Table 1 Summary of findings of previously conducted meta-analyses

Cai et al. (2012) [27]

Slow NAT2 genotype (without wild-type NAT2*4 allele) significantly increases the risk of ATDILI

CYP2E1 c1/c1 genotype significantly increases the risk of ATDILI among East Asian patients. This result is non-significant for the wider population

GSTM1 null genotype significantly increases risk of ATDILI

GSTT1 null genotype has no effect on ATDILI risk

Cai et al. (2015) [31]

–

–

The GSTM1 present genotype is significantly associated with decreased risks of ATDILI and schizophrenia

The GSTT1 present genotype is significantly associated with a high risk of schizophrenia, but not with ATDILI according to the fixed-effect model

Deng et al. (2012) [32]

–

CYP2E1 c1/c1 genotype significantly increases the risk of ATDILI. When stratifying for study population, significant results are observed in Chinese and Korean populations

–

–

Du et al. (2013) [33]

Slow acetylator genotype of NAT2 (without wild-type NAT2*4 allele) significantly increases risk of ATDH. When stratifying for study population, this result is significant for East Asian, South Asian, Brazilian and Middle Eastern populations, but not for Caucasian populations

–

–

–

Li et al. (2013) [34]

–

–

ATDH risk is significantly increased in GSTM1 null genotypes. Subgroup analyses show that the association is significant in patients receiving HRZE or HRZES, and East Asians. The opposite result is observed for patients treated with HR

The GSTT1 polymorphism is not associated with ATDH risk, except in patients receiving HRZ

Sheng et al. (2014) [35]

–

CYP2E1 c1/c1 genotype significantly increases risk of ATDH. CYP2E1 D/D genotype is not significantly associated with ATDH. CYP2E1 c1/c1 genotype patients who are NAT2 slow acetylators have significantly higher risk of ATDH than CYP2E1 c1/c1 genotype patients who are NAT2 fast or intermediate acetylators

–

–

Shi et al. (2015)[36]

Slow acetylator status is associated with increased risk of ATDILI compared with fast and intermediate acetylators when standard dose of isoniazid is administered

–

–

–

Sun et al. (2008) [25]

Slow NAT2 genotype (without wild-type NAT2*4 allele) does not significantly increase risk of ATDILI. A significant effect is demonstrated among Asian populations

CYP2E1 c1/c1 genotype is significantly associated with an increased risk of ATDILI

GSTM1 null genotype significantly increases risk of ATDILI

GSTT1 null genotype has no effect on ATDILI risk

Tang et al. (2013) [37]

–

–

GSTM1 null genotype is significantly associated with increased risk of ATDILI. Stratified analyses demonstrate significant associations in Chinese populations and when restricting to studies with case numbers >57

No statistically significant association is observed between GSTT1 null genotype, or GSTM1/GSTT1 interaction and risk of ATDILI

Wang et al. (2012) [28]

Individuals homozygous for rapid NAT2 acetylator alleles (NAT2*4, NAT2*11A, NAT2*12A, NAT2*12B, NAT2*12C, NAT2*13) are rapid acetylators; individuals homozygous for slow acetylator alleles are slow acetylators, and heterozygous individuals are intermediate acetylators. Slow acetylators have significantly higher risk of ATDILI than other acetylators

–

–

–

Wang et al. 2016 [38]

–

CYP2E1 c1/c1 genotype is significantly associated with an increased risk of ATDH. CYP2E1 D/D genotype is not significantly associated with ATDH. When stratifying for study population, significant results are observed in East Asian populations only

–

–