2020 Volume 84 Issue 10 Pages 1771-1778
Background: The complex clinical status of heart failure (HF) patients with concomitant cancer is gaining clinical interest. This study sought to explore the prevalence of cancer in patients with HF and its effect on outcomes using a nationwide inpatient database.
Methods and Results: In total, 447,818 HF patients who were admitted and discharged between January 2010 and March 2018 were studied and included in the Diagnosis Procedure Combination (DPC) database. The median age was 81 years; 238,192 patients (53.2%) were men and 25,951 (5.8%) had concomitant cancer. The prevalence of cancer peaked in patients aged in their 70 s and 80 s and increased with time. Patients with cancer were older and more likely to be male. Cigarette smoking was more common in patients with cancer. Patients with cancer more frequently had infectious complications during hospitalization. Advanced medical procedures were less frequently performed for patients with cancer. In-hospital mortality was higher in patients with cancer than those without (10.0% vs. 6.7%, P<0.001). Among patients with cancer, in-hospital mortality was higher in patients with metastasis than those without (18.9% vs. 9.4%, P<0.001). Multivariable logistic regression analysis, fitted with a generalized estimating equation, indicated cancer is associated with higher in-hospital mortality (odds ratio 1.51, 95% confidential interval 1.43–1.59, P<0.001).
Conclusions: Cancer was frequently observed in patients hospitalized for worsened HF, and its prevalence increased with time. The presence of cancer increased the risk of in-hospital death. Further studies are warranted to establish the optimal management strategy for HF patients with cancer in the field of cardio-oncology.
Cardio-oncology is an emerging field, and cancer complicating heart failure (HF) is a topic of increasing interest in this field.1 The populations of developed countries are aging rapidly. The incidence of both HF and cancer increases with age.2 Further, HF and cancer have several shared risk factors, such as cigarette smoking, obesity, and diabetes.3–5 Because the mortality due to cancer continues to decline due to prevention, early detection, and advances in treatment,6,7 the clinical significance of cardiovascular morbidity in cancer survivors will rise. Cancer survivors are susceptible to various cardiovascular diseases including HF for a variety of reasons; for example, chemotherapy, and radiation therapy.8 The number of HF patients with concomitant cancer is, therefore, likely to increase further, and this is an active area for scientific research. However, large-scale epidemiological data on the prevalence and prognostic effect of cancer in patients with HF is limited. In this study, we sought to explore the prevalence of cancer in patients hospitalized with HF, and further aimed to clarify the influence of cancer on in-hospital clinical outcomes in them using a nationwide inpatient database.
Editorial p 1689
We performed a retrospective cohort study using the Diagnosis Procedure Combination (DPC) database, a nationwide inpatient database in Japan.9,10 The database includes administrative claims data and clinical data for approximately 8 million hospitalized patients per year from participating hospitals including all 82 academic hospitals. The main diagnosis, comorbidities on admission, and complications during hospitalization are recorded using the International Classification of Disease and Related Health Problems 10th Revision (ICD-10) codes. We also obtained the information on cancer and the affected sites from the ICD-10 codes recorded during hospitalization. Our DPC database includes estimated total costs on the basis of reference prices in the Japanese national fee schedule that determine item-by-item prices for surgical, pharmaceutical, laboratory, nursing care and other inpatient services.
Study PopulationWe studied 466,921 patients aged ≥20 years with NYHA class ≥II, admitted and discharged between January 2010 and March 2018 with the main discharge diagnosis of HF defined by ICD-10 codes I50.0, I50.1, and I50.9. Exclusion criteria were as follows: (1) length of hospital stay ≤2 days (n=15,270); and (2) major procedures performed under general anesthesia (n=3,833). The final number of patients analyzed in this study was 447,818.
EthicsThis study was approved by the Clinical Research Review Board of The University of Tokyo [3501-(3)] on 25 December 2017. We conducted this study in accordance with the Declaration of Helsinki. Because of the anonymous nature of this database, the requirement for informed consent was waived. The opt-out procedure was conducted at each participating institute.
Statistical ProcedureContinuous and categorical data are presented as median (interquartile range) and number (percentages), respectively. We compared continuous data using the unpaired t-test. We performed the chi-squared test to compare categorical variables. We evaluated the serial changes in the prevalence of cancer using the Cochran-Armitage trend test. The association of cancer with in-hospital mortality of patients hospitalized for HF was evaluated using a multivariable logistic regression analysis with adjustment for other patient background information and medications used within 2 days after admission, while also adjusting for within-hospital clustering using a generalized estimating equation.11 We compared in-hospital mortality between patients with and without cancer in men and women separately. A probability value of <0.05 was considered to indicate a statistically significant difference. We performed statistical analysis using SPSS software (version 25; SPSS Inc., Chicago, IL, USA) and STATA (version 16; StataCorp LLC, College Station, TX, USA).
Cancer was diagnosed in 25,951 patients (5.8%) (ICD-10 codes for cancer are shown in the Supplementary File). The prevalence of cancer peaked in patients aged in their 70 s and 80 s (Figure 1A) and the prevalence of cancer increased with time (Figure 1B). Among the study population, 25,951 patients had a total of 28,342 diagnoses of cancers. Main sites of cancers are summarized in Figure 2.
Prevalence of cancer. Prevalence of cancer by age (A). Serial change in prevalence of cancer (B).
Main sites of cancers in patients with heart failure.
The characteristics of the study population are shown in Table 1. The median age was 81 years, and 238,192 patients (53.2%) were men. The patients with cancer were older, more likely to be male and smokers were more common in this group. Barthel Index was higher in patients with cancer than in those without. Medications administered within 2 days of admission were also summarized in Table 1.
| Cancer absent (n=421,867) |
Cancer present (n=25,951) |
P value | |
|---|---|---|---|
| Age (years) | 81 (72–87) | 81 (75–86) | <0.001 |
| ≥80 years | 227,967 (54.0) | 15,061 (58.0) | <0.001 |
| Male sex | 221,119 (52.4) | 17,073 (65.8) | <0.001 |
| Body mass index (kg/m2) | 22.1 (19.6–25.0) | 21.8 (19.4–24.5) | <0.001 |
| Hypertension | 284,271 (67.4) | 16,804 (64.8) | <0.001 |
| Diabetes mellitus | 132,746 (31.5) | 8,273 (31.9) | 0.164 |
| Chronic renal failure | 61,276 (14.5) | 4,227 (16.3) | <0.001 |
| Chronic liver disease | 15,781 (3.7) | 1,736 (6.7) | <0.001 |
| Chronic respiratory disease | 47,012 (11.1) | 3,729 (14.4) | <0.001 |
| Smoking | 133,630 (31.7) | 10,168 (39.2) | <0.001 |
| Myocardial infarction | 11,908 (2.8) | 609 (2.3) | <0.001 |
| Dilated cardiomyopathy | 32,402 (7.7) | 1,724 (6.6) | <0.001 |
| New York Heart Association | <0.001 | ||
| Class II | 122,118 (28.9) | 7,697 (29.7) | |
| Class III | 161,195 (38.2) | 10,244 (39.5) | |
| Class IV | 138,554 (32.8) | 8,010 (30.9) | |
| Barthel Index | 60 (10–100) | 65 (20–100) | <0.001 |
| Medications administered within 2 days of admission | |||
| β-blocker | 140,223 (33.2) | 8,163 (31.5) | <0.001 |
| Renin-angiotensin system inhibitor | 157,730 (37.4) | 8,628 (33.2) | <0.001 |
| Angiotensin-converting enzyme inhibitor | 67,860 (16.1) | 3,852 (14.8) | <0.001 |
| Angiotensin II receptor blocker | 92,930 (22.0) | 4,936 (19.0) | <0.001 |
| Mineralocorticoid receptor antagonist | 137,598 (32.6) | 8,017 (30.9) | <0.001 |
| Intravenous inotropic agent | 73,868 (17.5) | 4,368 (16.8) | 0.005 |
| Intravenous nitrate | 89,098 (21.1) | 4,467 (17.2) | <0.001 |
| Intravenous furosemide | 287,246 (68.1) | 17,426 (67.1) | 0.002 |
Data are expressed as median (interquartile range) or n (%).
The complications and procedures during hospitalization are summarized in Table 2. Pneumonia and sepsis occurred more frequently in patients with cancer. Occurrence rates of pneumonia, urinary tract infection, and sepsis by cancer site are summarized in Figure 3. Respiratory support, intubation, hemodialysis, and intra-aortic balloon pumping were less frequently performed in patients with cancer. Length of hospital stay was longer, and in-hospital mortality was higher in patients with cancer. In-hospital mortality by cancer site is summarized in Figure 4. Metastasis (ICD-10 codes of C78 or C79) was recorded in 1,689 patients among 25,951 patients with cancer (6.5%). Among patients with cancer, in-hospital mortality was higher in patients with metastasis than those without (18.9% vs. 9.4%, P<0.001).
| Cancer absent (n=421,867) |
Cancer present (n=25,951) |
P value | |
|---|---|---|---|
| Medications | |||
| β-blocker | 257,481 (61.0) | 15,117 (58.3) | <0.001 |
| Renin-angiotensin system inhibitor | 254,914 (60.4) | 14,348 (55.3) | <0.001 |
| Angiotensin-converting enzyme inhibitor | 112,874 (26.8) | 6,610 (25.5) | <0.001 |
| Angiotensin II receptor blocker | 154,775 (36.7) | 8,432 (32.5) | <0.001 |
| Mineralocorticoid receptor antagonist | 222,207 (52.7) | 13,375 (51.5) | <0.001 |
| Intravenous inotropic agent | 160,236 (38.0) | 9,473 (36.5) | <0.001 |
| Intravenous nitrate | 145,152 (34.4) | 7,320 (28.2) | <0.001 |
| Intravenous furosemide | 305,924 (72.5) | 18,859 (72.7) | 0.587 |
| Procedures | |||
| Respiratory support | 66,770 (15.8) | 3,884 (15.0) | <0.001 |
| Intubation | 11,759 (2.8) | 605 (2.3) | <0.001 |
| Hemodialysis | 10,838 (2.6) | 547 (2.1) | <0.001 |
| Intra-aortic balloon pumping | 3,072 (0.7) | 103 (0.4) | <0.001 |
| Extra-corporeal membrane oxygenation | 429 (0.1) | 21 (0.1) | 0.305 |
| Complications | |||
| Deep vein thrombus | 1,116 (0.26) | 85 (0.33) | 0.057 |
| Pulmonary embolism | 333 (0.08) | 28 (0.11) | 0.111 |
| Pneumonia | 11,364 (2.7) | 884 (3.4) | <0.001 |
| Urinary tract infection | 4,983 (1.2) | 300 (1.2) | 0.716 |
| Sepsis | 2,315 (0.5) | 196 (0.8) | <0.001 |
| Outcomes | |||
| Medical cost (JPY) | 735,720 (498,150–1,159,400) | 770,950 (519,710–1,220,280) | 0.071 |
| Medical cost (USD) | 6,695 (4,533–10,551) | 7,016 (4,729–11,105) | 0.071 |
| Length of hospital stay (days) | 17 (11–27) | 19 (12–30) | <0.001 |
| In-hospital death | 28,219 (6.7) | 2,603 (10.0) | <0.001 |
Data are expressed as median (interquartile range) or n (%).
Occurrence rate of infectious complications including pneumonia, urinary tract infection, and sepsis categorized by main sites of cancer.
In-hospital mortality categorized by main sites of cancer.
The multivariable logistic regression analysis, including other covariates, fitted with a generalized estimating equation for in-hospital mortality, showed that the concomitant presence of cancer was independently associated with higher in-hospital mortality of patients hospitalized for HF (Table 3).
| OR | 95% CI | P value | |
|---|---|---|---|
| Cancer | 1.51 | 1.43–1.59 | <0.001 |
| Age (years) | 1.04 | 1.04–1.04 | <0.001 |
| Sex | |||
| Female | Ref. | ||
| Male | 1.25 | 1.21–1.30 | <0.001 |
| Body mass index (kg/m2) | 0.95 | 0.95–0.96 | <0.001 |
| Hypertension | 0.52 | 0.50–0.55 | <0.001 |
| Diabetes mellitus | 1.03 | 0.99–1.06 | 0.111 |
| Chronic renal failure | 1.63 | 1.57–1.70 | <0.001 |
| Chronic liver disease | 1.46 | 1.37–1.57 | <0.001 |
| Chronic respiratory disease | 0.97 | 0.93–1.01 | 0.166 |
| Myocardial infarction | 1.60 | 1.49–1.72 | <0.001 |
| Dilated cardiomyopathy | 1.27 | 1.20–1.33 | <0.001 |
| Smoking | 0.88 | 0.84–0.91 | <0.001 |
| New York Heart Association | |||
| Class II | Ref. | ||
| Class III | 1.70 | 1.60–1.81 | <0.001 |
| Class IV | 3.45 | 3.19–3.73 | <0.001 |
| Barthel Index | 0.99 | 0.99–0.99 | <0.001 |
| Administration within 2 days of hospital admission | |||
| β-blocker | 0.94 | 0.91–0.97 | <0.001 |
| Renin-angiotensin system inhibitor | 0.62 | 0.59–0.64 | <0.001 |
| Mineralocorticoid receptor antagonist | 0.82 | 0.79–0.86 | <0.001 |
| Intravenous inotropic agent | 2.42 | 2.32–2.53 | <0.001 |
| Intravenous nitrate | 0.58 | 0.55–0.61 | <0.001 |
| Intravenous furosemide | 0.96 | 0.93–1.00 | 0.058 |
CI, confidence interval; OR, odds ratio.
The in-hospital mortality was higher in patients with cancer than those without cancer in both men (9.7% vs. 6.2%, P<0.001) and women (10.7% vs. 7.3%, P<0.001).
We analyzed a nationwide inpatient database and explored the association of cancer with the clinical outcomes of patients hospitalized for HF. The key findings of the study were as follows: (1) cancer was diagnosed in approximately 6% of patients hospitalized for HF, and the prevalence of cancer increased with time; (2) infectious complications were more frequently seen in patients with cancer; and (3) the presence of cancer was independently associated with higher in-hospital mortality.
The coexistence of cancer is common in patients with HF. In this study, approximately 6% of patients hospitalized for HF also had cancer. Further, the prevalence of cancer was highest in patients aged in their 70 s and 80s, and increased over time. The prevalence of cancer in our study (approximately 6%) was lower than that in the recent Japanese data obtained by Yoshihisa et al.12 Our result is an analysis of a nationwide in-patient database, whereas the study population included in the study by Yoshihisa et al12 comprised patients who were admitted to the Fukushima Medical University Hospital, which is the referral center hospital in the Fukushima prefecture with a population of 1.85 million. Therefore, we assume that many HF patients with severe HF and complications with multiple co-morbidities (including cancer) are transferred to the Fukushima Medical University Hospital. Such a background might have resulted in the difference in the prevalence of cancer between these 2 studies. Additionally, considering that the rate of cancer in patients with HF varies among reports (from 2% to 22%),13–15 the prevalence of cancer can differ depending on the characteristics of study population. Further data on the prevalence of cancer among HF patients according to the features of hospitals are needed.
Several possible linking mechanisms such as shared risk factors (aging, cigarette smoking, diabetes mellitus, etc.), chronic inflammation, oxidative stress, sympathetic nervous system activation, renin-angiotensin-aldosterone system activation, and genetic background are suggested for the association of HF with cancer.16 Epidemiological studies also showed that HF could increase the risk of cancer,13,14 and the number of HF patients with concomitant cancer is expected to increase further. The distribution of main sites of cancers in patients with HF shown in Figure 2 was almost similar to that found in the epidemiological data of the Japanese population.17
The association of cancer with outcomes in patients with HF is still uncertain. Banke et al indicated that HF patients with cancer had higher all-cause mortality.13 In contrast, Ram et al recently reported that a concurrent diagnosis of cancer was associated with a marginally lower in-hospital mortality.18 In this study, we conducted a multiple logistic regression analysis to identify the determinants of in-hospital death. Results of the multiple logistic regression analysis, such as the protective influence of hypertension and cigarette smoking (so called “smoker’s paradox”), were concordant with previous studies focusing on acute HF patients,19–22 and our results showed that cancer was independently associated with elevated in-hospital mortality in patients hospitalized for HF. However, none of these studies, or the present study, differentiate the status of the concurrent cancer, which affects the survival. As shown in Figure 4, in-hospital mortality could differ by sites of cancers. For example, prostate cancer and breast cancer are gender-specific cancers and, therefore, the difference in prognosis between these cancers might affect the gender difference in clinical outcomes of patients with HF. Furthermore, patients with metastasis had elevated in-hospital mortality than those without metastasis in patients with HF and cancer. Therefore, other statuses of cancer (stages, performance status, comorbidities, and treatment) could also influence the clinical outcomes. Additionally, the occurrence rate of infectious complication such as pneumonia and sepsis differed among the sites of cancer in this study, as shown in Figure 3. Further studies differentiating the detailed information on concomitant cancers are required to establish patient-specific approaches based on the condition of each patient.
Several factors could contribute to the association of concomitant cancer with worse in-hospital outcomes for patients hospitalized for HF.
First, patients with concomitant cancer were older and there were differences in background characteristics between patients with and without cancer. However, even after adjustment for these covariates, cancer was still associated with higher in-hospital mortality. Therefore, the higher in-hospital mortality of patients with cancer could not be simply explained by the differences in clinical presentations.
Second, cancer patients are known to be at high risk for several complications such as thrombotic events and infections.23,24 Although the difference in the occurrence rate of thrombotic complications between patients with and without cancer did not reach statistical significance in this study, infectious complications including pneumonia and sepsis during hospitalization were more common in patients with cancer. We should pay more attention to these life-threatening complications in hospitalized HF patients with cancer.
Third, advanced medical procedures including respiratory support, intubation, hemodialysis, and intra-aortic balloon pumping were less frequently performed for patients with cancer. However, detailed reasons for the lower implementation rate of these advanced procedures in patients with cancer could not be determined in this study. In real-world clinical practice, clinical decision on indication for advanced treatment is often difficult in patients with cancer. For example, the latest guidelines recommend several promising treatments such as implantable cardioverter-defibrillator, transcatheter aortic valve implantation, and ventricular assist device to be avoided for patients with cancer and limited life expectancy.25–27 Therefore, it is not easy to perform these advanced treatments for patients with cancer. However, given that the prognosis of patients with cancer continues to improve, indication for these advanced treatments should be made based on the careful prognostic assessment of each cancer patient. In contrast, the percentage of patients receiving treatment for cancer is also reported to be lower in patients with HF.28,29 Therefore, more clinical evidence and data are required to develop the treatment strategy for patients with HF and to ensure that patients would not miss an opportunity to receive optimal medical treatment for both HF and cancer. From this perspective, close collaboration between cardiologists and oncologists is necessary for the optimal management of patients with HF and cancer. Accordingly, a multidisciplinary approach including cardiologists, oncologists and other healthcare professionals is indispensable.
There are several limitations to this study. We could not analyze the difference in outcomes between de novo, prior-treated, and active status of cancer because of the lack of information on the stages and types of cancer and treatment for cancer. As described above, in-hospital mortality was higher in patients with metastasis than in those without metastasis among HF patients with cancer. Therefore, the status of cancer will no doubt affect the results. Further studies with more detailed information on the status of cancer are required. Similarly, our database lacked detailed information on diabetes mellitus such as HbA1c and diabetic duration, which could influence the risk of cancer. We performed multivariable logistic regression analysis with a generalized estimating equation, but residual bias due to unmeasured confounders could not be excluded. Because of the retrospective design, recorded diagnoses are commonly less well validated. Our database lacked data regarding blood pressure, HF etiology, left ventricular ejection fraction, and other comorbidities, which could affect the clinical outcomes of hospitalized HF patients. For example, it is well known that a poorer prognosis is associated with low blood pressure on admission.20 Although we used information on the percentage use of medications including inotropic agents as a surrogate for data on blood pressure, this is a major limitation of this study. Left ventricular ejection fraction is also important in determining the therapeutic strategy for HF patients and could influence the result. Data on the cause of death were unavailable in the DPC database. Although additional assessment evaluation and treatment for cancer could cause medical costs to rise for HF patients with concomitant cancer, we were unable to separate medical costs for the treatment of HF and those for the treatment of cancer. We had no information on cancer-related thromboembolism in our DPC database.
The analysis of a nationwide inpatient database including 447,818 hospitalized HF patients showed that the coexistence of cancer in patients with HF was common and was increasing with time. Further, concomitant cancer is associated with higher in-hospital mortality of patients hospitalized for HF, even after adjustment for covariates.
This work was supported by grants from the Ministry of Health, Labour and Welfare, Japan (19AA2007 and H30-Policy-Designated004) and the Ministry of Education, Culture, Sports, Science and Technology, Japan (17H04141).
Research funding and scholarship funds were received by H. Kaneko and K. Fujiu from Medtronic Japan Co., Ltd, Abbott Medical Japan Co., Ltd, Boston Scientific Japan Co., Ltd, and Fukuda Denshi, Central Tokyo Co., Ltd. I. Komuro is a member of Circulation Journal ’ Editorial Team.
Name of the ethics committee: the Clinical Research Review Board of The University of Tokyo [3501-(3)]. Reference number: 3501-(3).
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-20-0314
