Microcytic or Iron Deficiency Anemia as a Risk Factor for Cerebral Sinus Venous Thrombosis—Systematic Review and Exploratory Meta-analysis

Abstract

Background Cerebral sinus venous thrombosis (CSVT) is a rare but severe neurological condition. Iron deficiency anemia (IDA) has been suggested as a risk factor. Recent experimental data in iron-deficient rodents show a potential pathophysiological link between IDA and increased thrombus formation, resulting from iron-restricted erythropoiesis. Given the high prevalence of IDA, understanding its role in CSVT is of major clinical interest.

Methods We conducted a systematic review and exploratory meta-analysis, searching PubMed and Embase up to January 30, 2026, for original English-language articles investigating iron deficiency (ID), anemia, or IDA as risk factors for CSVT. The review followed PRISMA guidelines and assessed evidence quality using the Cochrane GRADE approach. The study was registered on PROSPERO (CRD42020205229).

Results A total of 62 studies comprising 2,877 patients met inclusion criteria, including four case–control studies, 18 case series, and 40 case reports. Reported odds ratios ranged from 1.1 to 4.4 in adults and 3.8 to 11.4 in pediatric studies. Exploratory meta-analysis yielded pooled ORs of 2.55 (95% CI: 0.89–7.33) in adults and 4.39 (95% CI: 1.16–16.57) in children. Both estimates were based on only two studies each, with limited sample sizes and substantial uncertainty. The pediatric estimate warrants particular caution, as some studies combined CSVT with other cerebrovascular outcomes. Overall, the included studies suggested an association between anemia and CSVT, although the certainty of evidence was low.

Conclusion Current evidence suggests a possible association of severe or microcytic anemia, indicating IDA, and an increased CSVT risk. However, available data are limited, heterogeneous, and largely observational, precluding conclusions on causality or precise effect sizes. These findings support heightened clinical awareness of IDA in CSVT patients and highlight the need for prospective population-based studies to determine whether IDA prevention or treatment reduces thrombotic risk.

Introduction

Iron deficiency anemia (IDA) is a public health concern, affecting more than 1.2 billion people worldwide. ¹ Although iron deficiency (ID) is the most frequent single cause for anemia and accounts for approximately 50% of cases, ² screening for ID is not conducted according to (inter)national standards. IDA counts among frequently overlooked diagnoses in high-income countries. ³ The negligence in diagnostics contradicts the high clinical relevance of ID for the general population. Neurodevelopmental deficits in children, ^{4 5} increased pregnancy morbidity, ⁶ and reduced cognitive performance in the elderly have been described. ⁷ Repeatedly, IDA has been suggested as a risk factor for cerebral sinus venous thrombosis (CSVT), primarily based on various case reports. CSVT incidence among adults is estimated to be 1.3–1.6 per 100,000/year. ^{8 9} Given its non-specific presentation, CSVT requires a high index of suspicion for diagnosis. ¹⁰ Mortality has been estimated to be around 5 to 10% and long-term sequelae are common. ¹¹ Due to IDA’s frequency and simple therapy by iron substitution or red blood cell transfusion, an association of IDA with CSVT would be of high clinical relevance.

Recently published experimental data provide new evidence for the association of IDA with CSVT by showing that IDA as well as associated reactive thrombocytosis are independent risk factors for venous thrombosis in a rodent model. ¹² The objective of this systematic review is to investigate the current available evidence on IDA as CSVT risk factor in humans.

Methods and Design

Design and Protocol Registration

This systematic review was designed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) 2020 statement. ¹³ The protocol was registered in the PROSPERO International prospective registry for systematic reviews (CRD42020205229).

Review Inclusion and Exclusion Criteria

Two independent researchers (H.K.; A.K. or F.C.), blinded for each other’s results, selected original peer-reviewed publications following a clearly defined search protocol. To not miss relevant publications in the initial screening, we kept the entry definition for IDA as wide as possible. Both prospective and retrospective studies were included without limits concerning study type or publication date. The systematic literature search was conducted in PubMed and Embase. Eligibility was restricted to peer-reviewed articles published in English, reflecting the predominant language of biomedical scientific publishing. Gray literature was not considered. In a second step, we excluded patients in whom anemia was associated with hemoglobinopathy, paroxysmal nocturnal hemoglobinuria, or leukemia.

Search Strategy and Research Screening

We searched PubMed and Embase through January 30, 2026, using MeSH terms: anemia, iron deficiency, sinus venous thrombosis, and cerebral vein thrombosis, combined with Boolean operators “AND” and “OR.” The final search string was: “((anemia OR iron deficiency) AND (sinus venous thrombosis OR cerebral vein thrombosis)).” Two independent reviewers screened titles and abstracts, excluded ineligible studies, and saved relevant full texts for detailed review.

Study Selection and Quality Assessment

Two reviewers (H.K.; A.K. or F.C.) independently screened titles and abstracts of all results. We resolved disagreements through a third reviewer (S.H.). Three independent reviewers (H.K., F.C., and S.H.) assessed the quality of full-length original articles by using the Cochrane Grading of Recommendations, Assessment, Development and Evaluation (GRADE) approach ¹⁴ and the critical appraisal tool checklists from Joanna Briggs Institute. ¹⁵

Data Analysis

We initially sought to combine source data to enable a meta-analysis on patient-level data. Accordingly, we contacted the authors of two case–control studies in adults. ^{16 17} However, the primary datasets were no longer available; one study had been conducted more than a decade earlier. Therefore, a pooled analysis of primary data could not be conducted. We subsequently performed an exploratory meta-analysis of the published data of the case–control studies. We stratified the studies to adult and pediatric studies, based on consistency of their inclusion criteria and used a random effects model with inverse variance weighting. ¹⁸ Statistical significance was assessed using a two-tailed p -value threshold of ≤0.05 and was considered supported when the 95% CI of the pooled OR did not include the neutral value “1.” The Higgins I ² metric was used to evaluate statistical heterogeneity, with 25–50% considered mild, 50–75% moderate, and >75% severe heterogeneity. ¹⁹ We ultimately performed a systematic review to critically appraise and synthesize all available evidence.

Results

Search Results

The initial literature search identified 295 records. After removing duplicates and screening titles and abstracts for inclusion criteria, we included 67 articles on full manuscript assessment. Data extraction and bias assessment were conducted on 62 publications meeting eligibility criteria ( Fig. 1 ). We included four case–control studies ^{16 17 20 21} and 18 case series ^{22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39} for the systematic review ( Table 1 ). A comprehensive list and summary, including 40 case reports, ^{10 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78} is given in Supplementary Tables S1–S3 (online only).

Fig. 1.

PRISMA flowchart of study selection for the systematic review.

Table 1. Case–control studies investigating the association of anemia or iron deficiency anemia (IDA) and cerebral sinus venous thrombosis (CSVT).

Reference	Stolz et al. 16	Coutinho et al. 17	Maguire et al. 20	Azab et al. 21
Study design; recruitment period, country of recruitment a	Prospective 2000–2005 Germany	Prospective 2006–2014 Netherlands	Retrospective 1992–2004 Canada	Retrospective 2009–2013 Egypt
Case definition	Adults with non-infectious CSVT aged 18–85 years	Adults with CSVT mean age 40 years	Children with stroke aged 12–38 months (5 children with CSVT, 9 children with AIS, 1 child with CSVT and AIS)	Children with stroke aged 12–72 months (10 children with CSVT, 11 children with AIS)
Total number of cases	121	152	15	21
Cases with CSVT	121	152	6 (1 of these also with AIS)	10
Matching of controls	Age, sex	Unmatched	Unmatched	Age, sex
Differences in matching parameters between cases and control group	None	CSVT patients younger than controls (mean age 40 vs. 48 years) CSVT patients had more women than in control group (74% vs. 53%)	No differences with respect to age and sex	None
n (controls)	120	2916	143	100
Definition of anemia, severe anemia, IDA	Anemia: Hb < 12 g/dL (females) Hb < 13 g/dL (males) severe anemia: Hb < 9 g/dL (arbitrarily, WHO definition of severe anemia: Hb < 8 g/dL)	Anemia: Hb < 12 g/dL (non-pregnant females) Hb < 11 g/dL (pregnant females) Hb < 13 g/dL (males) severe anemia: Hb < 8.1 g/dL (males and non-pregnant females) Hb < 6.9 g/dL (pregnant females)	IDA: Hb < 11 g/dL MCV < 73 fL serum ferritin <12 μg/L 83 84 85	IDA: Hb: <11 g/dL (6 months–5 years) Hb < 11.5 g/dL (6–11 years) MCV < 82 fL (age >11 years) 86 serum ferritin <12 μg/L
Specification of IDA as inclusion criterion?	No	No	Yes	Yes
Univariate odds ratio (with 95% confidence interval)	Anemia: OR 2.8 (1.4–5.9) b p = 0.005 severe anemia: OR 7.8 (1.7–35.1) p = 0.008	Anemia: OR 5.3 (3.6–7.9) p < 0.0001 severe anemia: not done as none of the control subjects had severe anemia	IDA: OR 11.4 (3.6–36.6) p < 0.0001 b (for CSVT and AIS) OR for CSVT group not calculated as all patients with CSVT had IDA b	IDA: OR 3.8 (1.3–11.2) p = 0.005 (for CSVT and AIS) OR for CSVT group: 6.6 (1.6–27.6) p = 0.0092 b
Multivariate odds ratio	Severe anemia: OR 1.1 (1.0–2.2) p = 0.03 OR 1.5 (1.0–2.2) b	Anemia: OR 4.4 (CI: 2.8–6.9)	IDA: OR 10 (3–33) (for CSVT and AIS)	IDA: OR 2.5 (1.9–2.3) c OR 2.5 (1.9–3.3) c p = 0.036 (for CSVT and AIS)
Confounders adjusted for in multivariate analysis	Pregnancy, malignancy, hypercholesterinemia, contraceptive use, corticosteroids, thrombophilia	Age, sex, malignancy, contraceptive use, pregnancy/puerperium	Platelet count	Thrombocytosis (platelet count >450× 10 9 /L)
Thrombophilia testing association with CSVT?	Factor V Leiden mutation, prothrombin polymorphism, protein S/protein C deficiency, lupus anticoagulant, combined defects multivariate analysis OR 1.22 (1.07–1.77), p = 0.003	Not done	Antithrombin, protein S, protein C deficiency, lupus anticoagulant, factor V Leiden mutation, G20210A prothrombin mutation, anticardiolipin antibodies, none of the CSVT patients had hereditary thrombophilia	Antithrombin, protein S, protein C deficiency, lupus anticoagulant, factor V Leiden mutation, G20210A prothrombin mutation, none of the CSVT patients had hereditary thrombophilia, 5 children with AIS had preceding varicella infection within the past 6 months

Abbreviations: AIS, acute ischemic stroke; CI, confidence interval; Hb, hemoglobin; OR, odds ratio.

Notes: ^a No information on ethnicity of patients provided.

^b Numbers in italics were not provided by the authors of the studies and were calculated by the authors of this review.

^c Upper limit of CI provided by the authors of the study is most likely a typo.

Case–Control Studies

The association between anemia or IDA and CSVT has been investigated in four case–control studies, ^{16 17 20 21} comprising a total of 289 patients and 3,279 controls. Two studies, including 273 cases, investigated adults with CSVT, while the other two, with 16 pediatric cases, also included patients with arterial cerebral thromboses. Reported OR from multivariate analysis ranged from 1.1 ¹⁶ to 10. ²⁰ The available data support a possible association between severe or microcytic anemia or confirmed IDA and CSVT in both children and adults.

As study design and patient characteristics differed substantially between the two adult and the pediatric case–control studies, we stratified the studies by age group for the exploratory meta-analysis ( Fig. 2 ). In the adult studies, we calculated a pooled OR of 2.55 (95% CI: 0.89–7.33); however, this effect size has to be interpreted with caution, as severe heterogeneity was detected, indicating considerable variation in effect magnitude. Since the 95% CI spans the null, the pooled estimate lacks statistical significance and remains hypothesis-generating rather than definitive. Moreover, OR for severe anemia calculated in our analysis (1.5) differed from the OR reported by the authors of one study (1.1) based on their published CI. ¹⁶ To clarify this discrepancy, we contacted the authors, but source data were no longer available. As the published OR is not consistent with the published CI, we used recalculated OR for the exploratory meta-analysis.

Fig. 2.

Forest plot for the association of anemia or iron deficiency anemia (IDA) and cerebral sinus venous thrombosis (CSVT) in the two case–control studies in adults and the two case–control studies in children. Exploratory meta-analysis per subgroup (1) adults and (2) children using a random effects model with inverse variance method to compare OR. The I ² value ( I ² ) indicates the percentage of variability among studies which arises from heterogeneity rather than random chance. CI, confidence interval; df, degrees of freedom; IV, inverse variance; OR, odds ratio; SE, standard error.

The two pediatric case–control studies included patients with both CSVT and arterial ischemic stroke (AIS) and reported only a small number of cases. As patient population and study design of both studies were comparable, we combined data in an exploratory meta-analysis. We calculated a pooled OR of 4.39 (95% CI: 1.16–16.57). Again, we detected substantial heterogeneity suggesting inconsistent effect sizes ( Fig. 2 ). Moreover, as the pediatric studies included other cerebrovascular outcomes than CSVT, isolated CSVT may not be represented with sufficient precision.

The authors of all four case–control studies provided population-based rates of IDA in the general population; the percentage of IDA in the respective control groups was comparable to published prevalence data (see Supplementary Appendix 1 (online only)). The two studies in adults did not specify anemia subtype as IDA due to lack of systematic iron metabolism analyses. However, the authors inferred IDA in cases of microcytic anemia (in 63% of patients with severe anemia, defined as hemoglobin <9 g/dL in one study, ¹⁶ and in 40% of anemic patients in the other ¹⁷ ). In the study by Coutinho et al., mean hemoglobin levels were significantly lower in patients than in controls ( p < 0.001) and anemia was more prevalent among patients (27% vs. 6.5%). Anemia was significantly associated with CSVT (OR 4.4), especially in microcytic anemia (OR 15.6). ¹⁷ Similarly, Stolz et al. demonstrated that severe anemia was independently associated with CSVT and significantly more common in patients than in controls (11.7% vs. 1.7%, p = 0.005). ¹⁶ Notably, Coutinho et al. did not conduct a separate analysis for severe anemia as none of the controls met the definition of severe anemia. ¹⁷

Stolz et al. specifically investigated the influence of platelet count, reporting a high prevalence of thrombocytosis (platelet count >470 × 10 ⁹ /L) in 81% of CSVT patients with severe anemia, compared with 5.8% in controls ( p = 0.81) and 16.7% in controls with anemia ( p = 0.38). ¹⁶ Although thrombocytosis emerged as prominent, it was not identified as an independent risk factor for CSVT in multivariate analysis. ¹⁶ Coutinho et al. did not provide platelet counts. ¹⁷

Stratification by sex in the study by Coutinho et al. revealed a stronger association between anemia and CSVT in men (adjusted OR, 9.9; 95% CI: 4.1–23.8) than in women (adjusted OR, 3.6; 95% CI: 2.1–6.0). The authors attribute this difference to the 2.5 lower incidence of CSVT in men compared with women. ¹⁷

Two case–control studies investigated the association between IDA and CSVT, ^{20 21} including a total of 16 children with CSVT and 236 controls. Notably, neither study exclusively examined CSVT. Both studies enrolled children with CSVT ( n = 16) as well as arterial ischemic stroke (AIS, n = 20), and no separate analysis was performed to assess the association between IDA and CSVT specifically. IDA was identified as a significant risk factor in both studies. Mean and median levels of mean corpuscular volume (MCV) and hemoglobin levels were significantly lower ( p < 0.01) in patients with CSVT and AIS compared with controls, ²⁰ while platelet counts were significantly higher in patients. IDA was present in 70% of patients with CSVT versus 26% of controls. ²¹ In the study by Maguire et al., all CSVT patients (100%, 6/6) showed IDA, versus 22% (2/9) of AIS patients, ²⁰ and one patient had both CSVT and AIS. Therefore, we did not calculate separate OR for the association between IDA and CSVT. In the study by Azab et al., no significant differences between stroke subtypes in children with IDA were found. However, also in this study, IDA was more common in CSVT patients (70%, 7/10) than in AIS patients (45%, 5/11). ²¹ The authors of this review calculated an OR of 6.6 (CI: 1.6–27.6) for CSVT in children with IDA from the data of Azab et al. ²¹

Thrombocytosis was more prevalent in patients with CSVT and AIS than in control subjects in both pediatric studies. In the study by Azab et al., thrombocytosis was significantly associated with IDA ( p = 0.02) and children with thrombocytosis were nine times more likely to have CSVT or AIS than those with normal platelet counts. ²¹ Maguire et al. reported an almost 12-fold increased prevalence of IDA in previously healthy children with CSVT or AIS compared with controls. This association remained significant after adjusting for platelet counts with an OR of 10 (CI: 3–33), suggesting that the elevated risk cannot be entirely attributed to secondary thrombocytosis, although thrombocytosis was five times more common in cases than in controls. ²⁰ In the study by Maguire et al., all patients with CSVT or AIS and IDA had thrombocytosis (>450 × 10 ⁹ /L) except for one patient who had both CSVT and AIS and normal platelet counts. Four patients with either CSVT or AIS and IDA had a concurrent infection (e.g., acute otitis media, upper respiratory tract infection, or acute gastroenteritis). Thus, secondary IDA and thrombocytosis due to infection cannot be excluded. ²⁰

Limitations Reported by the Authors of the Case–Control Studies

The authors of the case–control studies identified several limitations that could affect the validity of their findings. In adult studies, although sample sizes were large, authors acknowledged potential bias arising from the retrospective nature of data collection. Cases and controls were sourced from patient databases, and the association between anemia and CSVT was not a predefined research question. ¹⁷ Consequently, only patients with hemoglobin values available were included, potentially distorting results. Additionally, in both adult studies, patients and controls were not recruited during the same time period. The authors noted, however, that IDA prevalence has not changed substantially over time. ^{17 20}

In the study by Maguire et al., specific limitations were noted regarding the selection of pediatric controls. Children were excluded as controls if they had a recent history of breast milk or formula feeding, whereas no such exclusion criterion was applied to CSVT patients. ²⁰ Furthermore, children with a history of anemia or iron supplementation were excluded as controls in the Maguire et al. study, which could have led to underreporting of IDA prevalence in the control group. However, the authors stated that the rate of IDA in the large control group (9%) was consistent with population-based data for IDA (4–8%). ²⁰

Case Series

The majority of the included case series did not specifically investigate the association between IDA and CSVT as this was not the primary research objective. Nevertheless, anemia was frequently identified as a major contributing factor. Population-based reference values for hemoglobin were not provided in most studies, limiting comparability across cohorts.

Sébire et al. ²³ analyzed 42 children with CSVT and found anemia and microcytosis (defined as hemoglobin or MCV Z-scores <2 standard deviations below the mean for age) in >50% of patients. Microcytosis as proxy for IDA was predictive of CSVT rather than arterial disease in multiple logistic regression. This finding was based on a comparison between children with CSVT and a cohort of children with arterial stroke associated with arteriopathy (88 with ischemic, 24 with hemorrhagic stroke and arteriopathy).

Case series from India considering children and adults with CSVT reported anemia rates of 21 ²² and 18%, ²⁵ whereas Christerson and Strömberg found a markedly higher prevalence of anemia and thrombocytosis in a small Swedish pediatric cohort, with 75% of patients with CSVT affected. ²⁴ Other case series reported varying prevalence rates of IDA, ranging from 13.6% ²⁶ to 28% ²⁷ in adolescent and adult CSVT patients, with significantly higher rates in younger children (up to 75%). ²⁸ Maher and Amlie-Lefond suggested that anemia caused by excessive cow’s milk intake in toddlers could contribute to severe CSVT involving the deep venous system. ³⁵ Liu et al. ²⁹ investigated the clinical outcomes of anemic versus non-anemic patients with CSVT. Patients with anemia were found to have a higher risk of unfavorable outcomes (30.1% vs. 12.1%) and mortality (27.4% vs. 7.9%). Conversely, increased hemoglobin levels were associated with a lower risk of poor outcomes ( Supplementary Table S1 (online only)).

Case Reports

We report findings from 40 case reports, ^{10 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 75 76 77 78} including 51 patients aged 4 months to 63 years. Most patients were previously healthy without coagulation issues. CSVT risk factors included COVID-19 vaccination or infections, but mostly only IDA was present. IDA causes included poor diet, celiac disease, blood loss, and excessive milk intake in children. For summarized findings, see Supplementary Tables S2 and S3 (online only).

Quality Assessment

The available case–control studies provide evidence classified as low to moderate based on the Cochrane GRADE scoring system. Despite the limitations of the available studies, and significant heterogeneity between studies, the findings, including the large adjusted effect estimate reported by Coutinho et al., ¹⁷ are consistent with a possible association between CSVT and severe anemia, microcytic anemia, or IDA; however, the overall strength and consistency of the evidence remain uncertain. In the study by Stolz et al., an association between anemia severity and CSVT was observed, with severe anemia showing a strong association with CSVT, while mild anemia showed no significant association, ¹⁶ which we interpret as a dose–response relationship according to GRADE criteria. Coutinho et al. included anemia of any severity and reported a higher unadjusted OR of 5.3. ¹⁷ Using data from Stolz et al. cohort, they further calculated a comparable unadjusted OR of 2.8 for anemia. The pediatric case–control studies ^{20 21} included patients with CSVT as well as AIS, weakening the conclusions. Nonetheless, IDA was found to be a stronger risk factor for CSVT than for AIS, likely leading to an underestimation of the association. Additionally, in the pediatric case–control studies, children with a prior infection were not excluded, a potential confounder. The JBI critical appraisal checklist applied to the four included case–control studies indicates a low risk of bias (see Supplementary Appendix 2 , Supplementary Table S4.1 (online only)). For detailed critical appraisal tool tables assessing the bias of each included study, please see Supplementary Appendix 2 , Supplementary Tables S4.2 and S4.3 (online only).

Discussion

IDA as a Potential Risk Factor for CSVT

The available evidence from this systematic review suggests a potential association between severe anemia, microcytic anemia, inferring IDA, or confirmed IDA and an increased risk of CSVT in both adults and children. All four included case–control studies showed higher odds of CSVT in individuals with IDA, or microcytic anemia as a proxy for IDA. In adults, both studies reported robust and statistically significant associations that remained significant after adjustment for confounders; however, one study showed internal inconsistencies in the reporting of effect estimates and measures of uncertainty, which introduces uncertainty when interpreting the precision of the pooled evidence. In children, both studies indicated an increased risk; however, the strength of conclusions is limited by small sample sizes, clinical heterogeneity, and combined analyses of CSVT and AIS, likely attenuating effect estimates for CSVT and rendering precise risk assessment for isolated CSVT difficult.

While randomized clinical trials on this association are not available, the four published case–control studies provide the highest available evidence level. According to the Cochrane GRADE framework, this evidence was graded as low on formal grounds due to the study design. Nevertheless, the magnitude of the observed associations, together with the dose–response relationship reported in one study, suggests a possible association between severe or microcytic anemia or IDA and CSVT, while acknowledging the methodological limitations and heterogeneity of the available evidence.

A recently published systematic review explored the association between IDA and both arterial and venous cerebral thromboses. ⁷⁹ However, due to differing inclusion criteria, this review did not incorporate what we consider the most relevant case–control studies in adults. Furthermore, the pediatric studies included in that review did not differentiate between arterial and venous thromboses, limiting disease-specific conclusions. Finally, no quantitative analysis was conducted restricting the ability to synthesize effect sizes and to evaluate the strength of association for CSVT.

Strengths and Limitations of the Evidence

Our systematic approach ensured a rigorous assessment of study quality, bias, and potential confounding. Where necessary, we recalculated or derived effect estimates, enhancing accuracy and comparability of reported findings. Across all included studies, the direction of association between severe anemia, microcytic anemia, or IDA and CSVT was consistent, despite differences in study design and populations. Furthermore, our analysis accounted for context-specific IDA prevalence, particularly in low-income settings, enhancing relevance across diverse populations. Overall, we estimate a low overall risk of bias in the included studies.

Nonetheless, several limitations warrant consideration. First, the available evidence is insufficient to determine whether the observed association between anemia or IDA and CSVT is causal or generalizable. Variation in prevalence and severity of ID may modulate strength of the association, and this effect is likely population-dependent. Moreover, the association may be further modified by coexisting prothrombotic risk factors, or yet unidentified underlying conditions.

Second, although the two case–control studies in adults were similar in design and patient characteristics, effect sizes in the exploratory meta-analysis must be interpreted with caution. Discrepancies between reported and recalculated effect estimates in one study, the unavailability of source data, and statistically demonstrated heterogeneity limited the accuracy of the pooled estimates. Although the pooled analysis suggested a positive association between anemia and CSVT in adults, the confidence interval included unity and therefore did not indicate a statistically significant association. In addition, the studies used different definitions of anemia: Stolz et al. included only patients with severe anemia in their multivariate analysis, ¹⁶ whereas Coutinho et al. considered all levels of anemia. Furthermore, neither of these two case–control studies systematically assessed iron metabolism, precluding confirmation of iron deficiency as the underlying cause of microcytic anemia. These differences in anemia definition likely influenced effect size estimates and further limited comparability. ¹⁷

Third, pediatric evidence remains limited. The two case–control studies had small sample sizes and heterogenous populations, including children with CSVT and AIS. Meta-analysis yielded a summarized OR of 4.39, likely underestimating the specific effect of IDA on CVST, as the effect of IDA on CSVT appeared stronger than on AIS. We calculated OR for CSVT patients in the study by Azab et al. to be 6.6 (CI: 1.6–27.6), indicating a broad range of effect sizes.

Finally, case series and case reports provide supportive but low-quality evidence. Due to absence of control groups and the high IDA prevalence in the general population, these sources are prone to selection bias and add limited strength to the overall evidence.

Experimental Evidence Supporting a Causal Relationship between IDA and CSVT

Data from animal models complement clinical evidence, demonstrating that both IDA and secondary thrombocytosis independently promote thrombus formation and growth. In a rodent model of dietary ID, Jimenez et al. showed that iron-deficient rats developed larger thrombi with increased platelet activation and clot strength, and that this prothrombotic tendency was reversed with iron repletion. ¹² Thrombocytosis was consistently observed, and platelets expressed higher P-selectin levels, indicating enhanced platelet activation.

Rodent models thus provide an opportunity to explore pathophysiological mechanisms by which IDA leads to thrombocytosis and a prothrombotic state. ⁸⁰ Thrombocytosis is recognized as a key prothrombotic factor in ID. Iron regulates megakaryocytic/erythroid differentiation, and in states of ID, platelet production is promoted at the expense of erythropoiesis. ⁸¹

Beyond thrombocytosis, anemia itself may contribute to thrombosis by altering hemodynamics in blood vessels ⁴¹ : Decreased hemoglobin levels and resulting lower oxygen-carrying capacity of the blood may lead to insufficient tissue oxygenation, particularly in situations of increased metabolic demand. Furthermore, microcytic erythrocytes, characteristic of IDA, have reduced cell deformability increasing blood viscosity, as it is widely known for sickle cell disease. Compensatory mechanisms for hypoxia, such as higher blood flow velocity, can induce hemodynamic changes such as turbulences, which may damage the endothelium and promote inflammation. ⁸²

Together, these experimental findings support a causal relationship between IDA and CSVT. While applicability to humans remains to be fully established, the mechanistic plausibility strengthens the clinical evidence presented. The role of reactive thrombocytosis and altered flow dynamics in the development of thrombosis in IDA is biologically plausible and warrants further study. Evolutionarily, enhanced platelet production may have conferred hemostatic advantages during anemia-induced bleeding, while limiting erythroid expansion to preserve iron for essential physiological functions. Further research is needed to clarify the independent contributions of iron deficiency, anemia, and thrombocytosis to thrombotic risk.

Conclusion and Perspective

The available evidence supports a potential association between severe, microcytic anemia, IDA, and CSVT, but remains insufficient to establish causality. This hypothesis is supported by findings from case–control studies and by animal models that provide biologically plausible pathophysiological mechanisms. Although randomized trials are unlikely to be feasible given the rarity of CSVT and the established benefits of iron supplementation, large population-based studies are needed to better characterize the strength and nature of this association.

IDA is highly prevalent, particularly in low-income countries. ¹ However, it remains uncertain whether findings from case–control studies—mainly conducted in high-income settings—can be generalized to populations with higher IDA prevalence or different underlying etiologies. These differences also present challenges in identifying appropriate control groups for observational studies.

To determine whether IDA is an independent risk factor for CSVT, future studies should include well-characterized patient populations and appropriately matched controls. It is also critical to investigate whether correction of IDA modifies CSVT risk. While several case reports and case series have described CSVT in the absence of other identifiable prothrombotic risk factors, the current evidence remains insufficient to conclude that IDA alone can precipitate CSVT.

Given the high prevalence of IDA and the established benefits of diagnosing and treating ID, clinicians should consider IDA in patients with CSVT, particularly when no other risk factors are apparent. Whether systematic screening for, or treatment of, IDA can reduce the risk of CSVT or other thrombotic complications requires evaluation in adequately designed studies.

Declaration of GenAI Use

Artificial intelligence–based language support (ChatGPT, OpenAI) was used to assist with wording and stylistic refinement of the manuscript, with all scientific content generated and verified by the authors.