Cardiovascular Mortality Risk in Patients with Bladder Cancer: A Population-Based Study

Background: The purpose of this study was to evaluate the risk of cardiovascular mortality (CVM) among patients with bladder cancer (BC). Methods and Materials: Data were collected from the Surveillance, Epidemiology, and End Results (SEER) database for patients who were diagnosed with BC by pathology between 2000 and 2016. The standardized mortality rate (SMR) was calculated based on reference data from the general population. Nelson–Aalen cumulative hazard curves were used to assess the risk of experiencing CVM in BC patients. Multivariate competing risk models were performed. Results: In total, data from 237,563 BC patients were obtained from the SEER database for further analysis, of which 21,822 patients experienced CVM; the overall SMR for CVM in BC patients was 1.16 (95% CI: 1.14–1.17). Age, race, sex, year of diagnosis, histologic type, summary stage, surgery, marital status, and college education level were independent predictors of CVM in patients with BC. Conclusions: Patients with BC have a significantly increased risk of experiencing CVM compared to the general population. Pre-identification of high-risk groups and cardiovascular protection interventions are important measures to effectively improve survival in this group of patients.


Introduction
It is estimated that there are more than 500,000 new cases of bladder cancer (BC) worldwide each year, with approximately 40% of these resulting in death. However, the US alone accounts for 16% of all new cases worldwide each year [1][2][3].
Cardiovascular diseases (CVDs) are one of the primary causes of death worldwide. According to a study published in The Lancet in 2018, cardiovascular mortality (CVM) increased by 2.1% over a 10-year period from 2007 [4]. A total of 17.9 million people died from CVD in 2019, contributing to 32% of all deaths worldwide, while only about 10 million people died from cancer in 2019. In 2019, Kochanek et al. reported that in the US, over 0.64 million deaths were due to heart disease, while nearly 0.6 million deaths were attributed to primary cancers [5].
With the improvement in the quality of life and medical care, patients' life expectancy has increased, and mortality from primary cancers has gradually decreased, leading to the prominence of mortality factors for non-primary cancers, among which CVD is one of the leading causes of death for non-cancers [6]. Several published studies have indicated that cancer patients are at significantly higher risk of experiencing CVM than the general population for a variety of reasons [7][8][9].
A detailed literature search revealed no reports of CVM in patients with BC. Therefore, our discoveries may help to establish a more targeted follow-up strategy for BC patients as well as more effective CVM prevention measures.
A detailed literature search revealed no reports of CVM in patients with BC. Therefore, our discoveries may help to establish a more targeted follow-up strategy for BC patients as well as more effective CVM prevention measures.
CVM was the primary endpoint of interest, defined by the following six CVDs in the SEER database: (1) diseases of the heart, (2) hypertension without heart disease, (3) cerebrovascular diseases, (4) atherosclerosis, (5) aortic aneurysm and dissection, (6) other diseases of the arteries, arterioles, and capillaries [10,11], while competing events were deaths from BC, other cancers, and other non-cancer diseases.  Patients identified only by autopsy or death certificate and patients with incomplete data for certain variables (age, sex, race, etc.) were excluded.
CVM was the primary endpoint of interest, defined by the following six CVDs in the SEER database: (1) diseases of the heart, (2) hypertension without heart disease, (3) cerebrovascular diseases, (4) atherosclerosis, (5) aortic aneurysm and dissection, (6) other diseases of the arteries, arterioles, and capillaries [10,11], while competing events were deaths from BC, other cancers, and other non-cancer diseases.

Statistical Analysis
The SMR is the ratio of the number of observed deaths to the number of expected deaths of CVM [12]. We used an exact method to calculate the 95% confidence intervals (95% CIs) for all SMRs. Absolute excess risks (AERs) were also calculated, which are a proxy for the excess number of deaths per 10,000 person-years in different subgroups [10,12]. The Nelson-Aalen cumulative hazard curve was used to assess the risk of experiencing CVM in different subgroups of BC patients. Multivariate competing risk analyses were conducted to identify risk factors associated with CVM [13].

SMR and AER
The overall SMR for CVM was 1.16 (95% CI: 1.14-1.17), and the AER was 27.18/ 10,000 person-years in BC patients. Table 1 shows the baseline features and SMRs of CVM in patients with BC.

SMR and AER
The overall SMR for CVM was 1.16 (95% CI: 1.14-1.17), and the AER was 27.18/10,000 person-years in BC patients. Table 1 shows the baseline features and SMRs of CVM in patients with BC.     Figure 3A shows that the SMR of all causes of death of BC patients increased year by year, with BC having the highest SMR of all. Figure 3B,C show the SMR in CVD stratified by sex and all causes of CVM. Figure 4A shows that the SMR of all causes of death of BC patients decreased with increasing follow-up time. Figure 4B,C show the SMR for CVD stratified by sex and all causes of CVM.  Figure 3A shows that the SMR of all causes of death of BC patients increased year by year, with BC having the highest SMR of all. Figure 3B,C show the SMR in CVD stratified by sex and all causes of CVM.   Figure 4A shows that the SMR of all causes of death of BC patients decreased with increasing follow-up time. Figure 4B,C show the SMR for CVD stratified by sex and all causes of CVM.   Figure 5 illustrates the risk of CVM with increasing follow-up time for different factors, with the following subgroups associated with a higher risk of CVM: age over 71 years, male sex, White, carcinoma in situ and localized tumors, no surgery, Tcc and Nec pathological types, a college education level less than 50%, widowed, and a median household income less than USD 100,000. Figure 5 illustrates the risk of CVM with increasing follow-up time for different factors, with the following subgroups associated with a higher risk of CVM: age over 71 years, male sex, White, carcinoma in situ and localized tumors, no surgery, Tcc and Nec pathological types, a college education level less than 50%, widowed, and a median household income less than USD 100,000.  Figure 6A shows that the risk of all mortality factors in BC patients increased with the time to follow-up. Figure 6B shows that the risk of the six factors contributing to CVM increased progressively with the time to follow-up.  Figure 6A shows that the risk of all mortality factors in BC patients increased with the time to follow-up. Figure 6B shows that the risk of the six factors contributing to CVM increased progressively with the time to follow-up.    The risk of BCrelated death was highest in patients aged less than 70 years, followed by other cancers, other non-cancer diseases, and CVDs ( Figure 7A-C). In patients aged 71-80 years, the risk of CVD surpassed that of other non-cancer diseases and ranked third when the follow-up time exceeded 120 months ( Figure 7D). In patients older than 81 years, the risk of CVD-related mortality surpassed that of other cancers as the second leading risk factor at approximately 80 months of follow-up ( Figure 7E).

Predictors of Death from CVD
Risk factors associated with CVM in BC patients were identified using multivariate competing risk regression analysis (

Predictors of Death from CVD
Risk factors associated with CVM in BC patients were identified using multivariate competing risk regression analysis (  Figure 8 shows the CIF curves using Fine-Gray competing risk analyses.

Discussion
In this large population study based on the SEER database, we analyzed the longterm CVM of patients with BC. Although the number of diagnoses of BC is increasing every year, the number of people who develop CVM is decreasing. This may be related to the advancement of BC treatment strategies and the improved quality of comprehensive cancer management. At the same time, the treatment of CVD and the ability to cope with cardiovascular events have also improved, which has effectively reduced the incidence of CVM.
Published studies have illustrated that the risk of CVM varies considerably between patients with cancer at different primary sites [14][15][16]. In this study, we focused only on CVM in patients with BC. By studying 21,822 patients, we found that the risk of CVM in patients with BC was approximately 16% higher than that in the US general population (SMR [95% CI]: 1.16 [1.14-1.17]). Over the entire follow-up period, patients with BC had an increased risk of CVM from all causes. Our study identified age, race, sex, year of diagnosis, histologic type, summary stage, surgery, marital status, college education level, and median household income as independent predictors for the development of CVM in patients with BC.

Discussion
In this large population study based on the SEER database, we analyzed the long-term CVM of patients with BC. Although the number of diagnoses of BC is increasing every year, the number of people who develop CVM is decreasing. This may be related to the advancement of BC treatment strategies and the improved quality of comprehensive cancer management. At the same time, the treatment of CVD and the ability to cope with cardiovascular events have also improved, which has effectively reduced the incidence of CVM.
Published studies have illustrated that the risk of CVM varies considerably between patients with cancer at different primary sites [14][15][16]. In this study, we focused only on CVM in patients with BC. By studying 21,822 patients, we found that the risk of CVM in patients with BC was approximately 16% higher than that in the US general population (SMR [95% CI]: 1.16 [1.14-1.17]). Over the entire follow-up period, patients with BC had an increased risk of CVM from all causes. Our study identified age, race, sex, year of diagnosis, histologic type, summary stage, surgery, marital status, college education level, and median household income as independent predictors for the development of CVM in patients with BC.
Similar to the results previously published by Zaorsky et al. [17], we found that the risk of CVM in BC patients was highest in the first 10 months after diagnosis. Meanwhile, the study by Fang and Ye et al. suggested that newly diagnosed cancer may lead to psychological and emotional distress in cancer patients, which in part promotes the development of CVM [18,19]. Therefore, psychiatric assessment and psychological support are necessary for newly diagnosed BC patients. Using the Nelson-Aalen hazard curves, we found that the risk of CVM in BC patients gradually increased with increasing age at diagnosis. Primary cancer is the most common cause of death for most cancer patients. However, our study found that the risk of CVM ranked first when cancer patients were diagnosed at an age of >71 years. These results suggest that clinicians should focus not only on BC itself, but also on the risk factors for experiencing CVM in patients of advanced age.
Multivariate competing risk regression analysis was used to identify risk factors associated with CVM in BC patients. We also observed that patients with BC [17]. Men have a higher risk of CVM, which may be related to smoking, alcohol consumption, or higher work stress, all of which are independent risk factors for CVD [20][21][22][23] BC patients are at a higher risk of CVM, which may be associated with the fact that married patients are more likely to receive encouragement and support from their spouses, both emotionally and physically [24]. Additionally, some studies pointed out that marriage helps improve cardiovascular, endocrine, and immune function as well as cancer prognosis [25,26]. Patients with a lower socioeconomic status have been reported to be at higher risk for CVM [12,27], and our findings demonstrate that patients with lower levels of college education had a higher risk of CVM, in accordance with previous results.
In . This is probably explained by the fact that most patients who underwent RC surgery had an advanced tumor stage and did not have enough life expectancy to experience a CVM event (median survival time: 28 months for TURBT, 20 months for PC, 16 months for RC, and 11 months for no surgery). Our results show that patients with BC who are not treated with surgery have the highest risk of CVM, although the median survival time was only 11 months. One possible reason is that the diagnosis of BC often causes a longer period of psychological and emotional distress, and untreated BC often progresses rapidly, which would result in patients being at a higher risk of experiencing CVM [28,29].
Many risk factors are shared between cancer and CVD, such as smoking, radiation, air pollution, and metabolic syndrome [30]. Recent studies have shown that there is also a direct interplay between cancer and CVD, with anthracyclines having significant cardiotoxic effects that can lead to CVD such as heart failure or atherosclerosis during or years after anticancer treatment [31][32][33]. In addition to risk factors, the genetic background plays an important role in the interaction between cancer and CVD [30]. Studies have shown that mutations in age-related clonal hematopoiesis of indeterminate potential accelerate the development of CVD such as atherosclerosis or coronary artery disease [34][35][36], further worsening the prognosis of patients with heart failure [37]. Recent studies have demonstrated that heart failure can promote the transition to the pre-tumor stage and tumor growth [38][39][40]. In cancer patients, there is a high incidence of cardiomyopathy-related mutations, such as the DNA damage response/repair system, and mutations in the DNA damage response/repair system gene increase the risk of cardiotoxicity with anticancer therapy [41]. Meanwhile, pathophysiological alterations in hereditary cardiomyopathy can promote cancer development and progression and may further increase the cardiotoxic effects of anticancer therapies [30,42,43].
There are still some shortcomings in our study. First, information related to CVD, such as smoking, alcohol consumption, and the presence of congenital diseases, was not recorded in the SEER database. Second, there was no further analysis of the effects of chemotherapy, radiotherapy, and some other new therapeutic strategies on CVM. Meanwhile, some studies reported that the causes of CVM on death certificates might have been overestimated [44], which might have affected the accuracy of our study to some extent.

Conclusions
In summary, patients with BC have a significantly increased risk of developing CVM compared to the general population. This suggests that early screening for CVD and the assessment and monitoring of risk factors for CVM should be performed after the diagnosis of BC. It also provides important guidance on how BC patients should be followed up and educated about their associated health risks. In addition, further investigations are needed to understand the mechanisms by which BC patients develop CVD, and to design effective prevention and monitoring strategies.