ANC200009_34-39.

qxp 1/30/09 11:30 PM Page 34

ROSALIE O. MAINOUS, PHD, ARNP, NNP-BC • Section Editor

REGINA CUSSON, PHD, RN • Research Tutorial

Use of the Heated Gel Mattress and Its

Impact on Admission Temperature of
Very Low Birth-Weight Infants
Pamela G. Almeida, MS, RNC, NNP,1 Jane Chandley, MS, RNC, NNP,2 James Davis, PhD,3
Roseanne C. Harrigan, EdD3

ABSTRACT
PURPOSE: The purpose of this investigation was to describe the ability of a transport mattress (TransWarmer Infant
Transport Mattress), produced by Cooper Surgical (Trumbull, Connecticut), to reduce hypothermia in a group of very low
birth-weight infants.
SUBJECTS: Convenience sample of 115 infants weighing less than 1500 g who were born at Women & Infants’ Hospital,
Providence, Rhode Island, and admitted to the neonatal intensive care unit.
DESIGN: A quality assurance study using a nonrandomized experimental design. Infants placed on the TransWarmer
Infant Transport Mattress were compared with those treated with standard care.
METHODS: Charts were reviewed and data abstracted on the 115 very low birth-weight infants delivered at Women &
Infants’ Hospital, Providence, Rhode Island.
MAIN OUTCOME MEASURE: Hypothermia was significantly lower in the mattress group than for controls (52.5% vs
77.3% using a definition of hypothermia as body temperature less than 97.4⬚F, P ⫽ .01). The difference resulted despite
the fact that the mattress group was at increased risk based on various risk factors, lower mean age (26 vs 28.5 wk,
P ⫽ .001), a lower birth weight (876 vs 1091 g, P ⫽ .004), and a higher proportion of Apgar scores of less than 5 at 5
minutes (13.2% vs 6.4%, P ⫽ .29) compared with controls. A linear regression model adjusted for birth weight, gender,
and gestational age showed that the use of the heated gel mattress raised body temperatures by a mean of 0.7⬚F per
infant (P ⬍ .001).
PRINCIPAL RESULTS: The occurrence of hypothermia was significantly lower in the mattress group.
CONCLUSION: The results of this investigation suggest that further research of the TransWarmer Transport Mattress is
warranted using an adequately powered randomized controlled trial. Information on the safety and cost-effectiveness
is needed. Long-term follow-up evaluating admission temperatures and long-term outcomes is also warranted.
KEY WORDS: hypothermia, temperature, TransWarmer Transport Mattress, very low birth-weight infant

tion between hypothermia and increased morbidity

H
ypothermia of the preterm infant remains a
challenge to the 21st-century neonatal care
provider.1 Despite advances in technology,
observational studies have demonstrated an associa-
Address correspondence to Pamela G. Almeida, MS, RNC,
NNP, Department of Neonatology, Kaiser Foundation
Hospital, 3288 Moanalua Rd, Honolulu, HI 96819;
Pamela.G.Almeida@kp.org.
Author Affiliations: 1Department of Neonatology, Kaiser
Foundation Hospital, Honolulu, Hawaii; 2Department of
Pediatrics, Women & Infants’ Hospital, Providence, Rhode
Island; and 3Department of Complementary and Alternative
Medicine, John A. Burns School of Medicine, Honolulu,
Hawaii.
Copyright © 2009 by the National Association of
Neonatal Nurses.
and mortality in very low birth-weight (VLBW)
infants.2 One study suggested that hypothermia on
arrival to the neonatal intensive care unit (NICU)
inversely correlates with severity of illness.3
Increased severity of illness scores assist in projecting
long-term morbidities related to the birth of the
VLBW infant. Those infants with higher severity of
illness scores also suffered increased risk for long-
term morbidities such as cerebral palsy and chronic
lung disease.3
These VLBW infants continue to be at increased
risk for hypothermia even when treated according to
current recommendations to dry, remove wet linen,
and place on a radiant warmer.4 Different techniques
have been implemented to decrease hypothermia
after the birth of premature infants. Each technique
addresses a different mode of heat loss.

34 Advances in Neonatal Care • Vol. 9, No. 1 • pp. 34-39

ANC200009_34-39.qxp 1/30/09 11:30 PM Page 35
Heated Gel Mattress’s Impact on Admission Temperature of VLBW Infants
Infants lose heat through 4 mechanisms: radiant,
convective, evaporative, and conductive. Three ran-
domized controlled trials5-7 confirmed the efficacy of
plastic bags or plastic wrapping (food-grade, heat-
resistant plastic), used in conjunction with the cus-
tomary radiant heat. The plastic wrap was used to
decrease the evaporative and convective heat loss,
whereas the radiant warmer was used to decrease the
radiant heat loss. The admission temperatures of pre-
mature babies of less than 28 weeks’ gestational age
significantly improved when compared with the
admission temperatures of those given standard
care).4,5 The temperatures of infants placed in plastic
bags or plastic wrapping, however, must be moni-
tored closely because there is a small risk that this
technique may produce hyperthermia.7 There are
other techniques that have been used to maintain
temperature in the delivery room during stabilization
such as drying and swaddling, warming pads, and
skin-to-skin contact with covering.8 These techniques
address the mode of conductive heat transfer.
Conductive heat transfer occurs when there is a
temperature gradient between 2 objects that come in
direct contact with each other.9 An example is when
an infant is placed on a mattress at delivery.
Fourier’s Law states that conductive heat transfer
occurs when heat is transferred from the warmer
object to the cooler one.9 Therefore, an infant at
98.6⬚F placed on a mattress at 70⬚F will transfer heat
to that surface. As the infant’s temperature begins to
decline, the infant will begin to develop hypother-
mia. Hypothermia will lead to a cascade of events.
The infant’s body will attempt to produce heat
through nonshivering thermogenesis. Nonshivering
thermogenesis is the generation of heat by the body
through the use of brown fat deposits. With nonshiv-
ering thermogenesis and the consumption of brown
fat there will be an increase in a metabolic by-prod-
uct of lactic acid. The infant will eventually develop
lactic acidosis, leading to metabolic acidosis, hypo-
glycemia, and, if untreated, death. Thus, investiga-
tion of interventions to reduce hypothermia is of crit-
ical interest to reduce morbidity and mortality of
VLBW infants. We have particular interest in evalu-
ation of the heat gel mattress by reducing conductive
heat loss and providing conductive heat warming.
As described above, the use of a heated gel mattress
that is warmer than the infant will conduct heat to
the baby at delivery preventing the infant from
becoming hypothermic.
PROBLEM
There are limited data available on the use of the
TransWarmer Infant Transport Mattress, produced
by Cooper Surgical, to prevent hypothermia in the
delivery room. One study10 of 24 babies revealed
that the mattress was useful in decreasing hypother-
Advances in Neonatal Care • Vol. 9, No. 1
35
mia of the VLBW infant. There is additional evi-
dence from studies of full-term newborn babies in
nurseries11-13 and during transport14 suggesting that
heated mattresses reduce hypothermia and can
warm babies. In the previously mentioned study,10
the number needed to prevent 1 case of hypother-
mia was 2 (95% CI, 1–4).10,15 If this proves to be cor-
rect, large trials will not be needed to confirm her
findings, but numbers must be sufficient to establish
safety. One 1275-g baby in her investigation was
admitted with a temperature of 99.5⬚F, affirming that
temperature monitoring is needed to avoid over-
heating larger babies. During the investigation, the
infant’s positions on the mattresses were changed
after (every) 15 minutes to prevent burning.10 This
was important because the mattresses may reach
higher than specified temperatures if stored at a high
ambient temperature.16
PURPOSE
The purpose of this investigation was to evaluate the
ability of the TransWarmer Infant Transport Mattress
to reduce hypothermia in a group of VLBW infants.
METHODS
Design
A quasi-experimental quality assurance design of a
convenience sample was used. The VLBW infants
admitted to the NICU at Women & Infants’ Hospital,
Providence, Rhode Island, were assigned to either
the control group (standard therapy radiant warmer)
or the experimental group (radiant warmer and
TransWarmer Infant Transport Mattress). Data were
collected on 115 VLBW infants delivered at Women
& Infants’ Hospital, Providence, Rhode Island. There
were originally 120 subjects; of these, 5 had improper
charts or lacked documentation and were therefore
excluded from the study.
At the time this study was completed, the hospital
protocol indicated a neonatal fellow be present at all
deliveries less than 30 weeks’ gestation. The use of
the mattress for the delivery was initiated by the
neonatal fellow or NICU nurse. Approval was
granted by the institutional review board (IRB) and
the procedures were monitored by the Department
of Quality Improvement.
Equipment
The TransWarmer Infant Transport Mattress is
made of a combination of sodium acetate anhy-
drous and hydroxyethyl cellulose (Figure 1). The
mattress is activated by bending of a metal disk and
becomes warm within 1 minute. The mattress
reaches a maximum temperature of 104⬚F and
remains warm for up to 2 hours according to the
manufacturer’s description.
ANC200009_34-39.qxp 1/30/09 11:30 PM Page 36

36 Almeida et al

Quality Assurance mum temperature of 104⬚F and maintain heat for

Prior to the start of the study, a pilot study was
requested by the hospital’s IRB to look strictly at
the TransWarmer Infant Transport Mattress tem-
peratures that were generated at the surface of the
mattress. This was done to provide a quality assur-
ance measure because the mattresses are not Food
and Drug Administration (FDA) approved for use
in infants less than 36 weeks’ gestation. Ten mat-
tresses were activated and their surface tempera-
tures measured for a total of 4 hours. Interval tem-
peratures were taken every 15 minutes. These
times were chosen in accordance with the product
description that the mattress would obtain a maxi-
2 hours with a gradual decline in temperature. Of
the 10 mattresses tested, there were varying
temperature maximums (100.6⬚F–102⬚F). All 10
mattresses began dropping temperature from the
maximum, starting at 1 hour after activation, and
continued a gradual decline approximately every
30 minutes.
Upon completion of testing the mattresses, the IRB
felt confident that the there was no evidence of any
mattresses overheating and granted permission for
the mattresses to be trialed as a new device in the
admission process of admitting VLBW infants to the
NICU.

■ RESEARCH TUTORIAL

What is a Quality Assurance Study?

The purpose of quality assurance in healthcare is to assess and evaluate the quality of care provided and to implement
strategies to ensure that high quality care is actually provided. This movement began in the 1980s with the work of
Donabedian1,2 and Deming.3 The process utilized was a simple one initially, based on the scientific method, illustrated by
the acronym PCDA: plan, check, do, and act. In recent years, quality assurance has evolved into the quality improvement
movement, with a focus on evaluation of care using more sophisticated scientific methods.4
The authors of this study were interested in determining whether implementation of a change in practice to a new
method of transport warming could lead to improved infant outcomes, in this case, a decrease in hypothermia. The
authors’ stated purpose, to determine whether a heated transport mattress could decrease hypothermia in very low birth-
weight infants, is consistent with the goal of quality assurance to improve outcomes of care. The authors felt confident
in the safety of their study because of prior research findings with the heated gel mattress, but the mattress had not been
used in this more vulnerable group. Hence, they determined that a quality assurance study was needed. Quality assur-
ance has been criticized for not using rigorous experimental methods instead, implementing a change in practice and
then examining what happens after the change is made. This study illustrates the challenges associated with conducting
research in a clinical setting and the importance of using scientific methods to examine alternative caregiving strategies.

What is a nonrandomized experimental design?

A researcher who uses an experimental design studies 2 groups: one group has something done to them (the experi-
mental group) and the other group is simply a comparison group that does not receive the intervention that the exper-
imental group does (the control group).5 This is the only type of design that can truly establish causation, that is, that the
intervention caused the resulting outcomes. A nonrandomized design means that participants are not assigned to the
groups in a systematic, scientific fashion but rather by convenience. Randomization is the hallmark of the experimental
method. In this method, subjects are assigned to either the experimental group or the control group at random, so that
every subject has an equal chance of being in either group. Randomization eliminates any bias that could occur related
to characteristics of the subjects; therefore, the 2 groups are considered equivalent and any differences in results are then
caused by the manipulation the experimental group received. A nonrandomized experimental design, which is also
referred to as a quasi-experimental design, is not as powerful as a true experimental design. Because the subjects are not
assigned to groups randomly, differences in outcomes could be due to differences between the groups even before the
intervention. Researchers use the design when they are not able to randomly assign subjects to groups and often use
statistical methods to control differences between groups.
This study used a nonrandomized experimental design to determine whether infants transported on a heated gel mat-
tress would be less likely to develop hypothermia. Researchers use nonrandomized designs for many reasons but most
often when they simply are not able to randomly assign subjects to groups, as in this study. The researchers were exam-
ining a change in practice and comparing outcomes to the infants who received the standard treatment for transport
from the delivery room. They conducted a pilot study first to make sure that the heated mattress was safe to use. This
procedure is typical to protect the safety of vulnerable subjects when a potentially harmful intervention is used. They also
used statistical methods to control for the differences between the groups, such as gestational age, birth weight, and
Apgar scores, increasing the credibility of the findings, even in the absence of a true experimental design.

www.advancesinneonatalcare.org
ANC200009_34-39.qxp 1/30/09 11:30 PM Page 37
Heated Gel Mattress’s Impact on Admission Temperature of VLBW Infants
FIGURE 1.
Transwarmer Infant Transport Mattress.
Procedures
Data were collected from nursing documentation and
labor and delivery records. For all infants, radiant
warmers were set on manual heating mode at maxi-
mum heat output in the delivery rooms and operat-
ing rooms. For the mattress group, each gel mattress
was activated, weighed, and placed on the radiant
warmer. Each mattress was weighed so that the infant
would remain on the mattress from birth through
admission and would allow the infant to be weighed
on the mattress with his or her weight tared. Infants
in this group were placed directly on the heat mat-
tress, were dried with warm blankets, were resusci-
tated, had hat applied, had warm blankets wrapped
around the infants and the mattresses, and were
transported to the NICU in a heated incubator.
Those in the standard therapy group were placed
under the radiant warmer, were dried with warm
blankets, were resuscitated, had hat applied, were
TABLE 1. Details of Study Infantsa
TransWarmer Infant
Characteristics
No. of infants
Male (%)
Birth weight, g
Gestational age, wk
Apgar score of ⬍5 at 5 min (%)
Odds ratio delivery (%)
Values in parentheses are standard deviations.
a
Advances in Neonatal Care • Vol. 9, No. 1
37
wrapped in warm blankets, and were transported to
the NICU in a heated incubator. Temperatures
obtained on admission to the NICU were taken rec-
tally with a glass thermometer (B-D, Franklin Lakes
New Jersey), with subsequent temperatures taken by
the NICU’s temperature protocol. The protocol con-
sisted of taking temperatures every 15 minutes for
1 hour, then every 30 minutes for the second hour,
and then hourly for the next 2 hours. Hyperthermia
was prevented once the infant was admitted by using
the alarm settings on the radiant warmer set in servo-
control. All the infants were placed on warmers that
were set at 98.6⬚F and the warmer would alarm if the
infant’s temperature exceeded that temperature.
Data were collected for 4 hours after admission, using
the participant’s vital sign recordings. Timing to the
first measure was different in all admissions, with the
average being 15 minutes. Because of the lack of syn-
chronization of hospital clocks, there were 2 cases in
which the infants’ first readings were as early as 5
minutes of age. This did not seem likely as the infant
is normally still in the delivery room for the 5-minute
Apgar score.
ANALYSIS
Data were analyzed using SAS Enterprise Guide
2004 (SAS Institute, Inc, Cary, North Carolina.).
Descriptive statistics, multiple regression methods,
and ␹2 tests were used in the analysis of the data.
RESULTS
The control group was composed of 75 participants,
whereas only 40 participants were in the intervention
group. Further description of the study groups can be
found in Table 1. Because of the lack of true random-
ization, the groups were unevenly distributed. The
intervention group was of a younger gestational age,
smaller in weight, and had worse Apgar scores than
the control group. Theoretically, this made the risk of
hypothermia and hyperthermia higher in the
Transport Mattress Control P
40 75
14 33 .47
876 (234) 1091 (307) .0039
26 (2.3) 28.15 (4.4) .0001
13.2 6.4 .29
45.1 54.9 .4
ANC200009_34-39.qxp 1/30/09 11:30 PM Page 38
38 Almeida et al
TABLE 2. Incidence of Hypothermia on Admission to the NICU
TransWarmer Admission Hypothermia
Infant Transport Mattress (⬍
⬍97.4⬚F)
No 58 (77.33%)
Yes 21 (52.50%)
Total 79
intervention group. There were 2 instances of admis-
sion hyperthermia (T ⬎ 99⬚F), 1 in each group. None
of the participants had a major congenital anomaly.
Interestingly, despite the fact that the experimental
group was at increased risk for the occurrence of
hypothermia, it was significantly lower than for con-
trols (52.5% vs 77.3%, T ⬍ 97.4⬚F, P ⫽ .006) (Table 2).
In a linear regression model adjusted for gestational
age, gender, and birth weight, use of the heated gel
mattress was associated with the 0.7⬚F higher temper-
atures (P ⱕ.001).
DISCUSSION
Use of the TransWarmer Infant Transport Mattress
was beneficial in decreasing hypothermia of the
VLBW infant in our study. A primary benefit of the
mattresses was that these could be activated within 1
minute prior to the delivery, making them useful in
emergency deliveries. Our results, similar to
Brennan’s10 study, revealed that heated gel mattresses
were useful in decreasing hypothermia of VLBW
infants on admission to the NICU (Figure 2). We also
had several cases of hyperthermia in the experimen-
tal group, mimicking the results Brennan obtained.
Currently, use of the gel mattress is not a docu-
mented standard of care in perinatal guidelines or the
FIGURE 2.
Cumulative percentages of admission temperatures
of control versus mattress groups.
Admission Temperature
(ⱖ
ⱖ97.4⬚F) Total
17 (22.67%) 75
19 (47.50%) 40
36 115
American Academy of Pediatrics newborn resuscita-
tion program.4 Although used throughout the United
States and the United Kingdom, the TransWarmer
Infant Transport Mattress has not been FDA
approved for infants less than 36 weeks’ gestation.
When an electronic survey was conducted among
neonatal nurses asking how their hospitals used the
mattresses, there were many varying answers. Some
institutions had concern that infants could get burned
and did not use them at all. Other hospitals used
them with several blankets on top to prevent the
infant from coming in direct contact with the mat-
tress, while others followed the manufacturer’s rec-
ommendations and placed the infant directly on the
material side of the mattress. Much of the confusion
surrounding the use of the heat gel mattress needs to
be answered in true scientific methods.
LIMITATIONS
This study was done as a nonrandomized conven-
ience sample. Prospective randomized controls are
needed. A further limitation is that the mattress tem-
peratures were not measured as an oversight in the
original development of the research design.
Maximum temperatures may have differed among
the mattresses, as reported in the study of Carmichael
et al,16 causing differences in admission temperatures.
Moreover, risks for hyperthermia and burning need
to be addressed in further studies of VLBW infants
weighing less than 1500 g.
IMPLICATIONS FOR RESEARCH
AND PRACTICE
The results of this investigation suggest that further
research of the gel mattress is warranted using an ade-
quately powered randomized controlled trial.
Information on the safety, the cost-effectiveness, and
the contribution of this technology is needed. Further
studies examining admission hypothermia of VLBW
infants and their long-term morbidities also need to
be conducted. There is a need for FDA clinical trials
to establish efficacy and safety in these devices for
this vulnerable population of VLBW infants. With
documented clinical trials, more standardization of
care can occur that can improve current methods in
www.advancesinneonatalcare.org
ANC200009_34-39.qxp 1/30/09 11:30 PM Page 39

Heated Gel Mattress’s Impact on Admission Temperature of VLBW Infants 39

maintaining neutral thermal temperatures for VLBW
infants.
Acknowledgment
This article was supported in part by Cooper Surgical
and Grant #R25RR019321 “Clinical Research
Education and Career Development (CRECD) in
Minority Institutions.”
References
1. Watkinson M. Temperature control of premature infants in the delivery room.
Clin Perinatol. 2006;33:43-53.
2. da Mota Silveira SM, Goncalves de Mello MJ, de Arruda Vidal S, de Frias PG,
Cattaneo A. Hypothermia on admission: a risk factor for death in newborns
referred to the Pernambuco Institute of Mother and Child Health. J Trop Pediatr.
2003;49:115-120.
3. Richardson D, Shah BL, Frantz ID, Bednarek F, Rubin LP, McCormick MC. Perinatal
risk and severity of illness in newborns at 6 neonatal intensive care units. Am J
Public Health. 1999;89:511-516.
4. Kattwinkel J, ed. Newborn Resuscitation Textbook. 5th ed. Dallas, TX: American
Academy of Pediatrics; 2006.
5. Vohra SFG, Campbell V, Abbott M, Whyte R. Effect of polyethylene occlusive skin
wrapping on heat loss in very low birth weight infants at delivery. J Pediatr.
1999;134:547-551.
6. Knobel RB, Wimmer JE Jr, Holbert D. Heat loss prevention for preterm infants in
the delivery room. J Perinatol. 2005;25:304-308.
7. Vohra S, Roberts RS, Zhang B, Janes M, Schmidt B. Heat loss prevention
(HeLP) in the delivery room: a randomized controlled trial of polyethylene
occlusive skin wrapping in very preterm infants. J Pediatr. 2004;145:
750-753.
8. Sherman TI, Greenspan JS, St Clair N, Touch SM, Shaffer TH. Optimizing the
neonatal thermal environment. Neonatal Netw. 2006;25:251-260.
9. Conductive heat transfer. http://www.enigneeringtoolbox.com/conductive-
heat-transfer-d_428.html. Accessed July 20, 2008.
10. Brennan AB. Effect of Sodium Acetate Transport Mattress on Admission
Temperatures of Infants ⬍1500 g [dissertation]. Gainesville, FL: University of Florida;
1996.
11. Sarman I, Can G, Tunell R. Rewarming preterm infants on a heated, water filled
mattress. Arch Dis Child. 1989;64:687-692.
12. Green-Abate C, Tafari N, Rao MR, Yu K, Clemens JD. Comparison of heated
water-filled mattress and space-heated room with infant incubator in provid-
ing warmth to low birth weight newborns. Int J Epidemiol. 1994;23:1226-1233.
13. Sarman I, Tunell R. Providing warmth for preterm babies by a heated, water filled
mattress. Arch Dis Child. 1989;64(1, Spec No.):29-33.
14. L’Herault J, Petroff L, Jeffrey J. The effectiveness of a thermal mattress in stabi-
lizing and maintaining body temperature during the transport of very low-birth
weight newborns. Appl Nurs Res. 2001;14:210-219.
15. McCall EM, Alderdice FA, Halliday HL, Jenkins JG, Vohra S. Interventions to pre-
vent hypothermia at birth in preterm and/or low birthweight babies. Cochrane
Database Syst Rev. 2008, issue1. DOI:10.1002/14651858.CD004210.pub3.
16. Carmichael A, McCullough S, Kempley S. Critical dependence of acetate thermal
mattress on gel activation temperature. Arch Dis Child Fetal Neonatal Ed.
2007;92(1):F44-F45.

Advances in Neonatal Care • Vol. 9, No. 1