ABSTRACT
BackgroundAt our tertiary care public hospital, we saw COVID-19 presenting with thromboembolic phenomena, indicating a possible early thrombo-inflammatory pathology. ObjectivesWe documented patients with cardiac and neurological thromboembolic phenomena as a primary presentation of COVID-19, and compared a subset of COVID associated strokes against COVID-19 patients without thrombotic manifestations. MethodsWe included all COVID-Stroke and COVID-ACS (COVID-19, with ischemic arterial stroke/Acute Coronary Syndrome presenting prior to/simultaneous with/within 72 hours of systemic/respiratory COVID manifestations) admitted from April to November 2020. In the nested case control analysis, we used unpaired T-test and chi-square test to study differences between COVID-Strokes (case group) and non-thrombotic COVID controls. Results and ConclusionsWe noted 68 strokes and 122 ACS associated with COVID-19. ACS peaked in May-June, while stroke admissions peaked later in September-October, possibly because severe strokes may have expired at home during the lockdown. In the case-control analysis, cases (n=43; 12F:31M; mean age 51.5 years) had significantly higher D-Dimer values than controls (n=50; 9F:41M; mean age 51.6 years). Mortality was significantly higher in cases (51.2% vs. 26.0%; p = 0.018). We noted 7.5 times higher mortality in cases versus controls even among patients needing minimal oxygen support. Imaging in 37 patients showed both anterior and posterior circulation territories affected in seven, with almost half of Carotid territory strokes being large hemispherical strokes. Additionally, CT/MRI angiography in 28 strokes showed large vessel occlusions in 19 patients. Death in cases thus probably occurred before progression to intense respiratory support, due to severe central nervous system insult. Binary logistic regression analysis showed respiratory support intensity to be the sole independent predictor of mortality among cases. Respiratory distress could have been due to COVID-19 lung infection or aspiration pneumonia resulting from obtunded sensorium. In controls, mortality was predicted by increasing age, female sex, and respiratory support intensity.
ABSTRACT
Background and PurposeVarious neurological complications have been reported in association with COVID-19. We report our experience of COVID-19 with stroke at a single center over a period of eight months spanning 1 March to 31 October 2020. MethodsWe recruited all patients admitted to Internal Medicine with an acute stroke, who also tested positive for COVID-19 on RTPCR. We included all stroke cases in our analysis for prediction of in-hospital mortality, and separately analyzed arterial infarcts for vascular territory of ischemic strokes. ResultsThere were 62 stroke cases among 3923 COVID-19 admissions (incidence 1.6%). Data was available for 58 patients {mean age 52.6 years; age range 17-91; F/M=20/38; 24% (14/58) aged [≤]40; 51% (30/58) hypertensive; 36% (21/58) diabetic; 41% (24/58) with O2 saturation <95% at admission; 32/58 (55.17 %) in-hospital mortality}. Among 58 strokes, there were 44 arterial infarcts, seven bleeds, three arterial infarcts with associated cerebral venous sinus thrombosis, two combined infarct and bleed, and two of indeterminate type. Among the total 49 infarcts, Carotid territory was the commonest affected (36/49; 73.5%), followed by vertebrobasilar (7/49; 14.3%) and both (6/49; 12.2%). Concordant arterial block was seen in 61% (19 of 31 infarcts with angiography done). Early stroke (within 48 hours of respiratory symptoms) was seen in 82.7% (48/58) patients. Patients with poor saturation at admission were older (58 vs 49 years) and had more comorbidities and higher mortality (79% vs 38%). Mortality was similar in young strokes and older patients, although the latter required more intense respiratory support. Logistic regression analysis showed that low GCS and requirement for increasing intensity of respiratory support predicted in-hospital mortality. ConclusionsWe had a 1.6% incidence of COVID-19 related stroke of which the majority were carotid territory infarcts. In-hospital mortality was 55.17%, predicted by low GCS at admission.
