Topical Nitroglycerin Ointment for Treatment of Acute Hypertension in Hospitalized Inpatients

Abstract

Background:

Hypertension in the hospital setting is common; however, guidelines provide limited guidance specific to the inpatient setting. Acute antihypertensive treatment options can be limited in this setting by monitoring requirements of intravenous medications and patients’ inability to take oral medications. A possible treatment choice used to treat acute hypertension is nitroglycerin ointment. Nitroglycerin is not approved by the Food and Drug Administration for this condition, and limited evidence exists to support this indication.

Objective:

To evaluate the statistical and clinical effectiveness of nitroglycerin ointment as a treatment option for acute hypertension based on a 20 mm Hg or greater reduction in systolic blood pressure.

Methods:

A retrospective chart review at a large tertiary community teaching hospital was performed on all adult noncardiac inpatients with an episode of acute hypertension that resulted in the administration of nitroglycerin ointment.

Results:

Seventy-two patients met inclusion criteria with a total of 112 applications of nitroglycerin ointment. Of the 112 applications, systolic blood pressure decreased 20 mm Hg or more in 42% of occurrences with a median decrease of 16 mm Hg.

Conclusion:

Study results suggest possible efficacy of nitroglycerin ointment for the treatment of acute hypertension in noncardiac hospitalized patients.

Introduction

Hypertension is one of the most prevalent conditions in the United States, affecting 33% of adults age 20 and older. ¹ Increased blood pressure is a major risk factor associated with overall coronary heart disease morbidity and mortality. ² In fact, 16.5% of all deaths worldwide are attributed to hypertension. This includes 45% of deaths due to coronary artery disease. ¹

Although chronic hypertension is an established risk factor for cardiovascular, cerebrovascular, and renal disease, acute hypertension can result in end-organ damage. Patients with a systolic blood pressure greater than 179 mm Hg or a diastolic blood pressure greater than 109 mm Hg are usually considered to be having a “hypertensive crisis.” ³ Severe elevations in blood pressure in the presence of acute end-organ damage are classified as hypertensive emergencies. Essentially, hypertensive urgencies are hypertensive crises in the absence of acute target-organ involvement. ³

Inpatient hypertension is highly prevalent ranging from 50.5% to 72%, and studies suggest that 37% to 44% of these patients have uncontrolled hypertension with a blood pressure reading ≥140/90. ⁴ Hypertensive urgencies are managed in a monitored setting using oral antihypertensive agents to lower blood pressure in the next 24 hours to 2 to 3 days without specific goals of treatment. In contrast, hypertensive emergencies require intravenous antihypertensive drugs to lower blood pressure within the first few hours of presentation with the objective of reversion or reducing target organ damage. ⁵ Current consensus guidelines do not address the management of inpatient hypertension, particularly in patients with acutely elevated blood pressure hospitalized for reasons other than hypertension.

Topical nitroglycerin ointment is used by some practitioners for the treatment of acute hypertension (hypertensive urgencies) in hospital inpatients. Nitroglycerin ointment is approved in the United States for symptomatic relief of acute angina. However, given the venodilator properties of nitroglycerin, a common effect of the drug is a decrease in arterial blood pressure making its use for acute hypertension plausable. ⁶ Potential advantages of using nitroglycerin ointment for this indication is its rapid onset and offset as well as utility in patients who are unable to swallow medications. Additionally, many hospitals have nursing policies that may prohibit the use of intravenous vasoactive drugs in patients without cardiac telemetry monitoring, making nitroglycerin ointment a potentially attractive means for controlling acute hypertension.

Currently, there is a paucity of published data concerning the use of nitroglycerin ointment in noncardiac patients for the treatment of acute hypertension. There have been a few studies published regarding the use of nitroglycerin ointment for the prevention of acute hypertension associated with a procedure. ^{6 –8} Only 1 trial was found, published in the mid-1980s, that suggested transdermal nitroglycerin may be effective for severe outpatient hypertension. ⁹

Despite the lack of evidence demonstrating either efficacy or safety of nitroglycerin in this patient population, its use continues. A retrospective observational study was conducted to examine nitroglycerin ointment use for acute blood pressure control in general hospital inpatients.

Methods

A retrospective observational review of all patients admitted to a large tertiary care, community teaching hospital was conducted for those prescribed nitroglycerin ointment for the 5 months prior to starting data collection. Inclusion criteria included age 18 years or more and an episode of acute hypertension (≥140/90 mm Hg), resulting in administration of nitroglycerin ointment. At the time of the study, the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) guidelines defined hypertension as a blood pressure ≥140/90 mm Hg, study definition was based on this published cutoff. ¹⁰ Patient exclusion criteria included chief complaint of chest pain or angina, major surgery during hospital stay, diagnosis of coronary artery disease, nitroglycerin indicated for anal fissures, pregnant, or patient blood pressure was not rechecked ≤2 hours after the index administration of nitroglycerin ointment. The time restraint was based on evidence that nitroglycerin levels achieve steady state within an hour of ointment application. ⁶ The study was granted expedited institutional review board approval.

Study data were extracted from electronic medical records and included demographic and anthropometric information. Clinical data included primary diagnosis classified by admitting International Classification of Diseases, Ninth Revision (ICD-9) code, past medical history relating to hypertension and angina, cardiac medications, blood pressure reading prior to administration of nitroglycerin, blood pressure reading after administration of nitroglycerin, amount of nitroglycerin ointment applied (sq in), and documented side effects.

The primary study outcome was the evaluation of the statistical and clinical effectiveness of nitroglycerin ointment as a treatment option for acute hypertension based on a 20-mm Hg or greater reduction in blood pressure. This target decrease was based on several factors including (1) general consensus from physicians in the study health system would consider this a significant decrease; (2) the study using transdermal nitroglycerin listed previously found a mean decrease of 24 mm Hg in the active treatment arm ⁹ ; and (3) a small study in 2006 examining nitroglycerin and large artery stiffness revealed a reduction in systolic blood pressure of approximately 16 mm Hg after application. ¹¹ To authors’ knowledge, no official evidence-based guidelines exist concerning the goals of treatment of acute hypertension in hospitalized patients. Secondary study outcome was the quantification of documented adverse effects of nitroglycerin ointment therapy.

Descriptive statistics were performed with categorical data presented as counts with percentages and continuous data presented as measures of central tendency and dispersion (ie, medians and interquartile range [IQR]). The calculation of these quartiles for clinical data controlled for multiple contributions of patients by using normalized/relative weights. The median difference in patient blood pressures before and after administration of nitroglycerin ointment was presented with a 95% confidence interval (CI); calculated using a nonparametric bootstrap percentile interval, based on 10 000 resamples with replacement. ¹²

A hierarchical regression model was used to determine predictors of a 20-mm Hg or greater reduction in blood pressure after administration of ointment. The dependent variable in the model was binary and the model controlled for individual patient level random effects. Odds ratio was calculated for a 20-mm Hg or greater decrease in blood pressure along with 95% CI. Level of significance was set at .05 for candidate variables inclusion in the final model. Analyses were performed with SAS, version 9.3 (SAS Inc, Cary, North Carolina).

Results

Five hundred and one patients were prescribed nitroglycerin ointment at the study hospital during the 5-month period. Of these patients, 72 (14%) met inclusion criteria and 429 (86%) were excluded. Reasons for exclusion were past medical history of coronary artery disease (185 [43%]), chief complaint of chest pain and/or angina (130 [30%]), surgery (58 [14%]), blood pressure not rechecked within 2 hours after application of nitroglycerin ointment (43 [10%]), and incomplete data (13 [3%]).

Included study sample patients were predominantly female (64%) and had a median age of 67 years (IQR: 48.5-79.5). Sixty-nine percent of patients had a past medical history of hypertension and median weight was 75.2 kg (IQR: 63.3-86.8). Primary diagnosis was classified according to ICD-9 code by system, with 28% having an admitting diagnosis related to a circulatory system disease; 15% signs, symptoms, and ill-defined conditions; 15% injury and poisoning; and 14% disease of the digestive system (Table 1).

Table 1.

Patients Receiving Nitroglycerin Ointment Data.^a,b

Admitting ICD-9 Code by System
Diseases of the circulatory system	20 (28)
Injury and poisoning	11 (15)
Symptoms, signs, and ill-defined conditions	11 (15)
Diseases of the digestive system	10 (14)
Diseases of the genitourinary system	7 (10)
Diseases of the respiratory system	5 (7)
Diseases of the nervous system	2 (3)
Neoplasms	2 (3)
Diseases of the musculoskeletal system and connective tissue	1 (1)
Endocrine, nutritional, and metabolic disease, and immunity disorders	1 (1)
Infectious and parasitic diseases	1 (1)
Mental disorders	1 (1)

Abbreviation: ICD-9, International Classification of Diseases, Ninth Revision.

^an = 72.

^bData presented in counts (percentages).

The median number of blood pressure measurements per patient was 1 (IQR: 1-2). Within the 112 applications of nitroglycerin ointment, there were 47 (42%) occurrences when the systolic blood pressure decreased by 20 mm Hg or more after an application. In addition, systolic blood pressure went down by at least 5, 10, or 15 mm Hg after 74%, 70%, or 54% of applications, respectively. Table 2 depicts the treatment statistics for patients receiving nitroglycerin ointment. The median prenitroglycerin systolic blood pressure was 181 mm Hg (IQR: 168-192) with a median change after the administration of nitroglycerin of -16 (95% CI: -13 to -20) mm Hg. This median change in blood pressure represented a 10% decrease.

Table 2.

Blood Pressure and Nitroglycerin Ointment Treatment Data.^a,b,c

Variable
Pretreatment systolic blood pressure, mm Hg	181 (168-192)
NTG ointment amount, sq in	1 (1-1)
Time to follow-up blood pressure post-NTG ointment, minutes	72 (55-95)
Systolic blood pressure post-NTG ointment, mm Hg	160 (148-180)
Decrease in systolic blood pressure post-NTG ointment	16 (5-30)
Concomitant blood pressure medication	18 (16%)

Abbreviation: NTG, nitroglycerin.

^an = 114.

^bData presented with weighted medians (interquartile ranges) or counts(percentages).

^cResults included all blood pressure measurement data for the 72 study patients.

Eighteen (16%) patients received a concomitant blood pressure medication. Of these 18 patients, 3 patients received 2 concomitant blood pressure medications in addition to the administration of nitroglycerin ointment. In all, 3 patients received intravenous medications and 15 received oral medications. Table 3 depicts the concomitant blood pressure medications that were administered.

Table 3.

Concomitant Blood Pressure Medications.^a,b,c

Medication and Route
Clonidine PO	7
Amlodipine PO	3
Hydralazine PO	2
Hydralazine IV	2
Metoprolol PO	2
Metoprolol IV	1
Labetalol PO	1
Labetalol IV	1
Carvedilol PO	1
Propranolol PO	1

Abbreviations: PO, per oral; IV, intravenous.

^an = 18.

^bData presented in counts.

^cThree patients received 2 concomitant blood pressure medications.

The multilevel model conducted to predict a 20-mm Hg or greater decrease in blood pressure after nitroglycerin application revealed only pretreatment blood pressure as a significant predictor (P value = .0008). In particular, for every 1 mm Hg, the blood pressure was higher prior to treatment application, the odds ratio of a 20 or more unit decrease were 1.05 (95% CI: 1.02-1.08) times greater. Variables not found to be significant and excluded from the model were age, sex, weight, amount of nitroglycerin (sq in), amount of time between application and measurement (minutes), and concomitant blood pressure medication.

Of the 72 study patients, the adverse effect of headache was documented in 11 (15%) patients with 13 total occurrences. This was the only type of adverse drug event documented. Although several patients had an admitting diagnosis that may have contributed to headache, of these 11 patients, 2 had an admitting diagnosis of intracranial hemorrhage, 2 had a diagnosis of subdural hematoma, and another had a diagnosis of hypertensive emergency.

Discussion

The presented retrospective study showed a mean decrease in systolic blood pressure of 16 mm Hg. Although the application of nitroglycerin ointment may have caused a decrease in systolic blood pressure in most patients, its typical effect did not achieve the 20 mm Hg decrease. The clinical relevance of this difference in blood pressure is likely situation specific based on patient factors. In addition, treatment did not seem to be harmful, with headache being the only charted adverse effect in the study sample.

It was found that of the 112 applications of nitroglycerin ointment for the treatment of hypertensive urgencies, the systolic blood pressure went down 20 mm Hg or more in about a third of the cases with a blood pressure decrease of about 9%. Similar studies done with single doses of oral clonidine, on the other hand, demonstrated a much greater reduction with a mean decrease in systolic blood pressure of 50 to 56 mm Hg. ^13,14 A trial done with single doses of oral labetalol and nifedipine in the setting of acute hypertension showed a mean decrease in systolic blood pressure of 42.6 and 34.9 mm Hg. ¹⁵ Multiple studies have demonstrated that oral clonidine, labetalol, captopril, and nifedipine are all effective agents for the treatment of hypertensive urgencies. ¹⁶

Although evidenced-based guidelines for the outpatient treatment of hypertension are available from several sources, similar guidelines for treating acute hypertension (eg, hypertensive urgency) do not exist. Oral options for treating acute hypertension are often not feasible in patients with difficulty swallowing. Also many hospitals have policies that restrict the use of vasoactive drugs to patients monitored with cardiac telemetry. This leaves physicians in a conundrum on how to practicably treat some noncardiac patients with acute hypertension. At the study hospital, for at least 15 years, nitroglycerin ointment has been used to “fill the gap” of an antihypertensive drug with rapid onset and offset but is not given orally nor requires special monitoring for its use. This is in spite of a lack of information concerning nitroglycerin’s use in this context.

Acute hypertension in nonsurgical patients is commonly treated in Europe with intravenous nitroglycerin, while β-adrenergic blockers are more commonly used in the United States. ^17,18 Oral nifedipine was commonly used throughout the latter quarter of the 20th century for hospitalized patients with acute hypertension until increased reports of adverse effects with this strategy surfaced. ⁵ In general, topical nitroglycerin ointment would seem to possibly fill an important niche in the treatment of this commonly encountered problem.

Interestingly, a literature search of MEDLINE found no previously published reports on the use of this modality for acute hypertension in noncardiac inpatients with the exception of 3 studies on its use in patients undergoing procedures. Two studies observed the cardiovascular changes with and without nitroglycerin ointment before electroconvulsive therapy in 2 different sittings. In both of these studies, patients served as their own control. The nitroglycerin ointment was applied 45 minutes before electroconvulsive therapy. In both studies, the results suggest that nitroglycerin ointment as a premedication before electroconvulsive therapy significantly attenuates electroconvulsive therapy-induced hypertension. ^6,8 The third study observed the effects of nitroglycerin ointment versus saline applied intranasally prior to laryngoscopy and tracheal intubation in patients treated with a β-blocker. Nitroglycerin was applied 30 seconds prior to the procedure. The study concluded intranasal nitroglycerin is a safe, rapid, and convenient method of attenuating the pressor response to laryngoscopy and tracheal intubation in treated patients with hypertension. ⁷ All 3 articles suggested the use of a single dose of nitroglycerin ointment may be useful as a pretreatment medication for procedures that may induce hypertension. These studies used nitroglycerin ointment to prevent hypertension rather than for treatment of an acute elevation in blood pressure.

An area of inquiry that neither the studies mentioned previously nor our investigation had the ability to assess was the role of genomics in the development of hypertension or its response to treatment. Over 40 generic variants associated with the development of hypertension have been described. ¹⁹ The mechanisms surrounding these variations and their propensity for hypertension include abnormal regulation of endothelial nitric oxide synthase activity (CaMK4) and G-protein-coupled receptor kinases, which approximately one-third of all pharmacologic treatments for hypertension have activity for. ^20,21 To our knowledge, the role of these polymorphisms in acute hypertension remains to be described.

One point that should be mentioned is the notion that acute hypertension without symptoms of target organ damaged should not be treated at all. Experts have argued that the paucity of data suggesting benefit of treating these patients combined with the risk of adverse effects should deemphasize the desire to “treat the numbers” instead of assessing the whole patient. ²² Despite this prescribers are often reluctant to do just that. ²³

Another concern about the routine use of nitroglycerin ointment is the development of tolerance with continuous nitrate dosing. ²⁴ Further research is needed to determine whether a similar tolerance develops with the use of nitroglycerin ointment in acute hypertension.

Limitations

The presented study was not without limitations. Data were based on a retrospective review of existing records suspect to typical data quality concerns. Treatment was not standardized nor randomized within the sample, preventing the examination of normal blood pressure declines in a nontreatment group. The study was conducted at a single hospital, which may or may not be reflective of all hospitals of varying or comparable size. For each patient, there were limited blood pressure measurements. The study included 18 patients who received a concomitant antihypertensive agent, which may or may not have affected overall study results. The outcome assessed in the study, a blood pressure reduction of greater than 20 mm Hg, may or may not have been the goal of the treating physician. The sample size makes drawing conclusions on possible dosage effects of treatments difficult. Blood pressure cuff size as well as the use of automated versus manual reading could not be collected. Finally, higher baseline blood pressures were intuitively related to the ability to achieve a 20-mm Hg decrease likely due to indirect mathematical coupling and their greater distance from a possible floor effect.

Conclusion

Study results suggest possible efficacy of nitroglycerin ointment for the treatment of acute hypertension in noncardiac hospitalized patients. Further research including a randomized clinical study could help provide further information on the potential role the modality may have in a commonly encountered clinical situation. However, given the low likelihood of this occurring, the presented study may be the only such investigation in this area for the foreseeable future.