Metformin as firstline treatment for type 2 diabetes: are we sure?
BMJ 2016; 352 doi: https://doi.org/10.1136/bmj.h6748 (Published 08 January 2016) Cite this as: BMJ 2016;352:h6748All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Response to Ian D Watson :
We agree that all benefits and risks should be considered in the evaluation of drugs. The study by Gandini and col. is a meta-analysis of studies which includes observational studies, studies which have high risk of biases, and therefore a low level of evidence. The randomized controlled trials do not show any effect of Metformin on cancer.
Response to Michaël Rochoy :
In meta-analysis, the calculation of the RR includes a weighting on each trial’s variance’s inverse. It explains that the raw calculation doesn’t provide the same result.
Competing interests: No competing interests
Firstly, thank you for re-evaluation of the dogma of metformin and your significant questioning about the level of evidence of the current pharmacopoeia, including older drugs!
Secondly, I don't understand Table 1 and calculate risks based on the numbers presented here. For example, for cardiovascular mortality, I understand that there have been 163 deaths / 9167 patients on metformin (1.8%) patients versus 215/3268 in the control group (6.6%); however, the relative risk is 1.05.
Can you please explain your calculation?
Best regards,
M.R.
Competing interests: No competing interests
If there were to be a new trial to determine the efficacy of metformin in treating diabetics given the perceived biases in the UKPDS, it should also include the impact , or lack of it, of metformin on cancer in diabetic patients for which there appears to be some evidence (1).
1. Gandini S1, Puntoni M2, Heckman-Stoddard BM3, et al. Metformin and cancer risk and mortality: a systematic review and meta-analysis taking into account biases and confounders. Cancer Prev Res (Phila). 2014 Sep;7(9):867-85.
Competing interests: No competing interests
Re: Metformin as firstline treatment for type 2 diabetes: are we sure?
Boussageon and colleagues state that the UKPDS result with metformin has never been reproduced. The problem with that statement is that no such long study going on for well over 10 years has been undertaken to compare with the UKPDS and so the statement is misleading. The graphs published in the UKPDS 10 year follow up report, mentioned by Boussageon and colleagues, show that it is 3 years before these graphs, that demonstrate the reduced deaths and reduced myocardial infarctions with metformin, start to separate from those of the conventional therapy. This shows that short-term studies cannot be relied upon to give us the information we require.
Boussageon and colleagues ask for studies involving between 5,000 and 10,000 participants. There is often an implication that it is a weakness of the UKPDS result that it was found with the study involving only 342 patients in the metformin group and 411 in the conventional group. In fact it is a testimony to the strength of the metformin effect that it could be demonstrated in such a relatively small number of patients. Supposing it was 1922 and there were 20 patients about to die from type 1 diabetes. Supposing 10 were given the newly discovered insulin and 10 had “conventional” treatment. Supposing the 10 who had insulin survived and flourished, while the 10 who had “conventional” treatment died. Would we then go on to request a study involving 5,000 or 10,000 patients in order to believe that insulin stops people with type 1 diabetes from dying?
The case put by Boussageon and colleagues is insufficient to stop us using metformin as first line therapy in obese patients with type 2 diabetes.
Competing interests: No competing interests