A popular diabetes drug is still being prescribed two months after a safety body recommended its withdrawal amid concerns the drug can increase the risk of heart attacks, it has been revealed.

The British Medical Journal (BMJ) called for the immediate withdrawal of rosiglitazone, marketed as Avandia, saying the top-selling drug should never have been licensed.

An investigation found the Commission on Human Medicines advised an expert committee of the Medicines and Healthcare products Regulatory Agency (MHRA) in July to withdraw the drug as the "risks of rosiglitazone outweigh its benefits". They said it "no longer has a place on the UK market".

Rosiglitazone, which is manufactured by Glaxo SmithKline (GSK), was approved by the European Medicines Agency (EMA) in 2000 to help lower blood sugar levels in patients with type 2 diabetes.

GSK said their "extensive research" showed the drug was "safe and effective when it is prescribed appropriately".

But since its approval several studies have suggested the drug may lead to a small overall increase in the risk of heart attacks and the BMJ's investigations editor, Dr Deborah Cohen, said the European approval process was not rigorous enough.

She also raised concerns about the quality of the data used by GSK, the lack of publicly available trial results for independent scientific scrutiny, and failures to act swiftly on emerging safety fears.

The journal said doctors were advising that no new patients should start taking the drug and patients already using rosiglitazone should review their options. Those at higher risk of heart disease should be advised to stop taking it.

Professor Nick Freemantle at the University of Birmingham called for an overhaul in the standards of regulatory trials, saying: "In order to learn from our mistakes, we must improve the quality of safety data from clinical trials on all new health care interventions, not just antidiabetic drugs."

And Professor John Yudkin of University College London said: "Ten years after the release of rosiglitazone, we still cannot accurately quantify the harm to which we were exposing our patients. We need to be absolutely certain that our long term treatments for type 2 diabetes are not causing the very harm they are meant to prevent."