Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Anticoagulation for AF: RCTs versus clinical practice

Study
Results
Older versus younger patients
Comment

The number one frequently asked question in evidence-based medicine is "are the results of clinical trials applicable to clinical practice?" One of those simple questions for which simple answers are infrequently available from actual evidence. Because clinical trials often test a single technology and have defined inclusion and exclusion criteria, it can seem as if they could never apply to a clinic of octogenarians all with several other diseases.

It is unusual to have results from a meta-analysis of randomised trials compared with results of clinical practice, but for anticoagulation a group from London have turned up trumps [1], and another from Glasgow provide useful information about anticoagulation in the elderly [2].

Study


The main study [1] sought prospective cohort or retrospective case note reviews of anticoagulation for atrial fibrillation in clinical practice. Patients had to be in ordinary clinical practice settings unrestricted by age or other considerations. Anticoagulation had to be conducted in routine, not research, settings, and there had to be longitudinal data on stroke rates and haemorrhagic complications. Data from such studies was to be compared with a meta-analysis of randomised controlled trials.

Results


There were three eligible clinical practice studies performed in the USA, Canada, and England. They all had similar definitions for atrial fibrillation, risk, criteria for anticoagulation and for outcomes. In all there were 410 patients with 842 years of follow up, compared with 1225 patients and 1889 patient years in randomised trials.

Compared with randomised trials, patients in clinical practice were older, were more likely to be women, and have diabetes, previous stroke or heart failure, but less likely to have ischaemic heart disease.

Rates of ischaemic stroke, intracranial haemorrhage and major bleeding were similar for clinical practice and clinical trials (Figure 1). Minor bleeding occurred more frequently in clinical practice, though individual study rates varied markedly. Including an additional study on institutionalised patients with higher stroke and bleeding events did not change the overall conclusion.

Figure 1: Major events in clinical practice and clinical trials of anticoagulation for AF



Older versus younger patients


A retrospective follow up study in Glasgow [2] would not have fulfilled all the inclusion criteria for clinic studies in the London review, but was close. For instance, it recruited patients aged 60-69 years and those older than 75 years, but not in between.

It had 328 patients with 180 years of follow up, with similar average age but many more women than the randomised trials. There were two ischaemic strokes (1.1 per 100 patient years), four intracranial bleeds (2.2/100), five major bleeds (2.8/100 patient years) and four minor bleeds (2.2/100). The major difference was an excess of intracranial bleeds compared to the other practice studies and trials.

Analysis was by those aged 60-69 years (204 patients) and the 124 older than 75 years. There was no difference in number of high INRs, out of target INRs, INRs above 7 and haemorrhages in the two age groups. Two ischaemic strokes occurred in older women. Short duration follow up of less than 12 months was associated with more INR values above 7 and more haemorrhages.

Comment


Confidence that clinical trial results translate to clinical practice is important in implementing therapy, and fulfilling guidelines and service frameworks. People want to know that what they are doing for their patients will benefit them. Here we have evidence that a wet Thursday in Grimsby is much the same as anywhere else when it comes to using anticoagulation for atrial fibrillation. Older patients do as well as younger ones.

References:

  1. A Evans, L Kalra. Are the results of randomized controlled trials on anticoagulation in patients with atrial fibrillation generalizable to clinical practice? Archives of Internal Medicine 2001 161: 1443-1447.
  2. M Copland et al. Oral anticoagulation and hemorrhagic complications in an elderly population with atrial fibrillation. Archives of Internal Medicine 2001 161: 2125-2128.
previous or next story in this issue