Stopping Statins after Stroke Poses Mortality Risk

ROME -- Stroke survivors who discontinued statin therapy within a year of hospital discharge almost tripled their mortality risk, investigators here found.

Discontinuation of lipid-modifying therapy was the strongest predictor of one-year mortality, with a hazard ratio of 2.78 compared with stroke survivors who continued statin-based therapy, according to a study published online ahead of the October issue of Stroke.

And, the earlier a patient discontinued therapy, the greater the mortality risk, Furio Colivicchi, M.D., of San Filippo Neri Hospital here, and colleagues reported.

The authors placed much of the blame for early discontinuation of statin therapy on a lack of continuity in patient care.

"These findings suggest that patient care should be improved during the transition from a hospital setting to outpatient primary care," they wrote.

Discontinuation of statin therapy soon after acute myocardial infarction has been associated with an increased risk of clinical events, the authors noted. However, the clinical impact of stopping statin therapy after ischemic stroke had not been determined.

A study reported earlier this week in Neurology discussed the increased mortality risk of discontinuing statins in the acute phase of stroke. (See: )

In the current study, Dr. Colivicchi and colleagues prospectively followed 631 consecutive stroke survivors without clinical coronary artery disease.discharged from the hospital over a 4.5-year period. All the patients were on statin therapy at discharge and were followed for 12 months. The primary endpoint was all-cause mortality 12 months after discharge.

The presence of concurrent cardiovascular disease was ruled out during the index admission in all cases by a comprehensive clinical evaluation, which included history, physical examination, 12-lead ECG, and echocardiography.

Consequently, patients with any evidence of CHD, including a history of myocardial infarction, angina pectoris, or myocardial ischemia by stress testing, prior coronary artery bypass grafting or percutaneous coronary angioplasty, and abnormal coronary angiography, were preliminarily excluded.

Statin therapy was in the form of atorvastain 10-20 mg/d (N=409, 77.6%) or simvastatin 20-40 mg/d (N=222, 22.4%).

During 12 months of follow-up, 246 patients (38.9%) discontinued statin therapy. Additionally, 87 patients (13.7%) switched from their initial statin to a different drug in the same class.

The discontinuation rate was similar with atorvastatin and simvastatin, the authors reported. The principal reasons for discontinuation were side effects (71 of 246). In the remaining 175 cases, neither the treating physician nor the patient provided a specific medical reason for discontinuation.

They pointed out that "Because medication costs are covered by the Italian National Health Service, except for a small co-pay, cost cannot be related to these patients discontinuing their prescribed therapy."

Patients who discontinued statin therapy were older (71.4 versus 69.5 years, P=0.002) and more often female (56% versus 44% of male patients, P=0.004).

Patients with diabetes or a history of stroke were more likely to remain on statin therapy.

Overall, 116 of 631 patients (18.3%) died within 12 months of discharge, a figure consistent with mortality after acute ischemic stroke in Italy, the authors stated. In 93 cases (80.1%) the cause of death was cardiovascular in nature.

In multivariate analysis, discontinuation of statin therapy was the strongest predictor of one-year mortality. The earlier a patient discontinued therapy, the greater the mortality risk. In fact, the mortality hazard ratio associated with statin discontinuation decreased over time.

Other significant predictors of one-year mortality were:

• Discontinuation of antiplatelet therapy (HR 1.81, P=0.008)
• Stroke severity at admission (HR 1.11, P=0.002)
• Older age (HR 1.08 per year, P=0.001)

Discontinuation of antihypertensive medication, regardless of drug class, did not predict one-year mortality.

The results call into question the benefits of post-stroke statin therapy found in various clinical trials. Dr. Colivicchi and colleagues suggested that "additional research should be carried out to determine whether the benefits of statin therapy observed in randomized, controlled trials actually apply to the 'real world' of clinical practice."

The authors pointed out that "as in all observational investigations, we cannot exclude the possibility that our results might have been partially conditioned by some form of unmeasured confounding."