Foundations in Newborn Care. Occlusive Bags to Prevent Hypothermia in Premature Infants

Linda Ikuta, MN, RN, CCNS, PHN Section Editor Foundations in Newborn Care Occlusive Bags to Prevent Hypothermia in Premature Infants A Quality Improvement Initiative Kathleen Godfrey, DNP, NN P-BC, CPNP; Donna G. Nativio, PhD, CRNP, FAAN; Charles V. Bender, MD; Elizabeth A. Schlenk, PhD, RN ABSTRACT The aim of this quality i mprovement initiative was to improve the neonatal intensive care unit (NICU) admission rectal temperatures of premature infants less than 28 weeks gestation by placing them in an occlusive bag from the neck down immediately after birth. The historical control group consisted of a convenience sample of 46 very low-birth-weight infants from March 1, 2010, to August 31, 2010. A convenience sample of 35 very low-birth-weight infants from October 1, 2010, to April 30, 2011, was recruited during the prospective phase. A quasi-experimental design was used. A retrospective medical record review was performed to collect data on NICU admission rectal temperatures for the historical control group. During the prospective phase, infants were placed in a bag from the neck down immediately after birth and NICU admission rectal temperatures were recorded. In both groups, NICU rectal temperatures were measured immediately upon admission. Application of the bag resulted in a higher mean NICU admission rectal temperature in the intervention group compared with the historical control group. Occlusive bags applied at delivery decreased heat loss in premature infants. The results support previous findings and resulted in a change in clinical practice. Key Words: hypothermia, infant, low birth weight, occlusive wrap, premature The incidence of hypothermia in premature infants is estimated to be between 42% and 73%. 1 The body temperature of premature infants drops precipitously after delivery, which is associated with an increase in mortality and morbidity rates. 2 Improvements in survival rates and longterm outcomes have been demonstrated when premature infants are kept warm immediately after birth and during the early postnatal period. 3 At birth, premature infants are exposed to a colder temperature than they had experienced in the womb. Following birth, there is a significant drop in their surrounding temperature. Mance 4 identified multiple factors that contribute to heat loss in premature infants. Infants lose heat through conduction, convection, and evaporation. Evaporative heat loss is a substantial contributor to decreased body temperature in premature newborns immediately after delivery. Author Affiliations: Department of Health Promotion and Development, University of Pittsburgh School of Nursing (Drs Godfrey and Nativio); Department of Pediatrics and Obstetrics and Gynecology, University of Pittsburgh School of Medicine (Dr Bender); and Department of Health and Community Systems, Univers ity of Pittsburgh School of Nursing (Dr Schlenk), Pennsylvania. This quality improvement initiative was conducted at the University of Pittsburgh School of Nursing and Magee-Womens Hospital of University of Pittsburgh Medical Center, Pennsylvania. This quality improvement initiative was supported in part by a grant from the Leslie A. Hoffman Endowed Research Award. The authors thank Mary Kish, DNP, NNP-BC, who assisted with neonatal intensive care unit staff education and data collection. The authors declare no conflict of interest. Correspondence: Kathleen Godfrey, DNP, NNP-BC, CPNP, School of Nursing, University of Pittsburgh, 3500 Victoria St, Room 440, Pittsbu rgh, PA 15261 (godfreyk@pitt.edu). Copyright 2013 by The National Association of Neonatal Nurses DOI: 10.1097/ANC.0b013e31828d040a Advances in Neonatal Care Vol. 13, No. 5 pp. 311-316 311

312 Godfrey et al Early and effective interventions in the delivery room are an utmost priority if hypothermia is to be prevented. Preventive actions are directed toward reducing heat loss and providing warmth by utilizing external heat sources. At the time this quality improvement (QI) initiative was conducted, the 2005 neonatal resuscitation guidelines 5 offered recommendations for healthcare professionals responsible for the resuscitation of neonates at birth. These guidelines were updated later and continue to provide guidance for neonatal care. 6 Despite these general neonatal thermoregulatory guidelines, standard care has failed to prevent hypothermia at birth in premature infants. New interventions to improve the effectiveness of measures to prevent heat loss include providing barriers to heat loss. Barrier interventions focus on reducing evaporative heat loss and include the application of occlusive bags. In this regard, the 2005 neonatal resuscitation guidelines recommended placing premature infants less than 28 weeks gestation into a polyethylene bag, in addition to the standard postdelivery care, to prevent hypothermia. 5 PROBLEM Despite the current recommendations supporting this clinical practice, delivery room resuscitative practices at Magee Womens Hospital (MWH) of University of Pittsburgh Medical Center (UPMC) did not include the evidence-based practice of placing very low-birth-weight infants in an occlusive bag after delivery. The delivery room resuscitation protocol followed the neonatal resuscitation guidelines 5,6 ; however, the protocol did not include this particular suggestion for an additional heat loss preventiv e measure. A literature search was conducted to find evidence on the effectiveness of occlusive bags for the thermal management of premature infants immediately after birth to support a change in clinical practice. Three databases (PubMed, Embase, and CINAHL) were searched using the following terms: infant, premature; infant diseases and low birth weight; temperature and hypothermia; wraps, occlusive, and polyethylene. Seven relevant articles were identified. 1,3,7-11 These studies found that premature infants of various gestational ages, all less than 32 weeks, had significantly less heat loss initially after delivery and significantly less hypothermia upon admission to the neonatal intensive care unit (NICU) after being placed in bags from the neck down at birth. The quality of this evidence was graded high, 12 which signifies that the evidence included consistent results from well-designed, well-conducted studies in representative populations, assessed the effects of the intervention on health outcomes, supported the use of occlusive bags in the delivery room to prevent heat loss in premature infants, and provided conclusions that are unlikely to be strongly affected by the results of future studies. This QI initiative used this research evidence to support a change in clinical practice after documenting that hypothermia was a problem in premature infants born at less than 28 weeks gestation upon admission to the NICU. PURPOSE The purposes were to (1) improve the NICU admission rectal temperatures of premature infants born at less than 28 weeks gestation by placing them in an occlusive bag from the neck down immediately after birth, (2) implement and evaluate the clinical protocol, and (3) establish a change in clinical practice for the thermal management of very lowbirth-weight infants immediately after birth. METHODS Design A quasi-experimental design with a historical control group was used. Phase 1 consisted of a retrospective medical record review covering the period March 1, 2010, to August 31, 2010, and phase 2 consisted of prospective data collection from October 1, 2010, to April 30, 2011. Approval for this QI initiative was granted by the UPMC Total Quality Council. The University of Pittsburgh institutional review board approval was not required to conduct the QI initiative or publish its results. Setting and Sample The setting was a 71-bed level III NICU in a fullservice women s hospital, MWH of UPMC, which serves as the leading teaching facility for obstetrics, gynecologic oncology, and neonatology for the University of Pittsburgh. More than 10,000 babies are born at MWH each year. The NICU at MWH is the largest in Pennsylvania and treats more than 1600 seriously or critically ill newborns annually. Convenience sampling was used. For both phases, the inclusion criterion was premature infants born at less than 28 weeks gestation at MWH. For both phases, the exclusion criteria were abdominal wall defects, neural tube defects, blistering skin conditions, multiple congenital anomalies, and infants of mothers who received no prenatal care. Equipment A 50 50-cm sterile transparent polyethylene bag with a tie string closure was used. This bag was selected because it was familiar and available to the NICU staff being routinely used to cover newborn abdominal wall and neural tube defects at delivery. The bags were located in the NICU and additional bags were also stocked on the bottom of the transport www.advancesinneonatalcare.org

Hypothermia Prevention 313 incubators in the birthing center where the delivery and operating rooms are located. Procedures We modeled the procedures of our QI initiative after the Vohra et al. 10 study. In both phases, we assigned identification numbers to the infant and maintained the key of identification numbers and corresponding names and medical record numbers in a locked cabinet. During phase 1, we performed a retrospective medical record review to extract data on the infants: birth date and time, gender, gestational age, birth weight, Apgar scores, delivery site, presence of exclusion criteria, NICU admission time, NICU admission rectal temperature, and temperature date and time. The maternal temperature before delivery, along with temperature date and time, were also extracted. During the last month of phase 1, we conducted staff education to inform the NICU physicians (MDs), registere d nurses (RNs), neonatal nurse practitioners, and respiratory therapists of the QI initiative. Educating the staff members about the adverse effects of hypothermia and the implementation of the occlusive bag for all infants at less than 28 weeks gestation made them more aware of the role they play in keeping premature infants warm. The clinical protocol was presented to the NICU MDs at a monthly staff meeting. Several mandatory training sessions were arranged for RNs, neonatal nurse practitioners, and respiratory therapists who would attend deliveries where the bags were to be used. A copy of the educational presentation and reminder flyers were also posted throughout the NICU. Training for the nursing staff included both didactic and hands-on experiences. Didactic content about the causes and detrimental effects of hypothermia and the importance of preventing hypothermia in premature infants was reviewed. Research evidence was provided showing the effectiveness of bags on the temperature management of premature infants. Inclusion and exclusion criteria were discussed. Presentation of this information reinforced the need for a change in clinical practice. Hands-on training included a simulated delivery room scenario using an infant mannequin to demonstrate and return-demonstrate correct application of the bag. The training emphasized that the application and closure of the bags did not interfere with resuscitation. Staff members were told that if at any time during the resuscitation they believed that the bag was interfering with resuscitation or compromising patient safety, it should be removed and standard care provided. To avoid opening the bag in the delivery room, staff members were instructed to delay foot printing, banding, and weighing the infant until after the infant was transported and the bag was removed in the NICU. In addition to obtaining and applying the bag to eligible infants, RNs were responsible for completing documentation about application of the bags. They were trained in the correct method for documentation and informed that audits would be regularly conducted to ensure proper documentation. Staff members were reminded about the clinical protocol and documentation and their questions were answered as they arose. During phase 2, infants were placed in a prewarmed bag from the neck down immediately after birth before drying. Hats were then placed on their heads and resuscitation was continued as per the neonatal resuscitation guidelines. 5 When stabilized, the infants were transported to the NICU in a preheated transport incubator set at 36 C. Upon arrival to the NICU, the infants were taken from the transport incubator and placed on a preheated radiant warmer. The bag was removed. The infants rectal temperatures were taken and recorded immediately upon arrival in the NICU. In addition, during phase 2, the infants records were reviewed monthly to ensure compliance with documentation and to extract data on birth date and time, gender, gestational age, birth weight, Apgar scores, delivery site, presence of exclusion criteria, the use of the bag, NICU admission time, NICU admission rectal temperature, and temperature date and time. The maternal temperature prior to delivery, along with temperature date and time, were also extracted. Protocol deviations were recorded, including the bag not being applied to eligible infants and the bag being removed early; reasons for these deviations were also recorded. Measures NICU Admission Rectal Temperature The main outcome measure was NICU admission rectal temperature measured immediately upon admission using a MediChoice Water-Proof Digital Thermometer. On NICU admission, infants received a new thermometer for their personal use. According to the package insert, the MediChoice Water-Proof Digital Thermometers (Owens & Minor, Mechanicsville, Virginia) are accurate to ±0.1 C from 32 C to 42.9 C. We could not find any studies documenting reliability of this thermometer. Demographic and Clinical Factors and Potential Confounders Basic demographic (gender) and clinical (gestational age, birth weight, Apgar scores, and maternal temperature) factors and 2 potential confounders (delivery site and time from delivery to NICU admission rectal temperature) were collected by medical record review to describe the sample. Because delivery room temperatures were not available and operating room temperatures have been reported to be cooler than delivery Advances in Neonatal Care Vol. 13, No. 5

314 Godfrey et al room temperatures, 13 we recorded the delivery site (operating room for cesarean sections and delivery room/adult inte nsive care unit [AICU] for vaginal deliveries). Time from delivery to NICU admission rectal temperature was calculated using birth date and time and infant temperature date and time. Data Analysis SPSS version 19.0 (IBM, Armonk, NY) was used for data analysis and a P value less than.05 was considered statistically significant. Becuase this was a QI initiative, a power analysis was not conducted. An intention-to-treat approach was used, in which data on infants were analyzed regardless of protocol deviations. Data analysis was performed using analysis o f covariance (ANCOVA) to determine whether there was a statistically significant difference in mean NICU admission rectal temperatures between the infants who received the bag and the infants in the historical control group who did not receive the bag, controlling for any significant baseline group differences found in demographic (gender) and clinical factors (gestational age, birth weight, Apgar scores, and maternal temperature) and potential confounders (delivery site and time from delivery to NICU admission rectal temperature). A Cohen d value with 95% CI was computed to determine the clinical significance of the use of the bag on NICU admission rectal temperatures. Using the normal temperature range of 36.5 C to 37.5 C, 14 a chi- square test was conducted to examine the proportion of infants who were hypothermic, normothermic, and hyperthermic in both groups. Descriptive statistics were used to describe the 2 groups. Groups were compared on gender and delivery site, using a chi-square test and on gestational age, birth weight, Apgar scores, time from delivery to NICU admission rectal temperature, and maternal temperature using parametric independent t tests or nonparametric Mann-Whitney U tests. If group comparisons demonstrated statistically significant differences, the factors were controlled for in the ANCOVA. RESULTS From September 1, 2010, to September 30, 2010, we used a 12 15-inch, 14-mL, latex-free, plastic reclosable bag, which was consistent with the neonatal resuscitation guidelines. 5 The staff members reported difficulty applying and securing the bag and auscultating heart sounds through the bag. From October 1, 2010, to April 30, 2011, a second more pliable polyethylene bag with a drawstring neck closure was selected. The staff members reported no difficulty with the second bag. The results reported are on data collected using only the second bag. As shown in Table 1, the 2 groups were similar on all baseline characteristics except for delivery site. The TABLE 1. Sample Description Characteristic Historical Control (n = 46) Intervention (n = 35) t (P) or χ 2 (P) Birth weight, M (SD), g 780.78 (145.00) 822.20 (145.88) t (79) = 1.270 (.208) Gestational age, M (SD), wk 26.1 (1.1) 26.0 (0.9) t (79) = 0.213 (.832) Apgar score, M (SD) at 1 min 4.78 (2.99) 4.29 (2.49) t (79) = 0.795 (.429) at 5 min 6.67 (2.37) 6.69 (2.15) t (79) = 0.023 (.982) Time from birth to NICU admission rectal temperature, minutes, M (SD) 29.73 (20.36) 27.18 (14.43) t (75) = 0.612 (.542) Maternal temperature, 36.67 (0.49) 36.82 (0.65) t (72) = 1.178 (.243) M (SD), C Gender, n (%) Males 30 (65.20) 24 (68.60) χ 2 (80) = 0.101 (.751) Delivery site, a n (%) Operating room 16 (34.80) 22 (62.90) χ 2 (80) = 6.290 (.012) Delivery room/aicu 30 (65.20) 13 (37.10) Boldface indicates statistically significant. Abbreviation: AICU, adult intensive care unit; NICU, neonatal intensive care unit. a P <.05. www.advancesinneonatalcare.org

Hypothermia Prevention 315 historical control group had more vaginal deliveries in the delivery room/aicu (n = 30, 65.20%), whereas the intervention group had more cesarean deliveries in the operating room (n = 22; 62.90%) (χ 2 (80) = 6.290, P =.012). Thus, delivery site was controlled for in the ANCOVA. Unexpectedly, infant temperatures were not significantly different between delivery room/ AICU and operating room sites in the entire sample (M = 35.72 C, SD = 1.13 vs M = 35.72 C, SD = 0.78; t(79) = 0.001, P =.999), historical control group (M = 35.51 C, SD = 0.94 vs M = 35.45 C, SD = 0.70; t(44) = 0.212, P =.833), or intervention group (M = 36.21 C, SD = 1.40 vs M = 35.91 C, SD = 0.79; t(33) = 0.798, P =.431). There were 35 infants in the intervention group that weighed on average 822.20 g (SD = 145.88 g) at birth, were a mean of 26.03 (SD = 0.90) weeks gestation, and 68.60% (n = 24) were male with a mean Apgar score of 6.69 (SD = 2.15) at 5 minutes. The historical control group was composed of 46 infants who weighed on average 780.78 g (SD = 145.00 g) at birth, were a mean of 26.08 (SD = 1.11) weeks gestation, and 65.20% (n = 30) were male with a mean Apgar score of 6.67 (SD = 2.37) at 5 minutes. The application of the bag resulted in higher NICU admission rectal temperatures in the intervention group (M = 36.02 C, SD = 1.05 C, range 33.40 C-39.30 C) compared with the historical control group (M = 35.49 C, SD = 0.86 C, range 33.00 C-37.60 C) controlling for delivery site, F(2,78) = 3.46, P =.036. Cohen s d was 0.56 (95% CI: 0.112-1.008), indicating a medium effect of the bags on NICU admission rectal temperature. As shown in Table 2, chi-square test results showed that a smaller proportion of infants in the intervention group (74.3%; n = 26) compared with the historical control group (84.8%; n = 39) were hypothermic. A larger proportion of infants in the intervention group were normothermic compared with the historical control group (20.0%, n = 7 vs 13.0%, n = 6) as well as hyperthermic (5.7%, n = 2 TABLE 2. Neonatal Intensive Care Unit Admission Rectal Temperature Categories by Group Temperature Category Hypothermia (<36.5 C) Normothermia (36.5 C-37.5 C) Hyperthermia (>37.5 C) Historical Control (n = 46), n (%) Intervention (n = 35), n (%) 39 (84.8) 26 (74.3) 6 (13.0) 7 (20.0) 1 (2.2) 2 (5.7) vs 2.2%, n = 1). Although these results were in the expected direction, they were not significant (χ 2 (79) = 1.545, P =.462). Protocol Deviations and Adverse Events A few protocol deviations occurred during the prospective phase. Five of 35 infants did not have the bag applied because of forgetting (n = 4) and a critical event (n = 1). Of the 30 infants who had the bag applied, the bag was opened for resuscitation in 4 infants. There was 1 infant in the historical control group and there were 2 infants in the intervention group with NICU admission rectal temperatures higher than 37.50 C (37.60 C, 39.30 C, and 38.10 C, respectively). The infant in the historical control group was born to a mother with a temperature of 36.7 C. The first infant in the intervention group was born to a mother who was febrile with a temperature of 38.10 C and the other to a mother with a temperature of 37.40 C. Fetal temperature is higher than the maternal temperature by 0.5 C. 15 One would expect the 2 infants in the intervention group to have a warmer baseline temperature at delivery that, in conjunction with the bag, could account for their higher admission temperatures. Cramer et al 3 stated that in randomized controlled trials where infants were found to experience hyperthermia, the rise in temperature was associated with factors other than the use of the bag, including maternal infection and temperature. DISCUSSION The findings supported the clinical protocol for the thermal management of very low-birth-weight infants immediately after birth. The results supported previous findings that the use of bags placed immediately after birth on premature infants decreased heat loss upon admission to the NICU. 1,3,7-11 The 0.53 C difference in mean NICU admission rectal temperature between the groups was clinically significant and similar to what was reported in a metaanalysis. 3 Given that the optimal temperature for term infants only has a range of 1 C(36.5 C- 37.5 C), 14 moving a population mean by more than half a degree (0.5 C) is both clinically and statistically significant. 3 The education regarding the causes and detrimental effects of hypothermia in premature infants and the support provided reinforced the need for and the benefits of this simple, noninvasive, and practical intervention. Since the results were presented to the NICU staff, the use of bags became a documented standard of care in this hospital, consistent with recommendations of the neonatal resuscitation guidelines. 5,6 Based on staff feedback to facilitate use, bags are now located under the radiant warmers in all the Advances in Neonatal Care Vol. 13, No. 5

316 Godfrey et al delivery and operating rooms as well as in the NICU and under the birthing center transport incubator. LIMITATIONS A limitation of this QI initiative was the convenience sample. Another limitation was the use of retrospective data from the NICU database to measure variables for the historical control group. The clinical protocol was also conducted with nonrandomized groups; however, the groups were similar on infant characteristics except for delivery site, suggesting that there was minimal selection bias. The longitudinal nonrandomized design with historical controls meant that the higher NICU admission rectal temperatures observed in the intervention group might have been influenced by changes in skill mix and/or nursery management practices. However, during the time the protocol was implemented, there was no significant staff turnover and no substantial nursing care-related practice changes instituted. IMPLICATIONS FOR PRACTICE AND RESEARCH This QI initiative was simple to implement and could easily be replicated in similar settings. Future research could be aimed at expanding the population served by this intervention. Continued use of bags during early stabilization should be further explored since guidelines for early thermal management of very low-birthweight infants remain to be established. Information on the cost-effectiveness of the routine use of this particular bag in comparison to other commercially available occlusive bags or wraps is needed. References 1. Ibrahim CPH, Yoxall CW. Use of plastic bags to prevent hypothermia at birth in preterm infants: do they work at lower gestations? Acta Paediatr. 2009;98:256-260. 2. Watkinson M, BChir MB. Temperature control of premature infants in the delivery room. Clin Perinatol. 2006;33:43-53. 3. Cramer K, Wiebe N, Hartling L, Crumley E, Vohra S. Heat loss prevention: a systematic review of occlusive skin wrap for premature neonates. J Perinatol. 2005;25:763-769. 4. Mance MJ. Keeping infants warm: challenges of hypothermia. Adv Neonatal Care. 2008;8:6-12. 5. American Heart Association, American Academy of Pediatrics. 2005 American Heart Association (AHA) guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiovascular care (ECC) of pediatric and neonatal patients: neonatal resuscitation guidelines. Pediatrics. 2006;117:e1029-e1038. 6. Perlman JM, Wyllie J, Kattwinkel J, et al; on behalf of the Neonatal Resuscitation Chapter Collaborators. Neonatal resuscitation: 2010 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Pediatrics. 2010;126:e1319-e1344. 7. Bredemeyer S, Reid S, Wallace M. Thermal management for premature births. J Adv Nurs. 2005;52:482-489. 8. Knobel RB, Wimmer JE, Holbert D. Heat loss prevention for preterm infants in the delivery room. J Perinatol. 2005;25:304-308. 9. Mathew B, Lakshminrusimha S, Cominsky K, Schroeder E, Carrion V. Vinyl bags prevent hypothermia at birth in preterm infants. Indian J Pediatr. 2007;74:249-253. 10. Vohra S, Frent G, Campbell V, Abbott M, Whyte R. Effect of polyethylene occlusive skin wrapping on heat loss in very low birth weight infants at delivery: a randomized trial. J Pediatr. 1999;134: 547-551. 11. 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. 12. Agency for Healthcare Research and Quality. Guide to clinical preventive services, 2010-2011. Appendix A: How the U.S. Preventive Services Task Force grades its recommendations. AHRQ Pub. No. 10-05145. http://www.ahrq.gov/clinic/pocketgd1011/gcp10app. htm#apa. Published 2010. Accessed August 5, 2012. 13. Bhatt DR, White R, Martin G, et al. Transitional hypothermia in preterm newborns. J Perinatol. 2007;27:S45-S47. 14. Maternal and Newborn Health/Safe Motherhood Unit, Division of Reproductive Health, World Health Organization. Thermal Protection of the Newborn: A Practical Guide. Geneva, Switzerland: World Health Organization; 1997. 15. Rosenberg AA. The neonate. In: Gabbe SG, Niebyl JR, Simpson JL, eds. Obstetrics: Normal and Problem Pregnancies. 5th ed. Philadelphia: Churchill Livingstone; 2007:523-565. www.advancesinneonatalcare.org