INTRODUCTION

Asthma is one of the most common chronic respiratory diseases worldwide, characterized by airway inflammation, airway hyper-responsiveness, variable airflow obstruction, and allergy [1]. Although environmental exposures such as allergens may be risk factors for asthma development, both family- and twin-based studies indicated the importance of genetic components [2]. Therefore, the etiology of asthma is considered to be interactions between environmental factors and genetic characteristics of an individual [3]. It is not surprising that numerous studies have investigated the associations between genetic polymorphisms and susceptibility to asthma, and among them, the cluster of differentiation 14 (CD14) gene has been widely studied [3, 4].

The gene-encoding CD14 is located on chromosome 5q31, a region that is linked to asthma in genome-wide linkage studies [5]. Two protein forms are encoded by the CD14 gene: (i) a membrane receptor expressed on the surface of monocytes, macrophages, and neutrophils and (ii) a soluble form in serum [6]. CD14 is the principal receptor for bacterial lipopolysaccharides (LPSs); upon binding, this complex together with co-receptors toll-like receptor 4 can activate the immune system [3]. The subsequent release of pro-inflammatory cytokines can promote Th1 differentiation of naive T cells resulting in a lower prevalence of allergies such as asthma [3, 7].

Many studies have studied the associations between asthma susceptibility and polymorphisms in CD14 gene, especially the −159C/T polymorphism which was first investigated as a candidate one for atopy [8]. However, inconsistent results were observed. This inconsistency may be attributed to the relatively small sample size of the individual studies. A meta-analysis can provide more reliable evidence by combining the data from different studies. Although there have been three meta-analyses concerning the association between the CD14-159C/T polymorphism and asthma risk [1, 9, 10], further investigations are required for the following reasons. The first meta-analysis only included the earlier studies, and no subgroup analyses were presented [10]. The second one had some significant errors in including studies and abstracting data [9]. The last meta-analysis was performed in 2011, and a number of new studies with more data have been published since 2011 [8, 11–19]. In light of this, we performed an updated meta-analysis to re-investigate the association of the CD14-159C/T polymorphism with asthma susceptibility.

MATERIALS AND METHODS

Identification of Eligible Studies

Relevant studies were identified from PubMed, Embase, and two Chinese databases (Wanfang and China National Knowledge Infrastructure (CNKI)) up to 8 May 2015. The search strategy was as follows: “CD14” and “asthma” or “asthmatic” and “mutation” or “polymorphism” or “variant.” There was no restriction on languages or publication date. Also, the references of relevant reviews or meta-analyses were searched manually to identify additional studies. Studies that fulfilled the following criteria were incorporated into the meta-analysis: (1) evaluating CD14-159C/T polymorphism and asthma risk, (2) using a case-control design, and (3) providing enough data for calculating odds ratios (ORs) and 95 % confidence interval (CI). Exclusive criteria in this meta-analysis were (1) studies containing overlapping data; (2) design based on family or sibling pairs; and (3) some publication types such as comments, abstracts, reviews, and proceedings. For the studies with overlapping data, the one with the largest sample was selected.

Data Extraction

Two reviewers independently extracted the information from each included publication: author, year of publication, country of origin, ethnicity, age group, atopic status, sample size, asthma definition, and genotype distribution among cases and controls. Any discrepancy was resolved via discussing or consulting two additional authors.

Statistical Analysis

ORs and 95 % CI were employed to evaluate the strength of the associations between the CD14-159C/T polymorphism and asthma risk in the dominant (TT + TC vs. CC), recessive (TT vs. TC + CC), codominant (TT vs. CC), and allelic models (T vs. C). The pooled ORs were calculated using a fixed-effect or random-effect model according to between-study heterogeneity, which was checked by Q test and I 2 statistics [20]. If the P value for the Q test was > 0.10, the pooled OR was calculated with the fixed-effect model; otherwise, the random-effect model was used [9]. I 2 values were used to test the between-study heterogeneity quantitatively. I 2 values range between 0 and 100 %, and the values of 25, 50, and 75 % were nominally assigned as low, moderate, and high heterogeneity, respectively [21]. The statistical significance of the pooled ORs was analyzed by the Z test, and a P value of <0.05 was considered statistically significant. Subgroup analyses were conducted by ethnicity, age, and atopic status. For assessing the age—atopic and ethnicity—atopic associations, we further stratified the atopic asthma populations by age and ethnicity.

Hardy–Weinberg equilibrium (HWE) in controls was assessed with the chi-squared test (significance set at P < 0.05). Sensitivity analysis was performed to check the stability of the results using two methods: (i) removing studies that deviated from HWE in controls and (ii) excluding a single study each time if the statistically significant results were found. The potential publication bias was examined by visual inspection of Begg’s funnel plots and Egger’s test. All statistical tests were conducted with Revman 5.1 (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark) and STATA 12.0 (StataCorp LP, College Station, TX, USA) software.

RESULTS

Main Characteristics of Included Studies

Figure 1 provides a flowchart detailing the procedure for study identification. Finally, 36 case-control studies were included [7, 8, 11–19, 22–46]. The main characteristics of the included studies and genotype distributions of CD14-159C/T polymorphism are presented in Tables 1 and 2, respectively. Except one study without ethnic information [40], 15 studies were performed in Asians, 15 in Caucasians, one in Africans, and four in the populations of mixed ethnicity. Thirteen studies were performed in adults, 16 in children, and seven in undefined or mixed age groups. Eleven studies only included atopic asthma populations; moreover, 18 studies included both non-atopic and atopic asthma populations; however, data could not be extracted separately in nine studies of them. Seven studies did not provide the atopic status of subjects. This information was summarized in Table 1. Based on the same data set, the genotype frequency data for atopic asthma subgroup in one included study [43] were abstracted from another study that had been excluded [47]. The distribution of the CD14-159C/T polymorphism in controls was inconsistent with HWE in three studies [28, 29, 37] (Table 2).

Fig. 1
figure 1

Flow chart of included/excluded studies using the criteria before the meta-analysis.

Full size image
Table 1 Main Characteristics of Studies Included in the Meta-Analysis
Full size table
Table 2 Genotype and Allele Distributions of CD14-159C/T Polymorphism in Cases and Controls
Full size table

Meta-Analysis of CD14-159C/T Polymorphism and Asthma Risk in the Overall Populations

A total of 6954 cases and 7525 controls in 36 case-control studies were included in meta-analysis. As shown in Fig. 2, the between-study heterogeneity of the dominant model (TT + TC vs. CC) for all 36 studies was tested by Q test and the corresponding P value was 0.007. Thus, a random-effect model was chosen for data synthesis. The pooled OR for all 36 studies was 0.90 (95 % CI = 0.81–1.01, P = 0.08; Fig. 2), indicating no significant association between this polymorphism and asthma risk in the dominant model. Similarly, no statistically significant associations were observed in the recessive, codominant, or allelic model. Summary results of all the genetic comparisons are listed in Table 3.

Fig. 2
figure 2

Meta-analysis of the association between CD14-159C/T polymorphism and asthma risk among the overall populations in the dominant model, using a random-effect model. The squares and horizontal lines correspond to the odds ratios (ORs) and 95 % confidence interval (CI) in each study. The black diamond represents the pooled OR and 95 % CI.

Full size image
Table 3 Meta-Analyses of the Associations Between CD14-159C/T Polymorphism and Asthma Risk
Full size table

Stratified Analyses by Ethnicity, Age, and Atopic Status in the Overall Populations

For the subgroup analysis by ethnicity, no association was found between CD14-159C/T polymorphism and asthma risk among Caucasians (OR = 0.91, 95 % CI = 0.77–1.06, P = 0.22) or Asians (OR = 0.87, 95 % CI = 0.68–1.10, P = 0.25) in the dominant model; there were no significant results in the other models (Table 3). Stratification by age indicated no significant associations among adults (TT + TC vs. CC OR = 0.95, 95 % CI = 0.77–1.17, P = 0.63) or children (TT + TC vs. CC OR = 0.99, 95 % CI = 0.88–1.11, P = 0.81) in the dominant or other models (Table 3). The subgroup analysis by atopic status revealed no associations of this polymorphism with asthma risk among the atopic asthma patients, non-atopic asthma patients, or mixed atopic asthma patients in any model (Table 3).

Stratified Analyses in the Atopic Asthma Populations

Twenty eligible studies (3642 cases and 3806 controls) performed among the atopic asthma populations were identified for data analysis. Stratified analyses were conducted according to ethnicity and age. No matter which genetic comparison model we used, there were still no significantly statistical results in any subgroup analysis. Summary of the meta-analysis results is listed in Table 3.

Sensitivity Analysis

In the overall and stratified analyses, the conclusions of non-significance were stable in all the genetic models when removing studies that deviated from HWE. We did not sequentially exclude a single study each time because no statistically significant results were observed.

Heterogeneity and Publication Bias

Between-study heterogeneity was observed during most of the meta-analyses in the overall and subgroup populations, which was indicated by I 2 values (Table 3). We used Begg’s funnel plots to graphically estimate the publication bias, and found the relatively symmetric distribution of funnel plots in the dominant (Fig. 3) and other comparison models (figures not shown), which suggested the absence of publication bias. Moreover, the results of Egger’s test also indicated lack of publication bias in the dominant (P Egger = 0.116), recessive (P Egger = 0.774), codominant (P Egger = 0.524), or allelic model (P Egger = 0.274).

Fig. 3
figure 3

Begg’s funnel plot of CD14-159C/T polymorphism and asthma risk in the dominant model (TT + TC vs. CC).

Full size image

DISCUSSION

Although multifactors contribute to asthma susceptibility, genetic predisposition is regarded as one of the important determinants. Many promising candidate genes for asthma have been identified on chromosome 5q31.1 [48], and CD14 gene is such one. Variants in the promoter region of the CD14 gene may modify the structure of the CD14 protein and influence the CD14-LPS interaction [19]. Baldini et al. first reported the −159C/T polymorphism in the promoter region of CD14 gene [49], which was found to be associated with altered levels of soluble CD14 and IgE in various ethnic patients with asthma [15]. The above facts indicate that the CD14-159C/T polymorphism might be involved in the pathogenesis of asthma.

In order to investigate whether the CD14-159C/T polymorphism is associated with asthma risk, we have performed the current meta-analysis with 11 additional studies compared with previous meta-analyses [1, 9, 10]. The overall analysis and stratified analyses by ethnicity and age revealed no associations between this polymorphism and asthma risk. The findings were consistent with those in the previous meta-analyses. Maybe, the influence of CD14-159C/T polymorphism on soluble CD14 and IgE is not powerful enough so that the relationship between this polymorphism and asthma risk could not be detected. Another plausible explanation is that the solitary CD14-159C/T polymorphism merely influences the production of CD14 protein but does not play a key role in the subsequent pathway to asthma development.

When performing the stratified analysis by atopic status, there were still no significant results. However, the previous meta-analyses found a protective effect of the T-allele of −159C/T polymorphism for atopic asthma among the overall populations [1], Asians, and children [9]. Those results of those meta-analyses should be interpreted with caution because of a relatively small number of studies. Compared to previous meta-analyses [1, 9], the study inclusion of our meta-analysis is more comprehensive due to the largest sample size, suggesting more reliable results. Probably, this discrepancy reflects spurious relationship caused by the relatively small sample size. We speculate that the CD14-159C/T polymorphism contribute little to risk of atopic asthma. More genetic association studies concerning the relationship between this polymorphism and atopic asthma are warranted to validate our findings.

Heterogeneity is a major issue that needs to be mentioned when interpreting the meta-analysis results. Although we performed a meticulous literature search, precise data extraction, and strict data analysis, there was low to moderate level heterogeneity in the overall populations (Table 3). After performing subgroup analyses, the heterogeneity was not effectively reduced or removed in most of genetic models, indicating that ethnicity, age, or atopic status might not be the source of heterogeneity. The heterogeneity, together with the stable negative results in sensitivity analysis, probably reflects the unclear association of this polymorphism with asthma risk.

Some limitations of the meta-analysis should be considered. First, although no significant publication bias was detected, some relevant studies published in other databases or unpublished negative studies may have been missed. Thus, potential bias could not be excluded. Second, our ethnic-specific meta-analyses were performed in Asians and Caucasians, so our results are applicable only to these ethnic groups. Our findings should be optimized by including other ethnic populations such as Africans and Latinos. Third, data could not be stratified by disease severity, sex, environmental factors, smoking status, or other variables attributing to insufficient original information in the included studies. Fourth, the lack of corresponding data did not allow us to conduct a meta-analysis of the gene-gene or gene-environment interactions. For these reasons, the interpretation of our meta-analysis results should be taken carefully.

In summary, the current study is the most comprehensive and latest meta-analysis concerning the CD14-159C/T polymorphism and asthma risk to date. Our results indicate that there is no evidence of significant association between the CD14-159C/T polymorphism and asthma risk either in Asians or Caucasians or adults or children. Moreover, the negative results are not influenced by atopic status. Due to the limitations showed above, our results should be viewed with caution and future large-scale studies with different ethnic populations and clinical subphenotypes are required for clarifying the role of the CD14-159C/T polymorphism in the pathogenesis of asthma.