic dimensions, reduced systolic function and an increased ejection fraction of 50% (Table 1).

Table 1 Echocardiograms taken during the patient's second pregnancy and after delivery.

Full tableFigures & Tables indexDownload Power Point slide (231K)

Possible causes of the patient's reduced systolic function were investigated. Electrocardiography did not reveal any signs of arrhythmia, myocardial infarction or conduction delays. Laboratory blood tests demonstrated normal thyroid function, and antinuclear antibodies were absent. Levels of cardiac enzymes were all within the normal ranges. On physical examination, prominent S4 and S3 gallop rhythms, a jugular venous distention of 17 cm water and grade 2+ edema in the lower extremities were noted. A diagnosis of peripartum cardiomyopathy was made. Standard heart failure therapy was administered, including furosemide (40 mg daily), a -blocker (12.5 mg metoprolol succinate daily), an angiotensin-converting-enzyme inhibitor (10 mg lisinopril daily) and systemic anticoagulation with enoxaparin sodium injection in transition to warfarin therapy. The amount of salt in the patient's diet was limited to 2 g, and total fluid to 2 l, daily. During her stay in hospital the patient's weight decreased from 78.8 kg to 62.9 kg. As an outpatient, she sent weekly emails to her physicians, reporting daily weight measurements, medications used, exercise taken, fluid intake, blood pressure and pulse every morning and evening, and other symptom-related observations. After 1 month, the patient's weight had decreased from 60.9 kg to 59.8 kg, and 2 months later she weighed 57.0 kg. Anticoagulation therapy was discontinued when her ventricular function and level of physical activity had improved. Furosemide therapy was stopped when the patient had achieved her pre-pregnancy weight, and the lisinopril dose was also reduced and then discontinued because of low blood pressure and a drug-related cough. The patient continued to take low-dose -blocker therapy. Although the patient's S3 and S4 gallop rhythms initially converted to an S4 gallop, her heart sounds eventually returned to normal. The patient's last echocardiogram revealed normal diastolic and systolic dimensions, a normal ejection fraction of 60%, and normal right ventricular systolic pressure. Six months after delivery, her ventricular response to exercise stress echocardiography was normal, and the -blocker therapy was discontinued.

Discussion of diagnosis

Peripartum cardiomyopathy is defined as a heart muscle disorder that presents during the last month of gestation or up to 5 months postpartum.^{1, 2} Other diagnostic criteria include heart dysfunction of unknown cause and the absence of pre-existing heart muscle disease (Box 1). The diagnosis is not standard in the published literature, as some studies have included patients who developed heart failure at other times during gestation, particularly during the last trimester. Many, if not most, of these women had a pre-existing heart muscle disorder that was exacerbated by the challenge of volume adaptation during normal pregnancy. Pregnancy-associated heart failure is a more accurate description of such cases.

The prevalence of peripartum cardiomyopathy is unclear, but it could complicate 1 in every 1,300–4,000 deliveries in the US. Women of any race, age, or with any number of previous pregnancies can be affected. Several groups have an elevated risk of developing the condition, however, including older women, African Americans, multiple gravida and women who have twins or multiple embryos.^{3, 4, 5} Although the underlying cause of the condition is also unknown, several endomyocardial biopsy studies have revealed myocarditis (Table 2).^{6, 7, 8, 9, 10} A depressed immune system during pregnancy might predispose mothers to cardiotrophic viral infections that cause myocarditis, or the disorder could be a response to fetal or paternal antigens recognized during postpartum recovery of the immune system.

Table 2 Summary of studies describing myocarditis in patients with peripartum cardiomyopathy and the outcome of immunosuppression therapy on symptoms of congestive heart failure.

Full tableFigures & Tables indexDownload Power Point slide (232K)

Women with peripartum cardiomyopathy present with typical symptoms of congestive heart failure, which are often difficult to discern from common symptoms experienced by every pregnant mother as a result of physiological weight gain and volume expansion. Pregnant women should be referred for an echocardiogram if they gain excessive amounts of weight, have significant lower-extremity edema, or have jugular venous distension in the presence of gallop rhythms or cardiomegaly. Normal pregnancy is associated with an increased heart rate and left ventricular diastolic dimension; there should not, however, be elevations in diastolic dimensions beyond normal levels or reductions in ventricular function during gestation or the postpartum period.

Most women affected with peripartum cardiomyopathy present with symptoms within 1 month of delivery, with 80% coming to medical attention within 3 months.^{1, 2} Studies have reported remarkable variation in mortality. Compared with idiopathic dilated cardiomyopathy, however, the prognosis for women with peripartum cardiomyopathy is usually good.^{10, 11, 12} Women with suspected peripartum cardiomyopathy must undergo a careful medical and family history assessment and investigations to exclude pre-existing heart disease or intercurrent causes of left ventricular compromise. Electrocardiography can help to evaluate arrhythmias and myocardial infarctions, particularly those caused by pregnancy-related coronary artery dissection. Physical examination and echocardiography can rule out other primary valvular or pericardial disorders. Laboratory analyses should also be carried out to exclude hypothyroidism and hyperthyroidism and connective tissue disorders. Coronary angiography is not usually indicated, unless there is evidence of a previous or recent myocardial infarction and noninvasive studies indicate focal wall motion abnormalities. As symptoms spontaneously resolve in a high proportion of patients, endomyocardial biopsy is not recommended at presentation. If heart function does not improve within 2 weeks of standard therapy, however, it is useful to check for evidence of myocarditis, as the presence of inflammation can influence management decisions.

Treatment and management

Patients diagnosed with peripartum cardiomyopathy should receive standard heart failure therapy. Therapy regimens include diuretics to diminish volume overload (preload), afterload reduction with angiotensin-converting-enzyme inhibitors (postpartum only), and -blockers after signs and symptoms of congestion have improved. Pregnant patients should not be exposed to angiotensin-converting-enzyme inhibitors (Box 2). Anticoagulation therapy can be given to patients postpartum if symptoms are persistent or ventricular function is markedly diminished, as such individuals have a high risk of systemic and pulmonary embolism.

In my experience, heart failure symptoms are most likely to spontaneously and dramatically improve in women presenting with symptoms soon after delivery. Among 27 peripartum cardiomyopathy patients studied by Demakis et al., approximately half recovered normal heart size and function, whereas the remainder had persistent cardiomegaly at 6-month follow-up.^{1, 2} Patients in Demakis' study with persistent cardiomegaly had a similar prognosis to individuals with dilated cardiomyopathy; mortality was 85% and the average survival time was 4.7 years after the onset of heart failure. By contrast, only 14% of women whose heart function returned to normal died over an average follow-up of 10.7 years, and none had symptomatic cardiovascular disease. It is likely that early diagnosis and appropriate treatment has improved patient prognosis. Cole et al. used echocardiography to show that cardiac function improved in 13 out of 14 affected women following standard heart failure therapy.¹³ If symptoms do not respond to standard medical therapy, however, additional cardiovascular support might be required, such as intra-aortic balloon counterpulsation, left ventricular assist device and possibly cardiac transplantation. Indeed, the survival rate for peripartum cardiomyopathy patients following heart transplant is similar to that for other heart diseases.¹⁴

Most women with peripartum cardiomyopathy are initially admitted to hospital because of severe heart failure, but can be discharged once their symptoms have been stabilized. Patients are seen in follow-up at 2-week intervals. The echocardiogram should be repeated at these visits until ventricular function is no longer changing. Once stable, the outpatient visit and echocardiogram interval is determined by the patient's clinical state. With appropriate medication, diet and exercise, ventricular remodeling is usually complete within 6 months of initial presentation.

Women with a history of peripartum cardiomyopathy should seek medical advice before considering another pregnancy. If ventricular function fails to recover after therapy, several studies have reported worse pregnancy outcomes and maternal and fetal death (Table 3).^{1, 2, 15, 16} If ventricular function does return to normal, patients can be referred for exercise stress echocardiography at 6 months postpartum. If this strategy yields a normal response following withdrawal of medical therapy, additional pregnancies can be considered with careful monitoring by an obstetrician and cardiologist. Women should avoid pregnancy if they continue to have heart failure symptoms or an ejection fraction of less than 40%.

Table 3 Data showing elevated risk of several parameters in women with a history of peripartum cardiomyopathy who underwent repeat pregnancy.

Full tableFigures & Tables indexDownload Power Point slide (230K)

Conclusion

Peripartum cardiomyopathy is a distinctive form of heart muscle dysfunction, which can occur in pregnant women during the last month of gestation or within 5 months of delivery. Although the etiology of this condition is unclear, it might have an immunologic origin. As in the patient described here, affected individuals can be treated with standard heart failure therapy, which may restore normal ventricular function. If this treatment is ineffective, however, patients might benefit from more-aggressive ventricular support such as intra-aortic balloon counterpulsation, a left ventricular assist device or heart transplantation. Women with a history of peripartum cardiomyopathy are only advised to consider further pregnancies if their ventricular function returns to normal, both at rest and with exercise.

References

1. Demakis JG and Rahimtoola SH (1971) Peripartum cardiomyopathy. Circulation 44: 964–968 | PubMed | ChemPort |
2. Demakis JG et al. (1971) Natural course of peripartum cardiomyopathy. Circulation 44: 1053–1061 | PubMed | ChemPort |
3. Lampert MB and Lang RM (1995) Peripartum cardiomyopathy. Am Heart J 130: 860–870 | Article | PubMed | ChemPort |
4. Homans DC (1985) Peripartum cardiomyopathy. N Engl J Med 312: 1432–1437 | PubMed | ChemPort |
5. Reimold SC and Rutherford JD (2001) Peripartum cardiomyopathy. N Engl J Med 344: 1629–1630 | Article | PubMed | ChemPort |
6. Melvin KR et al. (1982) Peripartum cardiomyopathy due to myocarditis. N Engl J Med 307: 731–734 | PubMed | ChemPort |
7. O'Connell JB et al. (1986) Peripartum cardiomyopathy: clinical, hemodynamic, histologic and prognostic characteristics. J Am Coll Cardiol 8: 52–56 | PubMed | ChemPort |
8. Sanderson JE et al. (1986) Peripartum heart disease: an endomyocardial biopsy study. Br Heart J 56: 285–291 | PubMed | ChemPort |
9. Midei MG et al. (1990) Peripartum myocarditis and cardiomyopathy. Circulation 81: 922–928 | PubMed | ChemPort |
10. Felker GM et al. (2000) Underlying causes and long-term survival in patients with initially unexplained cardiomyopathy. N Engl J Med 342: 1077–1084 | Article | PubMed | ChemPort |
11. Felker GM et al. (2000) Myocarditis and long-term survival in peripartum cardiomyopathy. Am Heart J 140: 785–791 | Article | PubMed | ChemPort |
12. Whitehead SJ et al. (2003) Pregnancy-related mortality due to cardiomyopathy: United States, 1991–1997. Obstet Gynecol 102: 1326–1331 | Article | PubMed |
13. Cole P et al. (1987) Longitudinal changes in left ventricular architecture and function in peripartum cardiomyopathy. Am J Cardiol 60: 871–876 | Article | PubMed | ChemPort |
14. Rickenbacher PR et al. (1994) Long-term outcome after heart transplantation for peripartum cardiomyopathy. Am Heart J 127: 1318–1323 | Article | PubMed | ChemPort |
15. Elkayam U et al. (2001) Maternal and fetal outcomes of subsequent pregnancies in women with peripartum cardiomyopathy. N Engl J Med 344: 1567–1571 | Article | PubMed | ChemPort |
16. Lampert MB et al. (1997) Contractile reserve in patients with peripartum cardiomyopathy and recovered left ventricular function. Am J Obstet Gynecol 176: 189–195 | Article | PubMed | ChemPort |
17. Pearson GD et al. (2000) Peripartum cardiomyopathy: National Heart, Lung, and Blood Institute and Office of Rare Diseases (National Institutes of Health) workshop recommendations and review. JAMA 283: 1183–1188 | Article | PubMed | ChemPort |

Contact the journal about this article

Subject areas under which this article appears: Cardiomyopathy and heart failure