Dr Remy Boussageon and colleagues at Lyon University in France conducted a meta-analysis of data from 13 trials involving 13,110 patients.
They concluded that the ratio of risks and benefits in using metformin remain unproven.
Dr Boussageon and his team said they could not exclude a 25% reduction or a 31% increase in all-cause mortality or a 64% increase in cardiovascular mortality. They said that further studies were needed 'to clarify this problematic situation'.
'The specific efficacy of metformin to prevent death or cardiovascular events has not been proven,' they said. 'The number and quality of available studies are insufficient.'
The researchers added: 'We cannot exclude beyond any reasonable doubt that metformin use increases or decreases the risk of all-cause mortality or cardiovascular mortality.'
Dr Boussageon and his team said their findings suggested that metformin may not be the best comparator for evaluating new diabetes drugs.