This UK based cohort study (1998 and 2013; 77138 metformin initiators of which 25699 added or switched to sulfonylureas) suggests that the practice of continuing metformin when introducing sulfonylureas is safer than switching.
The results of the primary analysis remained consistent in sensitivity analyses, as well as after classifying sulfonylureas in two different groups - pancreas non-specific, long acting drugs (ie, glyburide and glimepiride) and pancreas specific, short acting drugs (ie, gliclazide, glipizide, and tolbutamide).