You are on page 1of 7

Association Between Anemia and Cerebral

Venous Thrombosis
Case–Control Study
Jonathan M. Coutinho, MD, PhD; Susanna M. Zuurbier, MD; Aafke E. Gaartman, BSc;
Arienne A. Dikstaal; Jan Stam, MD, PhD; Saskia Middeldorp, MD, PhD;
Suzanne C. Cannegieter, MD, PhD

Background and Purpose—Anemia is often considered to be a risk factor for cerebral venous thrombosis (CVT),
but this assumption is mostly based on case reports. We investigated the association between anemia and CVT in a
controlled study.
Methods—Unmatched case–control study: cases were adult patients with CVT included in a single-center, prospective
database between July 2006 and December 2014. Controls were subjects from the control population of the Multiple
Downloaded from http://stroke.ahajournals.org/ by guest on September 17, 2017

Environmental and Genetic Assessment of Risk Factors for Venous Thrombosis (MEGA) study. Anemia was defined
according to World Health Organization criteria: nonpregnant women hemoglobin <7.5 mmol/L, pregnant women
<6.9 mmol/L, and men <8.1 mmol/L. We used logistic regression analysis, adjusting for age, sex, malignancy, oral
contraceptive use, and pregnancy/puerperium.
Results—We included 152 cases and 2916 controls. Patients with CVT were younger (mean age, 40 versus 48 years) and
more often women (74% versus 53%) than controls. Anemia was more frequent in cases (27.0%) than in controls (6.5%;
P<0.001). Anemia was associated with CVT, both in univariate analysis (odds ratio, 5.3; 95% confidence interval [CI],
3.6–7.9) and after adjustment for potential confounders (adjusted odds ratio, 4.4; 95% CI, 2.8–6.9). Hemoglobin as a
continuous variable was inversely associated with CVT (adjusted odds ratio per 1 mmol/L change 0.53; 95% CI, 0.42–
0.66). Stratification by sex showed a stronger association between anemia and CVT in men (adjusted odds ratio, 9.9; 95%
CI, 4.1–23.8) than in women (3.6; 95% CI, 2.1–6.0).
Conclusion—Our data suggest that anemia is a risk factor for CVT.   (Stroke. 2015;46:2735-2740. DOI: 10.1161/
STROKEAHA.115.009843.)
Key Words: anemia ◼ case-control studies ◼ pregnancy ◼ risk factors ◼ sinus thrombosis, intracranial

C erebral venous thrombosis (CVT) is a rare cause of stroke


with an incidence of ≈1.3 per 100 000 among adults.1,2
Unlike arterial stroke, CVT mainly affects young to middle-
for CVT, it is inconclusive whether anemia is indeed a risk
factor by itself.
Only 1 controlled study has addressed the association
aged adults and children, and >90% are <60 years.3 Various between anemia and CVT in adults.13 After adjustment for
risk factors for CVT have been identified, such as local infec- potential confounders, mild anemia was not associated with
tions of the head, acquired and genetic thrombophilia, and CVT. The authors did find an association between severe ane-
cancer, especially hematologic malignancies. Because of mia and CVT, but it is was weak (odds ratio [OR], 1.1; 95%
women-specific risk factors (oral contraceptives, pregnancy, confidence interval [CI], 1.01 to 2.22). The aim of our study
and puerperium), CVT is 3× more common among women was to examine the association between anemia and CVT in a
than among men.4 large case–control study.
Anemia is also often considered to be a risk factor for
CVT,5–7 but the evidence for this assumption is mainly based Methods
on case reports and case series.8,9 In cohort studies, anemia is
indeed prevalent in patients who present with CVT, with esti-
Study Design and Patient Selection
Cases were adult patients with CVT who were included in a pro-
mates between 7% and 18% in adults10,11 and even higher per- spective database of consecutive patients admitted to the Academic
centages in children.12 However, because anemia could also Medical Center between July 2006 and December 2014.14 As con-
be associated with underlying diseases that increase the risk trols, we assessed subjects who were included in the control group of

Received April 22, 2015; final revision received July 10, 2015; accepted July 15, 2015.
From the Departments of Neurology (J.M.C., S.M.Z., A.E.G., A.A.D., J.S.) and Vascular Medicine (S.M.), Academic Medical Center, Amsterdam, The
Netherlands; and Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The Netherlands (S.C.C.).
Correspondence to Jonathan M. Coutinho, MD, PhD, Meibergdreef 9, 1105 AZ, Amsterdam, The Netherlands. E-mail j.coutinho@amc.nl
© 2015 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.115.009843

2735
2736  Stroke  October 2015

the Dutch Multiple Environmental and Genetic Assessment of Risk diagnosis of inflammatory bowel disease. We could not adjust for
Factors for Venous Thrombosis (MEGA) study. The MEGA study is these potential confounders in the model because these conditions
a case–control study performed in the Netherlands that included 4956 were not scored in the controls.
consecutive patients with a first deep-vein thrombosis of the leg or
pulmonary embolism between March 1999 and September 2004.15
Partners of patients were invited as control subjects and 3297 par- Results
ticipated. An additional 3000 controls were recruited by random digit Within the study period, 156 adult patients with CVT were
dialing. These participants were between the ages of 18 and 70 years admitted to our hospital. A total of 6297 control subjects par-
and had no history of venous thrombosis. ticipated in the MEGA study. After exclusion of subjects with
We excluded 4 cases and 3381 controls for whom no hemoglo- missing baseline hemoglobin, 152 cases and 2916 controls
bin concentration was available. The excluded control subjects were
mostly recruited after June 2002, when, for logistic reasons; blood were included in the analysis.
sampling was no longer performed in the MEGA study. In cases, di- Cases were younger (mean age, 40 versus 48 years), more
agnosis of CVT was confirmed with computed tomography-venog- often women (74.3 versus 52.6%), and more often had been
raphy, magnetic resonance imaging/magnetic resonance-venography, diagnosed with cancer (9.2 versus 3.7%) compared with con-
catheter angiography, or autopsy. Written informed consent was ob- trols (Table 1). Oral contraceptive use (69.7% versus 21.1%)
tained from cases if plasma and DNA of the patient were stored. In
patients for whom only clinical data were stored, informed consent and pregnancy/puerperium (5.3% versus 1.4%) were more
was not obtained because this is not required under Dutch law. All frequent in female cases. The baseline clinical manifestations
controls from the MEGA study provided written informed consent. of patients with CVT are provided in Table 1.
Mean hemoglobin concentration was lower in cases than
Downloaded from http://stroke.ahajournals.org/ by guest on September 17, 2017

Data Collection and Laboratory Measurements in controls (8.06 versus 8.68 mmol/L; P<0.001; Table 2).
Data on clinical manifestations, imaging, and outcome were recorded Anemia was present in 27.0% (41/152) of the cases, signifi-
with a standardized case report form for cases. We extracted hemo- cantly more frequently than in controls (6.5%; 189/2916; abso-
globin concentration and mean corpuscular volume (MCV) from the lute difference, 20.5%; 95% CI, 14 to 28). Severe anemia was
medical records. All laboratory measurements were done as a part of present in 2.6% of cases versus none in the controls (absolute
routine diagnostic work-up within 48 hours of admission. For labo-
ratory measurements, a Sysmex XE-5000 Automated Hematology
difference, 2.6%; 95% CI, 1.0 to 6.6). Increased hemoglobin
System (maximal imprecision ≤1.5%, inaccuracy, ±2%) was used for concentration was rare in both cases and controls (0.7 versus
cases, and a Coulter MD-II Series Analyzer (maximal imprecision 0.2%; absolute difference, 0.5%; 95% CI, −0.8 to 1.8).
≤2%, inaccuracy, ±2%) for controls. After adjustment for potential confounders, anemia was
Control subjects in the MEGA study completed a questionnaire significantly associated with CVT (adjusted OR, 4.4; 95%
with information on demographics, previous medical history, and
risk factors for thrombosis. Blood was collected into vacuum tubes
CI, 2.8 to 6.9; Table 3). Similarly, hemoglobin concentration
containing 0.106 mmol/L trisodium citrate and processed within 4 was inversely associated with CVT risk (adjusted OR per 1
hours. The final concentration of hemoglobin was multiplied by a mmol/L increase, 0.53; 95% CI, 0.42 to 0.66). Stratification
factor of 1.1 to adjust for the dilution by the trisodium citrate, as done by sex showed a stronger association between anemia and
previously.16,17 CVT in men (adjusted OR, 9.9; 95% CI, 4.1–23.8) than in
women (adjusted OR, 3.6; 95% CI, 2.1 to 6.0). Stratification
Definition of Anemia by MCV indicated that the risk of CVT was most clearly
We used the World Health Organization definitions for anemia: non-
pregnant women hemoglobin <7.5 mmol/L, pregnant women <6.9
mmol/L, and men <8.1 mmol/L (http://www.who.int/vmnis/indi- Table 1.  Baseline Characteristics and Risk Factors
cators/hemoglobin.pdf). Severe anemia is defined as hemoglobin
<5.0 mmol/L for men and nonpregnant women or hemoglobin <4.3 Cases (n=152) Controls (n=2916)
mmol/L for pregnant women. Anemia is categorized as microcytic Demographics
(MCV <80 fL), normocytic (MCV 80–100 fL), or macrocytic anemia
(MCV >100 fL).  Mean age, y (SD) 40 (14) 48 (12)
 Women, % 74.3 52.6
Data Analysis Risk factors for thrombosis, %
We analyzed categorical data with the χ2 test or Fisher exact test and  Oral contraceptives* 69.7 21.1
continuous data with a Mann–Whitney test or Student t test, which-  Pregnancy/puerperium* 5.3 1.4
ever was appropriate. The primary analysis was the difference in fre-
quency of anemia between cases and controls. Results are given as  Previous thrombosis 9.2 …
percentages with absolute differences and 95% CIs. We used multi-  Malignancy 9.2 3.7
variate logistic regression analysis to adjust for potential confound- Baseline characteristics CVT patients, %
ing variables. CVT was used as a dependent variable. Anemia and
hemoglobin concentrations (as a continuous variable) were used as  Headache 91.3 …
independent variables in separate models. In each model, we adjusted  Focal neurological deficits 63.3 …
for the following potential confounders: age, sex, malignancy, oral  Papilledema 25.7 …
contraceptive use, and pregnancy/puerperium. We stratified the anal-
ysis for sex and MCV value. We also performed an analysis in which  Seizures 46.1 …
we divided subjects into those with normal hemoglobin (reference  Intracranial hemorrhage 48.4 …
category), anemia, or increased hemoglobin. Increased hemoglobin
 Intracranial nonhemorrhagic 22.0 …
was defined as >10.2 mmol/L for women or >11.5 mmol/L for men.
lesion
In a subgroup analysis, we excluded patients with a history of malig-
nancy. We also did a subgroup analysis in which we excluded cases CVT indicates cerebral venous thrombosis.
with a recent infection or neurosurgical intervention, or who had a *Percentage of women.
Coutinho et al   Anemia and Cerebral Venous Thrombosis    2737

Table 2.  Laboratory Data CVT did not materially change (adjusted OR, 3.6; 95% CI,
2.2–5.9). Finally, we divided subjects into those with normal
Cases Controls Absolute
(n=152) (n=2916) Difference, % (95% CI) hemoglobin (reference category), anemia, or increased hemo-
globin. This stratification had no effect on the strength of the
Mean hemoglobin, 8.06 (1.4) 8.68 (0.7) 0.63 (0.40 to 0.85)
mmol/L (SD)
association between CVT and anemia (adjusted OR, 4.5; 95%
CI, 2.9 to 7.0). Furthermore, it did suggest that increased
Anemia 27.0% 6.5% 20.5 (14 to 28)
hemoglobin was also associated with CVT (adjusted OR, 9.6;
 Microcytic 10.5% 0.7% 9.8 (5.8 to 15.9) 95% CI, 1.0 to 90.7).
 Normocytic 10.5% 5.7% 4.8 (0.7 to 11.0)
 Macrocytic 1.4% 0.2% 1.2 (0.2 to 4.8)
Discussion
 Missing MCV 4.6% 0% In this case–control study, we found that CVT was sig-
Severe anemia 2.6% 0% 2.6 (1.0 to 6.6) nificantly associated with anemia. Accordingly, we found
Increased 0.7% 0.2% 0.5 (−0.8 to 1.8) a linear, inverse association between the risk of CVT and
hemoglobin* hemoglobin concentration. Subgroup analysis showed that
CI indicates confidence interval; and MCV, mean corpuscular volume. the association was stronger in men and in patients with
*Defined as >10.2 mmol/L for women or >11.5 mmol/L for men. microcytic anemia.
With a prevalence of 27%, the frequency of anemia
Downloaded from http://stroke.ahajournals.org/ by guest on September 17, 2017

increased in patients with microcytic anemia (adjusted OR, among patients with CVT in our study is higher than has been
15.6; 95% CI, 6.7 to 36.2). Exclusion of patients with a recent reported before. In the International Study on Cerebral Vein
infection, neurosurgical intervention, or inflammatory bowel and Dural Sinus Thrombosis (ISCVT), for instance, anemia
disease essentially did not change the results (adjusted OR, was present in 9.2% of patients.11 This discrepancy might be
3.9; 95% CI, 2.4 to 6.4). When we excluded patients with a explained by the fact that the prevalence of anemia was not a
history of malignancy, the association between anemia and specific research objective in ISCVT, and that measurement

Table 3.  Association Between Anemia and Cerebral Venous Thrombosis


Cases Controls OR (95% CI)
n % n % Unadjusted Adjusted
Entire study population
 Anemia 41 27.0 189 6.5% 5.3 (3.6–7.9) 4.4 (2.8–6.9)
 1 mmol/L increase in Hb* … … … … 0.42 (0.35–0.51) 0.53 (0.42–0.66)
Stratification by sex
 Men: anemia 10 25.6 46 3.3 10.0 (4.6–21.8) 9.9 (4.1–23.8)
 Women: anemia 31 27.4 143 9.3 3.7 (2.3–5.8) 3.6 (2.1–6.0)
 Men: increase in Hb* … … … … 0.44 (0.30–0.65) 0.49 (0.33–0.74)
 Women: Increase in Hb* … … … … 0.44 (0.34–0.57) 0.54 (0.40–0.72)
Stratification by MCV
 Microcytic anemia 16 10.5 19 0.7 19.4 (9.6–39.2) 15.6 (6.7–36.2)
 Normocytic anemia 16 10.5 165 5.7 2.2 (1.3–3.9) 1.9 (1.0–3.5)
 Macrocytic anemia 2 1.4 5 0.2 9.8 (1.9–51.2) 8.9 (0.93–85.6)
Potential residual confounders excluded†
 Anemia 33 25.8 189 6.5 5.0 (3.3–7.6) 3.9 (2.4–6.4)
 1 mmol/L increase in Hb* … … … … 0.44 (0.36–0.54) 0.59 (0.46–0.77)
Subjects with malignancy excluded
 Anemia 30 21.7 179 6.4 4.1 (2.6–6.3) 3.6 (2.2–5.9)
 1 mmol/L increase in Hb* … … … … 0.53 (0.44–0.65) 0.68 (0.54–0.85)
Stratification by decreased, normal, or increased hemoglobin
 Anemia 41 27.0 189 6.5 5.4 (3.6–7.9) 4.5 (2.9–7.0)
 Normal hemoglobin 110 72.4 2722 93.3 1.0‡ 1.0‡
 Increased hemoglobin§ 1 0.7 5 0.2 4.9 (0.57–42.7) 9.6 (1.0–90.7)
Unadjusted and adjusted OR for cases vs controls. In the multivariate model, we adjusted for age, sex, malignancy, oral contraceptive use, and pregnancy/puerperium.
CI indicates confidence interval; Hb, hemoglobin; MCV, mean corpuscular volume; and OR, odds ratio.
*OR for Hb is per 1 mmol/L increase in Hb.
†Cases with a recent infection or neurosurgical intervention, or with a history of inflammatory bowel disease were excluded in this subgroup analysis.
‡Reference category.
§Defined as Hb >10.2 mmol/L for women or Hb >11.5 mmol/L for men.
2738  Stroke  October 2015

of hemoglobin was not mandatory. This may have led to any of these conditions. Although attenuated, the association
under-reporting of anemia, especially of milder cases. A large between CVT and anemia remained clinically relevant and
registry from India where hemoglobin concentrations were statistically significant in this analysis. Nevertheless, we can-
routinely measured and which used a similar definition for not exclude the possibility that other residual confounders had
anemia as we did reported anemia in 18.4% of patients.10 an effect on the analysis.
We found a stronger association between CVT and anemia If a causal relation exists between anemia and CVT, it is
than a previous case–control study, despite a similar preva- interesting to speculate about the pathogenesis. Anemia, espe-
lence of anemia among cases.13 Stolz et al13 reported only a cially when caused by iron deficiency, can sometimes lead to
weak association with severe anemia (OR, 1.1) and no associ- thrombocytosis, which is a risk factor for venous thrombo-
ation with mild anemia. This difference is partly explained by sis.25,26 This mechanism would be in line with the fact that we
the fact that in their study, the authors calculated separate ORs observed a stronger association between CVT and microcytic
for mild and severe anemia, whereas in our main analysis, we anemia because iron deficiency is the most important cause
determined the OR of any anemia (mild or severe) versus no of microcytic anemia. Unfortunately, thrombocytes were not
anemia. Using our approach, the unadjusted OR for any ane- measured in the MEGA study, and thus we could not exam-
mia in the study by Stolz et al13 would have been 2.8, which ine whether the presence of thrombocytosis modified the
is in the same range as the unadjusted OR in our study (5.3). association between anemia and CVT. Iron deficiency has
It is important to compare the prevalence of anemia in also been associated with an increased concentration of fac-
our control population to other studies that included healthy tor VIII, which is also a risk factor for thrombosis.27 On the
Downloaded from http://stroke.ahajournals.org/ by guest on September 17, 2017

subjects. The US National Health and Nutrition Examination contrary, if either of these hypotheses were correct, anemia
Survey determined the prevalence of anemia in noninstitu- should probably also increase the risk of venous thrombo-
tionalized civilians.18 The prevalence in men aged 17 to 64 embolism (deep-vein thrombosis of the leg and pulmonary
years, comparable with the age range of our control group, embolism). Although there is some evidence supporting an
varied between 1.5% and 4.4%. The frequency of anemia association between venous thromboembolism and anemia,28
among male controls in our study (3.3%) falls within this esti- anemia is generally not considered to be a risk factor for this
mate. In women of similar ages, the prevalence was 6.8% to condition.29,30
12.2%, which is also comparable with the frequency we found It is uncertain if the results of our study can be extrapo-
(9.3%). A population-based study from Germany found a rate lated to pediatric CVT. Children with CVT have other risk fac-
of anemia of 3.2% in men and women aged 45 to 74 years,19 tors than adults, with an over-representation of infections and
which is even lower than in our study. Therefore, we think dehydration.31 A previous study indicated that there may also
that it is unlikely that an underestimation of the prevalence of be an association between iron deficiency anemia and CVT in
anemia among controls biased our results. children, but this study included only 6 patients with CVT.32
Our data indicate that anemia is a stronger risk factor for Our data also suggest that increased hemoglobin is associated
CVT in men than in women. The most likely explanation for with CVT. Causes of increased hemoglobin include dehydra-
this discrepancy is the difference in absolute incidence of tion and polycythemia vera, both of which have been reported
CVT, which is ≈2.5× higher in women than in men.1 If ane- in patients with CVT.31,33 However, the degree of uncertainty
mia adds a similar absolute risk for CVT to the baseline risk of this observation is high because there was only 1 case with
in both sexes, the relative risk associated with anemia will increased hemoglobin, and a larger sample size is required to
be higher in the group with a lower absolute baseline inci- confirm this finding.
dence (ie, men).20 Another contributing factor could be that Several limitations of our study warrant comment. First,
men with CVT more often have an infection or a recent opera- although all cases were enrolled in a prospective study, the
tion as a risk factor, and these conditions are also associated association between anemia and CVT was not a predefined
with anemia.4 Finally, a lower risk of CVT among women research question. As a result, hemoglobin concentrations
with anemia might be related to the high prevalence of female were missing in a small proportion (3%) of cases. Also,
patients who used oral contraceptives. Oral contraceptives are laboratory data that would have been helpful in determining
a well-known risk factor for CVT.21 Some studies have found the cause of anemia, such as ferritin, folic acid, and vitamin
that these drugs also decrease the risk of anemia, which could B12, were not available. However, we were able to catego-
result in a lower prevalence of anemia among women with rize anemia according to MCV values, which provides a crude
CVT.22,23 However, we adjusted for oral contraceptive use in indication of the underlying cause. Second, cases and con-
our analysis, which makes this last explanation less plausible. trols were not recruited in the same time period. The MEGA
Although our study shows an association between ane- study recruited control individuals between 1999 and 2004,
mia and CVT, this finding does not necessarily imply a causal whereas the CVT cases were collected from 2006 to 2014.
relationship. Anemia could be an epiphenomenon of another If the prevalence of anemia increased in those years, it could
underlying prothrombotic condition for which we were unable have biased the results. However, it seems unlikely that the
to adjust. Inflammatory bowel disease, infections, and neuro- extent of such a change in prevalence in such a short time
surgery, for instance, are all established risk factors for both period could be large enough to fully explain our findings. In
CVT and anemia.11,24 Because information on these condi- fact, there is evidence that the prevalence of anemia among
tions was not available for controls, we could not include healthy women has decreased in the past decades.34 Third, it
them in the multivariate model. To account for this limitation, would have been interesting to examine whether the associa-
we did a subgroup analysis in which we excluded cases with tion between anemia and CVT was influenced by the presence
Coutinho et al   Anemia and Cerebral Venous Thrombosis    2739

of thrombophilia. Unfortunately, this analysis was not possi- 12. Sébire G, Tabarki B, Saunders DE, Leroy I, Liesner R, Saint-Martin
C, et al. Cerebral venous sinus thrombosis in children: risk factors,
ble because routine testing for thrombophilia was not done in
presentation, diagnosis and outcome. Brain. 2005;128(pt 3):477–489.
cases. Finally, we were unable to calculate an adjusted OR for doi: 10.1093/brain/awh412.
severe anemia because none of the controls had severe anemia. 13. Stolz E, Valdueza JM, Grebe M, Schlachetzki F, Schmitt E, Madlener
The fact that 4 of the 152 cases had severe anemia, compared K, et al. Anemia as a risk factor for cerebral venous thrombosis? An
old hypothesis revisited. Results of a prospective study. J Neurol.
with zero of the ≈3000 controls, indicates an association, but 2007;254:729–734. doi: 10.1007/s00415-006-0411-9.
severe anemia may have been under-represented in controls. 14. Coutinho JM, van den Berg R, Zuurbier SM, VanBavel E, Troost D,
Severe anemia often leads to symptoms, and this may lead Majoie CB, et al. Small juxtacortical hemorrhages in cerebral venous
to reluctance to participate in a medical research study as a thrombosis. Ann Neurol. 2014;75:908–916. doi: 10.1002/ana.24180.
15. Blom JW, Doggen CJ, Osanto S, Rosendaal FR. Malignancies, pro-
control subject. Still, even if severe anemia is a risk factor for thrombotic mutations, and the risk of venous thrombosis. JAMA.
CVT, our data suggest that it is not a common risk factor. 2005;293:715–722. doi: 10.1001/jama.293.6.715.
In summary, we found that anemia is associated with the 16. Rezende SM, Lijfering WM, Rosendaal FR, Cannegieter SC.
Hematologic variables and venous thrombosis: red cell distri-
occurrence of CVT. Anemia is a risk factor for both sexes, but bution width and blood monocyte count are associated with an
the association was stronger in men than in women. increased risk. Haematologica. 2014;99:194–200. doi: 10.3324/
haematol.2013.083840.
17. Stuijver DJ, Debeij J, van Zaane B, Dekkers OM, Smit JW, Büller
Sources of Funding HR, et al. Levels of prolactin in relation to coagulation factors and
Dr Coutinho was supported by The Netherlands Organisation for risk of venous thrombosis. Results of a large population-based case-
Scientific Research, grant number 021.001.045 and the Dutch
Downloaded from http://stroke.ahajournals.org/ by guest on September 17, 2017

control study (MEGA-study). Thromb Haemost. 2012;108:499–507.


Thrombosis Society, grant number 2012-2. doi: 10.1160/TH12-04-0225.
18. Guralnik JM, Eisenstaedt RS, Ferrucci L, Klein HG, Woodman RC.
Prevalence of anemia in persons 65 years and older in the United States:
Disclosures evidence for a high rate of unexplained anemia. Blood. 2004;104:2263–
Dr Coutinho received a lecturing fee from Boehringer Ingelheim, 2268. doi: 10.1182/blood-2004-05-1812.
which was donated to the Stichting Klinische Neurologie, a local 19. Eisele L, Dürig J, Broecker-Preuss M, Dührsen U, Bokhof B, Erbel R,
foundation that supports research in the field of neurological disor- et al; Heinz Nixdorf Recall Study Investigative Group. Prevalence and
ders. The other authors report no conflicts. incidence of anemia in the German Heinz Nixdorf Recall Study. Ann
Hematol. 2013;92:731–737. doi: 10.1007/s00277-013-1697-1.
20. van Langevelde K, Flinterman LE, van Hylckama Vlieg A, Rosendaal
References FR, Cannegieter SC. Broadening the factor V Leiden paradox: pulmo-
1. Coutinho JM, Zuurbier SM, Aramideh M, Stam J. The incidence of cere- nary embolism and deep-vein thrombosis as 2 sides of the spectrum.
bral venous thrombosis: a cross-sectional study. Stroke. 2012;43:3375– Blood. 2012;120:933–946. doi: 10.1182/blood-2012-02-407551.
3377. doi: 10.1161/STROKEAHA.112.671453. 21. Martinelli I, Sacchi E, Landi G, Taioli E, Duca F, Mannucci PM. High
2. Bousser MG, Ferro JM. Cerebral venous thrombosis: an update. Lancet risk of cerebral-vein thrombosis in carriers of a prothrombin-gene muta-
Neurol. 2007;6:162–170. doi: 10.1016/S1474-4422(07)70029-7. tion and in users of oral contraceptives. N Engl J Med. 1998;338:1793–
3. Ferro JM, Canhão P, Bousser MG, Stam J, Barinagarrementeria 1797. doi: 10.1056/NEJM199806183382502.
F; ISCVT Investigators. Cerebral vein and dural sinus thrombo- 22. Mooij PN, Thomas CM, Doesburg WH, Eskes TK. The effects of
sis in elderly patients. Stroke. 2005;36:1927–1932. doi: 10.1161/01. oral contraceptives and multivitamin supplementation on serum ferri-
STR.0000177894.05495.54. tin and hematological parameters. Int J Clin Pharmacol Ther Toxicol.
4. Coutinho JM, Ferro JM, Canhão P, Barinagarrementeria F, Cantú 1992;30:57–62.
C, Bousser MG, et al. Cerebral venous and sinus thrombosis in 23. Grace E, Emans SJ, Drum DE. Hematologic abnormalities in adoles-
women. Stroke. 2009;40:2356–2361. doi: 10.1161/STROKEAHA. cents who take oral contraceptive pills. J Pediatr. 1982;101:771–774.
108.543884. 24. Cognat E, Crassard I, Denier C, Vahedi K, Bousser MG. Cerebral
5. Stam J. Thrombosis of the cerebral veins and sinuses. N Engl J Med. venous thrombosis in inflammatory bowel diseases: eight cases and lit-
2005;352:1791–1798. doi: 10.1056/NEJMra042354. erature review. Int J Stroke. 2011;6:487–492. doi: 10.1111/j.1747-4949.
6. Saposnik G, Barinagarrementeria F, Brown RD Jr, Bushnell CD, 2011.00620.x.
Cucchiara B, Cushman M, et al. American Heart Association Stroke 25. Park MJ, Park PW, Seo YH, Kim KH, Park SH, Jeong JH, et al. The rela-
Council and the Council on Epidemiology and Prevention. Diagnosis tionship between iron parameters and platelet parameters in women with
and management of cerebral venous thrombosis: a statement for health- iron deficiency anemia and thrombocytosis. Platelets. 2013;24:348–351.
care professionals from the American Heart Association/American doi: 10.3109/09537104.2012.699641.
Stroke Association. Stroke. 2011;42:1158–1192. doi: 10.1161/ 26. Ho KM, Yip CB, Duff O. Reactive thrombocytosis and risk of subse-
STR.0b013e31820a8364. quent venous thromboembolism: a cohort study. J Thromb Haemost.
7. Saadatnia M, Fatehi F, Basiri K, Mousavi SA, Mehr GK. Cerebral 2012;10:1768–1774. doi: 10.1111/j.1538-7836.2012.04846.x.
venous sinus thrombosis risk factors. Int J Stroke. 2009;4:111–123. doi: 27. Livesey JA, Manning RA, Meek JH, Jackson JE, Kulinskaya E, Laffan
10.1111/j.1747-4949.2009.00260.x. MA, et al. Low serum iron levels are associated with elevated plasma
8. Balci K, Utku U, Asil T, Büyükkoyuncu N. Deep cerebral vein throm- levels of coagulation factor VIII and pulmonary emboli/deep venous
bosis associated with iron deficiency anaemia in adults. J Clin Neurosci. thromboses in replicate cohorts of patients with hereditary haem-
2007;14:181–184. doi: 10.1016/j.jocn.2005.09.020. orrhagic telangiectasia. Thorax. 2012;67:328–333. doi: 10.1136/
9. Kinoshita Y, Taniura S, Shishido H, Nojima T, Kamitani H, Watanebe T. thoraxjnl-2011-201076.
Cerebral venous sinus thrombosis associated with iron deficiency: two 28. Hung SH, Lin HC, Chung SD. Association between venous throm-
case reports. Neurol Med Chir (Tokyo). 2006;46:589–593. boembolism and iron-deficiency anemia: a population-based study.
10. Narayan D, Kaul S, Ravishankar K, Suryaprabha T, Bandaru VC, Blood Coagul Fibrinolysis. 2015;26:368–372. doi: 10.1097/MBC.
Mridula KR, et al. Risk factors, clinical profile, and long-term outcome 0000000000000249.
of 428 patients of cerebral sinus venous thrombosis: insights from 29. Anderson FA Jr, Spencer FA. Risk factors for venous thromboem-
Nizam’s Institute Venous Stroke Registry, Hyderabad (India). Neurol bolism. Circulation. 2003;107(23 suppl 1):I9–I16. doi: 10.1161/01.
India. 2012;60:154–159. doi: 10.4103/0028-3886.96388. CIR.0000078469.07362.E6.
11. Ferro JM, Canhão P, Stam J, Bousser MG, Barinagarrementeria F; 30. Holst AG, Jensen G, Prescott E. Risk factors for venous thromboem-
ISCVT Investigators. Prognosis of cerebral vein and dural sinus throm- bolism: results from the Copenhagen City Heart Study. Circulation.
bosis: results of the International Study on Cerebral Vein and Dural 2010;121:1896–1903. doi: 10.1161/CIRCULATIONAHA.109.921460.
Sinus Thrombosis (ISCVT). Stroke. 2004;35:664–670. doi: 10.1161/01. 31. deVeber G, Andrew M, Adams C, Bjornson B, Booth F, Buckley DJ, et al;
STR.0000117571.76197.26. Canadian Pediatric Ischemic Stroke Study Group. Cerebral sinovenous
2740  Stroke  October 2015

thrombosis in children. N Engl J Med. 2001;345:417–423. doi: 10.1056/ splanchnic or cerebral venous thrombosis and without overt chronic
NEJM200108093450604. myeloproliferative disorders. J Thromb Haemost. 2007;5:708–714.
32. Maguire JL, deVeber G, Parkin PC. Association between iron-deficiency doi: 10.1111/j.1538-7836.2007.02424.x.
anemia and stroke in young children. Pediatrics. 2007;120:1053–1057. 34. Cusick SE, Mei Z, Freedman DS, Looker AC, Ogden CL, Gunter E,
doi: 10.1542/peds.2007-0502. et al. Unexplained decline in the prevalence of anemia among US
33. De Stefano V, Fiorini A, Rossi E, Za T, Farina G, Chiusolo P, children and women between 1988-1994 and 1999-2002. Am J Clin
et al. Incidence of the JAK2 V617F mutation among patients with Nutr. 2008;88:1611–1617. doi: 10.3945/ajcn.2008.25926.
Downloaded from http://stroke.ahajournals.org/ by guest on September 17, 2017
Association Between Anemia and Cerebral Venous Thrombosis: Case−Control Study
Jonathan M. Coutinho, Susanna M. Zuurbier, Aafke E. Gaartman, Arienne A. Dikstaal, Jan
Stam, Saskia Middeldorp and Suzanne C. Cannegieter
Downloaded from http://stroke.ahajournals.org/ by guest on September 17, 2017

Stroke. 2015;46:2735-2740; originally published online August 13, 2015;


doi: 10.1161/STROKEAHA.115.009843
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2015 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/46/10/2735

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:


http://www.lww.com/reprints

Subscriptions: Information about subscribing to Stroke is online at:


http://stroke.ahajournals.org//subscriptions/

You might also like