目的：探讨粒状头样2（grainyhead-like 2，GRHL2）基因多态性与噪声性听力损失（noise-induced hearing loss，NIHL）易感性的关系。方法： 使用病例对照研究方法，对某钢铁厂3 790名噪声暴露作业人员进行流行病学调查和听力学测试，通过相关标准和匹配条件，共确定286例病例作为病例组和286例作为对照组。根据《工作场所物理因素测量噪声》（GBZ/T189.8-2007）检测噪声暴露强度，抽取病例组和对照组人群空腹外周静脉血 2 mL，用于基因组 DNA抽提，通过中高通量SNP分型检测技术检测GRHL2基因的5个单核苷酸位点的多态性（single nucleotide polymorphisms，SNP）。分别采用t检验和卡方检验比较病例组和对照组之间计量资料和计数资料的差异，采用多元 Logistic 回归分析GRHL2基因的5个SNP与NIHL的关系，利用Haploview软件和Phase软件进行基因单体型的构建和分析。结果： 5个位点基因型频率分布符合Hardy-Weinberg平衡。在共显性模型下，调整累积噪声暴露量（cumulative noise exposure，CNE）、身高、吸烟、饮酒和血压状况后，与携带rs3735715位点GG基因型个体相比，GA基因型发生 NIHL的危险性降低，OR=0.644（95% CI:0.442～0.939，P=0.022）；通过对CNE进行分层分析发现，在CNE≥ 98 dB(A)组，rs3735715位点与NIHL的患病风险仍有关，与携带GG型人群相比，OR值为0.509（95% CI: 0.281～0.923，P=0.030）。单体型分析显示，四位点构成的单体型在病例组与对照组之间分布差异无统计学意义。结论： GRHL2基因单核苷酸多态性可能与NIHL易感性有关。

Objective： To investigate association between genetic polymorphism in the grainyhead-like 2 gene (GRHL2) and noise-induced hearing loss (NIHL) in the Chinese population. Methods: A matched case-control association study was employed, In which, 3 790 workers exposed to continuous and steady-state occupational noise in a steel factory participated. The questionnaires were adopted to collect individual features and audiometry tests performed. In the sstudy, 286 subjects were diagnosed as cases, Which were each designated on the basis of the matched criterion, and 286 paired samples were selected finally. Noise intensity was measured according to the standards given in ‘Measurement of Noise in the Workplace’(Occupational Health Standard of the People’s Republic of China, GBZ/T189.8-2007). Cumulative noise exposure (CNE) was calculated, according to monitoring data on A-weighed sound pressure level and employment time. Genomic DNA was obtained from peripheral blood samples using 2 mL DNA extraction Kit following the manufacturer’s protocol. Five single nucleotide polymorphisms (SNPs) of GRHL2 were genotyped by multiplex SNP genotyping kit. The continuous variables and categorical variables were analyzed by t-test and chi-square test respectively. Multivariate Logistic regression was used to test the association between genetic frequency and disease status, with adjustments for the possible confounding variables. The haplotypes were established and their frequencies in the two groups were assessed by haploview and phase softwares. Results: All the five SNPs (rs3735713, rs3824090, rs3735714, rs3735715 and rs611419) were in Hardy-Weinberg equilibrium (HWE) (P>0.05). The subjects carrying rs3735715 GG genotype had a higher NIHL risk than those carrying the GA genotype under the co-dominant model (OR=0.644, 95% CI: 0.442－0.939, P=0.022) after adjustment for height, blood pressure, drinking status and smoking status. After being stratified by CNE, in the CNE ≥ 98 dB (A) group, rs3735715 polymorphism was associated with the NIHL under the co-dominant model (OR=0.509, 95% CI: 0.281－0.923, P=0.026) after adjustment for height, blood pressure, drinking status and smoking status as well. However, no statistical significant difference was found in variant genotypes of the other SNPs between the case and control subjects. Four-locus (rs3735713, rs3824090, rs3735714 and rs3735715) haplotypes were constructed, and no risk or protective haplotypes was identified. Conclusion: It is suggested that GRHL2 polymorphisms may be associated with development of NIHL.