A Few Words about Stroke

Recently, there has been some interesting data about stroke that is worth mentioning.

There are some recent developments in the world of stroke, which are interesting:

The first manuscript is an evaluation of stroke risk and the stroke belt. The authors evaluated stroke risk measured over three decade census intervals for persons aged 30 to 80, based on birth or residence in the stroke belt, or both. The highest risk of stroke mortality was for persons both born in and living in the stroke belt (122/100,000). Birth in the stroke belt was the stronger factor: People born in the stroke belt who moved out had a risk of 115/100k. In comparison, those born outside of and living outside of the stroke belt had a mortality rate of 78/100k. The mortality risk elevation was true for both white and black persons, but overall risk for African Americans was higher. For persons both born in and living in the stroke belt, excess mortality compared to those with neither of these criteria was a 34%—55% odds elevation for black persons and a 29%–45% elevation for white persons. Overall stroke mortality risk declined significantly from 1980 to 2000, in agreement with other studies. So, the stroke belt is aptly named.

The second addresses antidepressant use and vascular risk in women. This was a post-hoc analysis of participants in the Women's Health Initiative study, evaluating a very large cohort. In the past, it was suspected that SSRIs reduced stroke risk due to effects on platelets. However, this study found otherwise. 5,496 women were taking antidepressants during the study, with 3,040 of these taking only selective serotonin reuptake inhibitors (SSRIs). In one statistical model, SSRI use was associated with an increase in stroke risk (HR,1.40; 1.09-1.80) and hemorrhagic stroke risk was particularly elevated (HR 2.12). All cause mortality was elevated 32%. Tricyclic antidepressants showed a similar risk elevation. Heart disease rates were not elevated in antidepressant users. These risks persisted in other models which adjusted for a large variety of other factors. It is important to note that, while this study suggest that the risk of stroke is elevated in antidepressant users, the mortality associated with untreated depression is much higher still. Also of note: Previous studies have shown a protective effect of antidepressant use on cardiac disease. This effect was not present in this large cohort study. What remains unclear is whether antidepressant use directly increases stroke and mortality risk, or whether it is a marker for other, unidentified risk factors.

Lastly, researchers in the UK looked at lowering blood pressure (BP) in the setting of acute stroke. The relative risks of hypertension versus iatrogenic hypotension in acute stroke treatment has long been controversial. The investigators evaluated 179 patients with systolic BP > 160 mm Hg. Participants were randomized to lisinopril, labetalol or placebo. The primary outcome measure was death or dependence at two weeks post-stroke. Only 73% of patients completed the 14 day protocol. No specific immediate therapeutic benefit was found: Death or dependency at two weeks occurred in 61% in the two treatment groups combined, v. 59% of the placebo group. However, a numerically significant (but marginally statistically significant) effect in three month mortality, a secondary outcome measure, was found: 10% mortality in the combined t reatment groups v. 20% in the placebo group (HR 0.4, 0.2—1.0; p=0.05). The authors attribute the absence of benefit of active treatment to lack of study power. Thus, despite material study limitations, this trial demonstrated a robust numerical decrease in three month mortality in stroke patients with acutely elevated BP who were treated with antihypertensives. Also, it does not appear that patients with hypertension treated acutely have worse outcomes than those not treated, at least suggesting that acute BP treatment is safe.

Stroke remains a common problem. It is more common in the stroke belt. And, conventional wisdom about treatment effects is not always correct.