BMJ study showed patients given low dose HRT patches had a lower risk of stroke than those treated with oral therapy. However, high dose patches were associated with increased risk.
Previous studies have examined the risk of stroke associated with HRT in post-menopausal women, but focused on oral delivery and did not compare different methods.
One study has suggested a lower risk of stroke with transdermal HRT compared with oral methods.
A team from Canada and Germany examined whether delivery method could impact the risk of side effects.
Researchers studied the risk of stroke associated with oral and transdermal HRT in 75,000 post-menopausal UK women.
Women were exposed to a variety of different HRT therapies, including estrogens only, estrogens plus progestogen, progestogen only and tibolone. Therapy was administered transdermally or orally, in high or low doses.
Overall, 15,710 cases of stroke occurred among all patients. These cases were then matched to 59,958 controls.
Patients receiving low-dose HRT patches had no increased adjusted risk of stroke compared with controls.
However, women using oral HRT had a 28 per cent increased stroke risk compared with those who had not used HRT. Courses of less than a year also reduced the risk of stroke for both delivery methods, compared with longer courses.
The researchers concluded: 'Our study suggests the use of transdermal estrogen replacement therapy containing low doses of estrogen could be associated with a lower risk of stroke than the oral administration.'
However, they said they could adjust for differences in menopausal age or socioeconomic status in the results, which may have confounded the results.
Authors said the results did not represent definitive evidence to promote the use of patches over oral administration for HRT. But they hoped it would encourage further research on the importance of delivery method.