Transient Neurological Symptoms in Patients with Intracerebral Hemorrhage

Research

Original Investigation Transient Neurological Symptoms in Patients With Intracerebral Hemorrhage

Sandeep Kumar, MD; Magdy Selim, MD, PhD; Sarah Marchina, PhD; Louis R. Caplan, MD

Supplemental content at IMPORTANCE Transient symptoms from an intracerebral hemorrhage (ICH) are not well jamaneurology.com recognized and have previously not been reported. CME Quiz at jamanetworkcme.com and OBJECTIVE To describe the clinical presentations and neuroimaging findings in a subset of CME Questions page 364 patients with ICH presenting with transient neurological signs and symptoms.

DESIGN, SETTING, AND PARTICIPANTS Clinical case series in which the hospital database of a large academic center in Boston, Massachusetts, was searched using International Classification of Diseases, Ninth Revision code 431 from June 1, 2000, to August 31, 2014, to identify patients with ICH who had transient deficits that resolved completely within 24 hours of symptom onset.

MAIN OUTCOMES AND MEASURES Clinical presentation, neuroimaging findings, and prognosis.

RESULTS Among 3207 consecutive patients with ICH initially screened, 17 fulfilled study criteria (median age, 65 years [interquartile range, 56-73 years]; 11 male). In most patients, recovery from neurological symptoms started within a few hours after onset. Most Author Affiliations: Stroke Division, hemorrhages in this cohort were small, with a mean (SD) hematoma volume of 17 (9.9) mL, Department of Neurology, Beth Israel and were subcortical in location. One patient died of hemorrhage recurrence. Deaconess Medical Center, Boston, Massachusetts (Kumar, Selim, Marchina, Caplan); Harvard Medical CONCLUSIONS AND RELEVANCE Patients with ICH can present with rapidly resolving deficits School, Boston, Massachusetts resembling transient ischemic attacks. Recognition of these instances is important to avoid (Kumar, Selim, Caplan). delays in investigations and to manage these cases appropriately. Corresponding Author: Sandeep Kumar, MD, Stroke Division, Department of Neurology, Beth Israel JAMA Neurol. 2016;73(3):316-320. doi:10.1001/jamaneurol.2015.4202 Deaconess Medical Center, 185 Published online January 4, 2016. Pilgrim Rd, Palmer 127, Boston, MA 02215 ([email protected]).

ntracerebral hemorrhage (ICH) is a particularly ominous senting symptoms and signs as well as imaging patterns, and stroke subtype that carries high rates of mortality and assess prognosis. I morbidity.1 Prompt recognition and appropriate management are necessary to affect outcomes in these patients.2 Usually, ICH presents with an abrupt onset of neurological Methods deficits that often progress over time; nausea, vomiting, impairment of consciousness, and significant hypertension In this clinical case series, we identified all patients with at presentation are common features.2 However, in a signifi- ICH using International Classification of Diseases, Ninth cant number of patients, the deficits are less pronounced Revision code 431 from our hospital database from June 1, and brain imaging is needed to differentiate ICH from an 2000, to August 31, 2014. Discharge summaries of all rec- ischemic stroke. Whereas temporary symptoms have been ords were individually reviewed to exclude patients with observed in patients with subdural hematomas and sub- traumatic ICH, subarachnoid and subdural hemorrhages, arachnoid hemorrhages due to amyloid angiopathy,3,4 tran- secondary ICH from an underlying neoplasm, and ICH from sient deficits resembling transient ischemic attacks (TIAs) hemorrhagic transformation of an ischemic infarct. We col- have not been reported, to our knowledge, in patients with lected and analyzed demographic information as well as ICH. clinical and imaging findings on all remaining patients with We encountered a patient in our practice who had devel- a spontaneous ICH. Patients who had transient deficits oped a transient language impairment from a left temporal (symptoms, signs, or both) from an ICH that had resolved hemorrhage mimicking a TIA. This prompted us to systemati- on a repeated examination within 24 hours were eligible. cally search the medical records at our institution to assess the The study was approved by the institutional review board at frequency of TIA-like presentations of ICH, analyze their pre- the Beth Israel Deaconess Medical Center. A waiver of

316 JAMA Neurology March 2016 Volume 73, Number 3 (Reprinted) jamaneurology.com

Downloaded From: https://jamanetwork.com/ on 09/30/2021 Transient Neurological Symptoms From Intracerebral Hemorrhage Original Investigation Research

informed consent was granted by the institutional review board owing to the low-risk, retrospective nature of this Key Points study. Question: Do transient signs and symptoms occur with intracerebral hemorrhages? Findings: In this case series involving 17 carefully selected Results hospitalized patients, rapidly resolving signs and symptoms resembling transient ischemic attacks were documented. All We identified 3207 patients with ICH using International patients had normal results on neurological examination within Classification of Diseases, Ninth Revision code 431. Among 24 hours after symptom onset. these, 2137 patients had a spontaneous ICH without any Meaning: Patients with rapidly resolving stroke-like symptoms coexisting subarachnoid or subdural hemorrhage, neoplasm, should undergo prompt brain imaging, even in the context of or secondary hemorrhagic transformation from a cerebral normal findings on neurological examination, to rule out intracerebral hemorrhage, especially prior to initiating any infarct. Of these, 34 patients had transient deficits, which antithrombotic or anticoagulant therapy. were defined as clinical symptoms and signs that had resolved within 24 hours. These records were further reviewed to ascertain details of history and imaging findings confounded the clinical presentation (5 cases). A cohort of and to discard records in which significant clinical informa- 17 patients was assembled (eAppendix in the Supplement). tion and details about the duration of symptoms were lack- The baseline characteristics of this cohort are summarized in ing (12 cases) or in which additional conditions may have the Table.

Table. Baseline Characteristics of the Patient Cohort

Hemorrhage Sex/Age, y Symptom Initial Examination Findings Duration Volume, mL Site Etiology M/mid-20s Right-sided paresthesias BP 120/90 mm Hg; sensory <24 h 11.7 Left subinsular Cavernoma loss; NIHSS score 2 region M/mid-70s Left foot numbness, arm BP 198/92 mm Hg; left 4 h 14.9 Right internal Hypertensive clumsiness, slurred speech pronator drift, extinction; capsule NIHSS score 2 F/mid-60s Headache, left tongue and BP 200/100 mm Hg; normal 30 min 7 Right basal Hypertensive lip numbness NIHSS score 0 ganglia M/early 60s Severe word-finding BP 220/98 mm Hg; normal 30 min 29 Left temporal Probable amyloid difficulties NIHSS score 0 angiopathy F/late 50s Nausea, vomiting, left-sided BP 120/80 mm Hg; mild left 30 min 32.6 Right internal Possibly hypertensive weakness, dysarthria facial droop, left pronator capsule drift; NIHSS score 2 M/early 70s Speech arrest BP 201/89 mm Hg; mild 15 min 14.8 Left temporal Hypertensive or anomia, right facial droop; amyloid angiopathy NIHSS score 3 M/early 70s Dizziness, slurred speech, BP 176/100 mm Hg; mild 30 min 10.7 Left basal ganglia Hypertensive gait unsteadiness dysarthria, left pronator drift; NIHSS score 2 F/early 80s Dizziness, near syncope BP 140/97 mm Hg; mild 5 min 8.5 Left internal Possible amyloid lethargy; NIHSS score 1 capsule angiopathy M/early 50s Dizziness, gait unsteadiness, BP 100/58 mm Hg; mild <1 h 9.8 Left basal ganglia Unclear mild confusion inattention, pronator drift; NIHSS score 1 F/mid-80s Right leg numbness BP 210/59 mm Hg; left leg 30 min 7.7 Left lateral Possibly hypertensive sensory loss; NIHSS score 2 thalamic M/mid-60s Right hemiparesis, BP 178/95 mm Hg; dysarthria, 2-3 h 24 Left basal ganglia Anticoagulation, dysarthria, gait unsteadiness right hemiparesis, dysmetria; possibly hypertensive NIHSS score 4 F/early 80s Left leg weakness BP 200/90 mm Hg; left leg 30 min 18 Right internal Anticoagulation, drift; NIHSS score 1 capsule possibly hypertensive M/late 60s Left arm clumsiness, facial BP 182/70 mm Hg; mild left 5 h 21 Right thalamus Hypertensive droop, slurred speech ataxia, dysarthria; NIHSS score 3 M/early 60s Dysarthria, hand clumsiness BP 141/93 mm Hg; mild Possibly 6-8 h 7 Left lentiform Hypertensive dysarthria, limb ataxia; NIHSS nucleus score 2 F/mid-60s Dysarthria, headache BP 197/93 mm Hg; left <12 h 9.4 Right basal Hypertensive hemiparesis; NIHSS score 5 ganglia M/late-40s Left hemiparesis BP 132/73 mm Hg; left <6 h 23 Right putamen Moyamoya disease hemiparesis; NIHSS score 4 M/mid-40s Slurred speech, right leg BP 220/120 mm Hg; mild right <8 h 40.4 Left putamen Hypertensive weakness, gait difficulty hemiparesis, aphasia, ataxia; NIHSS score 5

Abbreviations: BP, blood pressure; NIHSS, National Institutes of Health Stroke Scale.

jamaneurology.com (Reprinted) JAMA Neurology March 2016 Volume 73, Number 3 317

Downloaded From: https://jamanetwork.com/ on 09/30/2021 Research Original Investigation Transient Neurological Symptoms From Intracerebral Hemorrhage

Figure. Computed Tomographic Scans Demonstrating Characteristic Brain Hemorrhages in the Patient Cohort

A Patient with left-sided B Patient with left-sided C Patient with severe anomia hemiparesis hemiparesis

D Patient with right-sided E Patient with speech arrest F Patient with right-sided hemiparesis paresthesias and aphasia

A, Patient with left-sided hemiparesis that resolved within 6 hours. B, Patient with left-sided hemiparesis that resolved within 12 hours. C, Patient with severe anomia that resolvedin30minutes.D,Patient with recurrent right-sided paresthesias lasting less than 24 hours. E, Patient with speech arrest that largely resolved within 15 minutes. F, Patient with right-sided hemiparesis and aphasia that resolved within 8 hours.

The median age of the group was 65 years (interquartile own accord or per the advice of their physicians prior to range, 56-73 years); 11 of the 17 patients were men. All pa- obtaining any imaging studies. tients underwent a computed tomographic (CT) scan of the brain within 6 hours of symptom onset and had a docu- Imaging Findings mented detailed neurological examination at presentation as The prototypical imaging findings on CT scans at admission well as a subsequent progress note detailing the clinical course are shown in the Figure. Most of the hemorrhages were sub- during the preceding 24 hours including a follow-up neuro- cortical involving the basal ganglia or neighboring white mat- logical examination within 24 hours. ter tracts. Two involved the temporal lobes. No patients had intraventricular hemorrhage or hydrocephalus. All hemor- Clinical Presentation rhages were small, with a mean (SD) volume of 17 (9.9) mL as The trajectory of clinical improvement began early in all measured by the ABC/2 method.5 In addition to the initial patients, usually within several minutes. In 9 patients, the head CT scans, 11 patients underwent brain magnetic reso- deficits lasted less than 30 minutes, in 5 patients they lasted nance imaging and magnetic resonance angiography, 5 6 hours or less, and only 1 patient had symptoms lingering underwent CT angiography of the head, and 2 underwent more than 12 hours but less than 24 hours. The most com- conventional angiography. In 16 patients who underwent mon deficits were sensorimotor involving varying degrees of vascular imaging, only 1 had an abnormality consistent with a limb weakness, numbness, or incoordination. Three patients unilateral moyamoya-type pattern. Of the 11 patients who developed significant dysarthria and 2 others had major lan- underwent brain magnetic resonance imaging, 7 showed guage impairment at onset; 3 patients had dizziness with varying degrees of leukoaraiosis on fluid-attenuated inver- gait unsteadiness. The typical symptoms of ICH including sion recovery sequences; 2 showed scattered microbleeds on headache, nausea, vomiting, and reduction in levels of alert- gradient echo sequences; 1 displayed characteristic features ness were usually absent. Most patients had moderate to of a cavernoma; and 1 showed no other significant abnormal- severe hypertension at presentation, although their neuro- ity apart from the acute ICH. logical deficits were mild (mean [SD] National Institutes of Health Stroke Scale score, 2.3 [1.5]; range, 0-5) on initial neu- Etiology rological examination in the emergency department (Table). Etiology of the hemorrhage was determined based on dis- Five patients took antiplatelet medications either on their charge diagnosis and review of all discharge summaries and

318 JAMA Neurology March 2016 Volume 73, Number 3 (Reprinted) jamaneurology.com

Downloaded From: https://jamanetwork.com/ on 09/30/2021 Transient Neurological Symptoms From Intracerebral Hemorrhage Original Investigation Research

neuroimaging findings. In 8 patients the bleed was likely hy- referred to the emergency department.9,10 Previous reports pertensive; 2 patients had probable amyloid angiopathy; 1 pa- have shown that even in patients presenting to the emer- tient had moyamoya disease; 1 had bleeding from a cavern- gency department with symptoms of a TIA, CT scans are ous angioma; 2 cases were likely anticoagulant related; and a underused and only 50% to 70% of such patients undergo a definite etiology for hemorrhage could not be determined in scan.11 Delays in obtaining head CTs can substantially reduce 3 cases. the sensitivity of detecting a hemorrhage especially beyond a week.12 Third, in resource-limited settings, clinical scores Follow-up to diagnose ICH have been derived and validated.13,14 It is One patient died during hospitalization of recurrence of a mas- worth noting that only a small minority of our patients could sive ICH 4 days after admission. This patient had a pace- reliably be suspected of having an ICH based on their pre- maker and did not undergo brain magnetic resonance imaging senting features such as diminished consciousness, nausea, or vascular imaging. The etiology of her hemorrhage was un- vomiting, headache, and severe hypertension. Our findings clear. Detailed follow-up information was available in 9 pa- urge caution regarding the reliability of these clinical tools in tients. The follow-up varied between 1 and 8 months. In all identifying minor hemorrhages. these patients there was no recurrence of symptoms sugges- The biological underpinning driving such an early recov- tive of a stroke or TIA. ery after ICH is intriguing. The rapid resolution of symptoms may suggest a hemodynamic cause with spontaneous reduction in local mass effect from redistribution of blood along the Discussion tracts, especially in patients with subcortical bleeds. In 2 of our patients with lobar hemorrhages, focal seizures with postic- Whereas clinical deterioration after ICH has been shown in sev- tal deficits may have played a role. Neuroplastic changes with eral studies,6,7 to our knowledge no studies have reported rapid cortical reorganization are also a possibility, although the time improvement in clinical deficits in patients with ICH. We ad- course is less typical. opted a 24-hour threshold for labeling a deficit as transient as The major limitation of our work lies in its retrospective na- this time cutoff has traditionally been used for defining TIAs.8 ture, where documentation of a detailed neurological evaluation In most patients the improvement started early, usually within at predefined regular intervals was unavailable. Thus, we relied several minutes, and the symptoms and signs had resolved well on the admission and progress notes as well as discharge sum- before our adopted time threshold. This is likely an unusual maries to record timing of recovery. We meticulously excluded presentation of ICH, although rapid improvement of symp- cases in which this information was unavailable. The etiology of toms with relatively minor deficits may have contributed to a the hemorrhage was based on clinician impression, although we referral bias and led to an underascertainment of cases. Only independently reviewed all records and imaging studies for cor- a prospective population-based study with detailed early neu- roboration. Routine follow-up information was not available in rological assessment and close monitoring can estimate the true some of our patients and brain imaging was not repeated in many incidence. patients. Most patients with available information fared well, Our report carries significant clinical relevance. First, it although there was 1 death from recurrent ICH. shows that transient deficits from a minor ICH can mimic a TIA or minor infarct and thus require brain imaging for dif- ferentiation. Some of our patients received antiplatelet Conclusions medications prior to any imaging. While we did not observe any harm in this small group of patients, prescription of This study highlights a previously unrecognized presenta- these agents can be hazardous in this situation. Second, this tion of ICH and emphasizes its clinical relevance for patient study highlights a need to obtain prompt brain imaging in all care. Further studies are needed to better estimate its true in- patients with suspected TIAs. In many settings, patients cidence and understand the pathophysiological basis of this with TIAs are not routinely admitted for admission or presentation.

ARTICLE INFORMATION Administrative, technical, or material support: Council on Clinical Cardiology. Guidelines for the Accepted for Publication: November 4, 2015. Kumar, Marchina. management of spontaneous intracerebral Conflict of Interest Disclosures: None reported. hemorrhage: a guideline for healthcare Published Online: January 4, 2016. professionals from the American Heart doi:10.1001/jamaneurol.2015.4202. REFERENCES Association/American Stroke Association. Stroke. Author Contributions: Dr Kumar had full access to 2015;46(7):2032-2060. 1. van Asch CJ, Luitse MJ, Rinkel GJ, van der Tweel I, all of the data in the study and takes responsibility 3. Khirwadkar H, Thomas C, Fardy M, Hourihan M. for the integrity of the data and the accuracy of the Algra A, Klijn CJ. Incidence, case fatality, and functional outcome of intracerebral haemorrhage Postoperative subdural haematoma and data analysis. pneumocephalus mimicking TIA [published Study concept and design: Kumar, Selim. over time, according to age, sex, and ethnic origin: a systematic review and meta-analysis. Lancet Neurol. online August 30, 2011]. BMJ Case Rep. Acquisition, analysis, or interpretation of data: All doi:10.1136/bcr.03.2011.3965. authors. 2010;9(2):167-176. Drafting of the manuscript: Kumar. 2. Hemphill JC III, Greenberg SM, Anderson CS, 4. Kumar S, Goddeau RP Jr, Selim MH, et al. Critical revision of the manuscript for important et al; American Heart Association Stroke Council; Atraumatic convexal subarachnoid hemorrhage: intellectual content: All authors. Council on Cardiovascular and Stroke Nursing;

jamaneurology.com (Reprinted) JAMA Neurology March 2016 Volume 73, Number 3 319

Downloaded From: https://jamanetwork.com/ on 09/30/2021 Research Original Investigation Transient Neurological Symptoms From Intracerebral Hemorrhage

clinical presentation, imaging patterns, Clinical Cardiology; Council on Peripheral Vascular transient ischemic attack, 1992-2001. Acad Emerg and etiologies. Neurology. 2010;74(11): Disease. Guidelines for the prevention of stroke in Med. 2006;13(6):666-672. 893-899. patients with stroke and transient ischemic attack: 12. Wardlaw JM, Keir SL, Dennis MS. The impact of 5. Kothari RU, Brott T, Broderick JP, et al. The ABCs a guideline for healthcare professionals from the delays in computed tomography of the brain on the of measuring intracerebral hemorrhage volumes. American Heart Association/American Stroke accuracy of diagnosis and subsequent management Stroke. 1996;27(8):1304-1305. Association. Stroke. 2014;45(7):2160-2236. in patients with minor stroke. J Neurol Neurosurg 6. Moon JS, Janjua N, Ahmed S, et al. Prehospital 9. Goldstein LB, Bian J, Samsa GP, Bonito AJ, Psychiatry. 2003;74(1):77-81. neurologic deterioration in patients with Lux LJ, Matchar DB. New transient ischemic attack 13. Allen CM. Clinical diagnosis of the acute stroke intracerebral hemorrhage. Crit Care Med. 2008;36 and stroke: outpatient management by primary syndrome. Q J Med. 1983;52(208):515-523. (1):172-175. care physicians. Arch Intern Med. 2000;160(19): 2941-2946. 14. Poungvarin N, Viriyavejakul A, Komontri C. 7. Fan JS, Huang HH, Chen YC, et al. Emergency Siriraj stroke score and validation study to department neurologic deterioration in patients 10. Roebers S, Wagner M, Ritter MA, Dornbach F, distinguish supratentorial intracerebral with spontaneous intracerebral hemorrhage: Wahle K, Heuschmann PU. Attitudes and current haemorrhage from infarction. BMJ. 1991;302 incidence, predictors, and prognostic significance. practice of primary care physicians in acute stroke (6792):1565-1567. Acad Emerg Med. 2012;19(2):133-138. management. Stroke. 2007;38(4):1298-1303. 8. Kernan WN, Ovbiagele B, Black HR, et al; 11. Edlow JA, Kim S, Pelletier AJ, Camargo CA Jr. American Heart Association Stroke Council; Council National study on emergency department visits for on Cardiovascular and Stroke Nursing; Council on

320 JAMA Neurology March 2016 Volume 73, Number 3 (Reprinted) jamaneurology.com

Downloaded From: https://jamanetwork.com/ on 09/30/2021