Studies were eligible for inclusion if they:

Studies were excluded if they were editorials, commentaries, opinion pieces or animal studies.

There were no language restrictions. Where necessary, non-English articles were translated using Google Translate or professional translation services to maximise evidence capture, consistent with scoping review methodology.

A comprehensive literature search was conducted across five databases in PubMed, Google Scholar, Scopus, Web of Science and the Cochrane Library covering all records from inception to December 2025. In addition to electronic databases, we also search the reference lists of included articles (backward citation chaining) and use forward citation tracking via Google Scholar to identify additional relevant studies.

The search terms and synonyms were developed using Medical Subject Headings (MeSH) and piloted in PubMed. The final PubMed search string was:

((“Aspirin” [MeSH Terms] OR “low-dose aspirin” OR “acetylsalicylic acid”) AND (“Anemia” [MeSH Terms] OR “haemoglobin” OR “hemoglobin” OR “blood iron levels” OR “iron deficiency” [MeSH Terms]) AND (“Pregnancy” [MeSH Terms] OR “pregnant women” OR “maternal” OR “pregnancy complications” [MeSH Terms] OR “hypertensive disorder of pregnancy” [MeSH Terms] OR “preeclampsia” [MeSH Terms] OR “pregnancy outcomes” [MeSH Terms] OR “obstetric hemorrhage” [MeSH Terms] OR “postpartum hemorrhage” [MeSH Terms])). This search strategy was adapted for all other databases. The search was peer-reviewed by an information specialist according to the Peer Review of Electronic Search Strategies (PRESS) checklist to optimise sensitivity and precision.²⁶ To ensure comprehensive coverage, grey literature sources were included, such as ProQuest dissertations, theses and international guidelines or policies where published studies were limited.

All retrieved citations were imported into Zotero (v5.0.81) for reference management and duplicate removal using its built-in detection tool and manual verification. Title and abstract screening were performed using Rayyan software by two independent reviewers (Nokwethemba M. Ngcobo and Vinogrin Dorsamy) according to predefined inclusion and exclusion criteria.²⁷ Any disagreements arising during the screening process were addressed through mutual discussion or, if necessary, by seeking input from a third reviewer (Chauntelle Bagwandeen) to reach consensus. A calibration exercise was conducted prior to screening to ensure consistency and shared understanding of the eligibility criteria. Full-text articles deemed potentially relevant were retrieved and assessed for eligibility by two independent reviewers, and reasons for exclusion at this stage were recorded to maintain transparency.

Data were charted using a standardised extraction table developed in Microsoft Excel, guided by the JBI and PRISMA-ScR principles. Extracted information included citation details (author and year of publication), study design, country and population characteristics. Details of the intervention or exposure (LDA dose, timing and indication), comparator (if applicable) and outcomes measured – such as Hb levels, anaemia prevalence, ferritin concentration and related haematological indices – were recorded. Information on the indication for LDA use (e.g. PE or HDP) was extracted where reported. Contextual variables, including healthcare setting, comorbidities, nutritional status and other relevant sociodemographic factors, were also documented. Data extraction was conducted by one reviewer and independently checked by a second reviewer for accuracy and completeness. Prior to full charting, the data extraction tool was piloted on a subset of studies by two independent reviewers to ensure clarity and consistency.

Evidence was analysed using the two stages of the Arksey and O’Malley framework.²⁸ Firstly, the study selection process was mapped using a PRISMA-ScR flow diagram to provide an overview of the identification, screening and inclusion of sources of evidence. Secondly, a descriptive summary of the included studies was generated and presented in tabular form, detailing key study characteristics such as study design, setting, sample size, LDA dose and timing and reported outcome measures.

This review is part of a larger study investigating the impact of LDA on Hb levels during pregnancy, which obtained ethical approval from the University of KwaZulu-Natal (UKZN) Biomedical Research Ethics Committee (BREC reference BREC/00007156/2024).

Two studies reported on changes in maternal Hb levels. Both were secondary analyses of RCTs. Jessani et al., a secondary analysis of the ASPIRIN trial, enrolled nulliparous pregnant women from multiple LMICs between 6 and 13 weeks of gestation.³⁰ Ngcobo et al. included pregnant women of African ancestry (n = 249) with singleton, normotensive pregnancies attending a regional hospital in Durban, South Africa, enrolled between 12 and 20 weeks of gestation.³¹

Obstetric bleeding represents a potential pathway through which LDA may influence maternal Hb and anaemia risk; therefore, studies reporting maternal bleeding outcomes in relation to LDA use during pregnancy were examined. Eleven heterogeneous sources of evidence were identified, including case reports, observational cohort studies, two multicentre RCTs and systematic reviews. A single case report with a narrative literature review by Plancha et al. described a pregnant woman using LDA (75 mg/day – 150 mg/day) for PE prevention who developed life-threatening upper gastrointestinal bleeding, illustrating that serious haemorrhagic events can occur, although rarely, during LDA exposure.³² At the population level, Hastie et al. conducted a large registry-based cohort study in Sweden (approximately 315 000 pregnancies; 1.3% exposed to aspirin 75 mg/day – 160 mg/day) and reported an increased risk of bleeding during labour among aspirin-exposed pregnancies; anaemia outcomes were not assessed.³³ Similarly, White et al., in an observational cohort of women prescribed 81 mg LDA for early PE prophylaxis, reported a higher risk of postpartum bleeding compared with non-use.³⁴ In a multicentre propensity score–based cohort study, Souter et al. reported increased odds of placental abruption (OR 1.44) and postpartum haemorrhage (OR 1.21) among pregnancies exposed to LDA.³⁵

In contrast, high-quality RCT evidence demonstrates neutral bleeding safety profiles. Rolnik et al. conducted a multicentre double-blind RCT of 1776 high-risk pregnancies across 13 countries, randomising women at 11–14 weeks to 150 mg nightly aspirin until 36 weeks versus placebo and found postpartum haemorrhage rates of 4.4% (35/798) in the aspirin group versus 4.0% (32/798) in placebo (RR 1.09, 95% CI 0.69–1.73) with no significant increase in placental abruption (0.3% vs 0.1%).⁷ Similarly, Lin et al. reported findings from the largest Chinese multicentre RCT (n = 3236 high-risk women enrolled 12–20 weeks across 13 tertiary hospitals), comparing 100 mg daily aspirin until 34 weeks versus standard care, which showed no significant differences in postpartum haemorrhage or other bleeding complications between groups as safety endpoints.³⁶

Other observational studies and systematic reviews provide mixed findings. Zhang and Wang, in a retrospective cohort of high-risk pregnant women, found no significant difference in postpartum haemorrhage rates between aspirin-exposed and non-exposed groups.³⁷ Jiang et al. conducted a systematic review of 21 studies reporting postpartum haemorrhage (n = 373 926) and seven reporting blood loss (n = 10 163) among women receiving 60 mg/day – 150 mg/day aspirin and found a modestly increased risk of postpartum haemorrhage (OR 1.20, 95% CI 1.07–1.34) and greater blood loss in LDA users.³⁸ However, Roberge et al., in a systematic review of 45 RCTs (n = 20 909) in which LDA (≥ 100 mg/day) was initiated at or before 16 weeks of gestation, found no significant increase in placental abruption or antepartum haemorrhage.³⁹ Similarly, Henderson et al. reported no significant increase in haemorrhagic complications in high-risk pregnancies treated with 60 mg/day – 150 mg/day aspirin initiated at ≤ 16 weeks of gestation, while Duley et al., in a systematic review of 60 RCTs, reported a small increase in minor bleeding events but no clinically important excess in major haemorrhage or blood loss.^40,41

Overall, the available evidence primarily addresses obstetric bleeding outcomes – such as intrapartum bleeding, postpartum haemorrhage, placental abruption and rare gastrointestinal haemorrhage – without directly examining Hb or anaemia outcomes in relation to bleeding among pregnant women using LDA. The evidence base is methodologically diverse and occasionally discordant, with several observational studies and one systematic review suggesting modest increases in postpartum haemorrhage or blood loss, whereas two RCTs generally do not demonstrate a clinically meaningful increase in haemorrhagic events.^{7,33,34,35,36,37,38} Notably, few studies originate from LMIC settings with a high background prevalence of anaemia, and none explicitly link LDA-related bleeding to subsequent maternal Hb or anaemia outcomes during pregnancy or the postpartum period.^30,31,36,37

None of the included studies directly reported the incidence or prevalence of anaemia following LDA use during pregnancy. Although several studies evaluated related outcomes, such as Hb concentration or obstetric bleeding, anaemia was not measured or analysed as a primary or secondary endpoint in any study. Consequently, no quantitative or qualitative synthesis could be performed for this outcome.

As shown in Table 1, LDA is well established as an effective intervention for preventing PE, yet few studies have directly examined its impact on Hb levels or anaemia risk in pregnancy. One study reported no meaningful difference in Hb trajectories between aspirin and placebo groups, whereas another observed smaller declines in Hb among women receiving LDA, with larger Hb reductions associated with increased HDP risk.^30,31 Several large RCTs and systematic reviews have evaluated LDA for PE prevention in high-risk pregnancies; however, anaemia-related outcomes were infrequently reported.^7,36,41

Observational studies and one systematic review indicated a possible association between LDA and increased risks of postpartum bleeding or haemorrhage.^33,34,38 However, some large trials and other systematic reviews reported no clinically significant increase in obstetric haemorrhage.^7,39

For clinicians: Current evidence does not clearly establish whether LDA influences Hb or anaemia outcomes during pregnancy. For clinicians, this underscores the importance of interpreting LDA-related haematologic findings within the broader context of local anaemia prevalence, comorbidities and antenatal care practices.

For researchers: The review highlights substantial research gaps, particularly the absence of studies that systematically evaluate Hb or iron-related endpoints in pregnant women receiving LDA. Future work may benefit from incorporating standardised haematologic measures and exploring mechanistic pathways in diverse populations, including LMIC settings.

For policymakers: Policy frameworks may consider the need for more locally relevant evidence on anaemia and LDA use, especially in contexts where anaemia and infectious comorbidities are highly prevalent. Strengthening routine data systems and surveillance for both HDP and anaemia may help inform future guideline refinement.

As a scoping review, this work was designed to map the extent, nature and characteristics of the available evidence on LDA and anaemia-related outcomes in pregnancy, rather than to synthesise findings across studies, assess the risk of bias of individual studies or evaluate the certainty of evidence. Consequently, no formal quality appraisal, quantitative synthesis or assessment of causal effects was undertaken. In addition, the review may be subject to publication bias, selective outcome reporting and language bias, particularly given that anaemia-related outcomes were often not predefined endpoints in the included studies. The heterogeneity in study designs, populations, aspirin dosing, outcome definitions and reporting further limits direct comparison across studies.

Despite these limitations, this scoping review provides a descriptive overview of the existing evidence base and highlights critical gaps in the literature – particularly the paucity of pregnancy-specific data on Hb, iron status and anaemia outcomes in LMIC settings. These findings can inform the design of future primary studies and targeted systematic reviews aimed at evaluating the haematologic effects of LDA in pregnancy.