Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Editorial Board
Literature Review
Narrative review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Editorial Board
Literature Review
Narrative review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
View/Download PDF

Translate this page into:

15 (
4
); 945-951
doi:
10.1016/j.jor.2018.08.033

Association of GDF-5 rs143383 polymorphism with radiographic defined knee osteoarthritis: A systematic review and meta-analysis

, , , , , , ,
a Department of Radiology, Shahid Sadoughi University of Medical Sciences, Yazd, Iran
b Department of Orthopedics, Shahid Sadoughi University of Medical Sciences, Yazd, Iran
c Department of Emergency Medicine, Shahid Sadoughi University of Medical Sciences, Yazd, Iran
d Department of Biology, Science and Art University, Yazd, Iran
e Department of Anesthesiology, Shahid Sadoughi University of Medical Sciences, Yazd, Iran
f Department of Medical Genetics, Shahid Sadoughi University of Medical Sciences, Yazd, Iran

∗Corresponding author: Mohammad Reza Sobhan. mrsobhanardakani@gmail.com

Received: , Accepted: ,
Prof. PK Surendran Memorial Education Foundation
Disclaimer:
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.

Abstract

Abstract

To assess the association of GDF-5 rs143383 polymorphism with radiographic defined knee osteoarthritis (OA), a systematic review and meta-analysis was conducted.

A total of 17 relevant case-control studies with 7424 cases and 11,310 controls was collected from several electronic databases up to June 2018.

The pooled results showed that GDF-5 rs143383 polymorphism was significantly associated with radiographic defined knee OA in overall and stratified analysis by ethnicity, source of controls and genotyping techniques.

The GDF-5 rs143383 polymorphism might be used as a relevant risk estimate for radiographic defined Knee OA.

Keywords

Radiographic
Knee
Osteoarthritis
GDF-5 gene
Meta-analysis
1

1 Introduction

Osteoarthritis (OA), a severe joint disease, is one of the main leading causes of restricted activity and disability worldwide.1–3 In addition, persistent pain due to knee OA can be lead to a loss of function, reduced quality of life and increasingly heavy economic burden on social welfare and health care systems.4,5 Knee OA is a disease caused by multiple interrelated factors (biomedical and biomechanical) that contribute to a cascade leading to joint degeneration. Knee OA is a multi-factorial disease.6 Female sex, age, obesity, Knee Injury, heavy physical activity and occupational load are important risk factors for the incidence of knee OA.7

Several sets of diagnostic criteria have been proposed for knee OA. They usually rest upon radiographic findings, physical findings, clinical findings, or a combination of these.8 For example, to best define the presence of knee OA, the American College of Rheumatology (ACR) criteria combine outcomes from radiographs and physical examination.9 Radiographic assessments have long been considered as the primary method for imaging in the diagnosis and severity assessment of knee OA.10 The advantages of radiography are cost-effectiveness, safety, and excellent availability. However, it is rather insensitive at detecting early signs of knee OA. Moreover, there are several radiological classification systems for assessing the severity of knee OA, including those by Ahlbäck (1968), Nagaosa (2000), and Kellgren and Lawrence (K&L) (1957).11 For the classification of the structural changes associated with knee OA, K&L composed criteria for a 5-point grading scale using radiographic features. These K&L grades range from 0 (no OA) to 4 (severe OA), with grades ≥2 defined as definite knee OA.12

Increasing evidence has suggested that pain is poorly correlated with symptoms of radiologic knee OA and most risk factors for radiologic knee OA are not strong predictors of knee pain.13 A systematic review revealed that in 15–81% of patents the radiological knee OA has been associated with pain symptoms.14 Moreover, the radiographic severity of knee OA has been reported to have weak or no significant association with disability in knee OA patients. There is a large body of evidence demonstrating the role of genetic as a risk factor for knee OA. Though a number of genetic epidemiological studies have been conducted to explore the association of different single nucleotide polymorphisms with knee OA, there is no study evaluated the association of genetic polymorphisms with radiologic defined knee OA. Therefore, we have performed a meta-analysis to elucidate the association of growth and differentiation factor 5 (GDF-5) rs143383 polymorphism with radiographic knee OA.

2

2 Materials and methods

2.1

2.1 Search strategy

To identify all articles that examined the association of GDF-5 rs143383 polymorphism with radiographic knee OA and TKR, we conducted an electronic search in the PubMed, EMBASE, CNKI (China National Knowledge Infrastructure) and Chinese Biomedicine databases up to June 2018 using the combination of following MeSH terms and keywords (‘’knee osteoarthritis'’ OR ‘’radiographic knee osteoarthritis'’ OR ‘’RKOA″) AND (‘’Growth differentiation factor 5″ OR ‘’GDF-5″ OR ‘’BMP-14″ OR ‘’rs143383″ OR ‘’+104 T > C″) AND (‘’SNPs'’ OR ‘’polymorphism’’ OR ‘’genotype’ OR ‘’allele’’ OR ‘’variation’’). In addition, we also reviewed the references list of relevant reviews and eligible publications to find other potentially sources. The search was limited to human studies and English language published studies.

2.2

2.2 Inclusion criteria

All of the included original studies were selected according to the following selection criteria. Inclusion criteria: (1) case-control, family based study, linkage study, or cohort studies; (2) studies evaluating the association of GDF-5 rs143383 polymorphism with radiologic defined knee OA; (3) studies with sufficient data of genotype or allele frequency for examining an Odds Ratios (ORs) with 95% Confidence Intervals (CIs). Accordingly, the following exclusion criteria were also utilized: (1) not a case-control study or a cohort study; (2) case only and/or no controls; (3) no sufficient data reported; (4) abstracts, comments, case reports, letters, reviews, meta-analysis, and (5) duplicates of previous publications. If more than one article was published by the same authors using the same case series, the studies with the largest sample size or the most recently published case-control study was selected. In addition, in this meta-analysis different case-control groups in one publication were considered as independent case-control studies.

2.3

2.3 Data extraction

Two authors independently and carefully extracted data from all eligible studies according to the inclusion criteria listed above. Disagreements were resolved by discussion between the two investigators. If the two authors could not reach a consensus, then a third author was consulted to resolve the disagreements and a final decision was made. For each included study, the following information was collected: first author, year of publication, country, ethnicity, number of cases and controls, genotyping techniques, and p-value for Hardy–Weinberg equilibrium (HWE).

2.4

2.4 Statistical analysis

The strength of the association of GDF-5 rs143383 polymorphism with radiographic knee OA were estimated using ORs, with the corresponding 95% CIs. The significance of pooled ORs was tested by Z-test, in which P < 0.05 was considered significant. The risks (ORs) of knee OA and TKR associated with the GDF-5 rs143383 polymorphism were estimated for each study under five genetic models including allele model (C vs. G), homozygous model (CC vs. GG), heterozygous model (CC vs. GC), dominant model (CC + GC vs. GG), and recessive model (CC vs. GG + GC). The Q-statistic and the I2 statistic were used to assess between study heterogeneity in the meta-analysis. In addition, we used the I2 statics to quantify the between study heterogeneity, which ranges from 0 to 100% and represents the proportion of between-study variability attributable to heterogeneity rather than chance. I2 values of 25%, 50%, and 75% were nominally considered low, moderate, and high estimates, respectively.15,16 A p-value of less than 0.05 for the Q-statistic indicated presence of between study heterogeneity, so that the pooled OR estimate of each study was calculated by the random-effects model (the DerSimonian and Laird method). Otherwise, the fixed-effects model (the Mantel–Haenszel method) was used. For each study, we examined whether the genotype distribution in control groups was in agreement with Hardy-Weinberg Equilibrium (HWE) using the x2 test. One-way sensitivity analysis, by which a single study in the meta-analysis was omitted each time to reflect the influence of the individual data set for the pooled OR, was carried out to assess the stability of the results. Moreover, sensitivity analysis was performed by deletion those studies did not in agreement with HWE. To detect the presence of potential publication bias, visual inspection of Begg's funnel plot symmetry was used in which an asymmetric plot suggests a possible publication bias. In addition, Egger's linear regression test which measures funnel plot asymmetry using a natural logarithm scale of OR was also used to evaluate the publication biases statistically.17,18 When publication bias existed, the Duval and Tweedie non-parametric “trim and fill” method was used to adjust for it. All statistical tests were performed using Comprehensive Meta-Analysis (CMA) software version 2.0 (Biostat, USA). All P values in the meta-analysis were 2-sided, and P values less than 0.05 were considered significant, except for tests of between study heterogeneity where a level of 0.10 was used.

3

3 Results

3.1

3.1 Characteristics of studies

The literature search and study selection procedures are shown in Fig. 1. There were 170 studies retrieved through electronic database and hand search on the basis of the above criteria. Of those, 73 publication were excluded after title and abstract review. Of the remaining 97 studies, we further removed 80 articles due to reviews articles, case reports, case-only design, and insufficient data to calculate ORs. Finally, 17 case-control studies in 14 publications19–32 with 7424 cases and 11,310 controls were included in the meta-analysis. The detailed characteristics of the included studies were summarized in Table 1. All studies were case-control studies, the sample sizes of radiographic knee OA cases ranged from 50 to 933, and publication dates ranged 2007 to 2017. Among the 17 case-control studies, there were eight studies of Caucasians, eight studies of Asians, and one study of Africans. There were seven hospital-based studies, eight population based studies and two studies not stated. Different genotyping techniques were used, including TaqMan, allele-specific PCR (AS-PCR), restriction fragment length polymorphism (RFLP-PCR), mass spectrometry, direct sequencing, real-time PCR, and high resolution melt (HRM) analysis. The distributions of genotype and allele frequency in the cases and controls are presented in Table 1. The distribution of the genotypes in the control group was consistent with Hardy-Weinberg equilibrium (HWE), except for four case-control studies.

Studies identification, inclusion and exclusion.
Fig. 1 Studies identification, inclusion and exclusion.
Table 1 Characteristics of studies included in the meta-analysis.
First Author Country (Ethnicity) SOC Genotyping Technique Case/Control Cases Controls HWE
Genotypes Allele Genotypes Allele
TT TC CC T C TT TC CC T C
Miyamoto 200719 Japan(Asian) HB TaqMan 313/485 197 97 19 491 135 473 330 58 681 289 0.965
Miyamoto 200719 China(Asian) HB TaqMan 718/861 444 243 31 1131 305 244 193 48 1276 446 0.040
Tsezou 200720 Greece(Caucasian) HB DS 251/267 95 126 30 316 186 99 125 44 323 213 0.668
Yao 200825 China(Asian) PB RT 298/452 189 93 16 471 125 232 182 38 646 258 0.344
Chapman 200826 Netherland(Caucasian) PB MS 142/724 54 72 16 180 104 289 331 104 909 539 0.558
Vaes 200927 Netherland(Caucasian) PB TaqMan 667/2097 276 298 93 850 484 752 1014 331 2518 1676 0.723
Valdes 200928 UK(Caucasian) PB AS-PCR 259/509 126 98 35 350 168 181 244 84 606 412 0.907
Takahashi 201029 Japan(Asian) NS TaqMan 933/1225 566 313 54 1445 421 684 461 80 1829 621 0.844
Valdes 201130 Estonia(Caucasian) PB AS-PCR 65/427 32 24 9 88 42 168 179 80 515 339 0.010
Valdes 201127 Netherland(Caucasian) HB AS-PCR 867/758 413 361 93 1187 547 294 354 110 942 574 0.836
Valdes 201127 UK(Caucasian) PB AS-PCR 1141/536 467 511 163 1445 837 219 237 80 675 397 0.229
Tawonsawatruk 201131 Thailand(Asian) HB PCR-RFLP 103/103 35 41 11 117 63 33 47 23 113 93 0.424
Shin 201232 Korea(Asian) PB HRMA 725/1737 382 305 38 1059 381 942 689 105 2573 901 0.154
Mishra 201321 India(Asian) HB PCR-RFLP 300/300 124 130 46 378 222 54 160 56 328 272 0.002
Sabah-Ozcan 201722 Turkey(Caucasian) HB PCR-RFLP 94/279 37 43 14 117 71 74 153 52 301 257 0.083
Mishra 201723 India(Asian) NS PCR-RFLP 500/500 199 226 75 624 376 131 272 97 534 466 0.037
Elazeem 201724 Egypt(African) PB TaqMan 50/50 20 14 16 56 44 12 13 25 49 51 0.001
3.2

3.2 Quantitative data synthesis

3.2.1

3.2.1 Overall estimations

The results of meta-analysis of GDF-5 rs143383 polymorphism was listed in Table 2. When all the eligible studies were pooled into the meta-analysis of GDF-5 rs143383 polymorphism, a significant association was found with radiographic knee OA in overall population under all five genetic models, i.e., allele (C vs. T: OR = 0.792, 95% CI 0.732–0.857, p ≤ 0.001), homozygote (CC vs. TT: OR = 0.634, 95% CI 0.544–0.739, p ≤ 0.001, Fig. 2A), heterozygote (CT vs. TT: OR = 1.144, 95% CI 1.029–1.272, p = 0.013, Fig. 2B), dominant (CC + CT vs. TT: OR = 0.682, 95% CI 0.541–0.860, p = 0.001) and recessive (CC vs. CT + TT: OR = 0.772, 95% CI 0.999–0.853, p ≤ 0.001).

Table 2 Results of the association of the GDF-5 rs143383 polymorphism with radiographic defined Knee OA.
Subgroup Genetic Model Type of Model Heterogeneity Odds Ratio Publication Bias
I2 (%) PH OR 95% CI Ztest POR PBeggs PEggers
Overall C vs. T Random 58.36 0.001 0.792 0.732–0.857 −5.828 ≤0.001 0.174 0.048
CC vs. TT Random 44.45 0.025 0.634 0.544–0.739 −5.824 ≤0.001 0.127 0.065
CT vs. TT Fixed 0.00 0.809 1.144 1.029–1.272 2.489 0.013 0.149 0.090
CC + CT vs. TT Random 91.80 0.00 0.682 0.541–0.860 −3.237 0.001 0.266 0.114
CC vs. CT + TT Fixed 0.00 0.759 0.772 0.999–0.853 −5.064 ≤0.001 0.004 0.002
By Ethnicity
Caucasians C vs. T Fixed 39.69 0.114 0.835 0.760–0.917 −3.761 ≤0.001 0.901 0.548
CC vs. TT Fixed 0.00 0.504 0.727 0.630–0.839 −4.369 ≤0.001 0.710 0.355
CT vs. TT Fixed 0.00 0.906 1.122 0.974–1.292 1.598 0.110 0.386 0.338
CC + CT vs. TT Random 57.25 0.022 0.787 0.673–0.920 −3.004 0.003 1.000 0.716
CC vs. CT + TT Fixed 0.00 0.895 0.810 0.709–0.925 −3.107 0.002 0.386 0.200
Asian C vs. T Random 72.36 0.001 0.749 0.656–0.856 −4.266 ≤0.001 0.009 0.012
CC vs. TT Random 60.75 0.013 0.568 0.432–0.745 −4.072 ≤0.001 0.710 0.318
CT vs. TT Fixed 7.36 0.373 1.161 0.987–1.366 1.807 0.071 0.536 0.337
CC + CT vs. TT Random 96.05 ≤0.001 0.612 0.390–0.960 −2.140 0.032 0.265 0.123
CC vs. CT + TT Fixed 0.00 0.469 0.737 0.632–0.860 −3.867 ≤0.001 0.063 0.018
By Technique
PCR-RFLP C vs. T Fixed 0.00 0.979 0.538 0.448–0.646 −6.638 ≤0.001 0.734 0.555
CC vs. TT Fixed 0.00 0.669 0.682 0.602–0.772 −6.057 ≤0.001 0.734 0.384
CT vs. TT Fixed 0.00 0.696 0.459 0.353–0.598 −5.791 ≤0.001 0.734 0.853
CC + CT vs. TT Fixed 0.00 0.644 1.094 0.856–1.398 0.719 0.472 0.734 0.352
CC vs. CT + TT Fixed 0.00 0.546 0.716 0.568–0.903 −2.826 0.005 0.734 0.357
AS-PCR C vs. T Random 66.68 0.029 0.803 0.674–0.956 −2.468 0.014 1.000 0.596
CC vs. TT Fixed 43.24 0.152 0.701 0.533–0.923 −2.535 0.011 1.000 0.613
CT vs. TT Fixed 0.00 0.861 1.101 0.908–1.335 0.983 0.326 1.000 0.889
CC + CT vs. TT Random 71.13 0.016 0.738 0.571–0.954 −2.320 0.020 1.000 0.603
CC vs. CT + TT Fixed 0.00 0.545 0.809 0.675–0.969 −2.307 0.021 0.734 0.683
TaqMan C vs. T Fixed 30.77 0.228 0.792 0.723–0.868 −4.986 ≤0.001 0.734 0.431
CC vs. TT Random 67.45 0.027 0.569 0.367–0.882 −2.520 0.012 0.734 0.415
CT vs. TT Fixed 51.95 0.100 1.173 0.951–1.447 1.494 0.135 0.734 0.546
CC + CT vs. TT Random 97.98 ≤0.001 0.550 0.207–1.462 −1.198 0.231 0.734 0.578
CC vs. CT + TT Fixed 42.98 0.154 0.753 0.619–0.917 −2.821 0.005 0.308 0.116
By Source of Control
HB C vs. T Fixed 3.68 0.398 0.739 0.682–0.802 −7.284 ≤0.001 0.367 0.256
CC vs. TT Fixed 33.36 0.173 0.532 0.441–0.641 −6.616 ≤0.001 0.763 0.732
CT vs. TT Fixed 14.44 0.320 1.253 1.043–1.506 2.408 0.016 1.000 0.733
CC + CT vs. TT Random 96.15 ≤0.001 0.587 0.323–1.064 −1.756 0.079 0.763 0.369
CC vs. CT + TT Fixed 0.00 0.660 0.682 0.573–0.812 −4.314 ≤0.001 0.229 0.245
PB C vs. T Random 59.15 0.017 0.855 0.760–0.962 −2.605 0.009 0.536 0.207
CC vs. TT Fixed 9.29 0.358 0.767 0.659–0.892 −3.439 0.001 0.063 0.060
CT vs. TT Fixed 0.00 0.939 1.112 0.956–1.293 1.376 0.169 0.173 0.054
CC + CT vs. TT Random 71.21 0.001 0.804 0.666–0.970 −2.272 0.023 0.265 0.199
CC vs. CT + TT Fixed 0.00 0.765 0.829 0.720–0.956 −2.585 0.010 0.004 0.005
Forest plots for the GDF-5 rs143383 polymorphism with radiographic knee OA. A: the homozygote model (CC vs. TT); B: the heterozygote model (CT vs. TT).
Fig. 2 Forest plots for the GDF-5 rs143383 polymorphism with radiographic knee OA. A: the homozygote model (CC vs. TT); B: the heterozygote model (CT vs. TT).
3.2.2

3.2.2 Subgroup analyses

Table 2 lists the subgroup analyses results of this meta-analysis. Eight case-control studies with 3486 cases and 5597 controls were used to investigate the association of GDF-5 rs143383 polymorphism with radiographic knee OA in Caucasians. The results showed that the GDF-5 rs143383 polymorphism was associated with radiographic knee OA in Caucasians under four genetic models, i.e., allele (C vs. T: OR = 0.835, 95% CI 0.760–0.917, p ≤ 0.001), homozygote (CC vs. TT: OR = 0.727, 95% CI 0.630–0.839, p ≤ 0.001), dominant (CC + CT vs. TT: OR = 0.787, 95% CI 0.673–0.920, p = 0.003) and recessive (CC vs. CT + TT: OR = 0.810, 95% CI 0.709–0.925, p = 0.002). Similarly, a total of eight case-control studies with 3890 cases and 5663 controls were used to investigate the association of GDF-5 rs143383 polymorphism with radiographic knee OA in Asians. The results showed that the GDF-5 rs143383 polymorphism was associated with radiographic knee OA in Asians under four genetic models, i.e., allele (C vs. T: OR = 0.835, 95% CI 0.656–0.856, p ≤ 0.001), homozygote (CC vs. TT: OR = 0.727, 95% CI 0.432–0.745, p ≤ 0.001), dominant (CC + CT vs. TT: OR = 0.612, 95% 0.390–0.960, p ≤ 0.001) and recessive (CC vs. CT + TT: OR = 0.737, 95% 0.632–0.860, p ≤ 0.001).

The studies were further stratified on the basis of genotyping technique and source of control subjects. In the PCR-RFLP, AS-PCR and TaqMan group, a significant association between GDF-5 rs143383 polymorphism and radiographic knee OA were found (Table 2). Moreover, subgroup analysis based on the source of controls showed that there was a significant association between GDF-5 rs143383 polymorphism and radiographic knee OA in population based and hospital based studies (Table 2).

3.3

3.3 Heterogeneity and sensitivity analysis

There was obvious between study heterogeneity under three genetic models, i.e., allele (I2 = 58.36%, Phet = 0.001), homozygote (I2 = 44.45%, Phet = 0.025), and recessive (I2 = 91.80%, Phetp≤0.001). Thus, a subgroup analysis by ethnicity, genotyping method and source of controls was conducted to assess the source of heterogeneity. Heterogeneity dramatically decreased or disappeared when stratification analyses for Caucasians, PCR-RFLP and hospital based studies was conducted (Table 2). Moreover, we performed sensitivity analyses to assess the stability of our results, namely, a single study in the meta-analysis was omitted each time to reflect the influence of the individual study to the pooled OR. The sensitivity analysis indicated that no single study influenced the pooled OR and also between studies heterogeneity, suggesting that the results of this meta-analysis are stable. In addition, we excluded those HWE-violating studies for sensitivity analyses. However, the pooled ORs in overall was not statistically altered, indicating that the results was stable (data not shown).

3.4

3.4 Publication bias

The Begg's funnel plot and Egger's test were performed to examine the publication bias in this meta-analysis. There was evidence of significant publication bias two genetic models allele (PBegg's = 0.174 and PEgger's = 0.048, Fig. 3) and recessive (PBegg's = 0.004 and PEgger's = 0.002). This might be due to that the present meta-analysis was limited to radiographic knee OA published studies. Therefore, we have performed the Duval and Tweedie non-parametric ‘‘trim and fill’’ method to adjust the publication bias test results for the allele and recessive models. However, the results did not change GDF-5 rs143383 polymorphism with radiographic knee OA (data not shown), indicating that our results were statistically robust (Fig. 3).

Begg's funnel plot for publication bias (blue) and the Duval and Tweedie nonparametric “trim and fill” method's funnel plot funnel plot (red) for the meta-analysis of GDF-5 rs143383 polymorphism with radiographic knee OA under the allele genetic model (C vs. T).
Fig. 3 Begg's funnel plot for publication bias (blue) and the Duval and Tweedie nonparametric “trim and fill” method's funnel plot funnel plot (red) for the meta-analysis of GDF-5 rs143383 polymorphism with radiographic knee OA under the allele genetic model (C vs. T).
4

4 Discussion

Human GDF-5 gene (also called BMP-14), a well-characterized member of the transforming growth factor beta (TGF-β) signaling pathway, is promotes chondrogenesis and chondrocyte proliferation. Mutations in TGF-β superfamily members can lead to developmental disorders. GDF-5 is essential for joint formation during digit development and proper formation of adult nuclei pulposi.33 It has been demonstrated that the common polymorphisms in the GDF-5 gene contribute to variation in human height.31 Moreover, mutations in the GDF-5 gene result in severe chondrodysplasia and skeletal dysplasia such as brachydactyly type C and proximal symphalangism,34 while a milder variant that influences GDF-5 expression levels, results in a slightly elevated risk for OA.30 in 1996, the human GDF-5 gene is mapped to chromosome 20q11.2, contains two coding exons and spans approximately 4878 bp.35 The rs144383 (+104 T > C) C to T SNP located in the 5ʹUTR of GDF-5 gene, has been studied intensively.24,30

Previously published studies have indicated a significant association between GDF-5 rs143383 polymorphism and OA risk. However, the conclusions are still inconsistent. Furthermore, no meta-analysis has been evaluated the association of GDF-5 rs143383 polymorphism with radiographic defined knee OA. Therefore, our meta-analysis represents the first meta-analysis investigating the association of GDF-5 rs143383 polymorphism with radiographic knee OA. In the current meta-analysis, a total of 17 eligible case-control studies with 7424 radiographic defined knee OA cases and 11,310 controls were selected. In the total population, the pooled results of our meta-analysis indicated that there was an obviously significant association between the GDF-5 rs143383 polymorphism with radiographic knee OA under all five genetic models. In addition, subgroup analyses showed that the GDF-5 rs143383 polymorphism was significantly associated with radiographic defined knee OA among Caucasian and Asian populations.

Between studies heterogen7eity is a potential issue in a meta-analysis of genetic association studies.36,37 Ethnicity, sample size, study design, gender distribution, genotyping techniques, and so on, might be contribute to the between study heterogeneity.38–40 To explore the source of heterogeneity, subgroup analyses according to ethnicity, genotyping method and source of controls were conducted. The results showed the heterogeneity dramatically decreased or disappeared by subgroup analyses, suggesting that those studies on Caucasians, PCR-RFLP and hospital based was the major source of heterogeneity in the present meta-analysis. In addition, the problem of publication bias, which may influence the results of a meta-analysis, should also be explained. In this meta-analysis also there was a significant publication bias under two genetic models. It seems the limitation of search to the radiographic knee OA and English language published studies could bring bias in this meta-analysis. However, the results of ‘‘trim and fill’’ method did not adjust pooled ORs, indicating that our results were statistically reliable and robust.

The most important advantage of this meta-analysis is that this is the first meta-analysis to evaluate the association of GDF-5 rs143383 polymorphism with radiographic defined knee OA. Previous studies were all single case-control studies, which had limited statistical power due to the limited sample size. Meta-analysis is a very powerful tool for analyzing cumulative data of studies where the individual sample sizes are small and the statistical power is low. Therefore, the results of this meta-analysis were more robust and precise than that of previous individual case-control studies. However, several limitations may be considered for the interpretation of results. First, we performed a comprehensive search to obtain relevant studies through electronic databases, leading to a potential bias caused by missing of unpublished studies and the exclusion of one study which could not be included due to a lack of data. Second, predominantly studies on Caucasians and Asians were included in the present meta-analysis. Data on other ethnicities such as Africans and Latinos must be performed to evaluate the potential effects of ethnic variation on radiographic knee OA. Third, in the present meta-analysis a moderate to high heterogeneity was detected, which could influence the conclusions. However, when subgroup analysis was performed the heterogeneity obviously reduced or disappeared, suggesting that heterogeneity might be caused due to the differences in ethnicity background, genotyping method and source of controls. Fourth, Fourthly, in this meta-analysis publication bias existed because only published studies in English on radiographic knee OA were selected. Those studies not selected or published in other, which might contain other eligible studies and thus cause publication bias. Fifth, our analysis was based on ORs estimated without adjustment for several important confounding factors, such as age, gender, obesity, mechanical injury, severity of pain, and other factors owing to insufficient data in the primary reports, which are known to have significant effects on the development of knee OA. Thus, a more rigorous analysis should be assessed by stratifying other risk factors if data was available. Finally, a single gene polymorphism has only a moderate effect on knee OA development; hence, the GDF-5 rs143383 polymorphism may influence susceptibility of knee OA along with other polymorphisms in the GDF-5 gene, polymorphisms in the other genes, and environmental factors. However, enough data for further analysis is not available.

5

5 Conclusion

This systematic review and meta-analysis for the first time provided an overview on the association of GDF-5 rs143383 polymorphism with radiographic defined knee OA. Our meta-analysis showed that the GDF-5 rs143383 polymorphism is associated with radiographic defined knee OA in overall population and by ethnicity. Further studies with large sample size are needed to establish a more definitive conclusion. In addition, more studies should be performed in other ethnic groups such as African and Latinos populations are warranted to confirm this finding.

Conflicts of interest

The authors declare that they have no conflict of interest.

References

  1. , , , et al . Association of TNF-α-308 G > A and −238G > A polymorphisms with knee osteoarthritis risk: a case-control study and meta-analysis. J Orthop 2018
    [Google Scholar]
  2. , , , , , , . Association between aspartic acid repeat polymorphism of the asporin gene and risk of knee osteoarthritis: a systematic review and meta-analysis. Acta Orthop Traumatol Turcica. 2017;51(5)
    [Google Scholar]
  3. , , , et al . Association of ESRα gene Pvu II T > C, XbaI A > G and BtgI G > A polymorphisms with knee osteoarthritis susceptibility: a systematic review and meta-analysis based on 22 case-control studies. Archives of Bone and Joint Surgery. 2017;5(6)
    [Google Scholar]
  4. , . The epidemiology and impact of pain in osteoarthritis. Osteoarthritis Cartilage. 2013;21(9):1145-1153.
    [Google Scholar]
  5. , , , et al . Health economics in the field of osteoarthritis: an Expert's consensus paper from the European society for clinical and economic aspects of osteoporosis and osteoarthritis (ESCEO) Semin Arthritis Rheum. 2013;43(3):303-313.
    [Google Scholar]
  6. , , , et al . With a biomechanical treatment in knee osteoarthritis, less knee pain did not correlate with synovitis reduction. BMC Muscoskel Disord. 2017;18(1):347.
    [Google Scholar]
  7. , , , et al . Occupational Exposure to knee loading and the risk of osteoarthritis of the knee: a systematic review and a dose-response meta-analysis. Safety and Health at Work. 2017;8(2):130-142.
    [Google Scholar]
  8. , , , et al . Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and therapeutic criteria committee of the american rheumatism association. Arthritis Rheum. 1986;29(8):1039-1049.
    [Google Scholar]
  9. , , , et al . The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hand. Arthritis Rheum. 1990;33(11):1601-1610.
    [Google Scholar]
  10. , , , et al . Knee and hip osteoarthritis. current care guidelines. working group set up by the finish medical society duodecim and finish orthopedic association. 2012
    [Google Scholar]
  11. , , , , , . Radiographic osteoarthritis of the knee classified by the Ahlbäck and Kellgren & Lawrence systems for the tibiofemoral joint in people aged 35-54 years with chronic knee pain. Ann Rheum Dis. 1997;56(8):493-496.
    [Google Scholar]
  12. , , . Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957;16(4):494-502.
    [Google Scholar]
  13. , , , et al . Radiographic knee osteoarthritis and knee pain: cross-sectional study from five different racial/ethnic populations. Sci Rep. 2018;8(1):1364.
    [Google Scholar]
  14. , , . The discordance between clinical and radiographic knee osteoarthritis: a systematic search and summary of the literature. BMC Muscoskel Disord. 2008;9(1):116.
    [Google Scholar]
  15. , , , , , , . Association between Catechol-O-Methyltransferase Val158Met (158G/A) polymorphism and suicide susceptibility: a meta-analysis. J Res Health Sci. 2017;17(2)
    [Google Scholar]
  16. , , , , , . Association between MTHFR C677T polymorphism and risk of prostate cancer: evidence from 22 studies with 10,832 cases and 11,993 controls. Asian Pac J Cancer Prev APJCP. 2015;16(11)
    [Google Scholar]
  17. , , . Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994;50(4):1088-1101.
    [Google Scholar]
  18. , , , , . Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629-634.
    [Google Scholar]
  19. , , , et al . A functional polymorphism in the 5′ UTR of GDF5 is associated with susceptibility to osteoarthritis. Nat Genet. 2007;39(4):529-533.
    [Google Scholar]
  20. , , , , , . The growth differentiation factor 5 (GDF5) core promoter polymorphism is not associated with knee osteoarthritis in the Greek population. J Orthop Res. 2008;26(1):136-140.
    [Google Scholar]
  21. , , , , . AB0001Genetic polymorphism in GDF-5 gene as risk factor for development and progression of osteoarthritis. Ann Rheum Dis. 2013;71(Suppl 3)
    [Google Scholar]
  22. , , , et al . Polymorphisms in the growth differentiation factor 5 (GDF 5) gene in knee osteoarthritis. Journal of the College of Physicians and Surgeons-Pakistan : JCPSP. 2017;27(10):602-605.
    [Google Scholar]
  23. , , , , , . Expression of genes and their polymorphism influences the risk of knee osteoarthritis. J Nucleic Acids. 2017;2017:1-11.
    [Google Scholar]
  24. , , , . Genetic influence of growth and differentiation factor 5 gene polymorphism (+104T/C) on the development of knee osteoarthritis and its association with disease severity. European Journal of Rheumatology. 2017;4(2):98-103.
    [Google Scholar]
  25. , , . A single nucleid polymorphisms (SNP)in the 5′UTR of GDF5 is associated with knee osteoarthritis. Jiangsu Med J. 2008;34(3):1198-1199.
    [Google Scholar]
  26. , , , et al . A meta-analysis of European and Asian cohorts reveals a global role of a functional SNP in the 5’ UTR of GDF5 with osteoarthritis susceptibility. Hum Mol Genet. 2008;17(10):1497-1504.
    [Google Scholar]
  27. , , , et al . Genetic variation in the GDF5 region is associated with osteoarthritis, height, hip axis length and fracture risk: the Rotterdam study. Ann Rheum Dis. 2009;68(11):1754-1760.
    [Google Scholar]
  28. , , , , , , . Association of the DVWA and GDF5 polymorphisms with osteoarthritis in UK populations. Ann Rheum Dis. 2009;68(12):1916-1920.
    [Google Scholar]
  29. , , , , , , . Prediction model for knee osteoarthritis based on genetic and clinical information. Arthritis Res Ther. 2010;12(5)
    [Google Scholar]
  30. , , , et al . The GDF5 rs143383 polymorphism is associated with osteoarthritis of the knee with genome-wide statistical significance. Ann Rheum Dis. 2011;70(5):873-875.
    [Google Scholar]
  31. , , , , , , . A genetic association study between growth differentiation factor 5 (GDF 5) polymorphism and knee osteoarthritis in Thai population. J Orthop Surg Res. 2011;6(1):47.
    [Google Scholar]
  32. , , , et al . Genetic association analysis of GDF5 and ADAM12 for knee osteoarthritis. Joint Bone Spine. 2012;79(5):488-491.
    [Google Scholar]
  33. , , . Nuclei pulposi formation from the embryonic notochord occurs normally in GDF-5-deficient mice. Spine. 2011;36(24):E1555-E1561.
    [Google Scholar]
  34. , , , et al . Novel point mutations in GDF5 associated with two distinct limb malformations in Chinese: brachydactyly type C and proximal symphalangism. J Hum Genet. 2008;53(4):368-374.
    [Google Scholar]
  35. , , , , . Assignment of a new TGF-β superfamily member, human cartilage-derived morphogenetic Protein-1, to chromosome 20q11.2. Genomics. 1996;34(1):150-151.
    [Google Scholar]
  36. , , , , , , . Association between the p53 codon 72 polymorphism and primary open-angle glaucoma risk: meta-analysis based on 11 case-control studies. Indian J Ophthalmol. 2016;64(10):756-761.
    [Google Scholar]
  37. , , , et al . Association of XRCC2 rs3218536 polymorphism with susceptibility of breast and ovarian cancer: a systematic review and meta-analysis. Asian Pac J Cancer Prev APJCP. 2017;18(7)
    [Google Scholar]
  38. , , , et al . Association between thrombophilic genes polymorphisms and recurrent pregnancy loss susceptibility in the Iranian population: a systematic review and meta-analysis. Iran Biomed J. 2018;22(2):78-89.
    [Google Scholar]
  39. , , , et al . Association of Vitamin D receptor BsmI, TaqI, FokI, and ApaI polymorphisms with susceptibility of chronic periodontitis: a systematic review and meta-analysis based on 38 case-control studies. Dent Res J. 2018;15(3)
    [Google Scholar]
  40. , , , , , , . Lack of association of the fat mass and obesity associated (FTO) gene rs9939609 polymorphism with breast cancer risk: a systematic review and meta-analysis based on case - control studies. Asian Pac J Cancer Prev APJCP: APJCP.. 2017;18(4):1031-1037.
    [Google Scholar]

Fulltext Views
75

PDF downloads
8
View/Download PDF
Download Citations
BibTeX
RIS
Show Sections

Suggested read for related articles: