Preliminary studies with an experimental artificial kidney that incorporates living cells indicate that it can reduce deaths from acute kidney failure by 50 percent, Michigan researchers reported this week.

The device is meant for short-term use – up to three days – to allow a damaged kidney time to recover function.

The artificial kidney “may yield a better treatment for life-threatening acute renal failure, for which a high mortality rate has remained unchanged despite years of advances in conventional therapies,” said Dr. H. David Humes of the University of Michigan, the lead author of a report appearing Thursday in the Journal of the American Society of Nephrology.

“They achieved some pretty remarkable results with people … who were the sickest of the sick,” said Dr. Leslie Spry of Lincoln Nephrology & Hypertension in Nebraska and a spokesman for the National Kidney Foundation. “This would be a pretty good deal if it gets to be generally accepted.”

Acute kidney failure can arise from trauma, dehydration and other causes. It affects about 5 percent of hospitalized patients and a higher percentage of those in intensive care units. The mortality rate is 50 percent to 70 percent, even when patients receive the best care available.

The new artificial kidney, called a renal tubule assist device, or RAD, is a modified form of a cartridge filter normally used for continuous dialysis.

Humes and his colleagues devised a technique for plating human kidney cells, called renal proximal tubule cells, on the interior surfaces of the cartridge.

The cells, derived from donor kidneys not suitable for transplantation, restore vital electrolytes, salt, glucose and water to the filtered blood. Those components are normally removed during conventional dialysis. The cells also produce immune system molecules, called cytokines, that are important for fighting infections.

Humes and his colleagues studied 58 patients who were critically ill with acute kidney failure at 12 centers around the United States. Eighteen received conventional continuous dialysis and 40 received conventional dialysis plus up to 72 hours of treatment with the artificial kidney.

After one month, the team reported, 33 percent of those receiving RAD therapy had died, compared with 61 percent of those in the control group. The results were not statistically significant, however, because of the small size of the groups.

After 180 days, 76.5 percent of the patients receiving conventional therapy had died, compared with 46 percent of those who used the RAD. That difference was statistically significant.

The team will now undertake a larger trial required before the device can be approved for sale.

Humes noted that there were also many challenges to be overcome in mass-producing, shipping and storing a device containing living cells.