The perils of a “hypertensive” diagnosis

Abstract

Background

The current framework for approaching patients with severe hypertension assigns every patient with arterial pressure above an arbitrary threshold to a state of either “hypertensive emergency” or “hypertensive urgency.” By labeling every condition in these patients a “hypertensive” condition, this framework implies that elevated blood pressure is responsible for every vascular event. An extension of this idea holds that hypertension is even capable of causing symptoms directly, without any intervening pathology.

Case Presentation

A man in his early 60s presented to the Emergency Department with new hypertension, headache, vomiting and diplopia. His doctors confidently diagnosed his presentation as “symptomatic hypertensive urgency” — a direct hypertensive effect. However, when he reported back pain and leg weakness two days later, investigation resumed. Spinal MRI showed subacute hemorrhage extending along the spinal cord from T9-T12. Catheter angiography revealed a small fusiform aneurysm of the posterior spinal artery at T11. The aneurysm was surgically clipped. The patient made a complete recovery and his hypertension resolved.

Conclusion

Subarachnoid hemorrhage from a spinal aneurysm is a rare entity. Typical presenting symptoms of back pain and leg weakness help localize the lesion, but when intracranial symptoms such as headache or confusion predominate the diagnosis can be challenging. In this case, misconceptions about hypertension led clinicians to anchor on a diagnosis of “symptomatic hypertensive urgency.” When clinicians attribute signs and symptoms directly to elevated blood pressure they may fail to search for real disease processes.

Background

Patients with acute neurological symptoms often present to the Emergency Department with markedly elevated blood pressures. Several generations of clinicians have learned a framework for evaluating these patients in which every patient with blood pressure above an arbitrary threshold is in a state of either “hypertensive emergency” or “hypertensive urgency” [1, 2]. The framework begins with the implicit assumption that the current level of blood pressure is always the most important trigger and driver of every vascular event in these patients. This assumption can encourage diagnostic anchoring and over-attribution. The following case shows how easily clinicians can attribute signs and symptoms to hypertension itself and thereby fail to investigate real pathological causes.

Case presentation

A man in his early 60s presented to the Emergency Department with headache and new high blood pressure. His medical history included recurrent sinusitis but no hypertension or vascular disease. Three weeks before presentation he began to feel unwell with a mild sore throat followed by nasal congestion, cough, myalgias, generalized weakness and persistent headache. The headache was bilateral, throbbing, frontal, and retro-orbital. It was worse with bending forward, worse in the evening and sometimes woke him from sleep. A course of levofloxacin for sinusitis produced some improvement, but two days before presentation his headache worsened, and he vomited. On the day of presentation, he developed diplopia for distant objects.

In the ED his blood pressure was 222/115 mm Hg, heart rate 102 beats per minute, and temperature 36 °C. His general and neurological exam were considered normal. Intravenous metoclopramide brought only mild improvement in his headache. Non-contrast head computed tomography (CT) was unremarkable. Intravenous boluses of labetalol and hydralazine reduced his BP to 171/88 mm Hg.

The Emergency Medicine team, the Neurology consultant, and the admitting Internal Medicine team agreed that the headache and diplopia were caused by “symptomatic hypertensive urgency.” The consulting ophthalmologist speculated that the diplopia represented a chronic esophoria (inward eye deviation) which had “decompensated in the setting of elevated blood pressure” and which he would “monitor for improvement as BP is controlled.” An attending neuro-ophthalmologist found a right abducens palsy, which he thought was most likely ischemic and consistent with a condition of “hypertensive emergency.” He considered elevated intracranial pressure unlikely because of the absence of papilledema. The patient was admitted to the Medicine service for blood pressure control.

The next morning the patient’s blood pressure had fallen to 170/80 mmHg with oral metoprolol and lisinopril. His headache was somewhat improved, but he still felt “miserable.” Brain magnetic resonance imaging (MRI) showed T2 FLAIR hyperintensity in both precentral sulci, thought to be artifactual in the absence of restricted diffusion. Tests for secondary causes of hypertension such as plasma catecholamines and renal ultrasound were unremarkable.

That evening he complained that his legs “felt rubbery and gave way,” and reported that he had fallen twice while trying to walk to the bathroom. He mentioned that for several days he had low back pain, which he attributed to back strain from vomiting. His internist found bilateral quadriceps weakness and became concerned about a pathologic process in the lumbar spine.

On the third morning his legs were weaker, and he had trouble rising from a sitting position. Leg strength seemed normal, but the patient fell after taking several steps. Lower extremity reflexes were increased. Neurology found bilateral abducens palsies. That evening MRI of the spine showed intradural T1 and T2 hyperintensity along the cord from T9-T12, consistent with subacute hemorrhage (Fig. 1). He was lethargic and mildly confused and was transferred to the Neurological Intensive Care Unit.

Fig. 1.

Sagittal proton-density MRI image of the thoracic spine showing subacute hemorrhage (arrows)

The next day lumbar puncture revealed an opening pressure of 33 cm of water and xanthochromic cerebrospinal fluid with 3300 RBC/mm³ and 22 WBC/mm³. Brain magnetic resonance angiography showed mild multifocal narrowing. On day 6 catheter angiography of the brain and spine was performed. This revealed a small fusiform aneurysm of the ascending portion of the posterior spinal artery just above its origin at the T11 level (Fig. 2). There was no evidence of associated arteriovenous malformation (AVM). There was diffuse mild vasospasm of the distal middle and anterior cerebral arteries. By this time the patient felt better and was cognitively improved. Two days later the aneurysm was clipped surgically. By day 13 the patient was dramatically better, sitting up in bed solving a crossword puzzle, his headache gone. He was discharged on day 14. One year later he was asymptomatic and normotensive without medication.

Fig. 2.

Spinal angiogram showing fusiform posterior spinal artery aneurysm (arrow)

Discussion

Fewer than 1% of subarachnoid hemorrhages originate from a spinal source, most often from an AVM or neoplasm. Spinal artery aneurysms are uncommon and are usually asssociated with other vascular lesions. Isolated spinal artery aneurysms are quite rare [3, 4]. A recent review found only 21 published cases of isolated aneurysm of the posterior spinal artery [5].

Spinal subarachnoid hemorrhage most often presents with neck or back pain referable to the site of bleeding, frequently with weakness below the level of the lesion. This combination of findings can act as a “red flag” for clinicians and make localization easier. However, the presenting symptoms can be headache or confusion, either from an acute increase in ICP or the spread of blood products from the spinal to the intracranial compartment [6]. When intracranial symptoms predominate, as in this case, diagnosis can be more difficult.

The diagnostic challenge of this case was to find a process that would explain both new hypertension and the symptoms of headache, vomiting, and diplopia. Intracranial disorders such as stroke, subarachnoid hemorrhage, or meningitis should have been included in the differential diagnosis. In retrospect it is clear that subarachnoid hemorrhage caused not only the presenting symptoms, the sixth-nerve palsy and the brain MRI findings, but also the initial hypertension. Blood pressure elevation is common after subarachnoid hemorrhage even in previously normotensive individuals and is usually attributed to a Cushing response to elevated ICP [7, 8].

Several potential clues were set aside. Isolated sixth-nerve palsy is a well-known “false-localizing” sign which can be caused by alterations of ICP or pathologies in the subarachnoid space such as hemorrhage or infection [9]. The consulting neuro-ophthalmologist dismissed this possibility because of the absence of papilledema. However, papilledema can take a week or more to appear after an acute increase in ICP [10]. Subtle MRI-FLAIR hyperintensity can be a sensitive indicator of acute and subacute subarachnoid hemorrhage even when CT is normal [11]. Restricted diffusion in the same region is not necessary for confirmation but suggests ischemia of adjacent brain parenchyma. In this case, FLAIR hyperintensity in the bilateral precentral sulci was dismissed as artifactual because of the absence of abnormal diffusion and absence of hyperdense blood on the CT performed the previous day. The diffuse cerebral vasospasm found later on angiography suggests that the FLAIR hyperintensity did indeed represent real blood products.

Given the rarity of spinal hemorrhage, the gradual onset of the headache, and the late appearance of localizing back pain, it is not surprising that several days were required for a definitive diagnosis. However, it was misconceptions about the nature of hypertension that led specialists in emergency medicine, neurology, internal medicine, and ophthalmology alike to anchor on a diagnosis of “symptomatic hypertensive urgency” rather than pursue further investigation.

Four decades ago, the United States Joint National Committee on Prevention, Detection Evaluation, and Treatment of High Blood Pressure (JNC) first proposed a classification system for patients with marked hypertension [1]. In their fully developed framework, every patient with a blood pressure over 180/110 is in a state of “hypertensive crisis.” These crises are assigned to one of two categories. An “emergency” is any acute neurovascular or cardiovascular event, which they asserted requires immediate blood pressure reduction. An “urgency” is any other situation in which pressure can be reduced more slowly [2]. Unfortunately, this formulation has led to a variety of misconceptions.

First, by applying the prefix “hypertensive” to every possible clinical state in these patients, the JNC framework implies that above an arbitrary threshold the current level of blood pressure is the primary trigger and driver of every acute vascular event, and that every such event should therefore be treated by rapid blood pressure reduction. While there are several well-defined conditions such as aortic dissection and eclampsia in which blood pressure reduction appears to be beneficial, this is not uniformly true for all vascular emergencies. In ischemic stroke, for example, blood pressure is not unusually high before onset and acute blood pressure reduction is not beneficial once stroke occurs [12, 13].

A second misconception holds that all vascular events above an arbitrary blood pressure threshold share the same pathology – that behind the various conditions classified as “emergencies” there is a single pathological condition of “hypertensive emergency,” often taken to be a new name for “malignant” hypertension [14, 15]. This illness, in which arteriolar necrosis develops most prominently in the retina and kidney, appears only in a very small subset of patients with severe hypertension. There is no evidence that this is the mechanism of vascular events in the vast majority of hypertensive patients.

The assumption that hypertension must be responsible for every acute condition leads to a third misconception: that in the absence of any obvious pathology symptoms must be caused directly by elevated blood pressure itself. This is the syndrome of “symptomatic hypertensive urgency” [16, 17]. Despite the absence of evidence for such a condition [18, 19], many clinicians believe that hypertension can cause chest pain without myocardial ischemia, blurred vision without retinopathy, and neurological symptoms without infarction, hemorrhage, or edema. As this case demonstrates, misconceptions about hypertension can lead even experienced clinicians to view symptoms as direct blood pressure effects and thereby miss important diagnoses.

Abbreviations

mm Hg

Millimeters of mercury

CT

Computed tomography

BP

Blood pressure

MRI

Magnetic resonance imaging

FLAIR

Fluid-attenuated inversion recovery

RBC

Red blood cells

WBC

White blood cells

AVM

Arterio-venous malformation

JNC

Joint National Committee

Author contributions

R. L wrote and approved the manuscript.

Funding

NA.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval

NA.

Consent for publication

Written consent was obtained and is available on request.

Competing interests

The authors declare no competing interests.