Just when the message was getting through to the 15 million-odd type 2 diabetics in the USA , someone has called a big "Wait a minute!" Popular wisdom was that the lower diabetics could lower their glycosylated hemoglobin level - their HbA1c - the better. The National Heart, Lung, and Blood Institute (NHLBI) was running a clinical study to determine if intensive treatment (aiming at an HbA1c of below 6%) was superior to the 'standard' treatment (an HbA1c of 7.0% to 7.9%). There were roughly 10,000 patients in the study with type 2 diabetes and either heart disease or two risk factors for heart disease.

At a press conference on February 6, Dr Nabel, a NHLBI director, said there were 257 deaths in the intensive treatment group compared with 203 in the equal-sized standard treatment group, seen 4 years after starting the study. This was a big enough difference to stop the study in the intensive treatment group, 18 months before the planned end of the study. About half of all deaths were cardiovascular, but the excess deaths seen were clearly cardiovascular. This raised the question of whether glitazone-type drugs could be responsible - they have come under review because of an increased risk of heart failure in older patients taking them. But one of the directors of the trial has stated that they cannot find any link between one specific drug or drug type and the deaths.

Obviously this study raises concern about whether trying to lower HbA1c levels to below 7.0% is without risk to the patient. In the real, non-clinical-trial world, such efforts are probably rare, to date. However, a more positive result from the NHLBI trial might have led to a major campaign to treat diabetics more forcibly. We have been spared this - at least until another trial or two have been completed.