In recent years, Bayesian updating of structural reliability based on measured data has been attracting much attention because it can give more accurate predictions for structural reliability analysis is with the measured data increasing. In Bayesian analysis, the parameters of the probability distribution function of the basic random variables are considered as uncertain random variable with prior distribution and updated by combining with the likelihood function which is related to the measured data. Usually, since the dimension of the parameter space is too large, the evaluation of the integral cannot be done analytically. Recently, some Markov chain Monte Carlo simulation methods have been developed to solve the Bayesian multiple integral problems. However, in general, the efficiency of these proposed approaches is adversely affected by the dimension of the parameter space. In the present paper, the point-estimate method based on univariate-reduction integration is proposed to solve the high dimensional Bayesian integral problems. The Bayesian statistical framework for structural reliability analysis is firstly formulated, and then the point-estimate based on univariate dimension reduction integration is used to evaluate the high dimensional Bayesian integrals, from which the updated underlying random variables can be obtained. Finally, the updated random variables can be depicted by using three-parameter lognormal distribution, after which the reliability analysis can be easily performed. The efficiency and accuracy of the proposed methodology are demonstrated through a numerical example, where Markov chain Monte Carlo simulation methods are utilized for comparison.