The investigators will recruit participants from three tertiary care hospitals in Uganda into an observational cohort study of people living with HIV/AIDS (PLWHA) and HIV-uninfected persons matched for gender and residency, presenting with CT-confirmed stroke. We will collect socio demographic, clinical, laboratory, radiologic, cardiac, and clinical neurologic disease measures to investigate the effect of HIV-infection on 1) clinical and radiologic presentation, 2) risk factor profiles; and 3) stroke outcomes (death or disability).

The investigators will enroll 100 HIV-infected adults presenting to the emergency department with acute or sub-acute stroke (within 14 days of onset) confirmed by computed tomography (CT) scan at Mbarara Regional Referral, Mulago National Referral, and Nsambya Hospitals in Uganda. We will also enroll 100 HIV-uninfected controls, also presenting with acute or sub-acute stroke, matched by gender and county/district of residence. Participants will be followed for up to one year after enrollment. We plan to complete the following research aims:

Aim 1: To identify the impact of HIV-infection on outcomes after acute or sub-acute stroke in Uganda. Our primary outcome will be poor stroke outcome defined as mortality or poor functional status (modified Rankin Scale [mRS] > 2). The investigators will assess for mortality and functional status at one, three, six, and 12 months after presentation. We will estimate the relative risk of poor outcome by HIV serostatus in both unadjusted models and models adjusted for traditional stroke risk factors. We hypothesize that HIV-infected patients have higher rates of poor stroke outcomes compared to HIV-uninfected patients, and that this difference is, in part, attributable to the presence of concurrent central nervous system opportunistic infections.

Aim 2: To describe the clinical and radiologic characteristics of stroke in Uganda, with particular attention to differences by HIV serostatus. The investigators will describe the clinical presentation of stroke, including history and physical exam findings and presenting modified National Institutes of Health Stroke Scale (mNIHSS) score. The investigators will also compare radiologic characteristics, including location of the stroke and its classification as either ischemic or hemorrhagic. We hypothesize that: i) HIV-infected patients are predominantly younger at the time of presentation, ii) HIV infected-patients have more severe stroke presentations with higher mNIHHS scores, iii) HIV-infected patients are more likely to suffer from ischemic (versus hemorrhagic) stroke, and that iv) HIV-infected patients have a greater proportion of small vessel territory strokes due to the association with central nervous system infections and basal meningitis.

Aim 3: To identify the traditional and HIV-specific risk factors for poor outcome after stroke in Uganda. The investigators will fit regression models to identify predictors of poor outcomes as described in Aim 1, both in the total cohort, and restricted to the HIV-infected cohort. We hypothesize that age, presence of hypertension, dyslipidemia, smoking, and low Glasgow coma score are associated with PSO in Uganda and that HIV-specific predictors of poor outcome are low mid-upper arm circumference (MUAC), low CD4 cell count, high HIV viral load, and report of recent constitutional symptoms.

Inclusion Criteria:

• ≥ 18 years old
• In clinical care at the Mbarara Regional Referral, Mulago National Referral, or Nsambya Hospitals
• Hospital presentation with features of new sudden-onset signs of focal or global disturbance of cerebral function lasting >24 hours (unless interrupted by surgery or death)
• Stroke as presumed etiology of presentation
• CT confirmation of stroke

Exclusion Criteria:

• Presumed non-vascular etiology of neurodeficit (ie, trauma, tumor, brain abscess)
• Pregnant, as defined by a positive urine beta-hCG

The population of HIV-infected and uninfected individuals are both drawn from the three hospitals in geographically different parts of Uganda.

Data will be directly entered into a database and analysed