The objectives of the study were as follows: (i) to identify all studies published in Saudi Arabia reporting hematological parameters and (ii) to identify the reported hematological parameters’ RIs in healthy Saudi adults.

The literature search was performed using PubMed and Google Scholar to identify articles published in English from 1960 to March 2023. The search strategy employed Medical Subject Headings and keywords that expressed the terms “hematological parameters,” “complete blood count,” “hematological profile,” “hematological reference intervals,” “reference intervals,” “reference ranges,” “reference values,” and “reference parameters,” crossed with the operator AND and terms such as “adults” and “Saudi Arabia.” The search strategy for PubMed is summarized in Table 1. Full-text articles were retrieved after reviewing the title and abstract.

The authors (NS and AA) independently checked the titles and abstracts. The references were managed using Mendeley. The screening was performed to select articles published in English. The systematic review included primary research articles that reported a CBC test and included participants from all age groups. The meta-analysis was conducted for five studies only, with hematological parameters’ RIs data available for all CBC parameters in individuals aged ≥14 years (Table 2).

The selected studies were assessed for methodological quality by two independent reviewers (NS and AA) using standard critical appraisal instruments from the Joanna Briggs Institute (JBI) Critical Appraisal Checklist (18).

The methodological assessment criteria were based on the eight elements of the JBI Critical Appraisal Checklist (Figures 1a,b). A cut-off score of at least 60% was considered the inclusion criterion. The study complied with the Meta-analyses Of Observational Studies in Epidemiology (MOOSE) Checklist (19).

Data were extracted by the authors using a standardized data extraction sheet. The extracted data included the year of publication, the location of the study, sample size, hematological parameters’ RIs, and mean and standard deviation both for male and female participants (Table 3).

For continuous variables, weighted mean differences (SD) and 95% confidence intervals were estimated. Depending on the degree of study heterogeneity, a random-effects model was employed to obtain the pooled estimate of the results. The I² statistic and corresponding p-value were used to evaluate the degree to which statistical heterogeneity in the meta-analyses was caused by variations between the studies rather than by chance. Forest plots were used to represent the type of statistical heterogeneity. The data for CBC parameters were normally distributed, except for neutrophils. The mean values of the parameters were compared between male and female individuals using an independent samples t-test. All statistical analyses were performed using R software version 4.3.2.

Funnel plots are graphical tools commonly used in observational studies to assess publication bias and evaluate the presence of small study effects. These plots provide a visual representation of the relationship between effect size estimates from individual studies and their corresponding precision measures, such as standard errors or sample sizes (20). In a funnel plot, each point on the graph represents a study, with the effect size displayed on the horizontal axis and the precision measure on the vertical axis. In the current study, the funnel plot displays mean values on the horizontal axis and standard errors on the vertical axis. A vertical line through the center of the plot typically represents the true effect size. The shape of the funnel is determined by the inherent variability in effect sizes due to sampling error, with larger studies having smaller standard errors and thus being more precise (21). In the absence of publication bias, the points on the plot are expected to be symmetrically distributed around the true effect line. Studies in the upper portion of the plot with low standard errors are expected to cluster closer to the true effect line, resulting in a narrower funnel. In the lower portion of the figure, the funnel opens up, and with increasing standard errors, effect sizes are expected to scatter more widely to the left and right of the true effect. Supplementary Figures S1, S2 represent the funnel plots. Egger’s test was performed to quantify publication bias. Supplementary Table S1 shows the absence of publication bias, as the p-values were > 0.05 for all of the reported variables (21).

A total of 1,35,099 articles were identified after searching the PubMed database (Table 1). The details of exclusions are provided in Table 2. In total, 13 studies have reported hematological parameters’ RIs in Saudi Arabia from 1982 to present. A latest published study focused on methodology (22) and a pilot study published in 1982 (19) were excluded. A total of 12 studies reporting hematological parameters’ RIs under normal physiological conditions across all age groups were included in the systematic review. The published studies originated from all regions: Central (n = 5), Western (n = 5), Southern (n = 1), and Northern (n = 1). The majority of the studies, eight (66%), focused on adults, while four (33.3%) studies reported a sample of adolescents or children (Table 3).

The characteristics of the included studies are summarized in Table 3. A total of five studies were reported from Riyadh (Central region) (5, 23–26), five from Jeddah (Western region) (11, 15, 17, 27, 28), one from Aljouf (Northern region) (29), and one from Abha (Southern region) (30). All studies were cross-sectional and included both male and female individuals. The twelve studies were included in the systematic review. A total of seven studies were not included in the meta-analysis for the following reasons: three studies reported only WBC parameters, two studies had only abstracts available, and two studies involved newborns (Table 3).

The meta-analysis results of the data driven from the five included studies are summarized in Figures 2–4, presenting data for RBC, Hb, PLT, WBC, HCT, MCV, MCH, and MCHC. High heterogeneity was observed across all hematological parameters: RBC, Hb, or MCHC (I² = 100%); MCH or HCT (I² = 99%); PLT or MCV (I² = 98%); and WBC (I² = 90%). The RBC (p = 0.009) and Hb (p = 0.0006) values were higher in male individuals. The PLT values (p = <0.0001) were higher in female individuals. The remaining hematological parameters showed no statistically significant sex differences. The pooled mean estimates for all hematological parameters are summarized in Table 4.

Sensitivity analysis was not carried out as it was not applicable due to (i) the included studies having no differences in terms of quality and (ii) the absence of specific sub-groups.

The comparison across the studies was challenging due to variations in study settings, sample sizes, analytic conditions—including the technology used for blood counters—assessed parameters, the populations under study, the methods for selecting the study population, age ranges, and hematological parameters’ RIs, which were based on local laboratory standards for each study. Although the studies used terms such as “volunteers,” “healthy individuals,” and “participants coming for routine check-ups,” it was still challenging to identify a healthy individual in the Saudi population due to the high prevalence of disorders related to Hb, iron deficiency anemia, and endemic viral infections.

Several countries have established RIs based on their populations (41, 50). It is well documented in the literature that reported RIs vary across populations (51, 52). Therefore, we recommend establishing hematological parameters’ RIs based on the Saudi population on a broader scale. Special precautions should be taken when designing studies reporting hematological parameters’ RIs, taking into consideration factors such as geographical region, altitude, genetic factors, lifestyle, risk factors, and the baseline health status of participants. Future studies focusing on hematological parameters should also consider confounders such as smoking status and iron store assessments.

The CBC is a commonly requested laboratory test, but it can be misinterpreted. Caution should be exercised when interpreting CBC results as hematological parameters’ RIs vary across laboratories. It is important to recognize that inappropriate interpretation of hematological parameters’ RIs might lead to unnecessary further investigations or a failure to identify the underlying disease. This finding, in turn, can affect the quality of patient care and have an impact on the economy.