Introduction

The latest edition of the International Statistical Classification of Diseases and Related Health Problems (International Classification of Diseases) by the World Health Organization [1] introduced a new term "developmental learning disorder" instead of the current one: "specific developmental disorders of scholastic skills" [2]. This category of disorders includes, among others: developmental learning disorder with impairment in reading, commonly referred to as dyslexia [1]. Developmental dyslexia is a neurodevelopmental disorder identified in children with average intelligence and sensory capabilities. It is characterised by difficulties in reading including in accurate and/or fluid word recognition, spelling, and decoding skills across different languages. Dyslexia is considered a neurodevelopmental disorder, because it arises from the natural neurological variations in brain development, which deviate from the standard developmental path [3]. Ramus et al. [4] emphasise that some cases of dyslexia (a minority) might be explained by non-phonological deficits such as in the automaticity hypothesis [5] or visual theories [6–8].

However, since there are at least two empirically documented trends in the search for the causes of dyslexia (leading phonological and niche non-phonological), it should be recognised as a disorder stemming from various causes. Reading is a highly complex, multimodal cognitive skill. The act of reading must involve the coordinated action of many areas of the brain involved in both lower and higher order processing. Neuroimaging research has delineated a set of brain systems specialised in various sub-skills involved in reading that form a "reading network" [9, 10]. A problem in the operation or interaction of one or more parts of these systems seriously hampers the acquisition of fluent reading by people with dyslexia [11]. Thus, this complex network of systems may be susceptible to the disturbing influence of many factors throughout brain development, both in the pre- and perinatal period.

Furthermore, it has been established that genetics play a significant role in causing dyslexia [12–14]. However, a study on environmental factors focusing on inadequate oxygen supply during pregnancy or labour also presents an intriguing area of research for uncovering the pathomechanism of dyslexia [15]. Most perinatal injuries arise from hypoxia, which is defined as inadequate oxygenation of body tissues, and the conditions associated with it [16]. Various medical conditions can contribute to asphyxia. These conditions may lead to different types of hypoxic events (prenatal or neonatal, chronic or acute) and also to different outcomes [17, 18]. The diagnostic criteria for neonatal hypoxia are based on a set of markers including Apgar below 5 (at the fifth minute), need for intubation or cardiopulmonary resuscitation (CPR), umbilical artery pH below 7.00, and abnormal neurological signs such as hypotonic muscles or an absent sucking reflex [17]. Among impacted infants, a quarter experience profound and enduring neuropsychological consequences [19], encompassing conditions such as cognitive impairment, visual motor or visual perception issues, increased hyperactivity, cerebral palsy, and epilepsy [20, 21]. Hypoxic-ischemic encephalopathy (HIE) is a cerebral injury resulting from oxygen deprivation due to either hypoxic or anoxic incidents [22, 23]. Hypoxia signifies a reduction in blood oxygen reaching the brain, while ischemia denotes a decreased blood flow to the brain. The prevalence of HIE is estimated at 1–8 cases per 1,000 live full-term births [20, 24]. Some studies highlight that some patients with HIE may develop learning disabilities [21–23, 25–27]. Of course, the severity of neurological consequences depends not only on the severity of HIE but also on the therapeutic interventions employed. One of the most well-researched therapeutic methods is therapeutic hypothermia. Clinical studies have established that therapeutic cooling improves neurological outcomes following various acute brain injuries, including neonatal hypoxia-ischemia [28–30].

There are also studies indicating the relationship between smoking by pregnant mothers and reading difficulties in their school-age children [31–33] and it has been proven that one of the causes of fetal hypoxia is smoking by the mother during pregnancy [34, 35].

For example, a study conducted in China [15] on complicated natural childbirth and perinatal hypoxia confirmed a higher occurrence of dyslexic children in these populations compared to those without dyslexia (3.12% vs. 1.75% and 3.76% vs. 1.42% of the overall sample for difficult natural delivery and perinatal hypoxia, respectively).

Methods

Eligibility criteria

Studies were selected according to pre-specified eligibility criteria following the PICOS model presented in Table 1. The search was not restricted to articles on research conducted in the location of this study.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Eligibility criteria.

https://doi.org/10.1371/journal.pone.0308497.t001

Results

Results of the search

We performed the database searches on 1 April 2023, and identified 1,336 records in the National Medical Library (41), PsycINFO (50), Web of Science (527), EMBASE (714), and DARE (4). No studies were identified after searching the references lists of these records. Furthermore, 401 duplicates were found, including 241 automatically detected by EndNote and 160 manually removed. After excluding duplicates and clearly irrelevant papers, 3 pairs of reviewers examined the full-text articles of 11 potentially relevant studies.

The evaluating teams (pairs) received 309 items (team A), of which evaluator-1 included 1 item and evaluator-2 included 0 items. This discrepancy was assessed positively by reviewer-1, and the paper was included in further analyses. Out of the 309 items received by Team B, evaluator-1 included 7 items, and evaluator-2 also included 7 items. The assessments were consistent for 5 records, and 2 discrepancies were assessed positively by reviewer-2. Therefore, 7 papers were included in further analyses. Team C received 317 items, of which evaluator-1 included 4 items and evaluator-2 included 40 items. Furthermore, 36 discrepancies were found, of which 1 was assessed positively by reviewer-3. Thus, 1 paper was included in further analyses. As such, 9 papers were included in the next stage of evaluation. For one article (from 1987), neither the abstract nor full text could be found. Due to the lack of information on the corresponding author, the publishing house was contacted directly with a request to make the article available. The request to the publishing house was sent twice (20 July 2023 and 8 August 2023) without a response. At this stage, 4 more articles were rejected for the following reasons: 3 articles did not meet the "article type" criteria (1 thesis, 1 systematic review, 1 theoretical), and 1 was a duplicate. Two authors then independently screened the remaining 5 full reports, and decided whether they met the inclusion criteria and provided reasons in cases of rejection. Clinical psychology researchers with expertise in the area assessed the relevance of the studies.

Finally, no articles met the pre-specified eligibility criteria. Fig 1 shows the selection process.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Study flow diagram.

https://doi.org/10.1371/journal.pone.0308497.g001

The excluded studies are listed in Table 2. The primary reasons for exclusion of all these articles was lack of a control group. Additionally, other exclusion criteria include: for 3 articles (no. 1, 4, 5; see Table 2) lack of a formal dyslexia diagnosis, for 1 article (no. 2, see Table 2), the epidemiological nature of the study, and for 1 article (no. 3, see Table 2), lack of a group of children with dyslexia and hypoxia.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. List of studies excluded studies in the eligibility stage.

https://doi.org/10.1371/journal.pone.0308497.t002

Discussion

The aim of this systematic review was to evaluate the association between hypoxia and dyslexia by assessing the relationship between a health-related event at the stage of labour and the occurrence of an outcome among children of school age. Despite the extensive search method employed, we did not find any published studies on the subject matter. As there were no eligible studies, the association between hypoxia and developmental dyslexia remains unknown.

We were surprised to find that no studies meeting the established eligibility criteria could be located. In clinical practice, a connection is often indicated between perinatal abnormalities (including oxygen deprivation) and dyslexia. In addition, as dyslexia is a neurodevelopmental disorder and oxygen deprivation significantly influences nervous system development, the absence of research investigating the link between hypoxia and dyslexia appears to be a substantial gap.

The review conducted allowed the identification of several studies suggesting a correlation between perinatal hypoxia and dyslexia (or broader learning disorders). However, due to their epidemiological nature, lack of a control group, and absence of a formal diagnosis of dyslexia, these studies were not included in the systematic review. Nevertheless, the descriptive epidemiological study, conducted by Liu et al. [15], may provide valuable insight into the relationship between hypoxia and dyslexia. This research indicates that perinatal oxygen deprivation is a significant risk factor for dyslexia. Among the various perinatal abnormalities analysed (e.g. difficult vaginal delivery, birth injury, neonatal aspiration pneumonia, neonatal jaundice, neonatal infectious diseases), hypoxia most distinctly differentiated dyslexics from non-dyslexics. In total, 34,748 Chinese students (grades 3–6) were included in the study, among whom 1200 students with dyslexia were identified. Respectively, 3.76% of this group experienced neonatal asphyxia. In the non-dyslexic group, 1.42% of students experienced neonatal asphyxia. This implies that the occurrence of asphyxia is over 2.5 times higher in the group of students with dyslexia. A limitation of this study is its retrospective nature and the potential bias associated with the recall of outcomes.

In certain epidemiological studies where perinatal hypoxia is recorded, students with dyslexia are included in heterogeneous groups of children with neurodevelopmental disorders [45] or under the broader category of "learning disorders". Both these [45] prevent a precise determination of the risk factor’s relationship solely to dyslexia. Nevertheless, in all these studies, it emerges as a potentially significant risk indicator for this disorder. Hence, the conducted review provides a basis for further in-depth investigation into the role of perinatal hypoxia in shaping the risk of dyslexia through clinical control studies.

Research examining physiological variables like perinatal factors and their impact on the subsequent cognitive abilities of students tends to utilise broad qualitative categories such as learning disabilities rather than focusing on specific neurodevelopmental disorders like intellectual disability, dyslexia, or dyscalculia. Psychological research in this area often encounters a challenge in accessing medical records, which acts as a barrier. As a result, psychologists frequently rely on indirect methods like interviews to gather data. Therefore, it seems important that research on the relationship between perinatal burdens and learning disorders in the years of school education be conducted by interdisciplinary teams of specialists, including physicians, psychologists, and epidemiologists.

The existing studies that explore the presence of perinatal hypoxia in the developmental history of students with dyslexia primarily adopt an epidemiological approach. However, a gap exists in terms of clinical and case-control studies within this domain.

The lack of eligible studies might be due to the possibility of publication bias. Even in this study, we applied a comprehensive searching strategy for both an electronic and a manual search and did not include grey literature. However, data produced outside of traditional publishing including master and PhD theses, book chapters, books, reports, policy literature, government documents, speeches, and so on might be an important source of evidence. Therefore, it is important that the searching strategy consider non-published research findings. The lack of eligible records may also be because of restrictive inclusion criteria in terms of study design. We included only case-control and cohort studies, such as studies that included a control group.

Therefore, as asphyxia is a highly complex phenomenon, future research should also consider factors that characterize it, such as the timing of the insult or the distinctions between chronicity and acuity, among others, rather than solely focusing on its presence. This is crucial because these factors may significantly influence the observed outcomes [18].

Systematic reviews without results are called "empty reviews" and are prevalent in the literature. Yaffe et al. [46] reported that approximately 1 in 10 Cochrane reviews did not include results. Considering that no articles met the inclusion criteria, the current systematic review shows gaps in knowledge about the association between a health-related event—hypoxia during pregnancy or delivery and outcomes—and developmental dyslexia in school-age children at a current point in time. Due to the restriction of eligibility criteria, it might also be relevant to undertake a scoping review rather than systematic review. The protocol of the current systematic review was pre-registered with PROSPERO to minimise reporting bias. By pre-registering the protocol, readers could notice that the results of the systematic review do not present post-hoc amendments to the review methodology. We hope that the lack of evidence in this subject area will provide a basis for colleagues pursuing future research on this topic.

Conclusions

After a systematic search of the literature, we found no records that consider as an exposure the presence of hypoxia during pregnancy or delivery and consider as an outcome dyslexia in school-age children as defined by the ICD-11 classification of neurodevelopmental disorders. Thus, the association between a health-related event—hypoxia during pregnancy or delivery and outcomes—and developmental dyslexia in school-age children remains unknown in the developmental dyslexia aetiology.

Supporting information