Background/Purpose: Previous studies indicate that individuals
with SLE have a high risk of stroke but have not investigated when the highest
risk occurs. We aimed to estimate the risk of ischemic stroke (IS) and
hemorrhagic stroke (HS) in SLE compared to the general population by age, sex,
and time since diagnosis.

Methods: Adult patients (≥18y) with SLE were
identified from the National Patient Register (NPR) to create a
population-based and nationwide cohort of individuals with SLE. SLE was defined
as ≥2 SLE ICD codes in the inpatient (1969-2013) or outpatient (2001-2013)
registers with ≥1 code in a specialist clinic and ≥1 in outpatient
care. We included only those who received their first observed SLE ICD code Jan
2003 or later as a proxy for incident SLE. General population comparators were
selected from the Total Population Register alive at the SLE patient’s
estimated diagnosis date (index date) and were matched 5:1 to the SLE population
on age, sex, and county. Stroke was identified from the NPR and those with a
history of stroke at baseline were excluded. Follow-up was from index date
through first of death, stroke, emigration or end of study (Dec 2013). Age- and
sex-adjusted Cox models were used to estimate the hazard ratio (HR) for stroke comparing
SLE to non-SLE. Models were additionally stratified by sex, age at diagnosis,
attained age and time since diagnosis. Effect modification by stratification
variables was tested using the likelihood ratio test (LRT). In secondary analyses,
we estimated the HR for stroke among individuals with confirmed SLE from two
clinical cohorts.

Results: We identified 3 133 individuals with SLE and
15 504 general population comparators. The average age at SLE diagnosis was 49
and 85% were female. We observed 90 strokes in SLE (76 IS, 14 HS) and 198 in
the general population (160 IS, 38 HS). The HR for IS comparing SLE to the
general population was 2.4 (95%CI 1.9, 3.2) and the HR for HS was 1.9 (95%CI
1.0, 3.5). HRs were higher in analyses using confirmed SLE only (IS 4.2 (95%CI
2.2, 8.0), HS 4.4 (95%CI 1.2, 16.5).

The HR for IS was significantly higher in females than in males
(females HR 3.0 (95%CI 2.2, 4.1), males HR 1.5 (95%CI 0.9, 2.6); LRT p=0.01; Figure).
Younger individuals (18-49y) had the highest risk of stroke (IS HR 10.2 (95%CI
3.5, 29.7), LRT p=0.003; HS HR 7.7 (95%CI 2.2, 27.1), LRT p=0.008). The HR for
IS was similar over time since diagnosis whereas the HR for HS was higher the
further from diagnosis (LRT p=0.05), although this was based on small numbers.

Conclusion: Individuals with SLE are at high risk of both
ischemic and hemorrhagic stroke. The relative risk was especially high in
females and below the age of 50. Notably, the risk for IS was high already within
the first year of diagnosis and remained constant during the following decade.

Figure. Hazard ratios and 95% confidence intervals
for ischemic and hemorrhagic stroke stratified by sex, age and time since
diagnosis comparing individuals with SLE to the general population.