Conflict of Interest Declaration
The ILCOR Continuous Evidence Evaluation process is guided by a rigorous ILCOR Conflict of Interest policy. The following Task Force members and other authors were recused from the discussion as they declared a conflict of interest: none applicable
The following Task Force members and other authors declared an intellectual conflict of interest, and this was acknowledged and managed by the Task Force Chairs and Conflict of Interest committees:
- Authors Dawson and Kamlin have published studies on humidified gases used for resuscitation {Dawson 2014 24, McGrory 2018 47, Owen 2014 156} and are excluded from decisions about inclusion or bias assessment for these studies.
- Author Trevisanuto has published a study on use of a plastic cap during newborn resuscitation {Trevisanuto 2010 914} and is excluded from decisions about inclusion or bias assessment for this study.
- Author de Almeida has published a study that reported effects of various delivery room and subsequent interventions on temperature outcomes and is excluded from decisions about inclusion or bias assessment for this study. {de Almeida 2014 271}
- Author Trevisanuto has published a study about effective temperature under a radiant warmer {Trevisanuto 2011 720} and is excluded from decisions about inclusion or bias assessment for this study.
- Authors Trevisanuto and de Almeida wrote a recent review article on maintaining normothermia in newborn infants at birth. {Trevisanuto 2018 333}
- Author Ramaswamy is an author of a previous network meta-analysis of methods to maintain normal temperature in infants in the delivery room. {Abiramalatha 2021 e210775}
CoSTR Citation
Dawson JA, Ramaswamy VV, de Almeida MF, Trang J, Trevisanuto D, Nakwa F, Kamlin C, Weiner G, Wyckoff MH, Liley HG on behalf of the International Liaison Committee on Resuscitation Neonatal Life Support Task Force.
Maintaining normal temperature immediately after birth in preterm infants Consensus on Science with Treatment Recommendations [Internet] Brussels, Belgium:, [DATE]. Available from: http://ilcor.org
Methodological Preamble and Link to Published Systematic Review
A previous systematic review (NRP 599) conducted for ILCOR concluded that; “Among newly born preterm infants of less than 32 weeks of gestation under radiant warmers in the hospital delivery room, we suggest using a combination of interventions, which may include environmental temperature 23°C to 25°C, warm blankets, plastic wrapping without drying, cap, and thermal mattress to reduce hypothermia (temperature less than 36.0°C) on admission to NICU (weak recommendation, very low-quality evidence).
We suggest that hyperthermia (greater than 38.0°C) be avoided due to the potential associated risks (weak recommendation, very-low-quality evidence).
A companion review (NRP 589) Temperature Maintenance in the Delivery Room – Prognosis concluded that; "For the critical outcome of mortality, there is evidence from 36 observational studies of increased risk of mortality associated with hypothermia at admission (low-quality evidence but upgraded to moderate-quality evidence due to effect size, dose-effect relationship, and single direction of evidence)". {Perlman 2015 S204} The same systematic review concluded that "There is evidence of a dose effect on mortality, suggesting an increased risk of at least 28% for each 1° below 36.5°C body temperature at admission and dose-dependent effect size". {Perlman 2015 S204}
An update of this review was initiated following a priority list from the ILCOR NLS Task Force, and because it was a recommendation of an ILCOR Evidence Update of the previous review. {Wyckoff 2020 S185}
The review was led by a group of Task Force members. The previous systematic review focused on infants <32 weeks’ gestation. For this update, on the basis of the recommendation in the Evidence Update, it was decided to perform companion reviews; NLS 5100 which focused on late preterm and term infants ≥34 weeks’ gestation {Ramaswamy 2022 81} and the current review (NLS 5101) which focuses on infants born at <34 weeks’ gestation. The decision about the gestation threshold for each review was based on differences in vulnerability to certain critical and important outcomes at different gestations, and also on similar divisions that have been used for other NLS task force reviews.
The continuous evidence evaluation process for the production of Consensus on Science with Treatment Recommendations (CoSTR) started with a systematic review {Ramaswamy et al. 2023 Maintaining normothermia after birth in infants < 34 weeks’ gestation: Systematic review (citation to follow)} conducted by a team comprised of Neonatal Life Support Task Force members and content experts.
PICOST
The PICOST (Population, Intervention, Comparator, Outcome, Study Designs and Timeframe)
Population: Preterm infants (less than 34 weeks' gestation at birth)
Intervention: Increased room temperature ≥23.0°C or thermal mattress or plastic bag or wrap or hat or heating and humidification of gases used for resuscitation or radiant warmer (with or without servo control) or early monitoring of temperature or warm bags of fluid or swaddling or skin to skin care with mother or combinations of these interventions *
Comparators: Drying alone or with use of a plastic bag or wrap
* Note that comparisons between interventions or combinations of interventions will also be meta-analysed if there are sufficient trials and participants to draw meaningful conclusions.
Since we expect that most studies will compare bundles of interventions rather than single interventions, we will also focus on the following likely clustered interventions and comparators:
Intervention: Thermal mattress and wrap or bag with or without a cap or hat and radiant warmer
Comparator: Wrap or bag with or without a cap or hat and radiant warmer
Intervention: Wrap/bag and cap/hat and increased room temperature and radiant warmer
Comparator: Wrap or bag with radiant warmer
Intervention: Thermal mattress and wrap or bag with or without a cap or hat and radiant warmer and heated and humidified gases
Comparator: Wrap or bag with or without a cap or hat and radiant warmer
Outcomes: Primary outcomes:
- Survival to hospital discharge (critical)
- Rate of normothermia on admission to neonatal unit or postnatal ward (important)
Secondary outcomes:
- Body temperature (and rates of moderate hypothermia, cold stress, and hyperthermia) on admission to neonatal unit or before transfer to neonatal unit or postnatal ward, or at times ≤ 1 hour of age (as defined by authors).
- Response to resuscitation, e.g., need for assisted ventilation, highest FiO2
- Major morbidity: bronchopulmonary dysplasia (important), intraventricular hemorrhage all grades (important) and severe (critical), necrotising enterocolitis (important), respiratory distress syndrome (surfactant treatment for), late onset sepsis.
Study Designs: Randomized controlled trials (RCTs) and non-randomized studies (non-randomized controlled trials, interrupted time series, controlled before-and-after studies, cohort studies) are eligible for inclusion. Unpublished studies (e.g., conference abstracts, trial protocols) are excluded. For this review, cohort studies were considered eligible if there was a defined strategy to ensure that the participants were either all of those who received an exposure of interest in a defined population (e.g., infants born at a particular hospital between specified dates), or they were sampled in a way that they can be considered representative of such a population.
Timeframe: All years and all languages are included as long as there is an English abstract
No date restrictions were placed on the search because the question was modified since it was last run in November 2014. The literature search was updated to 20 July 2022
PROSPERO Registration Number CRD42021267301 31st August 2021.
Risk of bias and certainty of evidence was assessed per outcome. Where there was more than one study addressing the comparison of interest, but they have sufficiently different risk of bias as to preclude combining them, this is noted.
A search of MEDLINE (OVID interface), Embase (OVID interface), CINAHL, Cochrane Central Register of Controlled Trials, International Clinical Trials Registry Platform, US, Australian/New Zealand, and European clinical trials registries identified 4822 unique references. After review of titles and abstracts, 108 records were shortlisted for full text review. One additional article was found through hand searching of references lists of included articles and relevant systematic reviews. Seventy-four studies were identified (29 RCTs, 16 observational studies and 29 quality improvement studies) which addressed the PICOST question. Of these, 18 of the RCTs and 7 of the observational studies provided data that could be extracted to evidence tables (for various comparisons between interventions) for the review. {Ahmed 2013 169, Bhavsar 2015 23, Chantaroj 2011 S32, Chawla 2011 780, de Almeida 2014 271, Farhadi 2012 19, Ibrahim 2010 795, Knobel 2005 304, Lewis 2011 160, Mathew 2013 317, McCarthy 2013 e135, McCarthy 2011 1534, McGrory 2018 47, Meyer 2015 245, Pinheiro 2011 357, Reilly 2015 262, Reilly 2019 37, Simon 2011 33, Singh 2010 45, Smith 2013 235, te Pas 2010 e1427, Trevisanuto 2010 914, Vohra 1999 547, Vohra 2004 750} Among the 13 pairs of interventions from RCTs and 10 pairs of interventions from observational study for which evidence tables were developed, 5 comparisons (designated comparisons 2 to 6 below) were considered to provide sufficient data to allow the development of Evidence to Decision (EtD) tables that would inform the development of treatment recommendations. For 3 of these EtD tables (comparisons 2, 4 and 5), evidence from both RCTs and an observational studies was available and was included. For the other EtD tables only RCT evidence was used because either there were no observational studies, or the evidence from them was assessed as of such low certainty that they would not change the conclusions. The results for other comparisons for which evidence was found are summarized in a narrative.
Title and abstract screening, selection of articles from full text reports, risk of bias assessment and data extraction were performed by pairs of authors, with conflicts resolved by a third author or consensus of authors. These steps were performed in Covidence, (Veritas Health Innovation, Melbourne, Australia) except for data extraction. Analyses were performed in R {R Core Team 2019 } Development of evidence tables, GRADE assessment of certainty of evidence {Guyatt 2011 383} and development of evidence to decision tables were performed in GRADEPro.
For assessment of risk of bias, Cochrane Risk of Bias 2 was used for randomized and pseudorandomized trials, {Sterne 2019 i4898} and ROBINS-I was used for observational studies. {Sterne 2016 i4919} A risk of bias assessment was not performed for the QI studies, because all used multi-faceted interventions and presented data in a way that precluded assessment of the effect size of each component intervention.
The term standard hospital care for control groups appeared to have been variable in different studies. When this was uncertain or may have varied for different infants in the study, we used the term according to authors’ use. Few studies that were examining other interventions reported ambient temperatures or humidity.
Other definitions:
- Moderate hypothermia: Body temperature <36°C {WHO 1997 }
- Cold stress: Body temperature 360C to 36.4°C {WHO 1997 }
- Hyperthermia: Body temperature >37.5°C {WHO 1997 }
- Bronchopulmonary dysplasia (BPD): As defined by study authors but usually include a criterion of needing oxygen or respiratory support at greater than or equal to 36 weeks post-menstrual age.
- Intraventricular hemorrhage: intraventricular or periventricular hemorrhage (severe if Papile grade III or IV) {Papile 1978 529}
- Necrotizing enterocolitis (NEC): necrotizing enterocolitis (Bell stage III or greater) {Bell 1978 1}
- Respiratory distress syndrome: respiratory distress syndrome treated with endotracheal surfactant
- Late onset sepsis: positive blood or CSF culture for a pathogen after the first 7 days of life (or as defined by study authors)
Consensus on Science:
Treatment Recommendations
Comparison 1. Increased room temperature ≥23.0°C vs lower room temperature:
In preterm infants (<34 weeks' gestation), as for late preterm and term infants (≥34 weeks’ gestation), we suggest the use of room temperatures of ≥23°C compared to 20°C at birth in order to maintain normal temperature. (Weak recommendation, very low certainty evidence).
Comparison 2. Thermal mattress vs no thermal mattress:
In preterm infants (< 34 weeks’ gestation) immediately after birth, where hypothermia on admission is identified as a problem, it is reasonable to consider the addition of a thermal mattress, but there is a risk of hyperthermia. (Conditional recommendation, low certainty evidence).
Comparison 3. Plastic bag or wrap vs no plastic bag or wrap
In preterm infants (<34 weeks’ gestation) immediately after birth we recommend the use of a plastic bag or wrap to maintain normal temperature. (Strong recommendation, moderate certainty of evidence).
Temperature should be carefully monitored and managed to prevent hyperthermia. (Good practice statement).
Comparison 4. Cap vs no cap
In preterm infants (<34 weeks’ gestation) immediately after birth we suggest the use of a head covering to maintain normal temperature. (Strong recommendation, moderate certainty evidence).
It is reasonable to consider the use of a plastic cap unless another form of head covering is used. (Conditional recommendation, moderate certainty evidence).
There is currently little published evidence that head coverings of other materials are effective in preterm infants (< 34 weeks’ gestation), but they may also help maintain normothermia based on an observational study and studies in infants ≥ 34 weeks’ gestation.
Comparison 5. Heated and humidified gases compared to no heating and humidification:
In preterm infants (<34 weeks’ gestation) immediately after birth, we suggest heated and humidified gases for respiratory support in the delivery room can be used where audit shows that admission hypothermia is a problem and resources allow. (Conditional recommendation, very low certainty evidence)
Comparison 6. Radiant warmer - servo control vs manual mode
In preterm infants (<34 weeks’ gestation) immediately after birth there is insufficient published human evidence to suggest for or against the use of a radiant warmer in servo-controlled mode compared to manual mode for maintaining normal temperature. (Weak recommendation, moderate certainty evidence).
Comparison 7. Skin to skin care vs no skin to skin care:
In preterm infants (<34 weeks' gestation), after birth there is insufficient published human evidence to suggest for or against the use of skin to skin care immediately after birth. Skin to skin care may be helpful for maintaining normal temperature where few other effective measures are available. (Good practice statement).
Justification and Evidence to Decision Framework Highlights
In prioritizing an update of an ILCOR systematic review in 2015 of measures to maintain normal temperature after birth {Perlman 2015 S204} the NLS Task Force noted recent evidence from large observational studies supporting previous findings that hypothermia on admission, but also hyperthermia is associated with adverse outcomes. {de Almeida 2014 271, Wilson 2016 61}
A common theme of the Evidence to Decision tables for each comparison is that each included study examined the effectiveness of each intervention in the context of several other interventions aimed at maintaining normal temperature and the size of effect may differ in the presence of fewer or more cointerventions than those used in the included studies. A combination of several different interventions (which operate through different mechanisms) is likely to be needed for most preterm infants. This approach was the focus of most of the quality improvement studies, which in most cases used a bundle of care or staged introduction of different measures (e.g., using plan-do-study-act cycles). However, the review did not identify sufficient evidence for any specific bundle. The design of such bundles should consider the strength of international recommendations and the certainty of evidence for each component but should also depend on evaluation of local circumstances such as environmental and architectural considerations and available resources.
Risk of harm from hyperthermia is likely to be higher when multiple interventions are used concurrently. Another over-arching theme of the Evidence to Decision tables is that early measurement of temperature may detect when additional measures are needed for individual infants and regular audit is needed to ensure that strategies achieve maintenance of normal temperature for most infants.
A recent network meta-analysis of randomised and quasi randomised trials in infants <32 weeks’ gestation {Abiramalatha 2021 e210775} provides support for the finding of this systematic review, that the highest certainty evidence available is for use of a plastic bag or wrap. The network meta-analysis also found that the evidence supports certain bundles, notably the concurrent use with a plastic cap or heated and humidified gas, both for maintaining normal temperature and for reduced risk of brain injury or death. The authors also note the limitations of the available evidence.
Comparison 1: Increased room temperature ≥23.0°C vs lower room temperature:
This topic was prioritized by the NLS Task Force because of publication since the previous ILCOR systematic review of a large cluster-RCT {Duryea 2016 505.e1} in infants of all gestations that enabled development of a treatment recommendation for infants ≥34 weeks’ gestation. In addition, its inclusion in several QI studies assessing bundles of care suggests that the practice is regarded as feasible and is used commonly in some highly resourced sites. The Task Force considered that it was a potentially important part of a suite of interventions to maintain normal temperature immediately after birth in preterm infants.
However, the small number of preterm infants (<34 weeks’ gestation) in the few studies found for inclusion in the systematic review and the marked variation in study design (including temperatures in the intervention and control groups) precluded meta-analysis and yielded only very low certainty evidence. This very limited evidence suggests that ambient delivery room or operating theatre temperatures ≥23.0°C may improve body temperatures and reduce risk of cold stress and moderate hypothermia. Because of the limited evidence, an Evidence to Decision Table was not developed.
However, the Task Force notes that for preterm infants, the use of additional measures to maintain normal temperature (including prewarmed bedding and radiant warmers, plastic bags or wraps and other measures) may have resulted in a lower impact of ambient temperature on outcomes of preterm infants than in infants of higher gestation, as suggested by the authors of the cluster RCT. {Duryea 2016 505.e1}
Despite the sparse evidence for infants <34 weeks’ gestation, the Task Force considers that the treatment recommendation developed for infants ≥34 weeks’ gestation {Ramaswamy 2022 81} is likely to apply to all infants. Furthermore, changing room temperatures for infants of different gestations may not be feasible, so it is reasonable to suggest temperatures ≥23.0°C for all infants.
In low resource settings in hot climates, ambient temperatures may be very high and this could result in hyperthermia in some mothers and infants. A safe upper limit of ambient temperature was not identified from the available studies.
Comparison 2. Thermal Mattress vs no Thermal Mattress:
This topic was prioritized by the NLS Task Force to determine whether there were new RCTs published since those identified for the previous systematic review. {Perlman 2015 S204} However, no more recent studies were found. The analysis was conducted again because of the differences in protocol between the current and previous review.
The evidence from this systematic review is that there was moderate certainty evidence for a small improvement in the number of normothermic participants with use of a thermal mattress, and very low certainty evidence for a small improvement in mean body temperature. Of note, one of the two included RCTs was stopped early, although it is unclear from the paper whether this was because of a suspected efficacy or harm.
Possible harm was identified in regard to an increased risk of hyperthermia in infants exposed to the thermal mattress. To underly the importance for careful monitoring when a thermal mattress is used in conjunction with a radiant warmer the British Association of Perinatal Medicine (BAPM) has warned users to be aware of the risk of hyperthermia and skin burns. (Safety Issue - Transwarmer Mattresses | British Association of Perinatal Medicine (bapm.org) ww.bapm.org/articles/44-safety-issue-transwarmer-mattresses). Care should be taken to use a barrier layer of towelling or sheeting between the mattress and the infant to prevent skin burns and reduce the risk of hyperthermia, because they can reach temperatures of at least 42°C.
The effect sizes for both benefits and potential risks (such as hyperthermia) may be modified by the use of measures to maintain normal temperature other than those that were routine in the included studies.
There were no data to determine cost effectiveness or effects on equity. The cost of a thermal mattress varies between brands, but they may be unaffordable in low- and middle-income countries. However, using a thermal mattress was described in one of the articles as technically easy. Their use to promote thermoregulation is already standard care in some high resource settings described in publications reporting quality improvement activities. Many neonatal retrieval/transport services use thermal mattresses to maintain normal temperature during inter-hospital transport. The Task Force considered that thermal mattresses might have a specific role in settings where there is a particularly high risk of hypothermia, such as for out of hospital births where other forms of thermal support such as radiant warmers are unavailable. However, no studies were found to confirm this.
For these reasons, a conditional recommendation was made.
Comparison 3. Plastic bag or wrap vs no plastic bag or wrap
Plastic bags or wraps have been recommended by ILCOR as one of several measures for maintaining normal body temperature in preterm infants after birth, in order to prevent adverse outcomes including death. {Perlman 2015 S204} Several additional studies have been published since the last systematic review, and inclusion of these allowed upgrading of the overall certainty of evidence from very low to moderate. Plastic bags or wraps may prevent heat loss by reducing evaporative or convective heat losses.
The feasibility of using plastic bags or wraps is well established and their use is now considered standard of care in many neonatal services. It is important that the plastic bag or wrap is opened or lifted as infrequently as possible until the baby is transferred into a prewarmed humidified incubator, otherwise the benefits of the bag or wrap are likely to be reduced.
Plastic bags or wraps are likely to be an inexpensive method to improve thermoregulation in very preterm infants. However, the costs depend on whether bags or wraps specifically marketed for newborn care are used, or widely available products such as those used for food storage.
The Task Force also considered the environmental impacts of recommending widespread use of plastic bags or wraps. However, this must be weighed against benefits, and also compared with the widespread use of other disposables in clinical care.
Plastic bag or wrap vs no plastic bag or wrap was the only comparison in this review for which there were sufficient data to enable some of the prespecified subgroup analyses. These suggested that plastic bags or wraps are more efficacious at preventing hypothermia in lower gestation infants (<28 weeks) than those of higher gestation and in high income vs middle income countries, although there were no subgroup differences for other outcome measures. However, the included studies confirmed that there is a high risk of hypothermia in all preterm infants <34 weeks’ gestation. A plastic bag or wrap was still efficacious in the higher gestation subgroup.
The certainty of evidence was high for survival, the critical primary outcome of the review and moderate or low for all other outcomes, so a strong recommendation was developed.
Comparison 4. Cap vs no cap
A head covering (cap or bonnet) has been recommended by ILCOR as one of several measures for maintaining normal body temperature in preterm infants after birth, in order to prevent adverse outcomes including death. {Perlman 2015 S204} The scalp is a large proportion of body surface area in a preterm infant and hence will make a large contribution to evaporative, radiant, conductive or convective heat loss.
The systematic review found only one RCT eligible for inclusion. This may be because caps or hats are widely construed to be feasible and inexpensive and their use is now considered standard of care in many neonatal services, so studies may have not been considered necessary. They were specified as used for all infants in many of the RCTs and observational studies examining other interventions that were included in the review.
The one included RCT examined the use of a plastic cap. However, caps made of other materials might also be effective. Supporting this, a retrospective study that used logistic regression to identify factors associated with hypothermia in preterm infants identified that not using a linen or woollen cap (as opposed to using a cap) predicted hypothermia. {de Almeida 2014 271}
There is also indirect evidence from the companion ILCOR systematic review in late preterm and term infants: For the comparison woollen vs cotton cap, the review found one RCT enrolling 126 participants that examined two outcomes relevant to the review and found small differences in mean temperature and the rate of moderate hypothermia favouring the woollen cap group. {Lang 2004 843, Ramaswamy 2022 81}
The Task Force also considered the environmental impacts of recommending widespread use of plastic caps. However, this must be weighed against benefits, and also compared with the widespread use of other disposables in clinical care.
The single study that was included provided moderate certainty evidence for benefit for the important primary outcome of normothermia and other secondary outcomes, and found no undesirable effects. In view of this and the other supportive evidence, a strong recommendation was made.
Comparison 5. Heating and humidification of gases used for resuscitation, vs no heating and humidification
Use of warmed and humidified gases is regarded as routine in mechanically ventilated patients in intensive care units, to prevent damage to the lungs. {Sottiaux 2006 } Heating and humidification may also play a key role in maintaining normal body temperature, {Gillies 2017 Cd004711} particularly in preterm infants. The mean temperature of piped wall air has been measured as 23.4°C and mean relative humidity 5.4%. {Dawson 2014 24} However, clinical trials of the use of heated and humidified gases during initial resuscitation at birth have only been published recently.
The systematic review found 2 RCTs and 1 observational study that found improvements in some secondary outcomes of the review. For the important secondary outcome IVH > Grade 2 there was probable clinical benefit, albeit in two RCTs for which the combined number of participants fell well below the optimal information size (OIS) for this (and most other outcomes). The result of reduced IVH>grade 2 could easily be a type I error and would need to be replicated in larger studies.
Any potential harmful effect of using non-heated and non-humidified gases for resuscitation in the delivery room may be limited if respiratory support is only required for a short time. However, it is not established whether there is a safe duration for ventilation using non-heated and non-humidified gases before admission to a NICU.
The two studies included in the review took place in high resource settings. One of the studies estimated that the additional equipment required to provide heated humidified gas in the delivery room could cost US$50 for the single-use humidification circuit and chamber. {McGrory 2018 47} The true cost is likely to be higher as this estimate did not include the cost of the warmer, temperature probes and sterile water required to provide the intervention. Heated and humidified gases for newborn resuscitation are unlikely to be available in low- or middle-income countries or other settings where resources are limited. Where resources are limited, providing heated and humidified gases in the delivery room could divert resources away from the NICU. It is also very unlikely that heated and humidified gases would be available in an out-of-hospital setting.
For these reasons, the Task Force developed a conditional recommendation that suggests the use of heated and humidified gases in settings where audit shows that other measures have been insufficient to prevent hypothermia in many infants, and where resources allow.
Comparison 6. Radiant warmer - servo control vs manual mode
Infants in the neonatal intensive care unit are generally nursed under a radiant warmer or in an incubator. The temperature of the incubator or radiant warmer can be adjusted manually or by servo controlling warmer output to achieve a set neonatal body temperature measured at the site of a skin sensor. The previous ILCOR systematic review of Warming Adjuncts (and a subsequent evidence update) made no comment regarding use of manual or servo mode to control radiant warmers used in the delivery room.
For use in the delivery room to maintain normal temperature in preterm infants (<34 weeks’ gestation) the systematic review found no difference in primary outcomes of the review and possible evidence of harm (more infants mildly hypothermic in the group exposed to servo control). The servo control system was set at 37°C and manual mode was set at maximum output for ten minutes before the birth of each infant. We do not know the effect of specific manual mode settings and set temperatures in servo mode other than those used in the study on the outcomes of interest. The balance of benefits and risks might also vary by gestation and size of the infant.
Servo control was considered to have possible risk of harm because of the need for skin or rectal probes to continuously monitor temperature, for each of which published studies report uncommon or rare risks. {Mishra 2021 1627, Morley 1992 122}
The one included study was performed in a high resource setting, and no evidence was found to adequately assess costs. Servo control is not possible for all models of radiant warmer used in delivery rooms, particularly in low- or middle-income countries. Servo controlled radiant warmers are likely to be more expensive than manual mode radiant warmers. In addition to the cost of the device there is an additional cost for the sensors that need to be applied to an infant's skin. In a low resource setting providing servo control may be unaffordable.
In view of the single study which found lack of evidence for benefit but possible harm, a weak recommendation indicating that there is insufficient evidence to recommend for or against use of servo control was made.
Comparison 7. Skin to skin care vs no skin to skin care:
The systematic review found insufficient direct evidence to form a treatment recommendation. However, there is good evidence for the benefits of skin to skin care for maintaining normal temperature immediately after birth in late preterm and term infants {Ramaswamy 2022 81} and for maintaining subsequent normal temperature when used soon after birth for low and very low birth weight infants in low and middle income countries. {Arya 2021 2028} Therefore, a good practice statement was made suggesting use of skin to skin care for preterm infants (<34 weeks’ gestation), particularly when very few other measures are available.
Knowledge Gaps
For all the comparisons:
- Studies are needed comparing specific bundles of interventions to maintain normal temperature vs other specific bundles, to identify the balance or benefits and risks when combinations of interventions are applied.
- What is the effect of different interventions or bundles of interventions in the setting of different levels of ambient temperature and humidity?
- No cost-effectiveness studies were found for any intervention.
- Apart from some limited data for plastic bag or wrap, there is no information to determine whether there is any difference in safety and effectiveness by various subgroups (e.g., by gestation, by setting in a high-, middle- or low-income country or a high- or low-resource setting, by inborn vs outborn location of birth or by exposure to different forms of umbilical cord management.
Other knowledge gaps specific to individual interventions:
Comparison 1. Increased room temperature ≥23.0°C vs lower room temperature:
- What is the effect of various set temperatures (besides 20 or 23°C) for operating room or birthing room temperatures?
- Are the results found for operating room temperatures applicable to other birthing locations?
Comparison 2. Thermal mattress vs no thermal mattress:
- Is there a specific role for thermal mattresses for births in prehospital settings?
Comparison 4. Cap vs no cap
- What are the balance of risks and benefits when using head coverings composed of different materials?
Comparison 5. Heated and humidified gases compared to no heating and humidification:
- Does use of heated and humidified gases during resuscitation reduce lung injury?
- Does use of heated and humidified gases from birth reduce the risk of severe IVH in studies that meet the OIS for this outcome, and if so, what is the mechanism?
Comparison 6. Radiant warmer - servo control vs manual mode
- Is there a specific role of servo control in maintaining normal temperature in preterm infants requiring prolonged resuscitation?
- Are there servo-controlled devices that could be adapted for use during deferred cord clamping?
- Does position of the temperature sensor probe affect the outcomes?
Comparison 7. Skin to skin care vs no skin to skin care:
- For preterm infants (<34 weeks' gestation), which other interventions to maintain normal temperature are most efficacious and which can be safely adapted for use during skin to skin care?