A solution to an almost unrecognizable problem: how can we treat a patient before differentiating between hypertensive urgency and emergency?

KEYWORDS:

• Arterial hypertension
• hypertensive emergency
• hypertensive urgency
• emergency department

1. Introduction

Very high blood pressure (BP) is a common and significant problem in the emergency department (ED), affecting 4.6% of admitted patients [1]. Hypertensive urgency (HTN-U) is a diagnosis of exclusion [2] (for what time is needed) and the prognosis of HTN-U patients is not different from that of asymptomatic uncontrolled hypertension (HTN) population, with similar blood pressures as pointed out in the European Society of Cardiology (ESC) Council on hypertension position document [2]. Patients with HTN-U and with the asymptomatic uncontrolled HTN were not merged in our paper despite similar BP values, just as it was done in papers [3] and guidelines [4]. HTN-U definition we use is: “A clinical condition characterized by very high BP (≥180/120 mmHg) with acute symptoms but without an acute hypertension-mediated organ damage (HMOD)“. This definition is not changed essentially for a long time [5] except for diastolic BP – some guidelines prefer 120 mmHg (instead of 110 mmHg) as a cutoff value [6,7]. We suggested recently a small improvement to the definition of HTN-U: ‘not a life-threatening situation’ [8] because epistaxis is an obvious acute HMOD with high BP, but despite it, epistaxis is correctly classified as HTN-U, as it is (almost always) not a life-threatening situation [8]. The definition we used does not include asymptomatic patients with elevated BP, just as in the Dutch guideline for the management of hypertensive crisis – 2010 revision [2]. Additionally, we are not aware of a study that mixed the two types of patients (although the prognosis is the same or similar). Hypertensive emergencies (HTN-Es) are less frequent but more critical than HTN-Us [1,9]. The main difference is the presence of acute HMOD (formerly known as target organ damage – TOD) in HTN-Es [1,9,10].

Theoretically, it is easy to distinguish between HTN-U and HTN-E. In practice, it is often quite the opposite. One of the difficulties is the atypical or oligosymptomatic presentation of some HTN-Es, without enough indicative symptoms and signs to suggest a definitive diagnosis. Another difficulty is the individual perception of symptoms; some patients with less dangerous HTN-U have many more complaints than those with life-threatening HTN-Es. Moreover, some symptoms are present both in HTN-Us and HTN-Es, impeding a proper diagnosis on the clinical ground only. For example, headache is a prevalent symptom in HTN-U and even more so in hypertensive encephalopathy (which is an HTN-E) [1,2]. A reasonable approach for symptom-driven diagnostic strategy in HTN-E is recommended by the ESC council on hypertension [2].

In addition, within several minutes or hours, rare patients with HTN-U may suffer from an acute HMOD (and therefore undergo a ‘transformation’ from HTN-U into HTN-E) while waiting for the results of the examination or while waiting for the effect of an peroral antihypertensive agent. This antihypertensive drug is usually properly administered (according to recommendations in the guidelines) to the patient with HTN-U at the presentation, but it may not act fast enough to decrease BP. Aiming to prevent harm from very high BP in patients suspected of having an acute HMOD, it is logical to treat them as having HTN-Es until this is ruled out. Nevertheless, it is often very difficult to distinguish between HTN-U and HTN-E solely on the basis of symptoms because of the large overlap [11,12]. Therefore, the misclassification of HTN-U as being HTN-E occurred probably several million times in the past decades globally. Importantly, in numerous guidelines intravenous antihypertensive treatment is believed to be not needed for HTN-U [13] and that it may produce an excessive BP reduction in HTN-U [5,14]. To illustrate this, in a real-world scenario, a patient with severe headache and vomiting with BP of 221/121 mmHg can be suspected of having a stroke and can receive e.g. urapidil i.v. It is also recommended to take a blood sample for international normalized ratio (INR) and to send the patient to brain computerized tomography (CT) (according to ESC council on hypertension – hierarchical strategy to diagnose patients with HTN-E based on the presence of emergency symptoms) [2]. Eventually, there may be neither stroke nor hypertensive encephalopathy on CT, and since no HTN-E. As the final (post-CT) diagnosis is HTN-U, previously administered i.v. urapidil might be regarded as unappropriate treatment, particularly in the case of excessive BP fall with the potential to trigger ischemia in various vascular beds, including coronary and cerebral.

2. The problem of therapeutic uncertainty

We used PubMed, SCOPUS, and Google Scholar to search for articles with the recommendations on how to treat patients with very high BP if they have to wait for the results of the examinations to document the presence or the absence of an acute hypertensive HMOD. We noticed and formulated a problem of therapeutic uncertainty and a possible mistake in the choice of antihypertensives (peroral versus parenteral) during the differentiation between HTN-Us and HTN-Es [15]. This problem relates to the absence of recommendations on how to administer an antihypertensive (IV or by mouth) to the patient with very high BP in whom acute HMOD is suspected but yet unproven – while waiting for the results. It is a prevalent situation in the ED: the distinction between HTN-U and HTN-E is often not discernible without additional tests, such as laboratory and urinalysis, echocardiogram, CT scan, etc. There are no available evidence-based recommendations in guidelines to help physicians treat patients adequately during this period [15].

2.1. Two possible mistakes

There is a chance of correct identification of HTN-U and HTN-E before testing. On the other hand, two mistakes are also possible: to treat a patient as having HTN-U until tests demonstrate HTN-E and vice versa – to administer a parenteral antihypertensive until tests reveal no acute HMOD. Both mistakes carry the potential to worsen the prognosis, sometimes even with a fatal outcome.

Such one scenario is easy to understand in the example of painless aortic dissection (AoD). From 6.4% to 29% of AoD patients do not report chest pain [16]. Diagnosing patients with painless AoD lasts longer [16]. The exceedingly severe pain in typical AoD both indicates a diagnosis and forces physicians to act as quickly as possible. In the absence of such pain, particularly if an echocardiogram is suggestive, but not conclusive, there are two possibilities:

1. if a patient has a positive acute aortic dissection detection score (AADDS) a CT scan is needed;

2. with negative AADDS, D dimer ought to be measured, particularly in a patient with long-standing resistant or refractory HTN [17].

In both scenarios, time is needed for a CT scan, D dimer, or both. Nevertheless, an instant decision on an antihypertensive treatment is mandatory, if BP is very high. Without the proof of AoD, an intravenous antihypertensive drug is not indicated and a peroral drug ought to be administered [2]. After a period of approximately 30 to 60 minutes, the result of the CT scan will demonstrate AoD and then it becomes apparent that a parenteral β blocker is needed, often with a vasodilator [18]. The period spent waiting for the CT scan is lost for proper treatment, because peroral drugs cannot decrease BP adequately [18]. We know that AoD is a highly urgent situation – BP should be decreased immediately [2,10]. Therefore, 30–60 minutes of missed therapy is a life-threatening period.

The scenario in other HTN-Es is similar because the clinical picture is often not convincing at the presentation and other diagnostic procedures are needed to exclude or confirm an acute HMOD. Hence, we need recommendations on how to treat patients with very high BP but without obvious acute HMOD until conclusive results are available. For a valid answer, either randomized clinical trials (RCTs) and their meta-analyses or large real-life registries are required. Until then, an experts’ consensus is needed and we ask for it.

2.2. Possible solution(s)

It is known that peroral antihypertensives act too slowly and too weakly to decrease very high BP in HTN-Es. On the other hand, IV antihypertensive drugs may lower BP excessively in HTN-U. Therefore, the following approach might be adequate for patients while we are waiting for conclusive acute HMOD results: to decrease BP promptly, but not profoundly. This can be achieved by a slower rate of potent IV drug administration.

1. If the clinical picture indicates a certain HTN-E, but not conclusively, we may proceed to slowly administer an IV antihypertensive to avoid an excessive decrease of BP until we wait for the test to be performed and interpreted. For example, if a patient with very high BP and chest pain with ECG suggestive of non-ST-elevation acute myocardial infarction has to wait for laboratory confirmation of myocardial damage (troponin elevation), it may be wise to administer β blocker IV and/or nitroglycerin in the infusion. The rate of administration should then be (somewhat) slower as opposed to when the acute myocardial infarction is evident.

2. The other reasonable approach can be to give an antihypertensive agent [9,10] which is adequate for many HTN-Es (at a lower than usual rate). Numerous papers suggest nicardipine is a valid antihypertensive drug for patients with very high BP [19–21]. This is confirmed also in a systematic review [22,23]. Nicardipine is indicated in patients with very high BP with acute renal failure, HTN encephalopathy, acute cardiogenic pulmonary edema, acute AoD, eclampsia and severe preeclampsia/hemolysis, elevated liver enzyme levels, and low platelet levels (HELLP) syndrome [10], and stroke [2]. Urapidil is also useful in various indications in patients with very high BP [24,25]. Consequently, urapidil is recommended in various HTN-Es: very high BP with acute renal failure, ACS, acute pulmonary edema [10].

It is logical to question the adequacy of any particular drug in several HTN-Es, because of wide variations among patients as far as numerous characteristics are concerned: essential vs. renovascular/renoparenchimal/endocrine HTN, with or without obesity or salt excess, with or without psychical and physical stress, alcohol excess and recreational drugs that raise BP, comorbidities, prior antihypertensive treatment, etc [26]. Therefore, we should follow the guidelines in selecting the antihypertensive drug for IV administration once an acute HMOD is confirmed and the diagnosis of a particular HTN-E established. This paper focuses on the period of time until guidelines-recommended work-up is done to obtain the diagnosis of HTN-E and its type. We believe there is a case to carefully administer beta-blocker (BB), e.g. metoprolol or bisoprlol i.v. to a patient with very high BP and tachycardia (but without ECG signs of myocardial ischemia and without contraindications to BBs) who has strong chest and back pain while waiting for chest CT for the suspected acute aortic syndrome. Despite there is no evidence of acute aortic syndrome (and therefore no evidence of HTN-E) yet, the reason for such a treatment choice seems obvious – high probability of acute aortic syndrome. On the other hand, if the patient has several symptoms (e.g. chest pain, headache, and dyspnea), without a single clinically obvious diagnosis and we have a set of diagnostic tests to perform – it sounds reasonable to administer nicardipine i.v. because it has a guideline-recommended indication for numerous HTN-Es [2]. For that reason, despite the fact that a single drug cannot improve and control all mechanisms leading to high BP, it is important that several antihypertensive drugs (such as nicardipine and urapidil) are efficient in many types of HTN-E. Therefore, until there is an experts’ consensus (and even better – RCTs) about the optimal treatment, it may be wise to administer a parenteral drug, but at a slower rate than usual, to provide a fast, but not marked BP decrease.

3. Expert opinion

HTN-U is often a diagnosis of exclusion and it requires time for tests to exclude acute HMOD. Surprisingly, we do not have reliable data about how often such additional tests are needed and how long it usually takes to obtain them. Furthermore, it is needed to investigate how often a peroral versus IV antihypertensive drug is administered, how the BP responded to it, whether it was safe enough, and whether the mode of the initial antihypertensive treatment was adequate.

The best course would be an RCT to test the usual approach versus the slow-IV administration of nicardipine (or an evidence-based antihypertensive drug for suspected HTN-E) for patients with very high BP without clear clinical evidence of an acute HMOD at the presentation. This may help us to analyze how often the improper antihypertensive treatment is applied while waiting for the test results to diagnose HTN-E or HTN-U. It would be also helpful for the clinical practice to see which type of the initial antihypertensive treatment (IV or peroral) leads to better outcomes during the waiting period and while the patient is under observation or until hospital discharge. Moreover, slow IV infusion of small doses of e.g. nicardipine is not evaluated in RCT versus the usual care during the waiting period in the ED (until the evidence of HMOD presence is obtained). This study can directly influence the initial treatment choice in the ED.

There is a need for a large prospective registry to get a picture of the usual treatment pathways during this waiting period. Secondly, the advantages and disadvantages of reduced-dose, slow IV infusion of an antihypertensive with numerous indications (such as nicardipine) should be compared to the usual care (which commonly includes peroral antihypertensive drug) until the differential diagnosis is solved (HTN-U vs HTN-E).

Declaration of interest

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This work has been supported by the Serbian Ministry of Education and Science, Belgrade, Serbia through grants No.175092 and No. III41018.