Relation of Radiation Dose, Field, and Fractionation

There is a dose-dependent relation with the degree of cardiac injury and radiation. The increase in the risk of coronary events is proportional to the radiation dose to the heart (7.4%/Gy; 95% confidence interval 2.9 to 14.5; p <0.001), begins within the first 5 years after radiotherapy, continues for at least 20 years, and is similar in women with and without cardiac risk factors at the time of exposure. Moderate to severe myocardial fibrosis was evident only when the radiation dose was greater than 30 Gy. In addition, the type and location of the tumor treated influence the dose of radiation that the heart receives. Cardiac toxicity is greater for radiotherapy of left-sided breast cancer compared with right-sided cancer because of higher whole heart dose (0.9 to 14 vs 0.4 to 6 Gy), leading to higher cardiovascular mortality, mainly attributed to higher radiation dose to the left anterior descending artery with left-sided compared with right-sided cancer (21.8 to 51 vs 0.9 to 3 Gy). Tangential radiation fields using Cobalt-60 result in lower doses to the heart compared with anterior fields to the breast. Patients with Hodgkin disease historically received 35 to 45 Gy of mantle field irradiation to lymph nodes in the neck, mediastinum, and axillae, resulting in 27.5 Gy to the heart and >35 Gy to certain parts of the heart. Thickening of the aortic or mitral valve and most cardiovascular mortality typically occurred with a threshold dose of at least 30 Gy.

Strategies to minimize the radiation dose delivered to the heart for the treatment of breast cancer include the use of computed tomography planning to exclude the heart and internal mammary chain of lymph nodes from the field. Early randomized controlled trials showed a higher risk of coronary artery disease in patients treated with radiotherapy compared with patients treated only with surgery. However, the Danish Breast Cancer Cooperative Group 82b and 82c trials showed no increased risk of coronary artery disease in patients treated with radiotherapy. This may be explained by the use of electron fields, lesser heart volume irradiated, radiation protection blocks over the heart, and radiotherapy treatment planning using ultrasound measurement of chest wall thickness. Radiation protection blocks reduce the total cardiac dose to 15 Gy and the daily fraction size in patients with Hodgkin disease. Involved-node radiation therapy, which delivers radiotherapy to the involved nodes alone, reduces the average total heart dose compared with extended field (mantle) radiotherapy (7.7 vs 27.5 Gy). However, involved-node radiation therapy was still associated with considerable radiation dose to the proximal epicardial coronary arteries (18.2 and 16.2 Gy, respectively).

Fractionation of radiation doses may reduce the risk of cardiovascular complications. Pericarditis occurred more frequently in patients with Hodgkin disease when three 3.3 Gy fractions/week were used rather than four 2.5 Gy fractions/week. The reduction in fraction size coincided with the reduction in non–myocardial infarction cardiovascular mortality in patients with Hodgkin disease who received radiotherapy at a single center after 1972. However, a randomized trial for breast cancer comparing the current standard radiotherapy of 50 Gy given in 25 fractions (2 Gy/fraction) for 5 weeks with hypofractionated radiotherapy using a total dose of 42.5 Gy in 16 fractions (2.66 Gy/fraction) for 22 days reported no difference in cardiovascular mortality at 10 years. Although larger fractions of mediastinal radiotherapy for Hodgkin disease appear to be more detrimental to the heart, different fractionation does not appear to decrease cardiotoxicity in breast cancer.

Pericardial Disease

Increased vascular permeability and fluid extravasation and inflammatory cell infiltration were observed several months after irradiation and appear to coincide with the development of pericardial fibrin exudates and effusions. After 1 year, myocardial fibrosis, focal infarctions due to fibrointimal proliferation and occlusion, and coagulation necrosis were observed.

Pericardial involvement in patients treated with radiotherapy ranges from 70% to 100% and includes a wide spectrum of disease including fibrous thickening (65% to 94%), pericardial effusion with the potential for tamponade (57% to 75%), fibrinous pericardial adhesions (25%), acute pericarditis, pericardial constriction (36%), and obliterative pericardial fibrosis (25%). The onset of symptoms of acute pericarditis may occur immediately after radiotherapy to 2 years later and includes chest pain and a friction rub. Most of these patients recover, but a minority develop chronic pericarditis with effusion or pericardial constriction. Radiotherapy accounted for 13% to 31% of cases in a series of pericardial constriction. Symptoms may include pulmonary congestion or edema. Physical examination may often reveal elevation of jugular venous pressure, a pericardial knock, and hepatomegaly, especially in children. Electrocardiography may reveal low QRS voltage and T-wave abnormalities. The average time from irradiation to clinical presentation ranged from 14 to 88 months in several studies but can manifest >20 years later. Echocardiography or computed tomographic scanning may detect pericardial thickening. Invasive measurement of hemodynamics during cardiac catheterization may demonstrate impaired ventricular filling in mid and late diastole, leading to elevated end-diastolic pressure.

Treatment of acute pericarditis includes nonsteroidal anti-inflammatory drugs. Pericardiocentesis is the preferred treatment for patients symptomatic from large pericardial effusion. Pericardiectomy should be considered when extensive pericardial fibrosis significantly impairs diastolic filling. Five-year survival is significantly worse after pericardiectomy for radiation-induced pericardial constriction compared with nonradiation causes (64.3% vs 11.0%, p <0.001), possibly because of pancardiac involvement.

Pericardial Disease

Increased vascular permeability and fluid extravasation and inflammatory cell infiltration were observed several months after irradiation and appear to coincide with the development of pericardial fibrin exudates and effusions. After 1 year, myocardial fibrosis, focal infarctions due to fibrointimal proliferation and occlusion, and coagulation necrosis were observed.

Pericardial involvement in patients treated with radiotherapy ranges from 70% to 100% and includes a wide spectrum of disease including fibrous thickening (65% to 94%), pericardial effusion with the potential for tamponade (57% to 75%), fibrinous pericardial adhesions (25%), acute pericarditis, pericardial constriction (36%), and obliterative pericardial fibrosis (25%). The onset of symptoms of acute pericarditis may occur immediately after radiotherapy to 2 years later and includes chest pain and a friction rub. Most of these patients recover, but a minority develop chronic pericarditis with effusion or pericardial constriction. Radiotherapy accounted for 13% to 31% of cases in a series of pericardial constriction. Symptoms may include pulmonary congestion or edema. Physical examination may often reveal elevation of jugular venous pressure, a pericardial knock, and hepatomegaly, especially in children. Electrocardiography may reveal low QRS voltage and T-wave abnormalities. The average time from irradiation to clinical presentation ranged from 14 to 88 months in several studies but can manifest >20 years later. Echocardiography or computed tomographic scanning may detect pericardial thickening. Invasive measurement of hemodynamics during cardiac catheterization may demonstrate impaired ventricular filling in mid and late diastole, leading to elevated end-diastolic pressure.

Treatment of acute pericarditis includes nonsteroidal anti-inflammatory drugs. Pericardiocentesis is the preferred treatment for patients symptomatic from large pericardial effusion. Pericardiectomy should be considered when extensive pericardial fibrosis significantly impairs diastolic filling. Five-year survival is significantly worse after pericardiectomy for radiation-induced pericardial constriction compared with nonradiation causes (64.3% vs 11.0%, p <0.001), possibly because of pancardiac involvement.

Cardiomyopathy

Various cytokines and inflammatory cells play an important role in myocardial fibrosis after irradiation. Increased levels of type I and III collagens were observed in the ventricles, with a disproportionately higher type I collagen level in patients with pericarditis. Irradiation increases transforming growth factor-β and fibroblast growth factor-2 levels, which increase fibroblast activity and proliferation. Radiation-induced vascular injury may cause hypoxia, leading to late interstitial fibrosis. Radiation to the heart can cause endothelial damage when von Willebrand factor deposition occludes myocardial capillaries leading to myocardial fibrosis. Von Willebrand factor also invades the subendocardial layer, leading to myocardial ischemia, and the interstitium, where it adheres to the extracellular matrix, possibly leading to fibromuscular proliferation, migration, and myocardial fibrosis. Radiation also damages intercalated discs and inhibits cardiac mitochondrial respiration, resulting in elevated production of reactive oxygen species and myocardial dysfunction. The renin-angiotensin-aldosterone system is activated by oxidative stress, leading to the production of angiotensin II, a potent stimulator of fibrosis by way of upregulation of extracellular matrix proteins and transforming growth factor-β. Radiation stimulates the kallikrein-kinin system, which promotes inflammatory cell recruitment within the heart.

Radiotherapy can cause cardiomyopathy due to direct injury and myocardial fibrosis or ischemia. Histopathologic examination revealed myocytolysis in canines that developed congestive heart failure after exposure to mediastinal radiation. Autopsy studies revealed interstitial fibrosis of the myocardium in 50% to 63% of patients. Dyspnea is the primary symptom and magnetic resonance imaging may demonstrate myocardial fibrosis. Although the average ejection fraction was normal, patients with Hodgkin disease treated with radiotherapy had lower ejection fraction than that of healthy controls (56% vs 62%, p <0.05). Most patients (90%) were in the New York Heart Association class I, whereas the other 10% were in class II. Diastolic dysfunction secondary to restrictive cardiomyopathy may also develop after radiation because of endomyocardial fibrosis.

Patients with Hodgkin disease and breast cancer commonly receive both radiotherapy and anthracyclines. Mediastinal irradiation potentiates doxorubicin cardiotoxicity, even when they are administered during different time periods.

The treatment for congestive heart failure is similar to the conventional pharmacotherapy used in nonradiation-associated causes, including β blockers and angiotensin-converting enzyme inhibitors. There is a paucity of robust long-term data with orthotopic heart transplantation but it is an option in refractory cases especially if surgical revascularization or pericardiectomy is technically unfeasible because of severe diffuse coronary artery disease or severe fibrosis of the mediastinum is present. The Mayo Clinic reported no deaths in 4 patients at a mean follow-up time of 48 months. However, the Columbia University Medical Center reported a 33% perioperative mortality rate in 9 patients. Careful patient selection is warranted given the paucity of data and mixed results.

Valvulopathy

Ischemia is an unlikely cause of radiation-induced valvulopathy because of the lack of vasculature in valvular tissue. Aortic valve interstitial cells convert to osteoblast-like cells with increased expression of alkaline phosphatase, bone morphogenetic protein 2, and osteopontin when grown in a culture and exposed to 10 Gy of radiation. This likely represents a possible mechanism in the development of valvulopathy such as calcific aortic stenosis.

Autopsy often demonstrates diffuse fibrosis in all 4 valves. Valve thickening was observed in most patients (71% to 75%) treated with radiotherapy and can be detected even after 20 years. Most patients are asymptomatic. At the minimum, mild aortic regurgitation was present in 60% of patients, aortic stenosis in 16%, and mitral regurgitation in 52% ( Figures 1 and 2 ). Most patients (90%) have calcification of the aortic or mitral valves. Although regurgitation is more common than stenosis after radiotherapy, stenosis is more likely to be clinically relevant. The tricuspid and pulmonary valves are less commonly affected. Irradiation of the left atrium with >25 or >30 Gy to either ventricle increases the risk for valvular disease. Patients often develop symptoms of heart failure depending on the valve involved at an average of 22 years after radiation exposure.

A 42-year-old woman with a history of Hodgkin disease diagnosed in her early 20s was treated with radiotherapy, chemotherapy, and thoracotomy for debulking. Transesophageal echocardiography demonstrates aortic stenosis with a calculated aortic valve area of 0.9 cm² (A) and aortic regurgitation (B) . Cardiac magnetic resonance imaging shows poor leaflet opening of the aortic valve (C, D ) .

Transesophageal echocardiography of the same patient also demonstrates mitral stenosis (A) and mitral regurgitation (B) .

Valve replacement is the mainstay of treatment for severe valvulopathy. The perioperative (30-day) mortality rate is 12% and the 5-year survival rate is 66%. The presence of constrictive pericarditis was the highest independent predictor of perioperative mortality rate and was associated with a 30-day mortality rate of 40%. Fibrosis of the mediastinum (“frozen mediastinum”) can make the operation technically challenging and may explain the high mortality rate. Transcatheter aortic valve replacement may be preferred in patients with extensive mediastinal fibrosis or those who are poor surgical candidates. Transcatheter pulmonary valve replacement may also be a viable option for pulmonary stenosis in high-risk patients. Screening for valvular disease with echocardiography is advised at 10 years after radiotherapy.

Coronary Artery Disease

Endothelial injury and atherosclerosis may result when lysosomal enzymes within the intima and media are activated, leading to the formation of cholesterol plaques within days after exposure of coronary arteries to radiation as a result of increased endothelial permeability. These plaques are commonly inflammatory plaques, which contain infiltrates with macrophages and neutrophils with occasional plaque hemorrhages that are vulnerable to thrombosis. Radiation-induced arteriosclerosis and fibrosis of the intima, media, and adventitia are reported. Endothelial injury can also occur from the generation of reactive oxygen species. A prothrombotic state is created as radiation increases the adhesiveness of endothelial cells and promotes the release of von Willebrand factor from endothelial cells, decreasing the production of thrombomodulin. Radiation impairs endothelium-dependent vasodilation, which may contribute to small-vessel thrombosis.

Chest radiotherapy accelerates atherosclerosis, often leading to severe coronary artery disease. Although most patients who develop coronary artery disease have at least 1 cardiac risk factor, 38% of young patients (ranging from ages 15 to 33 years) who received >35 Gy to the heart had coronary artery stenosis >75% on autopsy. The distribution and extent of disease is related to the areas and dose of radiation delivered, respectively. Patients with left-sided breast cancer treated with radiotherapy have a higher risk of severe stenosis involving the mid and distal left anterior descending arteries and the distal diagonal branches. In patients who have received mediastinal radiotherapy, >75% of asymptomatic patients have coronary artery disease, with 30% of patients with severe multivessel disease, with the ostium of the left and/or right coronary arteries commonly involved ( Figure 3 ). Patients with Hodgkin disease who underwent mediastinal radiotherapy are at an increased risk of in-stent restenosis after percutaneous coronary intervention.

Coronary angiography of the same patient demonstrates diffuse atherosclerosis of the right coronary artery with a fractional flow reserve of 0.69 (A) . The left circumflex artery had a stenosis in the ostium (B) . The left anterior descending artery had diffuse atherosclerotic disease with a fractional flow reserve of 0.78 (C) . Angiography of the right brachiocephalic artery reveals a moderate stenosis (D) .

Angina is the most common clinical presentation but patients may also present with acute coronary syndrome and heart failure.

Electrocardiography and serum cardiac biomarkers such as troponin and creatinine kinase-MB are helpful when acute coronary syndrome is suspected. Noninvasive assessment of ischemia with myocardial perfusion imaging using technetium-99 m tetrofosmin has been validated in patients suspected of radiation-induced coronary artery disease. However, microvascular injury rather than epicardial artery disease may result in nonreversible defects on perfusion imaging or lesions that do not correspond to coronary artery territories. Computed tomography angiography with or without coronary artery calcium scoring is another technique to diagnose coronary artery disease. The gold standard for the diagnosis and localization of coronary artery disease remains coronary angiography.

The American College of Cardiology/American Heart Association guidelines for the management of stable coronary artery disease and acute coronary syndromes do not provide specific recommendations regarding radiation-induced coronary artery disease. Percutaneous and surgical revascularizations are reasonable options. Coronary artery bypass surgery may be more technically challenging because of mediastinal fibrosis and the potential for radiation-induced injury of the internal mammary artery, rendering it a suboptimal conduit for surgical revascularization. However, injury of the internal mammary artery was not detected in the 125 patients who were previously treated with radiotherapy.

Current guidelines have no recommendations regarding the screening for coronary artery disease in asymptomatic patients who have been exposed to radiotherapy. Proposed recommendations include screening asymptomatic patients who are aged ≥45 years 5 years after mediastinal radiotherapy ≥35 Gy or 10 years after radiotherapy with computed tomography angiography, coronary artery calcium scoring, stress electrocardiography, nuclear perfusion imaging, or stress echocardiography. Patients should also be closely monitored with modification of traditional risk factors such as hypertension, dyslipidemia, diabetes, and smoking. However, continued cancer surveillance is absolutely mandatory given that these patients usually die from secondary or tertiary tumors.