Anda di halaman 1dari 9


Subclinical Hypertensive Heart Disease in Black Patients With

Elevated Blood Pressure in an Inner-City Emergency Department
Phillip Levy, MD, MPH, Hong Ye, MS, Scott Compton, PhD, Robert Zalenski, MD, Timothy Byrnes, MD,
John M. Flack, MD, MPH, Robert Welch, MD, MS
From the Department of Emergency Medicine (Levy, Zalenski, Welch), the Division of Cardiology (Byrnes), the Department of Internal Medicine (Flack) and the
Cardiovascular Institute (Levy, Flack, Welch), Wayne State University School of Medicine, Detroit, MI; the Department of Radiation Oncology, William Beaumont
Hospital, Royal Oak, MI (Ye); and the Department of Emergency Medicine, UMDNJ–New Jersey Medical School (Compton), Newark, NJ.

Study objective: We examine the point prevalence of subclinical hypertensive heart disease in a cohort of urban
emergency department (ED) patients with elevated blood pressure.

Methods: A convenience sample of hypertensive (blood pressure ⱖ140/90 mm Hg on 2 measurements)

patients aged 35 years or older with no history of cardiac or renal disease who presented to a single urban ED
and were asymptomatic from a cardiovascular perspective (ie, no symptoms of dyspnea or chest pain) were
enrolled. All patients underwent a standardized evaluation (including echocardiography), and subclinical
hypertensive heart disease was defined by the presence of one or more of the following criterion-based
electrocardiographic findings: left-ventricular hypertrophy, systolic dysfunction, or diastolic dysfunction.

Results: A total of 161 patients were included. Mean age was 49.8 years (SD 8.3 years), 93.8% were black,
and 51.6% were men. Nearly all (93.8%) had a history of hypertension, and many (68.3%) were receiving
antihypertensive therapy at baseline. Mean systolic and diastolic blood pressures were 183.9 mm Hg (SD 25.1
mm Hg) and 109.5 mm Hg (SD 14.4 mm Hg), respectively. Subclinical hypertensive heart disease was found in
146 patients (90.7%; 95% confidence interval [CI] 85.2% to 94.3%), with most (n⫽131) displaying evidence of
diastolic dysfunction (89.7%; 95% CI 83.7% to 93.7%). Left-ventricular hypertrophy was also common (n⫽89;
61.0%; 95% CI 52.9% to 68.5%) and was often (but not exclusively) present in those with diastolic filling
abnormalities (n⫽75; 57.3%; 95% CI 48.7% to 65.4%).

Conclusion: In our largely black cohort of ED patients with elevated blood pressure, subclinical hypertensive
heart disease was highly prevalent, suggesting the need for coordinated efforts to reduce cardiac consequences
of hypertension in such inner-city communities. [Ann Emerg Med. 2012;60:467-474.]

Please see page 468 for the Editor’s Capsule Summary of this article.

A feedback survey is available with each research article published on the Web at
A podcast for this article is available at
0196-0644/$-see front matter
Copyright © 2012 by the American College of Emergency Physicians.

SEE EDITORIAL, P. 475. Importance

Unfortunately for many, such underlying hypertensive heart
disease is unlikely to be detected until advanced, clinically
apparent manifestations are present.5-8 With appropriate blood
Background pressure reduction, however, the deleterious effects of
Hypertension is commonplace in the United States, affecting subclinical cardiac disease can be mitigated, symptom onset
more than 76 million individuals older than 20 years.1 The (especially heart failure) can be forestalled, and future adverse
burden of hypertension is particularly prominent among blacks, events can be prevented.9-11 Early identification of subclinical
who experience a higher disease prevalence and, especially for cardiac disease has thus emerged as an important aspect of
men, poorer overall blood pressure control than their white and secondary cardiovascular disease prevention.12
Hispanic counterparts.1,2 As a result, blacks are at tremendous Despite the compelling benefits of directed intervention for
risk for pressure-related consequences of hypertension, patients with subclinical cardiac disease, the utility of routine
particularly premature onset of structural cardiac damage and screening in asymptomatic hypertensive patients remains
functional impairment.3,4 controversial.12-15 This may be attributable, in part, to uncertainty

Volume , .  : October  Annals of Emergency Medicine 467

Hypertension in the Emergency Department Levy et al

Editor’s Capsule Summary (Detroit Receiving Hospital, Detroit, MI) between April 2006
and July 2007 with elevated blood pressure. Detroit Receiving
What is already known on this topic Hospital is a tertiary care facility operating within a large urban
Emergency physicians treat many patients who have environment and is affiliated with the Wayne State University
suboptimally controlled hypertension. School of Medicine. The hospital ED has a total annual census
of approximately 98,000 adult patients, the majority of whom
What question this study addressed (⬇85%) are black. Study approval was obtained from the
Whether patients with asymptomatic hypertension Human Investigations Committee of Wayne State University
have subclinical heart disease. before patient recruitment.
What this study adds to our knowledge Selection of Participants
In a largely black 161-patient sample, the majority The purpose of this study was to evaluate the point prevalence
(91%; 95% confidence interval 85% to 94%) of of subclinical hypertensive heart disease in ED patients who had
patients with asymptomatic hypertension had elevated blood pressure but were asymptomatic from the
underlying cardiac dysfunction, most commonly perspective of potential acute or chronic end-organ cardiac damage.
diastolic dysfunction. As such, we targeted enrollment of patients aged 35 years or older
who met validated criteria for stage 1 or greater hypertension
How this is relevant to clinical practice (blood pressure ⱖ140/90 mm Hg on initial ED triage vital sign
The recognition that in some emergency assessment and at repeated measurement 1 hour later, both of
departments most patients with asymptomatic which were obtained with an automated, adult-sized blood pressure
hypertension have underlying cardiac dysfunction cuff27,29-31) and had neither a primary nor secondary presenting
should lead to improved vigilance in developing complaint potentially attributable to an acute hypertensive
coordinated care pathways to prevent further emergency or underlying cardiac disease (ie, chest pain,
deterioration. palpitations, dyspnea, syncope, focal neurologic deficits, and altered
mental status). To avoid conflict with patient care needs and to
minimize the potential influence of acute physiologic perturbations
about the prevalence of subclinical hypertensive heart disease, on echocardiographic findings, patients who required hospital
which, depending on operational definitions and the population admission for any reason were excluded from the study. Moreover,
studied, can vary from 0.9% to 50%.4,16-21 Such data, however, patients with a known cardiac condition (ie, heart failure, coronary
were compiled almost exclusively with community-based sampling artery disease, cardiomyopathy [dilated, idiopathic, or
methods and may not accurately depict the level of risk that exists hypertrophic], or valvular heart disease), those at risk for heart
for the many millions of patients who present with moderate or disease from a cause other than hypertension (ie, renal failure), and
severely elevated blood pressure each year to emergency those with previously documented evidence of abnormal cardiac
departments (EDs) across the United States.22,23 Of particular structure or function on echocardiography or other cardiovascular
concern, they may underestimate the prevalence of subclinical imaging study (ie, computed tomography, magnetic resonance
hypertensive heart disease in inner-city blacks, an especially high- imaging, or cardiac catheterization), whether hypertensive in cause
risk subset who, for lack of an alternative, rely on the ED for or not, were also excluded. Patients with diabetes mellitus, however,
detection or ongoing management of primary care amenable were eligible for inclusion.
conditions such as hypertension.24-27 Screening was performed in the ED by dedicated research
personnel. Written informed consent was obtained from
Goals of This Investigation patients who met eligibility criteria and were willing to
Because hypertension contributes more than any other factor to participate. Our initial intent was to recruit 24 hours a day, 7
racial differences in cardiovascular disease survival,28 much can be days a week; however, we encountered difficulty in having
gained through enhanced understanding of subclinical hypertensive patients return for study procedures once they were discharged
heart disease (a potentially modifiable preclinical disease state) in from the ED. To account for this, we adjusted our approach so
at-risk hypertensive patients. Accordingly, the objective of this that the entire study protocol (described in greater detail in the
study was to estimate the point prevalence of echocardiographically following section) could be completed in the ED at enrollment.
defined subclinical hypertensive heart disease in a cohort of To accommodate this change, recruitment had to be
predominantly black, inner-city ED patients with asymptomatic coordinated with the availability of key personnel (specifically,
yet profoundly elevated blood pressure. the study’s echocardiographic technician), limiting screening to
“usual” business hours (Monday through Friday, 9 AM to 4 PM).
MATERIALS AND METHODS Subjects included in the study cohort thus represent a
Study Design and Setting convenience sample of ED patients.
This was designed as a prospective cross-sectional study of To evaluate for potential bias in our sample, we compared
patients who presented to the ED of a single medical center patients who were enrolled in the study with a consecutive series

468 Annals of Emergency Medicine Volume , .  : October 

Levy et al Hypertension in the Emergency Department

of ED patients treated for hypertension during a Patients initially

contemporaneous period. Patients included in this consecutive consented for
series (n⫽742) were culled by applying study eligibility criteria participation in study
to a larger group identified by primary discharge diagnosis n=200

(International Classification of Diseases, Ninth Revision code Failed to return for

401.9 — unspecified essential hypertension) with billing data. outpatient echo
Once patients were enrolled, baseline data (demographics, n=20

medical and family history, and use of pharmaceutical agents)

were collected by trained research assistants, ECG was
Found to have prior
performed, urine was collected for dipstick testing, and serum documented history of
samples were obtained for measurement of blood urea nitrogen, cardiac disease
creatinine, and b-type natriuretic peptide (Biosite Incorporated, n=19
San Diego, CA). Echocardiograms were subsequently obtained
at the point of care in the ED by a single, certified
echocardiographic technician using a Phillips iE333 ultrasound
machine (Phillips Heathcare US, Andover, MA) with a 2.5-
MHz cardiac transducer and interpreted by one of 2 board- Included in final
certified cardiologists who were blinded to all demographic and study cohort
historical information, as well as ECG, urine dipstick, blood
urea nitrogen, creatinine, and b-type natriuretic peptide data. Figure. Flow chart outlining derivation of study cohort.
Echocardiography in enrolled subjects was performed expressly
for purposes of this study, and results were not used to inform Table 1. Breakdown of subclinical hypertensive heart disease
by left-ventricular dysfunction criterion in patients with and
clinical management.
without left-ventricular hypertrophy.
Outcome Measures Hypertrophy
Left-Ventricular Total, No.
Subclinical hypertensive heart disease, the primary outcome of Dysfunction Yes No (%)*
interest, was determined by demonstration of one or more of the
following externally validated 2-dimensional and Doppler None 12 0 12 (8.2)
Diastolic dysfunction alone 60 49 109 (74.7)
echocardiographic criteria32,33: left-ventricular hypertrophy, Systolic dysfunction alone 2 1 3 (2.1)
defined by left-ventricular mass indexed to height2.7 greater than or Systolic and diastolic 15 7 22 (15.1)
equal to 46 g/m2.7 for women or greater than or equal to 49 g/m2.7 dysfunction
for men; left-ventricular systolic dysfunction determined by Total, No. (%)* 89 (61.0) 57 (39.0) 146 (100)
Simpson’s biplane method, defined by an ejection fraction less than *Percentage of patients with subclinical hypertensive heart disease (n⫽146).
or equal to 50%; or left-ventricular diastolic dysfunction, defined
by diastolic velocity measured at the medial mitral annulus by tissue
Doppler imaging (e=) less than 8 cm/second. RESULTS
Two hundred patients were enrolled and echocardiograms
Primary Data Analysis were successfully obtained for 180 (90%). Despite contradictory
The primary purpose of this study was to provide a precise self-report, 19 (10.6%) patients were found on subsequent
estimate of the point prevalence of subclinical hypertensive heart comprehensive chart review to have a history of heart failure or
disease among asymptomatic, hypertensive ED patients. To coronary artery disease and were thus excluded, resulting in a
provide a sufficient sample population for an a priori prevalence final study cohort of 161 patients. A flow chart of study cohort
estimate of 10% with a 95% confidence interval (CI) of 6.5% to derivation is provided in the Figure.
13.5% (3.5%), screening of 254 patients was needed. Although enrollment fell short of our projected target,
Groupwise descriptive data were compiled, and proportions analysis of study echocardiograms revealed the presence of
or means and medians with corresponding measures of variance subclinical hypertensive heart disease in 146 patients, yielding a
(SD and interquartile range, respectively) were calculated. point prevalence that was far greater (90.7%; 95% CI 85.2% to
All data were entered into a Microsoft Office 2007 Access 94.3%) than our a priori estimate. The majority of patients with
program (Microsoft, Bellevue, WA) and subsequently analyzed subclinical hypertensive heart disease (Table 1) had diastolic
with SPSS (version 16.0; SPSS Inc, Chicago, IL). To ensure dysfunction (n⫽131; 89.7%; 95% CI 83.7% to 93.7%), a
data integrity, double entry was performed for all patients finding that was associated with concurrent left-ventricular
included in the final analysis, with adjudication by the principal hypertrophy in most but not all patients (n⫽75; 57.3%; 95%
investigator (P.L.) of any variables with demonstrated CI 48.7% to 65.4%). Twenty-five individuals (15.5%; 95% CI
discrepancy. 10.7% to 21.9%) had evidence of systolic dysfunction with an

Volume , .  : October  Annals of Emergency Medicine 469

Hypertension in the Emergency Department Levy et al

Table 2. Baseline data.

Subclinical Hypertensive No Subclinical Hypertensive
Overall, Heart Disease, Heart Disease,
Variable nⴝ161 nⴝ146 nⴝ15
Mean Age, y (SD) 49.8 (8.3) 50.0 (8.3) 47.3 (8.3)
Sex, No. (%)
Male 83 (51.6) 77 (52.7) 6 (40.0)
Female 78 (48.4) 69 (47.3) 9 (60.0)
Race, No. (%)
Black 151 (93.8) 137 (93.8) 14 (93.3)
White 6 (3.7) 6 (4.1) 0
Body mass index, kg/m2 (SD) 29.8 (7.1) 30.2 (7.0) 26.3 (6.7)
Insured, No. (%) 93 (57.8) 86 (58.9) 7 (46.7)
Reason for ED visit, No. (%)
High blood pressure 37 (23.0) 33 (22.6) 4 (26.7)
Headache 11 (6.8) 10 (6.9) 1 (6.7)
Epistaxis 3 (1.9) 3 (2.1) 0
Acute injury 13 (8.1) 12 (8.2) 1 (6.7)
Chronic pain 19 (11.8) 16 (11.0) 3 (20.0)
Acute infection 20 (12.4) 19 (13.0) 1 (6.7)
GI or GU complaint 32 (20.0) 30 (20.6) 2 (13.3)
Neurologic or psychiatric complaint 17 (10.6) 15 (10.3) 2 (13.3)
Metabolic issue 10 (6.2) 9 (6.2) 1 (6.7)
History of hypertension, No. (%) 151 (93.8) 137 (93.8) 14 (93.3)
Receiving antihypertensive therapy, No. (%) 110 (68.3) 98 (67.1) 12 (80.0)
Antihypertensive class, No. (%)
ACEI or ARB 25 (15.5) 21 (14.4) 4 (26.7)
␤-Blocker 20 (12.4) 16 (11.0) 4 (26.7)
Clonidine 30 (18.6) 30 (20.6) 0
Diuretic 35 (21.7) 30 (20.6) 5 (33.3)
Calcium-channel blocker 31 (19.3) 27 (18.5) 4 (26.7)
Other 9 (5.6) 8 (5.5) 1 (6.7)
Unknown 51 (31.7) 48 (32.9) 3 (20.0)
History of diabetes, No. (%) 30 (18.6) 26 (17.8) 4 (26.7)
Receiving diabetic medication, No. (%) 23 (14.3) 20 (13.7) 3 (20.0)
Insulin 11 (6.8) 9 (6.2) 2 (13.3)
Oral agents 12 (7.5) 11 (7.5) 1 (6.7)
History of stroke, No. (%) 15 (9.3) 15 (10.3) 0
Other medications, No. (%)
Statin 11 (6.8) 10 (6.9) 1 (6.7)
Aspirin 24 (14.9) 22 (15.1) 2 (13.3)
NSAID 22 (13.7) 18 (12.3) 4 (26.7)
Social history, No. (%)
Active cigarette smoker 78 (48.5) 72 (49.3) 6 (40.0)
Drinks alcohol 76 (47.2) 67 (45.9) 9 (60.0)
Uses cocaine 14 (8.7) 11 (7.5) 3 (20.0)
Uses heroin 8 (5.0) 8 (5.0) 0
Smokes marijuana 30 (18.6) 29 (19.9) 1 (6.7)
Physical examination findings, No. (%)
Elevated jugular venous pressure 1 (0.6) 1 (0.7) 0
Lower extremity edema 6 (3.7) 6 (4.1) 0
Ascites 1 (0.6) 1 (0.7) 0
S3 gallop 0 0 0
Pulmonary rales 0 0 0
Initial vital signs, mean (SD)
Systolic blood pressure, mm Hg 183.9 (25.1) 184.4 (25.3) 179.0 (23.0)
Diastolic blood pressure, mm Hg 109.5 (14.4) 109.7 (14.4) 108.3 (15.3)
Pulse rate, beats/min 85.1 (16.0) 84.6 (15.3) 89.9 (21.7)
Repeated vital signs, mean (SD)
Systolic blood pressure, mm Hg 175.1 (23.4) 174.9 (23.7) 177.7 (20.7)
Diastolic blood pressure, mm Hg 105.1 (15.1) 104.6 (14.7) 110.1 (18.6)
Pulse rate, beats/min 80.1 (14.4) 79.5 (14.1) 85.5 (17.2)

470 Annals of Emergency Medicine Volume , .  : October 

Levy et al Hypertension in the Emergency Department

Table 2. Continued.
Subclinical Hypertensive No Subclinical Hypertensive
Overall, Heart Disease, Heart Disease,
Variable nⴝ161 nⴝ146 nⴝ15
Renal function, No. (%) or mean (SD)
Proteinuria 94 (61.0) 86 (58.9) 8 (53.3)
Blood urea nitrogen, g/dL 13.5 (5.7) 13.4 (5.6) 15.0 (6.6)
Serum creatinine, g/dL 1.0 (0.3) 1.0 (0.3) 1.1 (0.8)
eGFR, mL/min 90.8 (25.9) 91.2 (26.0) 86.7 (25.8)
b-Type natriuretic peptide, mean (SD), pg/dL 56.2 (84.0) 56.0 (84.8) 57.9 (78.5)
Electrocardiography, mean (SD)
Cornell criteria, mV* 25.5 (11.8) 25.6 (12.1) 24.2 (8.6)
Cornell product, msec⫻mV* 2275.9 (1199.6) 2307.7 (1244.4) 1977.5 (584.6)
GI, Gastrointestinal; GU, genitourinary; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; NSAID, nonsteroidal anti-inflammatory agent;
eGRF, estimated glomerular filtration rate (calculated with the Modification of Diet in Renal Disease formula).
*Criterion for determination of left-sided ventricular hypertrophy: Cornell criteria⫽R in aVL⫹S in V3 (⫹8 mV if female); indicative of hypertrophy if greater than or equal
to 28 mV; Cornell product⫽Cornell criteria⫻QRS duration; indicative of hypertrophy if greater than or equal to 2,440 msec⫻mV.

Table 3. Echocardiographic measurements. had a history of hypertension and were aware of this diagnosis,
though not all (68.3%) were receiving antihypertensive therapy.
Subclinical Subclinical Diabetes (18.6%), cigarette smoking (48.5%), and at least
Hypertensive Hypertensive occasional alcohol use (47.2%) were frequently observed as well.
Heart Heart The most common reason for visiting the ED was elevated
Overall, Disease, Disease, blood pressure (23%) and, at triage, mean systolic and diastolic
Variable, Mean (SD) nⴝ161 nⴝ146 nⴝ15
blood pressures were 183.9 mm Hg (SD 25.1 mm Hg) and
Left atrial diameter, cm 3.7 (0.7) 3.7 (0.7) 3.3 (0.6) 109.5 mm Hg (SD 14.4 mm Hg), respectively. Proteinuria was
LV systolic diameter, cm 2.7 (0.6) 2.8 (0.7) 2.5 (0.3)
observed in 61% but renal function as assessed by estimated
LV diastolic diameter, cm 4.7 (0.7) 4.7 (0.7) 4.1 (0.5)
LV end-diastolic volume, mL 94.1 (33.0) 96.4 (33.2) 71.5 (20.2) glomerular filtration rate was normal (mean 90.8 mL/min [SD
LV end-systolic volume, mL 29.2 (13.8) 29.9 (14.0) 22.7 (10.4) 25.9 mL/min]). Despite the frequency with which subclinical
Stroke volume, mL 65.6 (27.0) 67.4 (27.5) 48.9 (14.7) hypertensive heart disease was encountered, mean brain
IVSD, cm 1.3 (0.3) 1.3 (0.3) 1.0 (0.2) natriuretic peptide was not elevated, at 56.2 pg/mL (SD 84.0
LVPWD, cm 1.2 (0.3) 1.2 (0.3) 1.0 (0.2)
pg/mL), and there was no electrocardiographic evidence of left-
Ejection fraction, % 62.4 (12.9) 61.9 (13.1) 67.2 (10.1)
E, cm/s 73.5 (19.9) 72.4 (19.4) 83.7 (22.2) ventricular hypertrophy by Cornell voltage or Cornell product
A, cm/s 77.9 (21.2) 79.2 (21.1) 65.6 (18.4) criteria. Mean systolic and diastolic blood pressures at the time
E/A 1.0 (0.4) 0.9 (0.3) 1.3 (0.4) echocardiography was performed were 167.4 mm Hg (SD 27.0
e=, cm/s 6.5 (1.8) 6.2 (1.5) 9.7 (1.6) mm Hg) and 101.1 mm Hg (SD 15.5 mm Hg), respectively. As
E/e= 11.8 (3.7) 12.1 (3.8) 8.9 (1.8)
shown in Table 3, echocardiographic findings reflect the
E deceleration time, ms 248.5 (80.5) 250.6 (82.3) 227.8 (58.7)
LV mass, g 223.2 (95.3) 231.8 (95.1) 139.6 (42.0) ubiquitous presence of subclinical hypertensive heart disease in
LV mass indexed to BSA, 112.1 (46.6) 116.2 (46.6) 71.9 (18.5) the overall study population and, by group, the operational
g/m2 definitions used in our study design.
LV mass indexed to 52.4 (23.8) 54.7 (23.8) 39.7 (6.9)
height2.7, g/m2.7
Relative wall thickness 0.52 (0.14) 0.52 (0.14) 0.49 (0.13) LIMITATIONS
LV, Left-ventricular; IVSD, interventricular septal diameter; LVPWD, left-ventricu- Our results were derived from a convenience sample of ED
lar posterior wall diameter; BSA, body surface area. patients in a single center and may have been influenced by
misclassification bias, particularly as it pertains to identification of a
truly asymptomatic state. More stringent exclusion criteria could
ejection fraction less than 50%, nearly all of whom (n⫽22) also have eliminated patients with subtle clinical manifestations but, in
had diastolic filling abnormalities. the absence of acute symptoms, it would be inherently difficult
Baseline data for the enrolled cohort are displayed in Table (and potentially erroneous) to attribute intermittent, nonspecific
2. Comparison of these individuals with the aforementioned features such as dyspnea, chest pain, or fatigue to underlying
consecutive series of nonenrolled hypertensive patients can be hypertensive heart disease without definitive (ie, echocardiographic)
found in the accompanying Appendix E1 (available online at assessment. In addition, because we sought to target Mean age of patients enrolled undifferentiated, asymptomatic hypertensive patients in the ED, a
was 49.8 years (SD 8.3 years), 93.8% were black, 51.6% were subpopulation typically neglected by the medical community,34 an
men, and 57.8% had health insurance. Most patients (93.8%) inclusive approach was needed to provide an accurate estimate of

Volume , .  : October  Annals of Emergency Medicine 471

Hypertension in the Emergency Department Levy et al

previously undetected subclinical cardiac disease. Moreover, mortality.37,38 Thus, knowledge of the subclinical hypertensive
although our intent was to evaluate the point prevalence of heart disease prevalence in our target population has therapeutic
subclinical hypertensive heart disease in a cohort of predominantly meaning that may help reduce the likelihood of adverse outcome.39
black, urban-dwelling patients, such an approach precludes Although racial differences in cardiac remodeling have been
extrapolation of our data to other racial or ethnic subgroups and previously demonstrated with an increased prevalence of left-
other clinical settings. Additionally, those included in the study ventricular hypertrophy across all age ranges in blacks with
experienced extremely poor blood pressure control, and our data hypertension,4,40 the unique aspect of our study is the sheer
may not apply to populations (irrespective of race) with better magnitude of previously unrecognized subclinical hypertensive
chronic hypertension management. As shown by comparison of heart disease. That a sizable proportion of our patients had
study subjects with a consecutive series of hypertensive individuals advanced abnormalities with diastolic and, in some cases, frank
evaluated in our ED, findings are, at the least, generalizable to systolic dysfunction provides some insight into the underlying
urban black patients in our geographic region. To some degree, this substrate that contributes to existing disparities in hypertension-
may reflect the relatively high rate of uninsured patients with mediated morbidity and mortality. Blacks progress from
hypertension in our community, and the applicability of our results asymptomatic to symptomatic stages of left-ventricular dysfunction
to more affluent but racially similar communities is unclear. more rapidly, and the mean age of blacks who are admitted to the
Although such a consideration would appear amenable to suggested hospital with heart failure is much lower than that of whites (63.6
improvements in outpatient management (especially efforts to [SD 15.4] versus 75.2 [SD 12.7] years; P⬍.001).41,42 Given that
enhance medication compliance and clinician adherence to existing the median life expectancy after an incident hospital admission for
guidelines), we are unable to project quantifiable benefits because heart failure is less than 2.5 years,43,44 there is remarkable potential
we did not collect data on existing primary care relationships. Last, for prevention of subsequent cardiovascular events12,45,46 in this
we did not include a control population of normotensive vulnerable population. Though generalizability may be limited by
individuals, making it difficult to determine how much race alone the included cohort, our data are applicable to the majority of
may have contributed to underlying cardiac structure and function individuals (ie, urban black patients with poorly controlled chronic
in study subjects. However, large-scale epidemiologic studies such hypertension) who, according to previously published reports,
as the Coronary Artery Disease Risk Development in Young Adults
typically present to EDs in the United States with elevated blood
have shown a low prevalence of left-ventricular hypertrophy in
pressure.34,47 Because hypertension contributes more than any
nonhypertensive black men (5%) and women (3%).19
other factor (⬇15%) to the overall racial difference in years of life
Furthermore, this study and the Dallas Heart Study, a probability-
lost to cardiovascular disease,28 initiatives targeting improved care
based random sample of Dallas County that included 1,335 blacks,
for those most at risk (especially within racially disparate locations)
found prevalent left-ventricular hypertrophy to be mostly mediated
are likely to be a high-yield endeavor.
by variation in blood pressure,19,35 suggesting that our results do
What are the specific implications of this research for
indeed reflect an independent relationship between blood pressure
emergency medicine? First and foremost, it highlights the
and subclinical cardiac disease.
unavoidable confluence of primary and acute care in the ED
DISCUSSION setting and suggests a need to be more proactive in the
In this study of asymptomatic, predominantly black individuals identification and referral of patients with uncontrolled or
who presented to our urban ED with elevated blood pressure, untreated hypertension. Since 2006, this approach has been
subclinical hypertensive heart disease was detected in 9 of every 10 endorsed in an official clinical policy on asymptomatic
patients. These data, which to our knowledge are the first to be hypertension by the American College of Emergency
compiled in such a population, greatly exceeded our estimated Physicians,31 yet according to a recent survey of ED directors,
point prevalence and underscore the disproportionate population- only 50% of US EDs routinely provide such a service.48 The
attributable risk of cardiovascular disease that hypertension imparts point is not to add further screening activities to the ED load
on the inner-city black community.28,36 Beyond epidemiologic but to enhance responsiveness to tasks that we currently carry
implications, the presence of subclinical hypertensive heart disease out (ie, check a blood pressure on every patient). Elevated blood
heralds an important branch point in clinical management because, pressure is common in the ED setting, even among patients
according to recently published guidelines by the International who lack overt evidence of end-organ damage, and knowing
Society of Hypertension in Blacks, such patients should be treated that up to 90% are at increased risk over time for death or other
in a manner that targets more aggressive blood pressure goals adverse events is actionable information. To be most effective,
(⬍130/80 mm Hg) than current guidelines based on the 7th however, the ED should function as a point at which a
Report of the Joint National Committee on Prevention, Detection, coordinated transition to enhanced outpatient management can
Evaluation, and Treatment of High Blood Pressure (⬍140/90 mm be initiated, not a destination for definitive care. Even then,
Hg).36 Appropriate antihypertensive therapy contributes clinician and patient inertia is common, and espousing a role
substantially to regression of left-ventricular remodeling and, in within the larger context of accountable care is important.
patients with subclinical hypertensive heart disease, decreases the Empowering patients for success through education, including
likelihood of subsequent cardiovascular disease morbidity and diet (especially sodium intake) and exercise advice, smoking

472 Annals of Emergency Medicine Volume , .  : October 

Levy et al Hypertension in the Emergency Department

cessation counseling, encouragement of medication compliance, Presented at the Society for Academic Emergency Medicine
and provision of information regarding blood pressure goals, is a annual meeting, May/June 2008, Washington, DC; and the
simple activity that can be accomplished during a relatively brief Society for Academic Emergency Medicine annual meeting,
May 2009, New Orleans, LA.
interaction. In patients with persistently poor blood pressure
control, up-titration of antihypertensive therapy from the ED or Address for correspondence: Phillip Levy, MD, MPH, E-mail
referral to a subspecialty hypertension clinic may be needed.
Our data also serve as a reminder that, in general, ED
patients with chronic disease are a high-risk subset, a finding
consistent with a recent analysis of National Health and 1. Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke
Nutrition Examination Survey data that showed a 4-fold greater statistics—2012 update: a report from the American Heart
odds of adverse cardiovascular outcome for hypertensive Association. Circulation. 2012;125:e2-e220.
patients who either lack a usual source of primary care or rely on 2. Egan BM, Zhao Y, Axon RN. US trends in prevalence, awareness,
treatment, and control of hypertension, 1988-2008. JAMA. 2010;
the ED for ongoing management of their blood pressure.49 303:2043-2050.
Simply put, because the ED functions as a site at which initial 3. Bibbins-Domingo K, Pletcher MJ, Lin F, et al. Racial differences in
contact with a health care provider is often made (particularly in incident heart failure among young adults. N Engl J Med. 2009;
the inner city) and blood pressures are measured at every 360:1179-1190.
4. Drazner MH, Dries DL, Peshock RM, et al. Left ventricular
encounter, emergency physicians are uniquely positioned to
hypertrophy is more prevalent in blacks than whites in the general
lessen the overall effect of chronic hypertension in at-risk population: the Dallas Heart Study. Hypertension. 2005;46:124-
communities. By rendering current screening activities more 129.
effective (ie, not just taking in new information but also acting 5. Agabiti-Rosei E, Muiesan ML, Salvetti M. Evaluation of subclinical
target organ damage for risk assessment and treatment in the
on it), we can substantively contribute to much-needed
hypertensive patients: left ventricular hypertrophy. J Am Soc
secondary disease prevention efforts.34 Nephrol. 2006;17:S104-108.
In summary, the point prevalence of subclinical hypertensive 6. Chalmers J. Enhancing risk stratification in hypertensive subjects:
heart disease in this cohort of asymptomatic, predominantly how far should we go in routine screening for target organ
black ED patients with elevated blood pressure is substantial. damage? J Hypertens. 2002;20:1255-1257.
7. Struthers AD, Davies J. Should we add screening for and treating
Efforts that seek to mitigate the adverse cardiovascular effects of left ventricular hypertrophy to the management of all patients
hypertension, especially subclinical structural and functional needing secondary prevention of cardiovascular disease? QJM.
cardiac abnormalities, in this and demographically similar 2003;96:449-452.
patient settings are warranted. 8. Suarez C, Villar J, Martel N, et al. Should we perform an
echocardiogram in hypertensive patients classified as having low
and medium risk? Int J Cardiol. 2006;106:41-46.
Supervising editor: Judd E. Hollander, MD 9. Davis BR, Kostis JB, Simpson LM, et al. Heart failure with
preserved and reduced left ventricular ejection fraction in the
Author contributions: PL conceived the study, designed the antihypertensive and lipid-lowering treatment to prevent heart
trial, obtained research funding, supervised the conduct of the attack trial. Circulation. 2008;118:2259-2267.
trial, oversaw data collection, and directed all data analyses. 10. Okin PM, Devereux RB, Harris KE, et al. Regression of
HY, SC, RZ, and RW assisted with design of the trial, provided electrocardiographic left ventricular hypertrophy is associated with
less hospitalization for heart failure in hypertensive patients. Ann
statistical advice on study design, and assisted with analysis
Intern Med. 2007;147:311-319.
of the data. TB oversaw interpretation of study 11. Verdecchia P, Angeli F, Cavallini C, et al. Blood pressure
echocardiograms. PL drafted the article, and all authors reduction and renin-angiotensin system inhibition for prevention
contributed substantially to its revision. PL takes responsibility of congestive heart failure: a meta-analysis. Eur Heart J. 2009;
for the paper as a whole. 30:679-688.
12. Schocken DD, Benjamin EJ, Fonarow GC, et al. Prevention of
Funding and support: By Annals policy, all authors are required heart failure: a scientific statement from the American Heart
to disclose any and all commercial, financial, and other Association Councils on Epidemiology and Prevention, Clinical
relationships in any way related to the subject of this article Cardiology, Cardiovascular Nursing, and High Blood Pressure
as per ICMJE conflict of interest guidelines (see Research; Quality of Care and Outcomes Research The authors have stated that no such Interdisciplinary Working Group; and Functional Genomics and
Translational Biology Interdisciplinary Working Group. Circulation.
relationships exist. Funding for this study was provided by the
Blue Cross Blue Shield of Michigan Foundation (BCBSM 13. Redfield MM, Rodeheffer RJ, Jacobsen SJ, et al. Plasma brain
Foundation grant 1072.ii). Design of the study, data natriuretic peptide to detect preclinical ventricular systolic or
interpretation, and development of the final article were diastolic dysfunction: a community-based study. Circulation.
conducted independent of the funding agency. 2004;109:3176-3181.
14. Vasan RS, Benjamin EJ, Larson MG, et al. Plasma natriuretic
Publication dates: Received for publication February 7, 2012. peptides for community screening for left ventricular hypertrophy
Revision received March 17, 2012. Accepted for publication and systolic dysfunction: the Framingham Heart Study. JAMA.
March 30, 2012. Available online May 31, 2012. 2002;288:1252-1259.

Volume , .  : October  Annals of Emergency Medicine 473

Hypertension in the Emergency Department Levy et al

15. Wachtell K. Prevention of congestive heart failure in high risk Chamber Quantification Writing Group, developed in conjunction
patients. Eur Heart J. 2009;30:638-639. with the European Association of Echocardiography, a branch of
16. Ammar KA, Jacobsen SJ, Mahoney DW, et al. Prevalence and the European Society of Cardiology. J Am Soc Echocardiogr.
prognostic significance of heart failure stages: application of the 2005;18:1440-1463.
American College of Cardiology/American Heart Association heart 33. Nagueh SF, Appleton CP, Gillebert TC, et al. Recommendations
failure staging criteria in the community. Circulation. 2007;115: for the evaluation of left ventricular diastolic function by
1563-1570. echocardiography. J Am Soc Echocardiogr. 2009;22:107-133.
17. Betti I, Castelli G, Barchielli A, et al. The role of N-terminal PRO- 34. Levy PD, Cline D. Asymptomatic hypertension in the emergency
brain natriuretic peptide and echocardiography for screening department: a matter of critical public health importance. Acad
asymptomatic left ventricular dysfunction in a population at high Emerg Med. 2009;16:1251-1257.
risk for heart failure. The PROBE-HF study. J Card Fail. 2009;15: 35. Kamath S, Markham D, Drazner MH. Increased prevalence of
377-384. concentric left ventricular hypertrophy in African-Americans: will an
18. Chapman JN, Mayet J, Chang CL, et al. Ethnic differences in the epidemic of heart failure follow? Heart Fail Rev. 2006;11:271-
identification of left ventricular hypertrophy in the hypertensive 277.
patient. Am J Hypertens. 1999;12:437-442. 36. Flack JM, Sica DA, Bakris G, et al. Management of high blood
19. Lorber R, Gidding SS, Daviglus ML, et al. Influence of systolic pressure in blacks: an update of the International Society on
blood pressure and body mass index on left ventricular structure Hypertension in Blacks consensus statement. Hypertension.
in healthy African-American and white young adults: the CARDIA 2010;56:780-800.
study. J Am Coll Cardiol. 2003;41:955-960. 37. Luque M, Martell N, Egocheaga I, et al. Left ventricular geometric
20. Taylor HA, Penman AD, Han H, et al. Left ventricular architecture patterns after 1 year of antihypertensive treatment. J Clin
and survival in African-Americans free of coronary heart disease Hypertens (Greenwich). 2005;7:333-338.
(from the Atherosclerosis Risk in Communities [ARIC] study). 38. Devereux RB, Wachtell K, Gerdts E, et al. Prognostic significance
Am J Cardiol. 2007;99:1413-1420. of left ventricular mass change during treatment of hypertension.
21. Wang TJ, Levy D, Benjamin EJ, et al. The epidemiology of JAMA. 2004;292:2350-2356.
“asymptomatic” left ventricular systolic dysfunction: implications
39. Bluemke DA, Kronmal RA, Lima JA, et al. The relationship of left
for screening. Ann Intern Med. 2003;138:907-916.
ventricular mass and geometry to incident cardiovascular events:
22. Nawar EW, Niska RW, Xu J. National Hospital Ambulatory Medical
the MESA (Multi-Ethnic Study of Atherosclerosis) study. J Am Coll
Care Survey: 2005 emergency department summary. Adv Data.
Cardiol. 2008;52:2148-2155.
40. Drazner MH. Left ventricular hypertrophy is more common in black
23. Pitts SR, Niska RW, Xu J, et al. National Hospital Ambulatory
than white hypertensives: is this news? Hypertension. 2004;43:
Medical Care Survey: 2006 emergency department summary. Natl
Health Stat Rep. 2008;7:1-38.
41. Dries DL, Strong MH, Cooper RS, et al. Efficacy of angiotensin-
24. Karras DJ, Ufberg JW, Heilpern KL, et al. Elevated blood pressure
converting enzyme inhibition in reducing progression from
in urban emergency department patients. Acad Emerg Med.
asymptomatic left ventricular dysfunction to symptomatic heart
failure in black and white patients. J Am Coll Cardiol. 2002;40:
25. Kramer H, Han C, Post W, et al. Racial/ethnic differences in
hypertension and hypertension treatment and control in the Multi-
42. Yancy CW, Abraham WT, Albert NM, et al. Quality of care of and
Ethnic Study of Atherosclerosis (MESA). Am J Hypertens. 2004;
17:963-970. outcomes for African Americans hospitalized with heart failure:
26. Shea S, Misra D, Ehrlich MH, et al. Predisposing factors for findings from the OPTIMIZE-HF (Organized Program to Initiate
severe, uncontrolled hypertension in an inner-city minority Lifesaving Treatment in Hospitalized Patients With Heart Failure)
population. N Engl J Med. 1992;327:776-781. registry. J Am Coll Cardiol. 2008;51:1675-1684.
27. Umscheid CA, Maguire MG, Pines JM, et al. Untreated 43. Ko DT, Alter DA, Austin PC, et al. Life expectancy after an index
hypertension and the emergency department: a chance to hospitalization for patients with heart failure: a population-based
intervene? Acad Emerg Med. 2008;15:529-536. study. Am Heart J. 2008;155:324-331.
28. Wong MD, Shapiro MF, Boscardin WJ, et al. Contribution of major 44. Setoguchi S, Stevenson LW, Schneeweiss S. Repeated
diseases to disparities in mortality. N Engl J Med. 2002;347: hospitalizations predict mortality in the community population
1585-1592. with heart failure. Am Heart J. 2007;154:260-266.
29. Baumann BM, Cline DM, Cienki JJ, et al. Provider self-report and 45. De Backer G. Prevention of cardiovascular disease in
practice: reassessment and referral of emergency department asymptomatic people. Heart. 2010;96:477-482.
patients with elevated blood pressure. Am J Hypertens. 2009;22: 46. Kahn R, Robertson RM, Smith R, et al. The impact of prevention
604-610. on reducing the burden of cardiovascular disease. Circulation.
30. Chobanian AV, Bakris GL, Black HR, et al. The seventh report of 2008;118:576-585.
the Joint National Committee on Prevention, Detection, 47. Karras DJ, Kruus LK, Cienki JJ, et al. Evaluation and treatment of
Evaluation, and Treatment of High Blood Pressure: the JNC 7 patients with severely elevated blood pressure in academic
report. JAMA. 2003;289:2560-2572. emergency departments: a multicenter study. Ann Emerg Med.
31. Decker WW, Godwin SA, Hess EP, et al. Clinical policy: critical 2006;47:230-236.
issues in the evaluation and management of adult patients with 48. Delgado MK, Acosta CD, Ginde AA, et al. National survey of
asymptomatic hypertension in the emergency department. Ann preventive health services in US emergency departments. Ann
Emerg Med. 2006;47:237-249. Emerg Med. 2011;57:104-108.e2.
32. Lang RM, Bierig M, Devereux RB, et al. Recommendations for 49. Ndumele CD, Baer HJ, Shaykevich S, et al. Cardiovascular
chamber quantification: a report from the American Society of disease and risk in primary care settings in the United States.
Echocardiography’s Guidelines and Standards Committee and the Am J Cardiol. 2012;109:521-526.

474 Annals of Emergency Medicine Volume , .  : October 

Levy et al Hypertension in the Emergency Department

Appendix E1. Comparison of study subjects with consecutive series of nonenrolled hypertensive patients.
Hypertensive Patients
Study Cohort Not Enrolled,
Variable nⴝ161 nⴝ742 Diff (95% CI)
Mean age, y (SD) 49.8 (8.3) 49.5 (11.9) 0.3 (⫺1.6 to 2.2)
Sex, No. (%)
Male 83 (51.6) 395 (53.2) ⫺1.6 (⫺10.1 to 6.7)
Female 78 (48.4) 347 (46.8) 1.6 (⫺6.7 to 10.1)
Race, No. (%)
Black 151 (93.8) 703 (94.7) ⫺0.9 (⫺6.0 to 2.4)
White 6 (3.7) 21 (2.8) 0.9 (⫺1.6 to 5.2)
Insured, No. (%) 93 (57.8) 442 (59.6) ⫺1.8 (⫺10.3 to 6.4)
Reason for ED visit, No. (%)
Blood pressure high 37 (23.0) 453 (61.1) ⫺38.1 (⫺44.8 to ⫺30.1)
Headache 11 (6.8) 185 (24.9) ⫺18.1 (⫺22.5 to ⫺12.3)
History of hypertension, No. (%) 151 (93.8) 635 (85.6) 8.2 (2.8 to 12.1)
Receiving antihypertensive medication, No. (%) 110 (68.3) 466 (62.8) 5.5 (⫺2.8 to 13.1)
Antihypertensive medication class, No. (%)
Angiotension converting enzyme inhibitors 25 (15.5) 188 (25.3) ⫺9.8 (⫺15.6 to ⫺2.8)
or angiotension receptor blockers
Beta-blockers 20 (12.4) 134 (18.1) ⫺5.7 (⫺10.8 to 0.9)
Clonidine 30 (18.6) 147 (19.8) ⫺1.2 (⫺7.2 to 6.1)
Diuretic 35 (21.7) 208 (28.0) ⫺6.3 (⫺12.8 to 2.4)
Calcium-channel blockers 31 (19.3) 122 (16.4) 2.8 (⫺3.2 to 10.0)
Other 9 (5.6) 28 (3.8) 1.8 (⫺1.3 to 6.7)
History of diabetes, No. (%) 30 (18.6) 105 (14.2) 4.5 (⫺1.4 to 11.6)
Receiving diabetic medication, No. (%) 23 (14.3) 78 (10.5) 3.8 (⫺1.4 to 10.3)
Insulin 11 (6.8) 30 (4.0) 2.8 (⫺0.6 to 7.9)
Oral agents 12 (7.5) 48 (6.5) 1.0 (⫺2.7 to 6.3)
History of stroke, No. (%) 15 (9.3) 37 (5.0) 4.3 (0.3 to 10.0)
Social history, No. (%)
Active cigarette smoker 78 (48.5) 342 (46.1) 2.4 (⫺6.1 to 10.8)
Drinks alcohol 76 (47.2) 217 (29.2) 18.0 (9.7 to 26.3)
Uses cocaine 14 (8.7) 52 (7.0) 1.7 (⫺2.3 to 7.3)
Mean initial blood pressure, mm Hg (SD)
Systolic 183.9 (25.1) 194.0 (23.5) ⫺10.1 (⫺14.2 to ⫺6.0)
Diastolic 109.5 (14.4) 110.1 (14.6) ⫺0.6 (⫺3.1 to 1.9)

Volume , .  : October  Annals of Emergency Medicine 474.e1