Table 2.

Multivariable (Cox) analysis; predictors of late mortality

	2.5 years censored (n = 403)	5 years censored (n = 323)	7.5 years censored (n = 267)	10 years censored (n = 243)	Full follow-up* censored (n = 238)
	HR	HR	HR	HR	HR
Age at hemorrhage (years)	1.06 (1.05–1.08)	1.08 (1.06–1.09)	1.07 (1.06–1.09)	1.08 (1.06–1.08)	1.08 (1.06–1.09)
Gender (male)	1.44 (1.09–1.83)	1.46 (1.15–1.86)	1.35 (1.08–1.68)	1.35 (1.09–1.67)	1.33 (1.07–1.64)
History of cerebrovascular disease	n.s.	n.s.	n.s.	n.s.	n.s.
History of cardiovascular disease	n.s.	n.s.	n.s.	n.s.	n.s.
History of hypertension	n.s.	n.s.	n.s.	n.s.	n.s.
History of diabetes mellitus	n.s.	n.s.	1.40 (1.07–1.84)	1.37 (1.05–1.79)	1.38 (1.06–1.79)
History of hypercholesterolemia**	0.68 (0.50–0.91)	0.75 (0.58–0.96)	0.64 (0.50–0.81)	0.67 (0.53–0.84)	0.69 (0.55–0.87)
History of peripheral vascular disease	1.70 (1.07–2.71)	1.78 (1.20–2.65)	1.81 (1.24–2.63)	1.87 (1.30–2.70)	1.87 (1.29–2.69)
NIHSS at admission	1.78 (1.52–2.09)	1.55 (1.35–1.78)	1.55 (1.36–1.77)	1.58 (1.39–1.79)	1.58 (1.39–1.79)
Known cause of hemorrhage	n.s.	n.s.	n.s.	n.s.	n.s.
Use of anticoagulants	1.56 (1.15–2.11)	1.47 (1.13–1.90)	1.56 (1.22–1.98)	1.45 (1.15–1.82)	1.43 (1.14–1.80)
First recurrence	n.s.	n.s.	n.s.	n.s.	n.s.
Development of early seizures	n.s.	n.s.	n.s.	n.s.	n.s.
Development of late seizures	n.s.	n.s.	n.s.	n.s.	n.s.
AED use***	n.s.	n.s.	n.s.	n.s.	n.s.

Hazard ratios (±95% CI) for significant covariates in the multivariate Cox survival analysis for the comparison of survival between patients that do not develop seizures (no seizures NS), patients that develop seizures within 7 days after the time of hemorrhage onset (early seizures), and patients that develop seizures more than 7 days after the time of hemorrhage onset (late seizures LS)

HR hazard ratio, n.s. non-significant, AED anti-epileptic drug

* Maximal follow-up up to 12 years (n = 27; 3.6%)

** No longer significant in any interval if statin use is included

*** Only patients who developed late seizures were given AED