Table 1. Characteristics of studies included in the meta-analysis.
| Study | Year | Country | Study Design | Diagnosticcriteria | Sample size | Matched factors | TNF-α Level | Methods | PCOS | Control | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| PCOS | control | ||||||||||
| Gonzalez[61] | 1999 | USA | Case-control | NIH | 12 | 20 | Age,BMI | ↑ | ELISA | 3.31±0.87 | 0.91±0.54 |
| Gonzalez[61] | 1999 | USA | Case-control | NIH | 22 | 20 | Age,BMI | NS | ELISA | 4.11±0.89 | 4.65±1.47 |
| Escobar-Morreale[16] | 2001 | Spain | Case-control | NIH | 21 | 27 | BMI | NS | CL | 7.3±3.7 | 5.6±4.0 |
| Araya[8] | 2002 | Chile | Case-control | NIH | 16 | 11 | Age | ↑ | CL | 6.73±2.34 | 4.82±1.15 |
| Escobar-Morreale[17] | 2003 | Spain | Case-control | NIH | 35 | 28 | BMI,Smoking | NS | CL | 3.74±1.18 | 3.53±1.32 |
| Sayin[9] | 2003 | Turkey | Case-control | NIH | 21 | 14 | Age,BMI | ↑ | CL | 23.67±25.57 | 7.58±6.66 |
| Tarkun[10] | 2006 | Turkey | Case-control | Rotterdam | 32 | 25 | Age,BMI,WHR | ↑ | CL | 5.27±1.59 | 4.48±0.81 |
| Vgontzas[18] | 2006 | USA | Case-control | NIH | 42 | 17 | BMI | NS | ELISA | 4.05±1.94 | 3.79±0.82 |
| Moran[19] | 2007 | Australia | Case-control | Rotterdam | 15 | 17 | BMI,Smoking | NS | ELISA | 6.0±5.1 | 5.8±3.5 |
| Olszanecka[20] | 2007 | Poland | Case-control | NIH | 39 | 34 | Age,BMI,Smoking | NS | ELISA | 5.5±2.0 | 6.8±3.5 |
| Jakubowska[29] | 2008 | Poland | Case-control | Rotterdam | 29 | 29 | Age,BMI,WHR | ↓ | RIA | 10.69±13.14 | 14.95±13.3 |
| Arika[21] | 2009 | Turkey | Case-control | Rotterdam | 39 | 30 | Age,BMI,WC | NS | CL | 11.52±5.68 | 13.84±10.72 |
| Samy[62] | 2009 | Egypt | Case-control | Rotterdam | 56 | 35 | Age,BMI | NS | ELISA | 3.72±1.26 | 3.66±1.02 |
| Samy[62] | 2009 | Egypt | Case-control | Rotterdam | 52 | 40 | Age,BMI | ↑ | ELISA | 6.87±1.12 | 3.76±1.04 |
| Soares[22] | 2009 | Brazil | Case-control | Rotterdam | 40 | 50 | Age,BMI,Waist | NS | CL | 10.08±7.38 | 12.25±6.54 |
| Ilie[23] | 2011 | Romania | Case-control | AES | 45 | 32 | Age,BMI | NS | ELISA | 8.2±5.49 | 7.26±2.54 |
| Victor[11] | 2011 | Spain | Case-control | Rotterdam | 39 | 43 | Age,BMI,WC | ↑ | MBIA | 5.1±0.9 | 3.2±1.8 |
| Xiong[12] | 2011 | China | Case-control | Rotterdam | 86 | 50 | Age | ↑ | CL | 2.312±1.762 | 1.751±1.725 |
| Choi[13] | 2012 | Korea | Case-control | Rotterdam | 37 | 33 | BMI,smoking | ↑ | ELISA | 3.19±1.19 | 1.79±0.78 |
| Wang[24] | 2012 | China | Case-control | Rotterdam | 35 | 35 | Age,BMI,WC | NS | ELISA | 4.1±0.3 | 3.1±0.2 |
| Lee[25] | 2013 | Korea | Case-control | NIH | 20 | 20 | Age,BMI | NS | RIA | 1.06±0.47 | 1.19±0.88 |
| Lee[25] | 2013 | Korea | Case-control | NIH | 20 | 20 | Age,BMI | NS | RIA | 0.98±0.52 | 1.05±0.59 |
| Li[14] | 2013 | China | Case-control | NIH | 16 | 18 | Age,BMI,WC | ↑ | ELISA | 68.66±35.92 | 30.69±16.67 |
| Pawelczak[26] | 2014 | USA | Case-control | Rotterdam | 23 | 12 | Age,BMI | NS | MBIA | 7.4±4.08 | 4.8±3.16 |
| Thathapudi[15] | 2014 | India | Case-control | AES | 204 | 204 | Age | ↑ | ELISA | 13.24±10 | 5.5±3.8 |
| Agacayak[27] | 2015 | Turkey | Case-control | Rotterdam | 15 | 15 | Age,BMI | NS | ELISA | 313±248 | 294±292.2 |
| Agacayak[27] | 2015 | Turkey | Case-control | Rotterdam | 15 | 15 | Age,BMI | NS | ELISA | 212±242.1 | 214.5±233 |
| souza[28] | 2015 | Turkey | Case-control | Rotterdam | 8 | 10 | Age,BMI | NS | CL | 7.65±4.83 | 6.37±1.38 |
| souza[28] | 2015 | Turkey | Case-control | Rotterdam | 12 | 10 | Age,BMI | NS | CL | 10.47±6.92 | 8.11±3 |
NIH, National Institutes of Health; AES, Androgen Excess Society; BMI, body mass index; WC, Waist circumference; WHR, Waist-to-Hip Ratio, NS, no significant difference; ELISA, enzyme-linked immunosorbent assay; CL, chemiluminescence; RIA, radioimmunoassay; MBIA, multiplexing bead immunoassay