Long-Term Risk of Cardiac Events Requiring Interventions in Survivors of Hodgkin's Lymphoma (HL) Treated With Mediastinal Irradiation

Article Outline

• Purpose/Objective(s)
• Materials/Methods
• Results
• Conclusions
• Copyright

Back to Article Outline

Purpose/Objective(s) 

To assess the incidence and actuarial risk of clinically-significant cardiac disease requiring intervention in HL patients treated with mediastinal irradiation.

Back to Article Outline

Materials/Methods 

Between 1968–1998, 1279 HL patients were treated with mediastinal irradiation at our institution. To quantify and analyze myocardial infarctions (MI) and procedure-based interventions, we retrospectively reviewed hospital records and patient questionnaires. Cardiac events were defined as: coronary artery disease (MI, percutaneous transluminal coronary angioplasty [PTCA] with or without stenting, coronary artery bypass graft surgery [CABG] or 75% stenosis on angiogram); arrhythmia (pacemaker, automatic implantable cardioverter defibrillator [AICD] implantation or significant disturbance documented by electrocardiogram or Holter monitoring); valve disease (valve repair or moderate or severe valve dysfunction on echocardiogram or angiogram); or pericardial disease (pericardial surgery or pericardiocentesis). In the estimation of incidence and actuarial risk over time, we focused on MI and procedure-based cardiac interventions. Multiple logistic regression models were used to explore factors associated with development of all cardiac complications.

Back to Article Outline

Results 

187 patients (15%) were identified as having a cardiac event. Median age at treatment was 25 years (range, 3–93). Median follow-up time for this cohort was 17.2 years (range, 1–41.5 years). 87% of patients had ≥5 years, 67% had ≥10 years and 48% had ≥15 years of follow-up. The cumulative incidence rates (adjusted for competing mortality) for all cardiac events were 2.1%, 4.5%, 9.6% and 15.4% at 5, 10, 15 and 20 years. Incidence and actuarial risk for MI and the four most common cardiac interventions are listed in Table 1. Older age at diagnosis (age 40–50 vs. age ≤20, p < 0.0001), longer follow-up (10–15 years vs. 15–20 years, p = 0.007) and male gender (p = 0.006) were independent predictors for cardiac events.

Back to Article Outline

Conclusions 

Increased risk of cardiac disease after mediastinal irradiation in HL survivors has been described in multiple studies. In this study, we specifically quantified cardiac events requiring interventions over time in this population. The results may help in guideline development for both the types and the timing of cardiac surveillance studies in survivors of HL.