Reproducibility should be a fundamental quality of published experiments.
If the published work under specifies the experiment such that it is unreproducible, that means the results can’t be reliably extrapolated because there are unstated conditions.