In this work an approximated path integral model describing the dynamics of a inextensible chain is presented. To this purpose, the nonlinear constraints which enforce the property of inextensibility of the chain are relaxed and are just imposed in an average sense. This strategy, which has been originally proposed for semiflexible polymers in statistical mechanics, is complicated in the case of dynamics by the extra dependence on the time variable and by the presence of nontrivial boundary conditions. Despite these complications, the probability function of the chain, which measures the probability to pass to a given initial conformation to a final one, is computed exactly. The Lagrange multiplier imposing the relaxed condition satisfies a complicated nonlinear equation, which has been solved assuming that the chain is very long.