Discussion
This study shows that pericardiectomy for constriction in patients with prior mediastinal irradiation is associated with poor long-term survival. The 1, 5- and 10-year survival was 74%, 53% and 32% respectively. Early mortality, though high (~10%) for the entire period, showed a modest, yet statistically nonsignificant, decreasing trend over time, with mortality rates over the second half of the period (approximately 4%; 2 deaths in 51 surgeries) being more comparable to those with non-radiation related constriction.
Radiation induced pericardial disease is one of the most common manifestations of RACD. This typically manifests itself several years after mediastinal irradiation. In the present study, the median duration from radiation to pericardiectomy was 28.5 years (IQR 20-34 years). This was a mean of 53 months as reported by Siefert and colleagues18, and a median of 11 years (range 2 to 30 years) as reported by Bertog and colleagues.11 Ling et al reported a median range of 13 years7, and McCaughan et al reported a range of 15 months to 44 years.19This may also reflect improved cancer cure rates, which have resulted in greater longevity and thus the likelihood of developing delayed cardiovascular sequelae.7 In previous studies, lymphomas have been the most common malignancy for which mediastinal radiation was received, followed by breast cancer.7, 11, 18, 19 Our experience has been similar, with lymphomas constituting 69% of the patients and breast cancer a further 15%.
Clinical manifestation of pericardial disease may vary in severity from asymptomatic pericardial effusion to debilitating constrictive pericarditis.3, 4, 20 Fibrosis is a common sequela of radiotherapy and leads to fibrous thickening of the pericardium. In addition to pericardial fibrosis, there is myocardial fibrosis with possible endocardial fibrosis as well.21 Brosius and colleagues found interstitial myocardial fibrosis in patients who had received prior irradiation.21 The pathogenesis of this myocardial fibrosis has been related to capillary endothelial cell injury, leading to a quantitative loss of capillaries, leading to ischemia.22 It is perceived that the increased filling pressures in these patients may be in part due to increased myocardial stiffness, which may not be relieved even with an adequate pericardial resection.
Pericardiectomy is also technically challenging in this subset of patients. The plane of dissection may not be easily accessible due to the dense mediastinal fibrosis. In the present study, a radical resection could be performed in only 50% of the patients, whereas 47% patients received a subtotal resection. This contrasts with previous reports from this institution where radical pericardiectomy was performed on 77% of patients presenting with constriction for causes other than radiation.10 Another indicator of the technical difficulty could be the use of cardiopulmonary bypass (CPB) in 96% of these patients. A previous report from this institution documented the use of CPB to be 51% in pericardiectomies performed for constriction for causes other than radiation.10 This maybe a reflection of the surgeon deeming it unsafe to perform a radical resection in some patients due to the degree of difficulty of the procedure and having a low threshold of going on CPB. Our experience was corroborated by the findings of Ni and colleagues, who concluded that a complete pericardiectomy may not be achieved in all patients with post-radiation constrictive pericarditis.13
Mediastinal irradiation damages the vascular endothelium and leads to radiation-induced vascular disease of peripheral, coronary and carotid arteries.23 When the coronary vasculature is involved, these patients may present with the classical features of coronary artery disease; however, some may also remain asymptomatic.24, 25 Valvular disease is frequently seen in this subset of patients as well. The aortic and mitral valves are more commonly involved, and the affected valves are typically fibrotic with focal dystrophic calcification and marked thickening.21, 26, 27 In the present study too, isolated pericardiectomy was performed in only 27% of the patients, with most of the patients having concomitant valve surgery (46%), coronary artery bypass grafting (4%), or a combination of both procedures (23%). This further illustrates the pan-cardiac involvement in these patients, further adding to the degree of complexity and justifies the higher need for CPB in this population.
Pericardiectomy for post-irradiation constrictive pericarditis has been reported to have high operative mortality with poor late results. Ni and colleagues collected data from a series of 46 patients across literature and reported an operative mortality of 22%.13 Bertog and colleagues also reported an operative mortality of 21% in this subset of patients in their large series on pericardiectomy for constriction.11 The general belief is that the effects of radiation on other cardiac structures, including coronary vasculature, valvular abnormalities and the associated myocardial fibrosis all contributes to the increased peri-operative mortality seen in this subset. In the present study, the operative mortality was 10%, which is higher than the 2.5% operative mortality reported by the same institution for pericardiectomy for constriction performed for etiologies other than mediastinal irradiation.10However, our data suggest an improvement over time, based on a decreasing rate in operative mortality that trended toward significance and an observed rate of 4% over the second half of the study. This maybe attributed to improved surgical and perfusion techniques, post-operative care, and better patient selection. Noteworthy, pericardiectomy was still associated with significant morbidity, with 13% of patients requiring post-operative dialysis.
The long-term outcomes of this patient population remain poor, with a 1, 5- and 10-year survival of 73%, 53% and 32% respectively. Previous reports from this institution have shown a 1, 5- and 10- year survival of approximately 87%, 77% and 58% respectively, for pericardiectomy performed for all other etiologies of constriction.10The only risk factor associated with increased mortality was advanced age which may reflect to medical complexities with high prevalence of comorbidities (coronary artery disease, valvular disorders, renal dysfunction). Despite the poor long-term outcomes, whether pericardiectomy in this population of patients is associated with significant improvements in functional capacity and quality of life requires further investigation.