a b s t r a c t

Introduction: The effect profile of differing antihypertensive agents is well studied, but minimal data regarding the interaction between Hemodynamic response and presenting Blood pressure exist.

Hypothesis: Achievement of target BP is less likely in patients with higher initial BPs.

Methods: This is a substudy of the multicenter safety and efficacy Evaluation of intravenous Cardene (nicardipine) and Labetalol Use in the Emergency department (CLUE) trial that randomized patients to Food and Drug Administration-recommended intravenous dosing of nicardipine or labetalol to reach a physician predefined systolic BP (SBP) and target range (TR) of +-20 mm Hg within 30 minutes. The proportion achieving TR was assessed as a function of initial SBP, and dichotomized comparisons were made using median SBP. Likelihood of a final BP within TR was modeled using logistic regression with forced inclusion of initial BP as a categorical variable.

Results: A total 223 patients were enrolled; 115 (51.6%) had an initial SBP greater than the median 202 mm Hg. The median SBP (interquartile range) of the high BP group was 218 (210-228) mm Hg vs the low BP group 190 (182-197) mm Hg (P b 0.0001). No groupwise differences existed except that the high group had higher mean (SD) serum creatinine level at baseline (3.1 [3.9] vs 1.9 [2.3], P = .008). The proportion of patients achieving SBP within TR at 30 minutes did not differ (85.2% [98 of 115] vs 88.9% [96 of 108], P = .42). Randomization to nicardipine (odds ratio = 2.85; 95% confidence interval, 1.16-7.01), but not initial SBP (odds ratio = 0.60; 95% confidence interval, 0.25-1.44), was associated with achievement of target SBP at 30 minutes.

Conclusion: Initial SBP is not a predictor of the ability to achieve a prespecified target range SBP within 30 minutes.

(C) 2014

Introduction

Significantly Elevated blood pressure (BP) is commonly seen in the emergency department (ED); and, when associated with acute target- organ damage, immediate intervention with intravenous (IV) antihyper- tensive therapy is warranted. Despite this, few clinical trials have studied the optimal pharmacological agent for ED treatment of such patients. Two agents commonly used for this purpose, nicardipine and labetalol, were recently compared in the Evaluation of intravenous Cardene (nicardipine) and Labetalol Use in the Emergency department (CLUE) study, a randomized comparison trial that evaluated BP control effects in ED patients with acute hypertension [1]. Using US Food and Drug Administration-recommended IV doses of nicardipine and labetalol, CLUE showed that patients were more likely to achieve a physician- determined prespecified target systolic BP (SBP) within 30 minutes when treated with nicardipine.

Although CLUE provides important insight into the general approach to management of elevated BP in the ED, there is tremendous variation in the degree of hypertension; and spectrum bias may exist

* Corresponding author.

E-mail addresses: [email protected], [email protected] (S. Farias).

with regard to antihypertensive response. However, whether higher BP is more difficult to acutely manage has been poorly studied. The identification of difficult-to-manage Hypertensive patients would have clinical Therapeutic implications for ED care. Accordingly, we sought to use the CLUE data set to determine if achievement of target BP is less likely in patients with higher initial BPs.

Methods

Using the CLUE database, we performed a post hoc analysis to determine therapeutic response based on a dichotomous division of hypertensive patients who received IV Antihypertensive therapy. The CLUE trial was a multicenter study performed at 13 US academic EDs. Its purpose was to compare the safety and efficacy of Food and Drug Administration-recommended dosing of IV nicardipine vs IV labetalol for management of acute hypertension. After a prespecified target BP was defined by the treating physician, patients were randomized to receive an open-label, IV titration of either nicardipine or labetalol to reach this target +-20 mm Hg within 30 minutes. Inclusion and exclusion criteria have been published elsewhere, but all study patients were required to be older than 18 years and have an SBP greater than 180 mm Hg on 2 consecutive cuff measurements

http://dx.doi.org/10.1016/j.ajem.2014.03.021

0735-6757/(C) 2014

834 S. Farias et al. / American Journal of Emergency Medicine 32 (2014) 833-836

Table 1

Demographic characteristics

	Total, N = 223	SBP b 202, n = 108	SBP >= 202, n = 115	P value
Age (mean +- SD)	52.4 +- 14.5	52.1 +- 14.3	52.8 +- 14.7	.720
Female (%)	52.9	49.1	56.5	.265
Body mass index (mean +- SD)	30.4 +- 8.3	30.4 +- 9.5	30.4 +- 7.2	.342

Race (%) African American	77.0	73.8	80.0	.275
White Social history	22.5	25.2	20.0	.351

History of cigarette smoking (%)	58.3	61.1	55.7	.409
Current smoker (%)	35.4	36.1	34.8	.836
History of stimulant use (%)	17.9	17.6	18.3	.897
Current stimulant use (%)	5.4	7.4	3.5	.194
Complaints at presentation Chest pain (%)	26.5	27.8	25.2	.665
Confusion (%)	3.6	5.6	1.7	.160
Headache (%)	47.5	46.3	48.7	.720
Syncope (%)	19.3	21.3	17.4	.460
Shortness of breath (%)	27.3	27.8	27.0	.891
Hematuria (%)	1.3	0.0	2.6	.247
Epistaxis (%)	0.9	0.9	0.9	1.000
Back pain (%)	13.4	13.9	13.0	.853

10 minutes apart [1]. Major exclusion criteria included contraindica- tions to giving either a ?-blocker or calcium channel blocker, or clinical scenarios where a preferred agent (eg, a ?-blocker in acute myocardial infarction) was indicated. Enrolled patients were ran- domized in a 1:1 ratio to receive either a nicardipine infusion starting at 5 mg/h and increasing by 2.5 mg/h until target SBP was reached (or a max rate of 15 mg/h was achieved) or labetalol 20 mg by IV bolus over 2 minutes, followed by boluses of 20, 40, or 80 mg every 10 minutes until the target SBP was achieved or a total of 300 mg had been administered. The primary end point was achievement of target SBP range within 30 minutes. Automated, Brachial cuff-monitored BP was obtained every 5 minutes over the 30-minute study period. To assess for delayed adverse effects, vital sign monitoring continued for 6 hours or until ED discharge after drug administration. Additional antihypertensive therapy (at the discretion of the treating physician) administered past 30 minutes was tracked along with laboratory and electrocardiographic results.

For this analysis, patients enrolled in the CLUE study were dichoto- mized using the median presenting SBP as the partition point. Individuals above and below the median were then evaluated as to the proportion achieving the primary outcome. Binary characteristics between the two cohorts were evaluated by ?² test; and continuous parameters, by Student t test. Likelihood of a final BP within TR was modeled using logistic regression with forced inclusion of initial SBP as a categorical variable.

Table 2

Medical history and current medications

Results

A total of 226 patients were randomized in the CLUE study; but 2 were excluded from the labetalol group and 1 from the nicardipine group after withdrawing consent, resulting in a final study cohort of

223 patients. The patients were 52.9% female and 77.0% African American (Table 1). Overall, the median (interquartile range [IQR]) SBP on presentation was 202 (190-219) mm Hg, with 115 (51.6%) above and 108 (48.4%) below this cut point. Median (IQR) SBP in the higher-BP group was 218 mm Hg (210-228 mm Hg) compared to 190 mm Hg (182-197 mm Hg) in the lower-BP cohort (P b .0001). Initial median heart rate was similar between groups. Although a greater proportion of those above the median SBP were randomized to treatment with nicardipine (53.9% vs 43.5%, P = .12), this difference was not statistically significant. Headache was the most common presenting chief complaint, followed by shortness of breath and chest pain, all of which occurred with similar frequencies between groups. Tobacco use, which was present in more than half of patients, and the use of stimulants, self-reported by nearly 20% of study participants, were also equal among subgroups.

As noted in Table 2, more than 95% of study participants had a prior history of hypertension with equivalence for this and all other comorbidities between study groups. Consistent with the high prevalence of hypertension, oral antihypertensive therapy use was common, with nearly 40% on ?-blockers at baseline, roughly 30% on an angiotensin-converting enzyme or angiotensin receptor blocker, about 25% on a diuretic, and close to 20% taking a calcium channel blocker, with no differences between stratified cohorts. There was also no

Medical history

Coronary artery disease (%)	13.6	12.3	14.9	.567
Diabetes (%)	27.9	23.1	32.5	.122
End-stage renal disease on	12.7	10.3	14.9	.301
dialysis (%)
Heart failure (%)	9.1	8.4	9.7	.733
Stroke (%)	7.3	6.6	8.0	.699
Hypertension (%) Prior medications	95.5	95.3	94.8	.852
Angiotensin-converting	24.2	26.9	21.7	.373
enzyme inhibitors (%)
Angiotensin receptor	10.3	10.2	10.4	.951
blockers (%)
?-Blockers (%)	40.4	38.9	41.7	.665
calcium channel blockers (%)	22.4	19.4	25.2	.302
Nitrates (%)	6.7	6.5	7.0	.887
Diuretics (%)	25.6	22.2	28.7	.268
Antiadrenergic agents (%)	9.4	5.6	13.0	.056

Total,

N = 223

SBP b 202,

n = 108 (%)

SBP >= 202,

n = 115 (%)

P value

difference in the number of Antihypertensive medications that patients were taking at baseline between groups (median [IQR]: 1 [0-2] vs 1 [0-3], P = .443). As shown in Table 3, with the exception of worse renal function among those with initial SBPs exceeding the median (mean blood urea nitrogen: 25.1 vs 17.7, P = .001) and mean creatinine (3.1 vs 1.9 mg/dL, P = .008), there were no groupwise differences in pretreatment laboratory or electrocardiographic findings.

A scatter plot of presenting SBP vs Percentage change in SBP during the treatment period is displayed in the Figure. Initial, randomized treatment resulted in similar achievement of SBP within the predefined target range between cohorts (85.2% vs 88.9%, P =

.415); but patients (Table 4) with an initial SBP above median took, on average, 3.5 minutes longer to reach target. Patients with higher initial SBP also required more medication titrations (2.3 vs 1.8, P = .004) and received additional antihypertensive medications after the initial treatment period with greater frequency (26.1% vs 12.0%, P = .008) than did the lower-SBP group. To evaluate the variability in BP control

S. Farias et al. / American Journal of Emergency Medicine 32 (2014) 833-836 835

Table 3

Laboratory and electrocardiographic data

Laboratory variable (mean +- SD)	Total, N = 223	SBP b 202, n = 108	SBP >= 202, n = 115	P value
% Hematocrit (mean +- SD)	39.8 +- 5.8	41.0 +- 5.8	38.8 +- 5.6	.009
Blood urea nitrogen (mg/dL) (mean +- SD)	21.7 +- 18.9	17.7 +- 14.6	25.1 +- 21.5	.001
Creatinine (mg/dL) (mean +- SD)	2.6 +- 3.3	1.9 +- 2.3	3.1 +- 3.9	.008
Sodium in (mEq/L) (mean +- SD)	138.0 +- 3.9	137.8 +- 4.0	138.2 +- 3.7	.127
B type natriuretic peptide (pg/mL) (mean +- SD)	1251 +- 1930	1243 +- 2229	1259 +- 1619	.191
Troponin I (ng/dL) (mean +- SD)	0.2 +- 0.8	0.2 +- 0.5	0.2 +- 1.0	.191
Electrocardiogram Normal (%)	27.4	26.6	28.3	.853
No known acute changes (%)	44.6	43.6	45.7
Abnormal (%)	28.0	29.8	26.1

during the treatment period, areas under the curve (AUCs) were calculated, multiplying time spent (in minutes) outside the pre- specified target BP range by the depth above or below the range in millimeters of mercury (Table 4). Both total time outside of target range (300.5 vs 123.2 mm Hg*min, P b .001) and time above the upper limit of target range (297.7 vs 111.4 mm Hg*min, P b .001) were substantially greater for those with a presenting SBP above the median; however, there was no difference in treatment overshoot as reflected by AUC below the lower limit of target range. On multivariate modeling, randomization to nicardipine (odds ratio = 2.85; 95% confidence interval, 1.16-7.01) but not initial SBP (odds ratio = 0.60; 95% confidence interval, 0.25-1.44) was associated with achievement of target SBP at 30 minutes.

Discussion

In this post hoc analysis of the CLUE data set, we found early achievement of target SBP to be independent of presenting SBP. That said, patients with higher SBP on presentation may require slightly more time with more aggressive medication administration or titration, and more medications to attain prespecified BP goals. Such information is important for clinicians managing individuals with suspected or confirmed Hypertensive emergencies, where rapid BP reduction may be needed to improve outcomes [2]. Further, it is generally recom- mended that, when treating hypertensive emergencies, the target for lowering BP should be 20% to 30% of the initial presentation BP, as overshoot of the target range could result in an iatrogenic adverse event (eg, watershed infarct in the setting of an ischemic stroke). Conse- quently, concern for target BP overshoot may result in less aggressive IV antihypertensive therapy. However, in this analysis, we found that

Fig. SBP vs percentage change in SBP during treatment period.

overshoot was uncommon, not severe, and relatively brief when occurring. Importantly, the probability of therapeutic iatrogenic hypotension was no more common in study participants presenting with a BP below the median than in those above.

Few other studies exist, especially within an ED environment, comparing the ability to achieve BP control for patients with acute hypertension based upon presenting BP. But a randomized trial did evaluate nicardipine vs labetalol administered to patients with acute stroke requiring BP management, and showed that a greater proportion of nicardipine-treated patients reached goal BP within 60 minutes (89% vs 25%, P b .001) [3]. This study was conducted in the ED, but the patient population included only acute stroke patients and did not differentiate levels of hypertension on presentation. Wallin et al [4] did demonstrate that IV nicardipine lowered BP more quickly (0.31 +- 0.13 vs 1.11 +- 0.36 hours) when started and maintained at higher IV drip rates (15 vs 4 mg/h). But this 18-patient study was small, did not categorize levels of hypertension on presentation, and was not carried out in the ED. A separate study of 139 postoperative hypertensive patients comparing IV nicardipine with sodium nitro- prusside concluded that these medications were equally effective for controlling elevated BP; but the study did not categorize levels of initial hypertension and was conducted on postoperative, not ED, patients [5].

This study has several limitations. Most importantly, this is a secondary analysis of the primary CLUE study, which was not specifically designed to determine potential therapeutic resistance as a function of presenting BP. Nonetheless, it is a well-matched cohort for this purpose as evidenced by the fact that the only statistically significant difference between participants in the higher- and lower-BP groups with regard to medical history, prior therapies, Presenting complaints, and initial laboratory results was that renal dysfunction occurred more commonly in the higher-BP group. This is not unexpected, as hypertension is a well-described risk factor for kidney disease and patients with poorer BP control are more prone to renal injury. In addition, SBP control was only evaluated for the first 30 minutes posttreatment; and we were unable to determine if a clinically relevant difference in BP control occurred later in the patient’s course. It should also be noted that the CLUE study did not include critically ill participants and thus excluded individuals with the potential for more significant complications from hypertension for whom acute BP control would be most warranted. Finally, nearly 80% of those enrolled were African American, reducing generalizability of these findings to the broader hypertensive population.

Conclusion

Presenting SBP does not appear to affect the ultimate ability to reduce BP for patients with marked, acute hypertension in the ED when treated with either IV nicardipine or IV labetalol.

836 S. Farias et al. / American Journal of Emergency Medicine 32 (2014) 833-836

Table 4

Blood pressure effects

	Total, N = 223	SBP b 202, n = 108	SBP >= 202, n = 115	P value
Initial SBP (median with IQR)	211.0 (198.0-226.0)	201.0 (192.0-211.0)	220.0 (208.0-238.0)	b.001
Initial heart rate (mean +- SD)	86.1 +- 17.2	86.2 +- 18.5	86.0 +- 16.0	.969
Met target within 30 min (%)	87.0	88.9	85.2	.415
Time to target (mean +- SD)	11.3 +- 7.8	9.5 +- 6.8	13.0 +- 8.3	b.001
AUC = SBP in mm Hg x time outside of target range
AUC, total outside target range in mm Hg*min (mean +- SD)	214.6 +- 264.0	123.2 +- 162.6	300.5 +- 309.1	b.001
AUC, above upper limit of target range in mm Hg*min (mean +- SD)	207.5 +- 266.0	111.4 +- 161.2	297.7 +- 310.5	b.001
AUC, below lower limit of target range in mm Hg*min (mean +- SD)	7.1 +- 40.5	11.8 +- 54.9	2.8 +- 18.2	.204
Number of titrations (mean +- SD)	2.0 +- 1.5	1.8 +- 1.5	2.3 +- 1.4	.004
Rescue medications Any rescue medication (%)	19.3	12.0	26.1	.008
1 Rescue medication needed (%)	76.7	84.6	73.3	.696
>=2 Rescue medications needed (%)	23.3	15.4	26.7	.696

References

1. Peacock WF, Varon J, Baumann BM, et al. CLUE: a randomized comparative effectiveness trial of IV nicardipine versus labetalol use in the emergency department. Crit Care 2011;15:R157.
2. Rhoney D, Peacock WF. intravenous therapy for hypertensive emergencies, part 1. Am J Health Syst Pharm 2009;66:1343-52.
3. Liu-DeRyke X, Levy PD, Parker Jr D, et al. A prospective evaluation of labetalol versus nicardipine for blood pressure management in patients with acute stroke. Neurocrit Care Aug 2013;19(1):41-7.
4. Wallin JD, Cook ME, Blanski L, et al. Intravenous nicardipine for the treatment of Severe hypertension. Am J Med Sci 1988;85:331-8.
5. Halpern NA, Goldberg M, Neely C, et al. Postoperative hypertension: a multicenter, prospective, randomized comparison between intravenous nicardipine and sodium nitroprusside. Crit Care Med Dec 1992;20(12):1637-43.