Internet-Capable Cell Phones Can Make Medical Devices Go Awry

AMSTERDAM, The Netherlands, Sept. 6 -- Cell phones with Internet capability and other bells and whistles can cause life-threatening electrical interference if they're used too close to medical devices, researchers here said.

Outside of the current standard safety margin of a meter, the new mobile phones appear to be safe, even though they were introduced without being widely tested in a medical setting, according to Erik Jan van Lieshout, M.D., of the Academic Medical Centre at the University of Amsterdam, and colleagues.

But closer than that in device testing, newer cell phone signals shut down ventilators or restated them at different rates. They also shut down syringe pumps without an acoustic alarm or with an incorrect alarm. Finally, a cell phone signal led to incorrect inhibition in a test of external pacemakers.

"Our work has real implications for present hospital restrictions of mobile phone use in patient areas," Dr. van Lieshout said.

The one-meter margin was set up to ensure the safety of the first generation of cell phones, which were used only for voice transmission, Dr. Van Lieshout and colleagues reported online in Critical Care.

But new type of cell phones -- using General Packet Radio Service (GPRS) and Universal Mobile Telecommunications System (UMTS) signals -- allow a range of other services, including data transmission, they said.

The researchers tested two different GPRS signals (dubbed GPRS-1 and GPRS-2) and one UMTS signal against 61 medical devices in 17 categories, including ventilators, syringe pumps, and critical care monitors.

All told, there were 48 incidents in 26 devices, the researchers found. Of those, 16 (33%) were classified as hazardous, 20 (42%) as significant and 12 (25%) as light.

For example, the researchers said:

• There were nine hazardous incidents in seven of the nine tested ventilators, ranging from total switch-off and restart to changes in the ventilation rate.
• There were two hazardous incidents in two of the seven syringe pumps tested -- complete stops without an acoustic alarm or with an incorrect alarm.
• One of the three external pacemakers tested developed incorrect inhibition of the pacemaker after the cell phone signal.

The median distance between cell phone signal and device when interference occurred was three centimeters, but one incident took place when the signal was 500 centimeters away from the device.

Most hazardous incidents took place when the signal was close to the device, but one occurred at a distance of 300 centimeters, the researchers reported.

The different types of signal had significantly different rates of interference, the researchers found:

• The GPRS-1 signal induced 25 of the 48 interference episodes.
• The GRPS-2 signal induced 15.
• The UMTS signal induced eight.
• The trend was significant at P