Skip to content

Advertisement

You're viewing the new version of our site. Please leave us feedback.

Learn more

BMC Research Notes

Open Access

Baseline characteristics in stroke patients with atrial fibrillation: clinical trials versus clinical practice

  • Christian Tanislav1Email author,
  • Sonja Milde2,
  • Sabine Schwartzkopff3,
  • Björn Misselwitz4,
  • Nicole Sieweke1 and
  • Manfred Kaps1
BMC Research Notes20158:262

https://doi.org/10.1186/s13104-015-1237-2

Received: 9 January 2014

Accepted: 17 June 2015

Published: 25 June 2015

Abstract

Background

When applying information gathered from medical research to the clinical setting, it is imperative that the sample of the investigated patients be representative of the clinical population. Because of this fact, it is necessary to closely examine the sample’s baseline characteristics in clinical trials.

Methods

We analysed baseline data of relevant trials investigating considerable proportions of patients with atrial fibrillation (AF) in the secondary stroke prevention: EAFT, SIFA, Active-W, BAFTA, RE-LY, AVERROES, ARISTOTLE and ROCKET AF. For comparing baseline data stroke patients with AF documented in a statutory stroke registry were considered. In a subgroup of patients (members of a large insurance) data on subsequent prescription for oral anticoagulants (oAK) were available.

Results

In the stroke registry (n = 15,886) the mean age was higher than in the selected clinical trials (mean 77.7 versus 70–72 years). Among insurance members (n = 1,828), those with a prescription for oAK (n = 827) were older than patients recruited in clinical trials (mean 75.1 versus 70–72 years). Results also showed that the male sex was overrepresented in clinical trials (59–63% versus 46%). The distribution of vascular risk factors in recent clinical trials was comparable to proportions in the registry (hypertension: 77–85% versus 80%; diabetes mellitus: 20–26% versus 27%).

Conclusions

The majority of stroke patients with AF in the clinical setting are considerably older than those included in clinical trials. While the distribution of vascular risk factors in clinical trials corresponds to proportions observed in clinical practice, an overrepresentation of the male sex in clinical trials is evident.

Keywords

StrokeOral anticoagulationAtrial fibrillationCare deliveryNew anticoagulants

Background

Evidence-based medicine, identified in clinical trials, is crucial to facilitate appropriate decision-making in patient care. However, patients in clinical trials inevitably represent a selected sample of the population. As a result, physicians might challenge the external validity especially in specific subgroups such as elderly patients. Additionally, different sources of bias might distort results. One of the most important and most often discussed confounding variables is selection bias, which occurs when considering patients to participate in a study [1]. Applying information derived from medical research into the clinical setting depends on the generalizability of the investigated patients to the population of interest.

A large number of studies have shown the efficacy of oral anticoagulation (OA) in atrial fibrillation (AF) [210]. These studies evaluated primary and secondary stroke prevention with results demonstrating the substantial benefit of OA [210]. Therefore, it is crucial to determine how the baseline characteristics of study samples compare to those of patients encountered in clinical practice and nation-wide populations.

In the presented investigation, we aimed to analyse baseline characteristics in relevant clinical trials of atrial fibrillation when compared to stroke victims documented in a larger, more representative stroke registry sample.

Methods

We evaluated the baseline data of previous clinical trials with substantial proportions of stroke patients with atrial fibrillation, including: EAFT, SIFA, Active-W, BAFTA, RE-LY, AVERROES, ARISTOTLE and ROCKET AF [210]. Characteristics such as age, sex, vascular risk factors including hypertension and diabetes mellitus were investigated. For RE-LY, ARISTOTLE and ROCKET AF baseline data and frequencies calculated in the stroke subgroups were collected for analysis.

For comparison, we gathered baseline data from ischaemic stroke or TIA patients with AF documented (2004–2010) in the stroke registry (n = 15,886) of the Institute of Quality Assurance Hesse (Geschäftsstelle für Qualitätssicherung, GQH) (Figure 1) [11]. Patients with a haemorrhagic stroke or intracranial bleeding were excluded from this study sample. Patients with moderate disability (modified Rankin scale ≤3 as assessed on discharge) were selected because these patients might most meet the criteria for measures in the secondary stroke prevention such as intake of oral anticoagulants. This selection might further provide certain structural equality regarding the disability status when comparing patients in clinical trials versus daily practice.
Figure 1

Patient selection within the stroke registry data set. TIA refers to transient ischaemic attack, CVT refers to cerebral vein thrombosis, mRs refers to modified Rankin scale.

Among this sample, we selected a subgroup of patients (n = 827) from a large health insurance consortium with evidence of a prescription for oral anticoagulants (oAK). For this subgroup insurance data between 2005 and 2007 was used. In order to identify pertinent members of this insurance, we examined patients matching a set of criteria including date of birth, date of hospital admission and the admitting hospital. This data was linked in a pseudonymous manner. For our analysis, we selected patients without further hospitalisation in a 30-day period within 90 days after discharge (n = 1,828). We defined evidence of a prescription for oral anticoagulants (including phenprocoumaron, warfarin and coumadin) as a marker of anticoagulation. This information was gathered from the insurance claims; 827 patients were identified. For the categorical variables, data was presented in proportions. A Chi squared test was used to compare proportions between the entire sample identified in the Hessian stroke registry, the EAFT and SIFA’s samples, and the subgroups of patients with a previous stroke or TIA in RE-LY, ARISTOTLE and ROCKET AF.

The protocol of the present study was reviewed and approved by the ethical committee of the medical faculty of the Justus Liebig University Giessen.

GQH

The GQH database is a mandatory nationwide hospital-based registry spanning more than 95% of all ischaemic strokes, transient ischaemic attacks (TIA), and intracerebral haemorrhages in more than 6 million residents of Hesse, Germany. The GQH includes data of acute inpatient treatment, as well as factors proven to be relevant for the course and the prognosis of a stroke. For quality assurance purposes, the acquisition of this data is regulated by law and implemented as a guideline, which is elaborated by the Federal Joint Committee for hospital quality assurance in accordance with Volume V of the Social Insurance Code (§137 SGB V and §135a SGB V). Based on this regulation, the Hesse State Hospital Law contains a provision that allows the GQH to record such data legally. The publication of aggregate quality assurance data has been approved by the Hesse Data Protection Commissioner, so no data protection problem arises here [1214].

Results

In the GQH registry sample (n = 15,886) the mean age was 77.7 years. Apart from the BAFTA trial (mean age 81.5 years), the mean age in the selected trials and subgroups of patients with a previous stroke or TIA ranged between 70 and 72 years. In the insurance subgroup the mean age was comparable to the registry sample (77.6 versus 77.7 years) (Table 1). In the insurance subgroup of patients with a prescription for oAK, the mean age was 75.1 years and in those without evidence of a prescription the mean age was 79.8.
Table 1

Comparison of baseline characteristics between the registry cohorts versus relevant studies providing evidence on secondary stroke prevention in atrial fibrillation

 

Hessian stroke registry (2004–2010) n = 15,886

Insurance (total Cohort) n = 1,828

Insurance (subgroup of patients with prescription for AK) n = 827

EAFT (1993) n = 669

SIFA (1997) n = 916

Active-W (2006) n = 6,706

BAFTA (2007) n = 973

AVERROES (2011) n = 5,599

RE-LY (2010) n = 18,113

ARISTOTLE (2011) n = 18,201

ROCKET AF (2012) n = 14,264

Age (years) mean, ±SD

77.7 (±9)

77.6 (±8)

75.1(± 8)

71 (±7)

72 (± 8)

70.2

81.5 (±8)

70 ± 9

70 (±9)a

70.1 (±9.7)a

71 (median)a

(IQR 64–77)

Sex

 Male

7,312 (46%)

764 (41.8%)

398 (48.1%)

395 (59%)

430 (47%)

4,430 (66%)

531 (55%)

3,277 (59%)

2,279 (63%)a,†

2,152 (63%)a,†

4,538(61%)a,†

 Female

8,574 (53%)

1,064 (58%)

429 (51.9%)

274 (41%)

486 (53%)

2,276 (34%)

442 (45%)

2,322 (41%)

1,344 (37%)a,†

1,284 (37%)a,†

2,930(39%)a,†

Risk factors

 Hypertension

12,679 (80%)

1,444 (79%)

660 (79.8%)

294 (44%)

506 (55%)

5,522 (82%)

528 (54%)

4,837 (86%)

2,783 (77%)a,†

2,858 (83%)a,†

6,343 (85%)a,†

 Diabetes mellitus

4,371 (27%)

530 (29%)

240 (29%)

87 (13%)

144 (16%)

1,429 (21%)

129 (13%)

1,096 (20%)

816 (22%)a,†

902 (26%)a

5,695 (24%)a,†

Proportion of Strokes or TIAs within the study population

100%

100%

100%

100%

100%

15% (n = 1,006)

13% (n = 124)

14% (n = 764)

20% (n = 3,623)

19% (n = 3,436)

52% (n = 7,468)

Comparison of agents

Warfarin vs. placebo and aspirin vs. placebo

Indobufenb vs. warfarin

Aspirin + clopidogrel vs. warfarin

Warfarin vs. aspirin

Apixaban vs. aspirin

Dabigatran vs. warfarin

Apixaban vs. warfarin

Rivoroxaban vs. warfarin

aData refers to the subgroup of patients with a previous stroke or TIA.

bIndobufen is a reversible inhibitor of platelet cyclooxygenase activity.

p value <0.001 calculated with a Chi square test (comparison between the entire cohort determined in the hessian stroke registry (n = 15,886) and the EAFT cohort, the SIFA cohort and the stroke subgroups in RE-LY, ARISTOTLE and ROCKET AF).

The proportion of males in the registry sample was 46%, while in the insurance subgroup only 41% of patients were male. The proportion of males increased in the insurance subgroup of patients with an oAK prescription to 48.1%. Apart from SIFA (proportion of male patients 47%) there was a majority of males reaching 55–66% in the selected clinical trials (Table 1).

The proportion of hypertension and diabetes mellitus in the registry population, in the selected trials, and in the subgroups of patients with a previous stroke or TIA were similar (hypertension 80, 79 and 79.8%); diabetes: 27, 29 and 29%) (Table 1). Earlier investigations (EAFT, SIFA and BAFTA) focused on patients with a moderate vascular burden and had lower proportions of these risk factors (hypertension: 44–55% of the patients; diabetes mellitus: 13–16% of the patients). In recent trials (Active-W, RE-LY, AVERROES, ARISTOTLE and ROCKET AF) these factors are of similar proportions when compared to the registry population (hypertension: 77–86% of the patients; diabetes mellitus: 20–26% of the patients).

Using a Chi squared test, we compared the proportions of these characteristics among the entire study sample, which was composed of the Hessian stroke registry and the EAFT and SIFA populations and the subgroups of patients with stroke or TIA in RE-LY, ARISTORLE and ROCKET AF. Results showed that in the majority of the cases p values were <0.001 (Table 1). However, no relevant differences were identified in the comparison between proportions of patients with diabetes mellitus in the registry study group and the stroke or TIA subgroup in ARISTOTLE (27 versus 26%, P = 0.1317) and the gender comparison with SIFA (males: 46 versus 47%, P = 0.5889).

Discussion

In the registry population patients were on average 7 years older than those included in the selected studies [210]. Apart from age, the comparison of baseline characteristics revealed an overrepresentation of male individuals in clinical trials. The proportions of vascular risk factors such as hypertension and diabetes mellitus were comparable in recent clinical trials with those seen in clinical practice.

It is also important to note that factors such as age might impact a patient’s inclusion in clinical trials on OA. In these clinical trials, there are strict inclusion criteria that determine whether or not a patient is included. The factor age may bias the inclusion in studies due to the increased risk of bleeding in the elderly and as a result elderly patients may be less selected for inclusion in clinical trials for OA. In our GQH registry sample the mean age within the subgroup of patients with prescriptions for oAK was lower than in the total registry sample (75.1 vs. 77.7 years respectively); however, it still ranged above the mean age observed in clinical trials (75.1 vs. 70–72 years).

The BAFTA trial, which included oAK for individuals with AF over 75 years of age, reported a 2.4% absolute risk reduction per year and a similar risk for major bleeding when compared to aspirin (1.9 versus 2.0%) [8]. While distributions of vascular risk factors in the recent trials are comparable to proportions calculated in the registry population, participants in the BAFTA-trial were less affected by vascular risk factors (hypertension: 80 versus 54%; diabetes mellitus 27 versus 13%) [8]. The burden of leukoaraiosis, microbleeds and silent brain infarctions, which correlates with the presence of vascular risk factors, substantially increases beyond 75 years of age; this results in a higher risk for intracerebral haemorrhage [1520]. In the BAFTA selection process there was a preference for elderly individuals without considerable vascular burden. Therefore, those patients with supposedly higher risk for bleeding were less preferred for inclusion. Furthermore, regarding secondary stroke prevention, the BAFTA trial is of limited benefit, as only a small proportion of patients who have had a previous stroke were investigated (13%) [8].

ROCKET AF claims to address more elderly patients (median 73 years) [10]. However, in the subgroup of patients with a previous stroke, the median age decreases (median 71 years) corresponding with the subgroups in RE-LY (mean 70.5 years) and ARISTOTLE (mean 70.1 years) [7]. In this case the additional burden of a previous cerebrovascular event a further bias to younger patients.

In many clinical trials males represent the majority of the participants [24, 21, 22]. In contrast, in nation-wide data collections, males do not exceed 50% [2325]. In line with this finding, the proportion of male participants in our nationwide registry was also lower than in clinical trials (46% versus 55–66%). This might also be a result of the increased age of the participants as females suffer from stroke later in life than males [18]. Consistent with this assumption, the slight decrease in age within the insurance subgroups (mean 77.6 versus 75.1 years) corresponded with a higher proportion of male subjects (41.8–48.1%).

Even though EAFT and SIFA included a small number of patients with a relatively moderate burden of vascular risk factors, we considered these two ground-breaking trials, because both specifically addressed stroke patients in secondary prevention [2, 9]. In these trials the significant advantage of oAK over a therapy with platelet inhibitors was demonstrated. In contrast, the vast majority of AF trials provided evidence for the safety and efficacy of OA in the general population of both stroke and non-stroke patients. The percentage of stroke patients ranged from 13 to 25% in these trials. Despite methodological limitations, specifically the small number of patients, these subgroup analyses indicated the efficacy of OA in patients with AF after having a stroke [4, 5, 22]. Indisputably, the participation of stroke patients in AF trials is crucial because they face higher risks of both recurrent stroke and bleeding complications [18]. Because of this higher risk of a worse outcome, it is difficult to extrapolate the benefit of OA in secondary prevention from primary prevention studies. Trials designed to specifically investigate patients within secondary stroke prevention would be ideal.

Conclusions

Despite clear evidence of the efficacy of OA, there is an age-dependent variable when considering stroke patients for this treatment. There is a considerable difference in age of patient populations in clinical trials when compared with clinical practice. Results have shown that in clinical trials younger patients (70–72 years) are preferable; however, in clinical practice, patients selected for OA are older (75.1 years) than those in trials, but also younger than those who are selected for alternative therapy (79.8 years). With regard to the vascular risk factors of hypertension and diabetes mellitus, we found that the proportions in recent clinical trials are comparable to those in clinical practice. Both patients and clinicians would benefit from further investigations into secondary stroke prevention when selecting patients suitable for OA. Currently evidence is generally collected from studied in the primary stroke prevention with considerably younger patients.

Abbreviations

OA: 

Oral anticoagulation

AF: 

atrial fibrillation

GQH: 

stroke registry of the Institute of quality assurance Hesse (Geschäftsstelle für Qualitätssicherung, GQH)

TIA: 

transient ischaemic attacks

oAK: 

oral anticoagulants

mRS: 

modified Rankin scale

Declarations

Authors’ contributions

MK and CT designed the study. CT and SM calculated the results. CT, SM, BM, SS, NS and MK were involved in data analysis and interpretation. They proved und contributed for important intellectual content. All authors read and approved the final manuscript.

Disclosures

All authors report no disclosures related to the manuscript.

Compliance with ethical guidelines

Competing interests The authors declare that they have no competing interests.

Study sponsorship

There was no funding nor any funders involved in the study.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Authors’ Affiliations

(1)
Department of Neurology, Justus Liebig University
(2)
Dresden International University
(3)
AOK Division of the Federal State of Hesse
(4)
Geschäftsstelle Qualitätssicherung Hessen (GQH)

References

  1. Miller KD, Rahman ZU, Sledge GW Jr (2001) Selection bias in clinical trials. Breast Dis 14:31–40PubMedGoogle Scholar
  2. EAFT (European Atrial Fibrillation Trial) Study Group (1993) Secondary prevention in non-rheumatic atrial fibrillation after transient ischaemic attack or minor stroke. Lancet 342:1255–1262Google Scholar
  3. Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A et al (2009) Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 361:1139–1151PubMedView ArticleGoogle Scholar
  4. Diener HC, Connolly SJ, Ezekowitz MD, Wallentin L, Reilly PA, Yang S et al (2010) Dabigatran compared with warfarin in patients with atrial fibrillation and previous transient ischaemic attack or stroke: a subgroup analysis of the RE-LY trial. Lancet Neurol 9:1157–1163PubMedView ArticleGoogle Scholar
  5. Easton JD, Lopes RD, Bahit MC, Wojdyla DM, Granger CB, Wallentin L et al (2012) Apixaban compared with warfarin in patients with atrial fibrillation and previous stroke or transient ischaemic attack: a subgroup analysis of the ARISTOTLE trial. Lancet Neurol 11:503–511PubMedView ArticleGoogle Scholar
  6. Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M et al (2011) Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 365:981–992PubMedView ArticleGoogle Scholar
  7. Hankey GJ, Patel MR, Stevens SR, Becker RC, Breithardt G, Carolei A et al (2012) Rivaroxaban compared with warfarin in patients with atrial fibrillation and previous stroke or transient ischaemic attack: a subgroup analysis of ROCKET AF. Lancet Neurol 11:315–322PubMedView ArticleGoogle Scholar
  8. Mant J, Hobbs FD, Fletcher K, Roalfe A, Fitzmaurice D, Lip GY et al (2007) Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet 370:493–503PubMedView ArticleGoogle Scholar
  9. Morocutti C, Amabile G, Fattapposta F, Nicolosi A, Matteoli S, Trappolini M et al (1997) Indobufen versus warfarin in the secondary prevention of major vascular events in nonrheumatic atrial fibrillation. SIFA (Studio Italiano Fibrillazione Atriale) Investigators. Stroke 28:1015–1021PubMedView ArticleGoogle Scholar
  10. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W et al (2011) Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 365:883–891PubMedView ArticleGoogle Scholar
  11. Foerch C, Misselwitz B, Sitzer M, Berger K, Steinmetz H, Neumann-Haefelin T (2005) Difference in recognition of right and left hemispheric stroke. Lancet 366:392–393PubMedView ArticleGoogle Scholar
  12. Stolz E, Hamann GF, Kaps M, Misselwitz B (2011) Regional differences in acute stroke admission and thrombolysis rates in the German federal state of Hesse. Dtsch Arztebl Int 108:607–611PubMed CentralPubMedGoogle Scholar
  13. Jauss M, Schutz HJ, Tanislav C, Misselwitz B, Rosenow F (2010) Effect of daytime, weekday and year of admission on outcome in acute ischaemic stroke patients treated with thrombolytic therapy. Eur J Neurol 17:555–561PubMedView ArticleGoogle Scholar
  14. Tanislav C, Milde S, Schwartzkopff S, Sieweke N, Kramer HH, Juenemann M et al (2014) Secondary stroke prevention in atrial fibrillation: a challenge in the clinical practice. BMC Neurol 14:195. doi:10.1186/s12883-014-0195-y.:95-0195 PubMed CentralPubMedView ArticleGoogle Scholar
  15. Schmidt R, Ropele S, Ferro J, Madureira S, Verdelho A, Petrovic K et al (2010) Diffusion-weighted imaging and cognition in the leukoariosis and disability in the elderly study. Stroke 41:e402–e408PubMedView ArticleGoogle Scholar
  16. Kalra L, Yu G, Perez I, Lakhani A, Donaldson N (2000) Prospective cohort study to determine if trial efficacy of anticoagulation for stroke prevention in atrial fibrillation translates into clinical effectiveness. BMJ 320:1236–1239PubMed CentralPubMedView ArticleGoogle Scholar
  17. Darkow T, Vanderplas AM, Lew KH, Kim J, Hauch O (2005) Treatment patterns and real-world effectiveness of warfarin in nonvalvular atrial fibrillation within a managed care system. Curr Med Res Opin 21:1583–1594PubMedView ArticleGoogle Scholar
  18. Fang MC, Chang Y, Hylek EM, Rosand J, Greenberg SM, Go AS et al (2004) Advanced age, anticoagulation intensity, and risk for intracranial hemorrhage among patients taking warfarin for atrial fibrillation. Ann Intern Med 141:745–752PubMedView ArticleGoogle Scholar
  19. Hylek EM, Evans-Molina C, Shea C, Henault LE, Regan S (2007) Major hemorrhage and tolerability of warfarin in the first year of therapy among elderly patients with atrial fibrillation. Circulation 115:2689–2696PubMedView ArticleGoogle Scholar
  20. Jeerakathil T, Wolf PA, Beiser A, Hald JK, Au R, Kase CS et al (2004) Cerebral microbleeds: prevalence and associations with cardiovascular risk factors in the Framingham Study. Stroke 35:1831–1835PubMedView ArticleGoogle Scholar
  21. Connolly SJ, Eikelboom J, Joyner C, Diener HC, Hart R, Golitsyn S et al (2011) Apixaban in patients with atrial fibrillation. N Engl J Med 364:806–817PubMedView ArticleGoogle Scholar
  22. Diener HC, Eikelboom J, Connolly SJ, Joyner CD, Hart RG, Lip GY et al (2012) Apixaban versus aspirin in patients with atrial fibrillation and previous stroke or transient ischaemic attack: a predefined subgroup analysis from AVERROES, a randomised trial. Lancet Neurol 11:225–231PubMedView ArticleGoogle Scholar
  23. Palm F, Kleemann T, Dos SM, Urbanek C, Buggle F, Safer A et al (2012) Stroke due to atrial fibrillation in a population-based stroke registry (Ludwigshafen Stroke Study) CHADS(2), CHA(2) DS(2)-VASc score, underuse of oral anticoagulation, and implications for preventive measures. Eur J Neurol 20:117–123PubMedView ArticleGoogle Scholar
  24. Giralt-Steinhauer E, Cuadrado-Godia E, Ois A, Jimenez-Conde J, Rodriguez-Campello A, Soriano C et al (2013) Comparison between CHADS2 and CHA2 DS2-VASc score in a stroke cohort with atrial fibrillation. Eur J Neurol 20:623–628PubMedView ArticleGoogle Scholar
  25. Hylek EM, Evans-Molina C, Shea C, Henault LE, Regan S (2007) Major hemorrhage and tolerability of warfarin in the first year of therapy among elderly patients with atrial fibrillation. Circulation 115:2689–2696PubMedView ArticleGoogle Scholar

Copyright

© Tanislav et al. 2015

Advertisement