Insurance status may impact on survival from malignant cardiac tumours

Cardiac neoplasms may develop as either a primary malignant disease or as metastases from an extra-cardiac location1. The incidence of cardiac malignancies is very low and only about 10% of primary cardiac tumours are malignant2. The most frequent benign cardiac tumour is myxoma, while the majority of primary malignant cardiac tumors (PMCTs) are sarcomas, consisting of various histological subtypes3. In general, an aggressive biological behaviour is typical of primary cardiac tumours4, and is associated with tumour size and location5. Survival outcomes remain poor even if detected early and aggressive surgical resection is the main primary curative option6,7. Survival is only around 10% at one year for tumours managed without surgery8,9. To date, the combination of surgery with systemic chemotherapy remains the best treatment for malignant cardiac tumours9. No consensus currently exists concerning the value of peri-operative chemotherapy and radiotherapy10,11,12. Palliative chemotherapy should be considered for patients with inoperable or metastatic disease.

In the current study we aimed to assess long-term overall survival (OS) differences based on insurance status in patients with malignant cardiac tumours using the National Cancer Database (NCDB). The NCDB is an oncology database that is sponsored by the American Cancer Society and the American College of Surgeons. We included data from the NCDB from 2004 to 2017. Overall survival and operative mortality were the primary and secondary outcomes, respectively.

Our study cohort included 699 patients that were stratified by insurance type: 412 (58.9%) had private insurance, 243 (34.8%) had governmental insurance and 44 (6.3%) were uninsured. Overall, operative mortality was 8.5%: 11.1% in the Uninsured, 2.6% in Medicaid, 13.9% in Medicare and 7% in Private Insurance/Managed Care groups, respectively (p=0.09). Histopathological details of the included patients can be seen in Table 1. Angiosarcoma, leiomyosarcoma and fibrosarcoma were the most prevalent variants in this cohort. Figure 1 describes the type of insurance over the considered timeframe. Private insurance/manages was always the most common form of insurance.

At univariable Cox regression those who had Medicaid showed a 28% mortality risk reduction (p=0.036) and those with private insurance had a 46% mortality risk reduction (p<0.001), suggesting a significantly better survival when compared to Medicare after median follow-up of 64.1 months. At multivariable Cox regression, higher comorbidity index showed a 92% increase in mortality risk (p<0.001), and angiosarcoma and cancer stage III/IV showed 33.8% and 34% increases in mortality risk (p=0.002 and p=0.003), respectively, while private insurance was associated with a 33% mortality risk reduction (p=0.002), year of diagnosis with a 4% mortality risk reduction in recent years (p<0.001), and surgery and chemotherapy with 57% and 40% mortality risk reduction (p<0.001 and p<0.001) respectively.

The main limitation of the study lies in the possibility of unknown confounders bias due to its retrospective nature.

Presenting symptoms were not analyzed. However, a recent meta-analysis2 showed that the most commonly presenting symptom in patients affected by cardiac tumours was dyspnoea. A previous study also reported heart failure as the most frequent symptom13. Over 60% of the current patient population underwent surgery, which has been established as the most effective treatment for PMCTs3. Sarcomas are not highly radiosensitive and this may explain the fact that less than 20% of our cohort received radiation therapy14. Most of the present literature is composed of limited single-centre studies, excluding the large national databases, confirming the rarity of data regarding this topic. Hence, more consistent data are warranted and essential in understanding the clinical evolution of this disease.

A recent study15 reported a significant increase in early mortality following cardiac valve surgery in uninsured, Medicaid and Medicare patients compared with patients with private insurance. Opposing opinions exist regarding the influence of insurance status on different outcomes among various cancers16. Another series reported that short-term mortality was affected by insurance status in major surgical operations17, however in patients with cardiac tumors this difference was not statistically significant.

In conclusion, this study provides increased awareness of insurance status survival disparity among patients with PMCTs, however further studies are warranted to explore in depth this association.