Systolic BP Predicts Mortality in Acute Heart Failure

http://www.medscape.com/viewarticle/547410_print

News Author: Steve Stiles

Release Date: November 8, 2006
Systolic hypertension is not only common in patients hospitalized with acute heart failure, but it may also help protect against death while in the hospital and for several months after discharge, regardless of left ventricular (LV) systolic function at admission, suggest data from a huge multicenter registry published in the November 8 issue of JAMA. The adjusted in-hospital mortality for more than 48,000 patients hospitalized with heart failure varied inversely with their admission systolic blood pressure (SBP) and was about 4 times higher when the SBP was lower than 120 mm Hg as compared with higher than 161 mm Hg.

"Hypertension is very frequent in patients hospitalized with heart failure, including those with reduced systolic function as well as those with preserved systolic function," Gregg C. Fonarow, MD, of the University of California, Los Angeles, Medical Center, a coauthor of the analysis, told heartwire. Half of the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients with Heart Failure (OPTIMIZE-HF) registry cohort — more than a third of patients with LV systolic dysfunction and more than half of those with preserved LV function — had an admission SBP of more than 140 mm Hg.

The inverse relationship between admission SBP and in-hospital mortality, Dr. Fonarow observed, applied to both groups of patients. Postdischarge mortality also rose significantly with declining admission SBP, regardless of any treatment with vasodilators or inotropic agents, in a subgroup of the cohort that was followed up for 2 to 3 months.

The OPTIMIZE-HF analysis by Mihai Gheorghiade, MD, of Northwestern University in Chicago, Illinois, and colleagues, not only suggests that admission SBP is independently prognostic in patients hospitalized with heart failure, they have implications for therapy, observed Dr. Fonarow. They suggest, he said, that "with the characteristics and outcomes so different among the patients by varied SBP levels, management will need to vary. Rather than grouping all of these patients together, we'll potentially need to stratify them."

The report states that "Elevated SBP appears to signal specific pathophysiological processes that differ from the underlying processes in patients with low SBP. Because the characteristics and outcomes are different among patients with heart failure with varying SBP levels, management may need to vary according to SBP at admission."

The in-hospital outcomes analysis included patients admitted with heart failure at 259 academic and community-based US hospitals "of all sizes and from all regions of the country." In the overall cohort and the subgroup of 5791 patients followed up after discharge, admission SBP was a significant predictor of mortality after controlling for a long list of demographic, hemodynamic, renal functional, clinical, and drug treatment criteria.

For admission SBP readings lower than 160 mm Hg, the hazard ratio for in-hospital death went up 21% (95% confidence interval [CI], 1.17 - 1.25) for every 10-mm Hg drop; the risk did not vary for SBP values higher than 160 mm Hg. Also for every 10-mm Hg decrease in SBP, the postdischarge-mortality hazard ratio climbed 18% (95% CI, 1.10 - 1.26), and the hazard ratio for the composite of mortality or rehospitalization rose 5% (95% CI, 1.03 - 1.07).

Table. Outcomes by Admission SBP Quartile and LV Systolic Dysfunction (All Trends Across Quartiles, P < .001)*

Endpoint by Patient Group	<>	120 - 139 mm Hg	140 - 161 mm Hg	> 161 mm Hg
In-hospital Mortality
Overall, % (n = 48,612)	7.2	3.6	2.5	1.7
LV systolic dysfunction,%†	6.6	3.1	2.5	1.6
No. of LV systolic dysfunction, %†	6.2	3.2	2.0	1.4
Postdischarge Mortality
Follow-up cohort, % (n = 5791)	14.0	8.4	6.0	5.4
LV systolic dysfunction, %‡	13.0	6.8	6.3	4.1
No. of LV systolic dysfunction, %‡	14.9	10.0	4.7	4.7

*SBP indicates systolic blood pressure; and LV, left ventricular.
^†LV function measured in 41,267 patients overall.
^‡LV function measured in 4959 patients in the follow-up cohort.
Source: JAMA. 2006;296:2217-2226.

The average hospital length of stay also declined with rising admission SBP regardless of LV systolic functional status; in the overall group, it decreased from 6.5 days for patients with SBP lower than 120 mm Hg to 5.1 days for those with pressures higher than 161 mm Hg (P < .001). The rate of rehospitalization 60 to 90 days after discharge did not vary significantly overall or in the 2 LV systolic-function subgroups.

"Interestingly, we put so much emphasis on LV ejection fraction," Dr. Fonarow said, "and yet it's not a major determinant of clinical outcomes in these patients, whereas admission systolic blood pressure, which has not received a lot of focus, is a much larger predictor of mortality in these patients."

Disclosure statements for all coauthors are listed in the article, which also states that GlaxoSmithKline funded the OPTIMIZE-HF registry, was involved in its design and conduct, and reviewed the manuscript prior to submission but "was not involved in the management, analysis, or interpretation of data or the preparation of the manuscript."

JAMA. 2006;296:2217-2226, 2259-2260.

News Author

Steve Stiles
is a journalist for Medscape. He has been reporting on cardiovascular medicine since 1984 and for the past 3 years has been a journalist for theheart.org, a website acquired by WebMD. Steve is a graduate of Kenyon College and has an MS from the journalism department at Boston University. He can be contacted at SStiles@webmd.net.

Disclosure: Steve Stiles has disclosed no relevant financial relationships.