Session Information
Session Type: Abstract Submissions (ACR)
Background/Purpose:
Little is known about the prevalence of systemic autoimmune rheumatic diseases (SARDs) in patients with congestive heart failure (CHF) and its contribution to long-term adverse events. Therefore, we documented the prevalence of SARDs in an incident cohort of patients hospitalized with CHF in the province of Alberta, Canada and examined the association between SARDs and 1-year mortality after adjusting for traditional cardiovascular risk factors.
Methods:
This retrospective cohort study examined all Alberta residents, aged 20 years and older, hospitalized with incident CHF between April 1, 1999 and December 31, 2008. Definitions of CHF, SARDs and other comorbidities were based on established ICD-9 &-10 codes. SARDs included rheumatoid arthritis, systemic lupus erythematosus, inflammatory myositis, systemic sclerosis, Sjogren’s syndrome, overlap syndrome and other connective tissue diseases. Hospitalization records in the five years prior to the incident CHF hospitalization were examined to identify the presence of SARDs and other comorbidities. Baseline characteristics and comorbidity rates of SARDs/non-SARDs patients were described. The independent association of SARDs and mortality after adjusting for demographic and traditional cardiovascular risk factors was calculated with logistic regression. Kaplan-Meier analysis and the log-rank statistic examined the unadjusted one-year mortality between SARDs/non-SARDS patients.
Results:
SARDs prevalence was 3.1% (1208 patients) out of 38,668 patients hospitalized with CHF. Patients with SARDs were younger, more likely female, and had lower rates of diabetes, hypertension, COPD, anemia and renal disease (Table 1). After multivariate adjustment, SARDs was associated with higher odds of 1-year mortality (adjusted Odds Ratio 1.3 (95% Confidence Interval 1.2-1.5) (Table 2). Kaplan-Meier analysis showed greater 1-year mortality in SARDs versus non-SARDs hospitalized CHF patients (not shown in abstract).
Conclusion:
Significant mortality risk exists among SARDs patients hospitalized with CHF despite lower rates of factors such as diabetes and hypertension. Agressive recognition and management of CHF in SARDs patients may improve survival rates. Further work is needed to examine the outpatient prevalence of SARDs in this population.
Table 1. Baseline characteristics of SARDs versus non-SARDs CHF patients
|
CHARACTERISTIC |
SARDS |
Non-SARDs |
p-value |
|
|
Age (mean (STD)) |
74.8 (13.2) |
72.9 (12.7) |
<0.01** |
|
|
Male sex |
18,842 (50.3%) |
347 (28.7%) |
< 0.01 |
|
|
Diabetes |
12624 (33.7%) |
331 (27.4%) |
< 0.01 |
|
|
Hypertension |
28432 (75.9%) |
885 (73.3%) |
0.04 |
|
|
Dementia |
4832 (12.9%) |
108 (8.9%0 |
< 0.01 |
|
|
COPD |
17756 (47.4%) |
628 (52.0%) |
< 0.01 |
|
|
Anemia |
14560 (38.9%) |
724 (59.9%) |
< 0.01 |
|
|
Cerebrovascular disease |
8016 (21.4%) |
259 (21.4%) |
0.95 |
|
|
Renal disease |
6181 (16.5%) |
248 (20.5%) |
< 0.01 |
|
|
Cancer |
7042 (18.8%) |
216 (17.9%) |
0.40 |
|
|
PVD |
6780 (18.1%) |
236 (19.5%) |
0.22 |
|
|
Atrial fibrillation |
12549 (33.5%) |
393 (32.5%) |
0.46 |
|
|
1-year mortality |
11238 (30.0%) |
422 (34.9%) |
< 0.01 |
|
|
|
|
|
|
|
|
**p-value was based on Kruskal-Wallis one-way analysis of variance |
|
|||
|
P-values of categorical variables are based on Chi-square test |
|
|||
|
|
|
|
|
|
TABLE 2. Logistic regression of 1-year mortality in incident CHF
|
|
|
|
Effect |
OR (95% Confidence Interval) |
|
Male |
1.2 (1.1 – 1.2) |
|
Age |
1.0 |
|
SARDs |
1.3 (1.2 – 1.5) |
|
Hypertension |
0.7 (0.6 – 0.7) |
|
Dementia |
1.8 (1.7 – 2.0) |
|
COPD |
1.1 (1.0 – 1.1) |
|
Anemia |
1.3 (1.2 – 1.3) |
|
Cerebrovascular disease |
1.3 (1.2 – 1.4) |
|
Renal disease |
1.7 (1.6 – 1.8) |
|
Cancer |
2.5 (2.4 – 2.6) |
|
PVD |
1.2 (1.1 – 1.2) |
Disclosure:
S. O. Keeling,
None;
A. Bissonauth,
None;
B. Leung,
None;
P. Kaul,
None.
ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-impact-of-systemic-autoimmune-rheumatic-disease-on-one-year-mortality-in-congestive-heart-failure-patients-a-population-level-analysis/