1. Maximum likelihood estimate of odds ratios (OR) for the association between genotypes and case status were calculated with their exact mid-p 95% confidence interval (CI) using Epi6 software.20 OR heterogeneity among studies included in a meta-analysis was tested by the Breslow-Day test.2" Pooled Mantel-Haenszel weighted OR was computed when heterogeneity was not statistically significant (p<0.05). The attributable fraction, representing the proportion of all cases in the target population that is attributable to the presence of the mutation(s), was estimated using the equation22: (AF= [fca (OR-1)]/OR), where fc. is the fraction of cases with the genotype under study and OR is the odds ratio
2. between the homozygosity for the C677T mutation and spina bifida was causal, the proportion of all spina bifida cases that in the whole target population could be attributed to the presence of the homozygous C677T mutation was 8.5% or 10.8% depending on the comparison group used for the risk estimate (T/T v C/C only or T/T v C/C plus C/T);Assuming that the association
3. Table 3 summarises the results of seven available studies including the present one.9'13 23 When possible, only data for spina bifida were considered (in the UK study'2 the NTD type was not specified) since four studies,9 10 23 including this one, did not include anencephaly, one found a higher risk for spina bifida than for other NTDs," and one found the 3 T/T genotype only in 31 spina bifida cases and not in 12 anencephaly cases.'3 In other words, studies on the possible association between the MTHFR C677T mutation and NTDs seem to be based more on spina bifida than on other NTDs