Prospective cohort study to determine if trial efficacy of anticoagulation for stroke prevention in atrial fibrillation translates into clinical effectiveness
BMJ 2000; 320 doi: https://doi.org/10.1136/bmj.320.7244.1236 (Published 06 May 2000) Cite this as: BMJ 2000;320:1236All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Sir-Kalra et al (1) reported in a recent publication that
anticoagulation in non-valvular atrial fibrillation (NVAF) yielded
comparable rates of stroke and major complication in clinical practice to
those obtained from trials. Pooled data from randomized controlled studies
in patients with atrial fibrillation showed that anticoagulation reduced
ischemic stroke by two thirds (2). Despite of such a dramatic benefit,
warfarin is underused in patients with atrial fibrillation in clinical
practice (3). Moreover, this particularly occurred in elderly rural
patients (4).
We asked reasons why warfarin was not prescribed for NVAF patients
with risk factors (previous stroke, hypertension, diabetes mellitus, etc.)
for stroke in clinical setting. A questionnaire was sent to 40 general
practitioners in 30 clinics in Kochi, Japan; 30 (75%) doctors in 20
clinics responded. In 10 (50%) clinics of responders, the majority of
which provide primary care with no beds for admission or less than 20,
rapid anticoagulant test in their own facilities were not available.
Almost half of the doctors (14/30) prescribed warfarin for only 10 % of
NVAF patients or less and only 10% (3/30) of doctors did anticoagulation
for 50% of NVAF patients and more. Conversely, antiplatelet therapy was
substantially utilized; more than half of doctors (16/30) prescribed
aspirin or ticlopidine for 50% of NVAF patients and more. Target
therapeutic range for anticoagulation by more than 80% of doctors was 1.5-
2.5 in INR of prothrombin time and 10-30% in thrombotest. Many doctors
were reluctant to prescribe warfarin to NVAF patients with risk factors
for stroke in this survey, and the reasons were as following; Unstable
anticoagulation due to bad compliance of warfarin in patients (53%),
doctorsÕ fear of bleeding complications (40%), Uncontrolled dose of
warfarin due to unavailable rapid anticoagulant test (30%), etc. PatientsÕ
refusal to anticoagulation (13%) or reliable efficacy of antiplatelet
therapy for stroke prevention (13%) was a minor opinion.
Many doctors in small clinics ascribed, in part, underuse of warfarin
to the lack of their own anticoagulant laboratories in our survey. If they
manage NVAF patients, anticoagulant test are requested for laboratory
services and the results are provided a few days later; the time lag would
make strict control of anticoagulation difficult. Because atrial
fibrillation is a common disease which is detected in 5-10% or more of
elderly people (5) and many cases are asymptomatic, not only specialists
in hospitals but also general practitioners in primary care are to take
responsibility of managing patients, especially in rural areas apart from
hospitals. Anticoagulant laboratory services should develop to be able to
provide rapid results for general practitioners.
References
1. Kalra L, Yu G, Perez I, Lakhani A, Donaldson N. Prospective
cohort study to determine if trial efficacy of anticoagulation for stroke
prevention in atrial fibrillation translates into clinical effectiveness.
BMJ 2000;320:1236-1239.
2. Atrial Fibrillation Investigators. Risk factors for stroke and efficacy
of antithrombotic therapy in atrial fibrillation. Analysis of pooled data
from five randomized controlled trials. Arch Intern Med 1994;154:1449-
1457.
3. Brass LM, Krumholz HM, Scinto JM, Radford M. Warfarin use among
patients with atrial fibrillation. Stroke 1997;28:2382-2389.
4. Flaker GC, McGowan DJ, Boechler M, Fortune G, Gage B. Underutilization
of antithrombotic therapy in elderly rural patients with atrial
fibrillation. Am Heart J 1999;137:307-312.
5. Feinberg WM, Blackshear JL, Laupacis A, Kronmal R, Hart RG. Prevalence,
age distribution, and gender of patients with atrial fibrillation.
Analysis and implications. Arch Intern Med 1995;155:469-473.
Competing interests: No competing interests
EDITOR - The paper by Kalra et al 'Prospective cohort study to
determine if trial efficacy of anticoagulation for stroke prevention in
atrial fibrillation translates into clinical effectiveness'1 extends our
knowledge of how effectively research into stroke prevention in atrial
fibrillation (AF) can translate into clinical practice. However, it
researched patients recruited in district general hospitals (incorrectly
stated in the accompanying Editorial by Connolly to be patients from
general practices), and whose anti-coagulants were controlled in hospital
out-patient clinics. As a result there remain major differences from the
whole population of patients with AF. While the cohort of patients
studied by Kalra was significantly older than that in the pooled AF
studies, it is still younger than those patients who make up primary
health care teams' (PHCT) population of patients with AF (48% aged >75
cf 58-72% aged >75 in primary care based studies 2,3 ).
Further, the
information given by Kalra about the frequency with which INR was measured
can be calculated to imply a mean value of approximately 27 hospital tests
per patient over two years. This will have excluded from analysis all
those who were unwilling or unable to meet this commitment, including many
of the frailest patients about whom general practitioners are especially
concerned when it comes to deciding whether to initiate anticoagulation.
Pragmatic studies of effectiveness must be based on the whole population
of patients with AF and the inevitable implication that blood for INRs
will have to be obtained frequently in GPs surgeries or patients' homes
and the results applied by PHCTs.
A literature is beginning to develop which suggests that in unselected
primary care populations about half of all patients with AF will be
considered for anticoagulation and will accept that advice 2,4,5. This
may be the level against which performance should be judged. If so, it
sets a limit on our ambitions and may help us to be more realistic about
the likely effectiveness of applying the evidence. It also means that,
although there is undoubtedly scope for identifying more people who should
be offered treatment, current levels of anticoagulation represent an
existing penetration of evidence twice as great as the raw figures
suggest.
References:
1. Kalra L, Yu G, Perez I, Lakhani A, Donaldson N. Prospective cohort
study to determine if trial efficacy of anticoagulation for stroke
prevention in atrial fibrillation translates into clinical effectiveness.
BMJ 2000;320:1236-1239
2. Oswald N, Bateman H. Applying research evidence to individuals in
primary care: a study using non-rheumatic atrial fibrillation. Family
Practice 1999;16:414-419
3. Lip GYH, Golding DJ, Nazir M, Beevers DG, Child DL, Fletcher RI. A
survey of atrial fibrillation in general practice: the West Birmingham
Atrial Fibrillation Project. Br J Gen Pract 1997;47:285-289
4. Howitt A, Armstrong D. Implementing evidence based medicine in
general practice: audit and qualitative study of antithrombotic treatment
for atrial fibrillation. BMJ 1999;318:1324-1327
5. Sudlow M, Thomson R, Thwaites B, Rodgers H, Kenny RA. Prevalence
of atrial fibrillation and eligibility for anticoagulants in the
community. Lancet 1998;352:1167-1171
Nigel Oswald
Professor in Primary Health Care
Universities of Teesside and Newcastle upon Tyne,
Primary Care Resource & Development Centre,
Stokesley Road,
Nunthorpe,
Middlesbrough
TS7 OPN
Competing interests: No competing interests
Sir - Kalra et al in his study of warfarin for patients with high
risk AF (accurately identified by specialists, with access to
echocardiography) shows that 296 patient years of warfarin treatment
reduced ischaemic stroke risk by an amount comparable to clinical trials.
Connolly’s editorial says that meta-analysis of such trials shows the risk
of ischaemic stroke is reduced by two thirds, and asks why warfarin is
under-used.
In Kalra et als study six ischaemic strokes occurred. Assuming a risk
reduction of two thirds then presumably around twelve were prevented.
However five major bleeds occurred and sixteen ‘minor’ bleeds , the
latter including epistaxes, bruising, rectal bleeding, haematuria, and
haemarthrosis all of which presumably required investigation and
treatment.
GPs are frequently asked to initiate, take over, and otherwise manage
anticoagulation. Many areas have poorly developed anticoagulant clinics,
and patients in rural areas, residential care etc.. may be unable (or
unwilling)
to travel. This will encompass a significant number of ‘high risk’
patients.
An INR in these patients may need a home visit by the GP as many district
nursing services are hard pressed, and many will not enter nursing
homes.Assuming seven INRs are needed to stabilise a patient at initiation
,a further twenty three per patient will be needed over the two years at
the
thirty one day intervals used by Kalra et al , making a total of thirty.
Assuming ten minutes each this is five hours of GP/nurse time, not
including travel , ‘while you’re here doctor’ consultations, dealing with
complications and the problems leading to drop out(Kalra et al recruited
167 patients, of whom 18%dropped out),repeat prescriptions, telephone
calls
about co-prescriptions etc...
By going through the above on about twelve patients for two years we
prevent 1 ischaemic stroke, and generate 1.7 bleeding complications, in
this well-defined, high risk group identified by specialists( in everyday
practice the therapy may not be so well targeted, so results may be less
impressive). This utilises 60 hours taking blood, plus other time as above
which can only be guessed at. Despite this I am an enthusiastic supporter
of warfarin for carefully selected patients. As Moss in her editorial
points
out however, the UK has fewer doctors and nurses per capita than other
developed nations, and GPs are not short of imposed targets in
hypertension, diabetes, prescribing, ischaemic heart disease prevention,
and elderly care
to name just a few. The danger is that GPs will be berated by the
Department of Health for yet another ‘failure’, when the problem is an
under-resourced system of care which forces them to make very difficult
choices about how to
allocate their time, which some of our specialist colleagues seem to be
genuinely unaware of. The problem of how to practically implement the new
NSF for cardiovascular disease, given the audit demands it imposes on GPs,
is going to be a major one.
M. D. Oliver MB ChB MRCP
General Practitioner,
Browning Street Surgery,
Stafford.
No competing interests.
Competing interests: No competing interests
We read with interest the paper of Kalra et al. entitled “Prospective
cohort study to determine if trial efficacy of anticoagulation for stroke
prevention in atrial fibrillation translates into clinical effectiveness”,
recently published on BMJ.(1) The authors, by demonstrating that their
rates of stroke and major haemorrhage after anticoagulation were
comparable from pooled data of randomized controlled studies, confirm that
warfarin is more effective than aspirin for stroke prevention in chronic
atrial fibrillation. The authors also claim that this treatment is
feasible, safe and effective in clinical practice even in elderly people,
indirectly suggesting that the reasons of warfarin under-use are strongly
related to physicians’ malpractice. Because barriers to anticoagulation
implementation are not conclusively addressed, responsibility of warfarin
under-use could be blamed principally to physicians’ incompetence about
anticoagulation benefit. We are not persuaded that this is true in all
cases. Oral anticoagulation therapy is a complex and labor-intensive
therapeutic modality, where not only good dosing decisions, but attention
to detail and communication can make difference between success and
failure.(2) Physicians may have some concerns to start this treatment when
the risks linked to an inadequate management are so large to overhelm the
benefit of stroke prevention. (3-5)
Recent European studies showed that prevalence of dementia among
patients aged 65 or more is around 6.0% , increasing twice over each
decade. On the basis of these data, we could expect that nearly 30% of
elderly subjects with 80 years or more may have cognitive impairment
conditioning the performance in activities of daily living. In this case,
a strong support by caregivers and prompt availability of laboratory
service are needed to carry out the regular testing of patients’
international normalized ratio. If this is not possible, physicians may
legitimately opt for a less effective but more manageable
thromboprophylaxis.
In their study, Kalra et al. found that treatment with warfarin was
stopped in 18% of patients, reasons including death, major bleed,
intracranial haemorrhage, disabling stroke, patients’ choice, poor
compliance and interactions with other drugs. Unfortunately, because the
authors did not assessed the cognitive and functional status of their
patients, we can only speculate that warfarin withdrawal was related in
some cases to inadequate management of long.-term anticoagulation. We
suggest that a comprehensive geriatric evaluation become a routine
physician procedure before starting this therapy.
References
1. Kalra L, Yu G, Perez I, Lakhani A, Donaldson N. Prospective cohort
study to determine if trial efficacy of anticoagulation for stroke
prevention in atrial fibrillation translates into clinical effectiveness.
BMJ 2000; 320:1236-9.
2. Ansell JE. The quality of anticoagulation management. Arch Intern Med
2000;160:895-6.
3. Bellelli G, Barbisoni P, Gusmeri A, Sabatini T, Rozzini R, Trabucchi M.
Underuse of anticoagulation in older patients with chronic atrial
fibrillation: malpractice or accuracy? J Am Geriatr Soc 1999;47:1034-5.
4. Rozzini R, Sabatini T, Trabucchi M. Risk assessment and anticoagulation
in atrial fibrillation in the elderly: malpractice or accuracy? Stroke
1999;30:2239-40.
5. Rozzini R, Sabatini T, Bellelli G, Trabucchi M. Anticoagulation in
elderly patients with chronic atrial fibrillation: the need for geriatric
assessment. Age Ageing (in press).
Competing interests: No competing interests
ARTIFICIAL REALITY: Can we create reality in clinical trials?
Dear Editor,
Kalra et al(1) report a two year prospective cohort study of patients
with atrial fibrillation at high risk of stroke. They compare prophylaxis
with warfarin in clinical practice with the pooled data of five randomised
controlled trials (RCTs). They conclude that their results are
comparable, and believe their study strengthens the case for 'wider but
judicious use of anticoagulation'.
The authors conducted their study because they believed that 'major
randomised trials were considered unrepresentative of clinical practice'.
Despite drawing patients from their clinics, they based the inclusion and
exclusion criteria on the previous RCTs. This was intended to allow
comparisons to be made. However, we would argue that this resulted in the
use of another artificially selected population, potentially not truly
representative of day to day clinical practice.
The aim of the study concerned primary prevention of stroke, and
therefore patients who had suffered a stroke in the past six months were
excluded. However, patients with strokes occurring more than six months
ago were included. Treatment in these patients would therefore be
considered as secondary prevention.
We were also concerned with the reliability of some of the data. The
International Normalised Ratio (INR), as a measure of anticoagulation,
appeared poorly controlled, only being within the target range 63% of the
time. Long intervals between assessment of INR mean that the actual time
within target range may be even less than this. As anticoagulation is the
main intervention its measure needs to be accurate to ensure validity.
In conclusion, it is important to know if the results of RCTs can be
applied to clinical situations, and the authors have attempted to address
this issue. The conclusions they draw from their results are limited by
difficulties in reproducing a comparable clinical setting. The paper does
provide some supportive evidence for anticoagulation in atrial
fibrillation patients at high risk of stroke, but we feel the data should
be viewed with a degree of caution.
Tara Brown, Rumana Chowdhury, Josephine Flowers, Deepta Vijayaratnam,
Al-Hashem Abdulmohsen,
3rd year medical students,
Department of
Epidemiology and Public Health, the Medical School, University of
Newcastle.
J.R.Flowers@ncl.ac.uk
(1) Kalra L, Yu G, Perez I, Lakhani A, Donaldson N, Prospective
cohort study to determine if trial efficacy of anticoagulation for stroke
prevention in atrial fibrillation translates into clinical effectiveness.
BMJ 2000;320;1236-9.
Competing interests: No competing interests