Aggressively treating diabetes doesn't prevent heart problems and deaths any better than standard treatment for lowering blood sugar, say Australian researchers.

It's the second large study, involving thousands of patients, to show no heart benefit from drastically lowering diabetics' blood sugar levels. Experts said doctors should stick to the recommended target levels.

Heart disease is the cause of death for two-thirds of diabetics. Researchers tried pushing blood sugar down to near-normal levels to see if that would protect the hearts of high-risk patients with Type 2 diabetes.

But the Australian study showed no difference in the number of heart attacks, strokes and heart-related deaths between groups who received intensive or standard care. A U.S. study that was stopped earlier this year also showed no benefit and in addition reported an unexplained higher number of deaths among those who were aggressively treated.

The Australian study released Friday showed one positive result — a one-fifth reduction in kidney problems, a common complication of diabetes, compared to normal care.

"Both studies are important contributions to the field but do not provide a definitive answer," Dr. William Cefalu, of Louisiana State University, wrote in an editorial in the New England Journal of Medicine. He said other ongoing studies should provide clarification.

Both studies were released Friday in the journal and are being presented at an American Diabetes Association meeting in San Francisco. Partial results of the U.S. research were released in February when it was halted.

An estimated 21 million Americans and 250 million people worldwide have diabetes, meaning their bodies can't properly regulate their blood sugar, or glucose. Most have Type 2 diabetes. High levels of blood sugar can cause damage to the heart, blood vessels, kidneys and eyes.

Instead of trying aggressive measures, experts say there should be more focus on other strategies known to lower heart risks — diet, exercise and medications such as aspirin, cholesterol-lowering statins and blood pressure drugs.

Blood glucose levels tracked

In the two studies, researchers used a test that tracks average glucose levels over two to three months. For diabetics in the U.S., the recommended level is below a rating of seven. People without diabetes have levels around five.

Both studies targeted diabetes patients middle-aged or older who had a heart problem or other heart-risk factors. Doctors used a variety of diabetes drugs and insulin to try to get blood sugar levels down — to less than seven in the U.S. study and 6.5 or lower in the international study.

The Australian study had more than 11,000 participants from Asia, Australia, Europe and Canada; the U.S. study had 10,000.

The U.S study was stopped after 3 1/2 years because of more deaths in the aggressively-treated group: 257 deaths compared to 203 for standard care. The researchers said they haven't found a reason for the difference. Everyone was switched to standard treatment, and the researchers are continuing to follow them.

In a statement, Dr. Elizabeth Nabel, director of the National Heart, Lung, and Blood Institute, said that severely lowering blood sugar appears to be too risky for diabetes patients at higher risk for heart problems. Her institute helped pay for the study.

Dr. Alvin Powers of Vanderbilt University said the Australian study was reassuring because it showed blood sugar levels could be safely lowered below the current targets, in contrast to the U.S. results. He said reducing kidney complications is significant because it might mean fewer people needing dialysis or a kidney transplant.

"I think this affirms that seven (blood sugar level) should remain our goal — but most people don't reach that goal," he said.

The Australian study was funded by the government and diabetes drugmaker Servier. The U.S. study was paid for by National Institutes of Health, and various companies provided diabetes drugs.

A number of researchers in both studies report receiving grant support or fees from drugmakers.