Elimination of Admission Hypothermia in Preterm Very Low-Birth-Weight Infants by Standardization of Delivery Room Management
Perm J 2013 Summer; 17(3):8-13
Context: Temperature instability is a serious but potentially preventable morbidity in preterm infants. Admission temperatures below 36ºC are associated with increased mortality and late onset sepsis.
Hypothermia is a serious but potentially preventable morbidity in preterm infants. Studies have shown that for each 1ºC decrease in admission temperature below 36ºC, there is an increase in mortality by 28% and in late-onset sepsis by 11%.1 Hypothermia also is associated with hypoglycemia, respiratory distress, and metabolic acidosis. The incidence of hypothermia is between 31% and 78% for those with birth weight less than 1500 g.2 Infants at less than 28 weeks' gestation have the highest incidence of hypothermia. The skin temperature of an exposed preterm infant will drop at a rate of approximately 0.5ºC to 1.0ºC per minute.3 Both physical characteristics and environmental factors predispose the preterm infant to hypothermia.
The physical characteristics include a large surface area-to-volume ratio, immature skin with minimal stratum corneum, a thin layer of insulating fat, poor vasomotor control, and lack of nonshivering thermogenesis. The environmental factors include low ambient air temperature in delivery rooms and neonatal intensive care unit (NICU) admission rooms, and low surface temperature of beds used at admittance and during resuscitation. Delivery rooms are usually kept cool for the comfort of the mothers and the staff, but the cool air causes the preterm infants to lose heat through conduction and convection. Because of these factors, preterm newborns are unable to maintain equilibrium between heat gain and loss and are dependent on their caregivers to provide that balance.
In 1997, the World Health Organization (WHO)4 provided the following definition of hypothermia of newborns:
• Potential cold stress: 36.0°C to 36.5°C; cause for concern
• Moderate hypothermia: 32.0°C to 36.0°C; danger, immediate warming of the baby needed
• Severe hypothermia: less than 32.0°C; outlook grave, skilled care urgently needed.
The American Society of Heating, Refrigerating and Air-Conditioning Engineers5 and WHO recommend that the temperature in all standard single-patient rooms (labor and delivery, recovery, and postpartum) be 24ºC to 25ºC and that the nursery be 24ºC to 25.6ºC. The recommendations further state that delivery room temperatures should never be below 20ºC.
In 2006, after identification of a high rate of hypothermia in infants with very low birth weight (VLBW), we initiated a quality-improvement (QI) effort to prevent hypothermia in preterm infants admitted to our NICU. Our primary goal was to prevent hypothermia in newborn VLBW infants by preventing excessive heat loss in the delivery room. Our target admission temperature was an axillary skin temperature between 36ºC and 37ºC. We achieved this goal by standardizing delivery room management and NICU admission processes for all VLBW infants.
Santa Clara Valley Medical Center is a county hospital in San Jose, CA, with a 40-bed Regional Level III NICU that admits 50 to 75 VLBW infants annually, 80% of whom are inborn.
In December 2005, a preterm infant born at 24 weeks' gestation with a birth weight of 715 g was admitted to our NICU with a temperature of 33.2ºC. Despite maximal interventions to correct the hypothermia, the temperature did not increase to above 36°C until after 16 hours of life. At that time, all VLBW infant deliveries were attended by a multidisciplinary team composed of a physician, a registered nurse, and a respiratory therapist. Immediate postdelivery thermal management included placing the infant on a prewarmed resuscitation bed, drying skin surfaces, placing a hat, and wrapping the infant in a warm blanket. Frequently the infant was left exposed for assessment during the initial resuscitation.
We gathered baseline data for the next 6 months (January 2006 to June 2006) and recognized that the rate of hypothermia was 68% in VLBW (1000 to 1500 g) infants and was even higher—at 84%—in extremely low-birth-weight (< 1000 g) infants. Root cause analysis of the cases of hypothermia identified multiple issues, including the following:
• A lack of close monitoring of the infant's temperature after the initial assessment
• A thermoregulation policy that did not address the special considerations for preterm infants
• A knowledge gap regarding the neonatal neutral thermal environment among staff
• No standard for the effective use of thermal equipment (eg, combination incubator-warmer [Giraffe OmniBed, Ohmeda Medical, now GE Healthcare, Chalfont St Giles, UK], chemical warming mattress, isolette, resuscitation bed)
• Limited staff training in the use of special thermal equipment (only 6 selected registered nurses, called "Super Users," were trained in the use of the Giraffe OmniBed)
• Ambient air temperatures in the delivery room and admission room were adjusted for staff comfort and were frequently less than 20ºC.
All preterm neonates who were born at our Medical Center between January 2006 and December 2011 at less than 33 weeks of gestation with a birth weight under 1500 g were the target population for this QI effort and received standardized delivery room and NICU admission management. We obtained institutional review board approval for data collection and publication of the results of our QI effort.
We formed a multidisciplinary Thermoregulation Committee consisting of nurses, physicians, and nurse practitioners to research, develop, and implement new guidelines for the management of thermoregulation. Our thermoregulation guidelines were based on a review of the medical literature,6 best practice recommendations from the California Perinatal Quality Care Collaborative (CPQCC),7 evidence-based principles from the Neonatal Resuscitation Program,8 and recommendations from WHO for ambient air temperatures in the delivery room.4 We implemented a QI initiative using the rapid-cycle Plan-Do-Study-Act approach (PDSA) improvement model.
After the guidelines were developed, Thermoregulation Committee team members were responsible for staff education regarding thermoregulation and the new treatment guidelines, and for ensuring multidisciplinary team compliance with the guidelines. The timeline of the planning and implementation is shown in Figure 1.
The thermoregulation guidelines involved the use of multiple modalities divided into three phases: predelivery preparation, resuscitation, and NICU admission. Key equipment and supplies used throughout included a prewarmed resuscitation bed (a servo-controlled radiant warmer), polyethylene occlusive wrap, chemically activated warming mattress (Figure 2), warming lights (used in the initial phase of implementation), newborn hat, warm blankets, and an axillary thermometer.
The predelivery preparation included the following:
• Set the resuscitation bed on manual control set at maximum heat output
• Increase ambient temperature in the delivery room to 25ºC or higher
• Turn on the warming lights
• Activate the chemical warming mattress according to the manufacturer's instructions
• Place polyethylene occlusive wrap over the warming mattress.
A scripted, standardized, multidisciplinary approach to resuscitation, called the Smart Start protocol, was implemented for delivery room management in 2007 (Table 1 available online at www.thepermanentejournal.org/files/Summer2013/hypothermia.pdf). This included delayed umbilical cord clamping as an intervention to prevent intraventricular hemorrhage.9 After delivery, the infant was handed to the pediatric clinician who held the baby in a sterile warm blanket with the chemical warming mattress underneath to prevent hypothermia. The infant was held approximately 25 to 40 cm (10 to 15 in) below the level of introitus for vaginal deliveries and below the level of the incision for cesarean delivery. The pediatric staff counted the time after delivery out loud, and the obstetrician clamped and cut the cord after 30 to 60 seconds.
Smart Start included specific, predefined tasks and responsibilities for all staff members. One registered nurse was assigned to temperature management, which involved the following tasks at each time point.
• Without drying, immediately wrap the infant with polyethylene wrap
• Place the infant on the chemical warmer mattress
• Dry the infant's head and place a hat on the head
• Assess the infant through the polyethylene wrap
• Take and record the infant's axillary temperature every 5 minutes during resuscitation
• Attach the servo-control temperature probe with insulating cover.
Before and at NICU admission:
• Increase the ambient air temperature in the admission room to 23ºC to 25ºC before admission
• Prewarm the admission bed (Giraffe OmniBed) to 36.5ºC and water-circulating heating pad to 40ºC
• Prewarm all surfaces with which the infant will come into contact
• Take the infant's axillary temperature with the polyethylene wrap on and before moving the infant to the admission bed
• Leave the polyethylene wrap on until the temperature is 36.5ºC or higher
• Dry the infant with a warm blanket after the polyethylene wrap is removed.
A nursing champion initiated the QI effort to prevent hypothermia in 2006. While developing the thermoregulation guideline, she provided education to increase staff awareness of neonatal hypothermia by:
• Displaying a "thermoregulation" poster board with mechanisms of heat loss and the prevention of hypothermia in newborn infants
• Introducing, displaying, and demonstrating the effective use of the chemical warmer and polyethylene wrap for immediate postdelivery thermal management of all extremely low-birth-weight infants
• Introducing a one-hour self-study module on neonatal thermal physiology and the complications and prevention of hypothermia and hyperthermia.
Interested nursing staff on the Thermoregulation Committee were responsible for presenting a two-and-a-half-hour mandatory training on thermoregulation for all staff who attend deliveries. In addition, another two hours of thermoregulation education was provided as part of resuscitation simulation training for the delivery room.
Data Collection and Analysis
We developed a prospective data collection form for all VLBW deliveries. Data collected included the following: date and time of birth, birth weight, gestation, specific thermal equipment used (eg, chemical mattress, polyethylene wrap), axillary temperature taken in the delivery room and on admission to the NICU; ambient air temperatures in the delivery room/operating room and admission room, and surface temperatures of the resuscitation and admission beds.
We evaluated the success of implementation of the standardized care using process measures, including use of polyethylene wrap and a chemical warming mattress in the delivery room, serial temperature measurements in the delivery room, and temperature measurement on admission to the NICU. The primary outcome was prevention of hypothermia, defined as an axillary temperature below 36ºC on admission to the NICU. We evaluated hyperthermia (> 37.5ºC) on admission as a balancing measure to capture any inadvertent adverse effect result from our intervention. We also evaluated, as a secondary outcome during this period, survival without serious morbidity including necrotizing enterocolitis, chronic lung disease, nosocomial infection, severe intraventricular hemorrhage, and severe retinopathy of prematurity.
We monitored the percentage of preterm VLBW infants with hypothermia and hyperthermia every year from 2006 to 2011. We submitted data on preterm infants who had a birth weight ranging from 401 to 1500 g, and were at 22 0/7 to 29 6/7 weeks of gestation, to the CPQCC, which is a state collaborative that focuses on benchmarking and facilitating QI work throughout California. We compared the primary and secondary outcomes in VLBW infants less than 33 weeks of gestation born in our unit with outcomes at similar California Children's Services-level NICUs in the CPQCC. The purpose was to benchmark our outcomes after risk adjustment for gestational age, small for gestational age, congenital anomalies, 5-minute Apgar, multiple gestations, male sex, maternal race, and no prenatal care. Figures 3 and 4 were obtained from the CPQCC report Web site with permission.10
The mean gestational age was 28 weeks (standard deviation (SD) = 2.4 weeks), and mean birth weight was 1075 g (SD = 280 g) in our overall study population. Of these, 69% were delivered by cesarean and 95% were exposed to antenatal steroids. Table 2 describes the gestational age, birth weight, and percentage of newborns who received antenatal steroids and who were delivered by cesarean in our VLBW population each year from 2006 to 2011.
Process, Outcome, and Balancing Measures
The process measures were used in 100% of the preterm births, indicating that we were successful in implementing standardized delivery room care for preterm infants (Table 3). We achieved 0% hypothermia within 2 years after implementation of the Smart Start protocol in 2007. We had 2 cases of admission temperatures of 35.6ºC and 35.8ºC, which led to a 6% hypothermia rate in 2010. We noticed an inadvertent increase in hyperthermia as the admission temperatures improved. This balancing measure was addressed in real time, with a change in the thermoregulation guidelines to remove the polyethylene wrap once the temperature was above 36.5ºC.
For primary outcome, the overall percentage of hypothermia in VLBW infants was reduced from 44% in early 2006 to 0% by 2009. There was a slight increase to 6% in 2010, but the rate returned to 0% in 2011. Figure 3 shows the yearly risk-adjusted trend of hypothermia for our unit, and Figure 4 shows the 3-year aggregate (2009 to 2011) risk-adjusted inborn hypothermia rate in our unit compared with the hypothermia rate in same-level regional NICUs that participate in the CPQCC. The 3-year aggregate percentage of hypothermia for our center was substantially lower than the average inborn hypothermia rate for regional NICUs (1.5% vs 15.6%).
Figure 5 shows the yearly risk-adjusted trend of survival without serious morbidities (secondary outcome) for our unit, and Figure 6 shows 3-year aggregate (2009 to 2011) risk-adjusted rate in all regional NICUs in CPQCC.
We succeeded in preventing hypothermia in VLBW infants using standardized delivery room care, including a portable chemical warmer, polyethylene wrap, and a dedicated staff to monitor temperature every 5 minutes in the delivery room. Even before eliminating hypothermia, we noted a major improvement in the admission temperatures, with elimination of extremes of low temperature below 35.5ºC within a year of initiating this QI project. Subsequently, we were able to eliminate the use of warming lights in the delivery room because we had better servo-controlled radiant warmers for resuscitation and we began maintaining the ambient temperature above 25°C. We had an increase in the percentage of infants with axillary temperature in the hyperthermic range (> 37.5ºC) as our admission temperatures were improving. This was addressed by a slight modification in the protocol of reducing the set servo-control temperature on the resuscitation bed from 37.5°C to 37°C and removing the polyethylene wrap after the admission temperature reached 36.5ºC. Because we followed the temperatures prospectively, we were able to address the hyperthermia before it became a major problem. Real-time feedback to the staff, including case reviews and monitoring the real-time hypothermia trend in monthly QI meetings, reinforced staff awareness and adherence to the thermoregulation protocol.
Prospective data collection, feedback to individual clinicians, and a review of the percentage of hypothermia cases monthly along with state and national benchmarks have been key to the overall sustainability of this QI effort. This requires strong nursing support for staff education, data collection, real-time feedback to clinicians by the "temperature police," and both nursing and medical leadership support to provide the time for the staff.
One of the potential confounding factors of this study is the simultaneous implementation of our "Standardized Scripted Approach to Delivery Room Resuscitation of Preterm Infants—Smart Start." This approach was implemented after two days of simulation-based training. The thermoregulation protocol was one of the aspects that were included in this training. Interventions such as delayed cord clamping and multidisciplinary team readiness for resuscitation of all high-risk deliveries, also part of the Smart Start protocol, may have had a protective effect in the thermoregulation and positively confounded the effect on admission temperatures.
We monitored survival without serious morbidities as a secondary outcome because hypothermia has been shown to be associated with late-onset sepsis, intraventricular hemorrhage, and death.1,11 Concomitant with the near elimination of hypothermia was a trend toward improved survival without severe morbidity. It is reassuring that not only did our intervention prevent hypothermia but it also potentially had a positive impact on the overall outcome of the premature infants. However, this impact on survival without serious morbidities could have been confounded by our multiple other QI efforts that were ongoing in our unit, such as the central line infection prevention collaborative, intraventricular hemorrhage prevention bundle, comprehensive bundled care of all VLBW infants including early parenteral nutrition immediately after birth, and a major focus on breast milk nutrition.
We were able to implement a bundled approach to prevent hypothermia in preterm infants despite varying levels of experience among those responding to deliveries (eg, pediatric house staff, hospitalists, neonatal nurse practitioners, and neonatologists). All centers with high-risk deliveries are required to have a resuscitation team ready and available at all times. After implementation of this QI effort at our center, we were able to replicate the approach at an affiliate hospital with a community-level NICU (data not presented).
The addition of our thermoregulation intervention to the delivery room management of preterm births was a successful strategy at our institution and may have improved overall outcomes. We can potentially eliminate hypothermia in VLBW infants by standardizing their care with evidence-based interventions. Optimizing thermoregulation of the preterm neonate requires consistent use of multiple kinds of thermal equipment and close monitoring of temperature. This requires extensive staff education and real-time feedback to individual caregivers to enforce adherence to the thermoregulation guidelines. Finally, the rapid-cycle PDSA model for improvement was critical to implementing this QI effort. v
The author(s) have no conflicts of interest to disclose.
We would like to thank all delivery room staff, particularly the NICU nursing staff, neonatal nurse practitioners, hospitalists, neonatologists, and the pediatric house staff for their enthusiastic participation in achieving and sustaining this improvement work. We could not have done this without the support of our nursing, medical, and administrative leaders or the funding support from the VMC Foundation, San Jose, California, First 5 of Santa Clara County, and a Patient Safety Grant from the Cardinal Health Foundation, Dublin, OH. We would like to thank the National Association of Public Hospitals and Health Systems, Washington, DC, for recognition of this work with the Gage Award Honorable Mention.
Kathleen Louden, ELS, of Louden Health Communications provided editorial assistance.
1. Laptook AR, Salhab W, Bhaskar B; Neonatal Research Network. Admission temperature of low birth weight infants: predictors and associated morbidities. Pediatrics 2007 Mar;119(3):e643-9. DOI: https://doi.org/10.1542/peds.2006-0943
2. Bhatt DR, White R, Martin G, et al. Transitional hypothermia in preterm newborns. J Perinatol 2007 Dec;27 Suppl 2:S45-7. DOI: https://doi.org/10.1038/sj.jp.7211842
3. MacDonald MG, Mullett MD, Seshia MMK, editors. Avery's neonatology: pathophysiology and management of the newborn. 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005 Apr 8.
4. Thermal protection of the newborn: a practical guide [monograph on the Internet]. Geneva, Switzerland: Maternal and Newborn Health/Safe Motherhood Unit, World Health Organization; 1997 [cited 2013 Feb 26]. Available from: http://whqlibdoc.who.int/hq/1997/WHO_RHT_MSM_97.2.pdf. p 64.
5. American Society of Heating. 2003 ASHRAE handbook: HVAC applications. Atlanta, GA: American Society of Heating, Refrigeration and Air-Conditioning Engineers, Inc (ASHRAE); 2003. Chapter 7, p 7.5-7.7.
6. McCall EM, Alderdice FA, Halliday HL, Jenkins JG, Vohra S. Interventions to prevent hypothermia at birth in preterm and/or low birthweight babies. Cochrane Database Syst Rev 2005 Jan 25;(1):CD004210. DOI: https://doi.org/10.1002/14651858.CD004210.pub2
7. Bell R, Finer N, Halamek L, et al. Delivery room management: quality improvement toolkit [monograph on the Internet]. Stanford, CA: California Perinatal Quality Care Collaborative; 2011 Jul [cited 2013 Feb 26]. Available from: www.cpqcc.org/quality_improvement/qi_toolkits/delivery_room_management_toolkit_revised_july_2011.
8. International Liaison Committee on Resuscitation. 2005 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations. Part 7: neonatal resuscitation. Resuscitation 2005 Nov-Dec;67(2-3):293-303. DOI: https://doi.org/10.1016/j.resuscitation.2005.09.014
9. Rabe H, Diaz-Rossello JL, Duley L, Downswell T. Effect of timing of umbilical cord clamping and other strategies to influence placental transfusion at preterm birth on maternal and infant outcomes. Cochrane Database Syst Rev 2012 Aug 15;8:CD003248. DOI: https://doi.org/10.1002/14651858.CD003248.pub3
10. CPQCC Reports [Web site on the Internet]. Stanford, CA: California Perinatal Quality Care Collaborative; c2008 [cited 2013 Feb 26]. Available from: www.cpqccreport.org (password protected).
11. Miller SS, Lee HC, Gould JB. Hypothermia in very low birth weight infants: distribution, risk factors and outcomes. J Perinatol 2011 Apr;31 Suppl 1:S49-56. DOI: https://doi.org/10.1038/jp.2010.177