Cause of Device-Related Incident
Device factors; User errors

Clinical Specialty or Hospital Department
CCU / ICU / NICU; Nursery; Nursing

Device Factors
Device interaction

Document Type
Hazard Reports

External Factors
*Not stated

Mechanism of Injury or Death

Support System Failures
*Not stated

Tampering and/or Sabotage
*Not stated

User Errors
Inappropriate reliance on an automated feature

Blankets, Infant [10-417]; Warmers, Radiant, Infant [13-249]

Plastic Thermal Blankets

Hazard [Health Devices Aug 1984;13(10):261-3]


We received a report of an infant death associated with the use of a "bubble blanket" on an infant being cared for under a radiant warmer. The hospital reports that the 12-day-old, very-low-birthweight (VLBW) infant had previously experienced major problems but had been stable for six days. When routine vital signs were taken, the infant's temperature was normal. However, one hour later, the infant was extremely hyperthermic (axillary temperature greater than 42° C). The infant died shortly thereafter, reportedly as a direct result of hyperthermia. At the time that the infant was found to be hyperthermic, hospital personnel noted that the radiant warmer's temperature sensor was adhering to the plastic thermal blanket, rather than the infant's skin.


Plastic blankets (e.g., Saran wrap, bubble packing material) have been found to reduce insensible water loss and can help provide a neutral thermal environment for infants who require very high air temperatures in infant incubators or high power outputs from infant radiant warmers. However, plastic blankets can interfere with the thermal performance and safety features of radiant warmers.

Overheating of infants by radiant warmers as a result of sensor dislodgment has been discussed in the literature. In previous Health Devices evaluations and in our work completed for the U.S. Food and Drug Administration (FDA) in The Development of Standards for Infant Warmers and Incubators, we noted that, when the sensor was dislodged, some radiant warmers can raise infant temperatures dangerously without activating an alarm. In response to our work and manufacturers' recognition of the risk of infant overheating, some radiant warmer manufacturers have designed their temperature sensing, control, and alarm circuits to alarm if a sensor falls or is pulled off the infant's skin.

Usually, sensor dislodgment is detected as a low skin temperature and the radiant warmer will activate a low-skin-temperature alarm. (The radiant warmer in the reported incident had high- and low-skin-temperature alarm functions that act, in some circumstances, as sensor dislodgment alarms.) However, most warmers still cannot recognize when a sensor is partially detached (and thus sensing a temperature between that of the infant's skin and the ambient air temperature) or when a sensor slowly detaches from the infant's skin over a considerable period of time.

The activation of the low-skin-temperature alarm occurs primarily because sensors react quickly to the large thermal gradient between contact with the infant's skin (e.g., 36° C) and the ambient air (e.g., 23° C). For example, abrupt sensor temperature changes of about 3° C (less than 1/4 of the overall thermal gradient) can activate a low-skin-temperature alarm. However, if the magnitude of the thermal gradient is reduced, as might occur if the contact temperatures were 36° C for the infant's skin and 35° C for the plastic thermal blanket, the sensor temperature may not decrease enough to activate the alarm, since the radiant warmer output initially increases to try to maintain "infant" temperature. We believe that the use of the plastic thermal blanket in the reported incident contributed to the lack of an alarm and may have partially obscured personnel's view of the dislodged sensor.


  1. Avoid routine use of thermal blankets or heat shields with radiant warmers. Thin plastic thermal blankets (e.g., Saran wrap) are less likely to interfere with radiant warmer function than thicker, plastic bubble blankets. Plastic heat shields placed well above the infant may cause some changes in radiant warmer performance but are less likely to come into direct contact with the sensor's temperature-sensing surface.
  2. If it is necessary to use a plastic thermal blanket or heat shield to provide thermal support, check infant temperature, radiant warmer function, and sensor placement at least every 15 minutes. This policy should be added to the standard operating procedures for any area where radiant warmers and plastic thermal blankets are used.
  3. If plastic thermal blankets or heat shields must be used, consider using a separate temperature monitor with properly set high- and low-temperature alarm functions to monitor infant skin or core temperature. Thus, if the sensor detaches or if radiant warmer function is significantly affected, the temperature monitor will alarm. It is likely that the two measurements (i.e., indicated skin temperature on the radiant warmer and indicated skin temperature on the temperature monitor) will not agree exactly. However, if proper sensor attachment methods are used (i.e., placement on the skin surface facing the radiant warming source, use of a heat reflecting patch), any offset should be within acceptable limits (about 0.7° C). The nursing staff should be made aware of these recommendations, procedures to be used, problems that may be encountered, and operating anomalies that are likely to occur.


  • Blankets, Infant [10-417]
  • Warmers, Radiant, Infant [13-249]

Cause of Device-Related Incident

Device factor: Device interaction

User error: Inappropriate reliance on an automated feature

Mechanism of Injury or Death


[Home]    [About]    [Help]    [Site Map]
Copyright © 2021 ECRI
All rights reserved