Factors influencing hypothermia in very low/extremely low birth weight infants: a meta-analysis

Introduction

Neonatal hypothermia is characterized by an abnormal thermal state in which a newborn’s body temperature falls below 36.5 °C (97.7 °F). Its prevalence ranges from 31% to 78% worldwide (Mank et al., 2016; Caldas et al., 2018). Based on core body temperature, the World Health Organization (WHO) categorizes hypothermia into three groups: cold stress (36.0 °C to 36.4 °C, 96.8 °F to 97.5 °F), moderate hypothermia (32.0 °C to 35.9 °C, 89.6 °F to 96.6 °F), and severe hypothermia (<32.0 °C, <89.6 °F) (Feldman et al., 2016). Very low/extremely low birth weight (VLBW/ELBW) infants have a severely limited ability to maintain core body temperature compared to healthy newborns, and so have limited vasoconstriction capacity, higher surface to mass ratio, and less brown fat deposition, which is essential for non-fermentative thermogenesis (Fairchild et al., 2011; Fawcett, 2014). Hypothermia occurs soon following birth in preterm infants, especially in VLBW and ELBW infants, and remains a worldwide problem that can lead to a variety of adverse consequences (Nemeth, Miller & Bräuer, 2021). It has been reported that every 1 °C drop in neonatal body temperature could increase the risk of death by 28% (OR = 1.28, 95% CI [1.16~1.42]) (Laptook et al., 2018) and the occurrence of neonatal hypothermia is significantly associated with the incidence of necrotizing entero-colitis (NEC), intraventricular hemorrhage (IVH), and late-onset sepsis (LOS) (Jakuskiene, Daugeliene & Vollmer, 2009; Shah et al., 2014; Cavallin et al., 2020). Therefore, it is urgent to provide theoretical guidance to improve the survival rate, quality of survival, and administrated direction for VLBW/ELBW infants. Although several works (Almeida et al., 2009; Ali, Osman & Mustafa, 2022; Wilar & Lestari, 2022) have focused on influencing factors, the factors associated with hypothermia of VLBW and ELBW infants remain under-evaluated due to limited prospective data and the inconsistency of the study population. Therefore, there is still a lack of clinical consensus on the risk factors associated with hypothermia due to few existing published studies and recommendations to support healthcare professionals’ decision-making. In this study, we conducted a systematic search for studies on factors influencing the occurrence of hypothermia in VLBW and ELBW infants. Additionally, we performed a meta-analysis of the relevant factors in order to obtain scientifically valid results to guide clinical practice.

Materials and Methods

Results

Initially, 202 relevant papers were retrieved, 31 duplicates were excluded by Endnote software, and 137 papers were excluded after reading the titles and abstracts and according to the inclusion and exclusion criteria set for this study. A secondary screening of the remaining 34 papers was performed, and 10 papers were finally included. Reasons for exclusion included unavailability of full text (n = 5), NOS score <6 (n = 1), systematic evaluation and meta-analysis (n = 3), unadjusted OR values (n = 1), language discrepancy (n = 2), and no clearly defined influence factors or OR values (n = 12). The literature screening process is shown in Fig. 1. Because the P value of Begg’s test was greater than 0.05, this demonstrated that the literature included in this study had no publication bias. This study has been registered on the PROSPERO platform under registration number CRD42022342782.

Body weight

Of the included papers, six involved the effect of body weight on the occurrence of hypothermia (Zhou et al., 2010; Miller, Lee & Gould, 2011; Akter, Parvin & Yasmeen, 2013; Lee et al., 2019; Yu et al., 2020; Gao et al., 2022), and the overall effect (P = 0.32, 95% CI [1.00–1.00]) of body weight on the occurrence of hypothermia was determined to be not statistically different by meta-analysis (Fig. 2).

Discussion

Failure to keep warm in time involves low room temperature, prolonged time in the room, and no systematic use of hypothermia prevention. Studies have confirmed that for every 2–3 °C drop in deep body temperature, skin temperature drops by 4–6 °C within 30 min of birth. Therefore, to prevent rapid evaporation of heat from newborns, WHO recommends that room temperature should be maintained at 25 °C (Russo et al., 2014). Prolonged time in the room is a risk factor for hypothermia in VLBW/ELBW infants, and may lead to heat evaporation during routine nursing operations and resuscitation due to varying levels of proficiency of health care workers and failure to keep VLBW/ELBW infants warm in a timely manner (Reilly et al., 2015). Hypothermia prevention strategies mainly include the use of polyethylene or polyurethane plastic bags (caps) with variable heat mattresses, and studies have shown that the implementation of these measures can reduce evaporation and transepidermal water loss and maintain constant body temperature by maintaining skin moisture in VLBW/ELBW infants (Sharma et al., 2020).

Neonatal resuscitation includes resuscitation needed, intubation at delivery room, and history of neonatal asphyxia. Neonatal resuscitation may cause hypothermia due to the lack of cerebral blood flow, causing damage of brain tissue and the thermoregulatory center. On the other hand, the main focus of the medical staff during the resuscitation of an infant with asphyxia is on saving the infant’s life, but this may result in prolonged exposure and increased heat evaporation (Zayeri et al., 2007; Ogunlesi et al., 2008).

It has been confirmed that VLBW/ELBW infants have a significant difference in body weight compared to full term infants due to their younger gestational age (Pinheiro et al., 2014). Body weight is closely related to heat production, storage, and thermoregulation in VLBW/ELBW infants. In terms of heat production, these infants have lower birth weight and less brown fat, which is the main source of heat production in infants, and therefore, they do not produce enough heat compared to normal weight infants. The immature development of organs leads to the imperfect development of the thermoregulation center, which aggravates the occurrence of hypothermia (Boo & Guat-Sim Cheah, 2013; Tay et al., 2019).

In the literature we examined, maternal combined complications included gestational hypertension, gestational diabetes mellitus, and chorioamnionitis. Maternal combined complications are often associated with a risk of preterm delivery and related medications, such as hormones and magnesium sulfate, that are used more frequently and can easily lead to muscle relaxation and respiratory depression, increasing the risk of resuscitation and exposure of the child (Greenberg et al., 2011).

The results of this meta-analysis indicated that Cesarean delivery is one of the risk factors for the development of hypothermia in VLBW/ELBW infants, which may be caused by the fact that maternal heat can be generated during the second stage of labor during vaginal delivery by exertion, which promotes the increase of neonatal body temperature. At the same time, the use of anesthetic drugs during Cesarean delivery is not conducive to the recovery of maternal and neonatal thermoregulatory center functions after surgery (Johannsen, Vochem & Neuberger, 2017).

Our meta-analysis suggested that the 5 min Apgar score was a factor influencing the occurrence of hypothermia in VLBW/ELBW infants. Hypothermia may be a result of the neonatal thermoregulatory center being impaired due to asphyxia or other complications. It is also possible that hypothermia is caused by neglecting the application of warming measures during resuscitation (Croop et al., 2020).

Although this study was based on an evidence-based correlation approach to analyze the factors influencing hypothermia in VLBW/ELBW infants, there were several limitations in this meta-analysis. Above all, there was a large heterogeneity across some of the factors. Though we attempted to explore the sources of heterogeneity using subgroup analysis, the effect was minimal. After research and discussion, the reasons may be as follows: first, there were differences in the diagnosis of hypothermia among studies. Although WHO proposed a threshold of 36.5 °C for hypothermia, some of the studies used 36.0 °C or 35 °C as the threshold, which increased the heterogeneity among studies due to different inclusion criteria. In addition, there was inconsistency in the way neonatal temperature was measured. Among the included studies, neonatal temperature measurement methods included rectal, axillary and/or skin, and the inconsistency of temperature measurement methods may have led to differences in temperature reporting between studies, thus increasing the heterogeneity of the results of this meta-analysis. On the whole, there were large differences in temperature variation among different ethnic groups. The American Collaboration for Advancing Pediatric Quality Measures (CAPQuaM) confirmed by analyzing data on newborns in New York State that there were differences in temperature variation among different ethnic groups of newborns, and black infants were more likely to have hypothermia compared with white infants (Lawrence, 2016). In the literature included in this study, the studies covered Asian, European, and American populations, and the differences in body mass of different ethnic groups may have affected the OR values of influencing factors in risk prediction models, thus leading to greater heterogeneity. Second, although the overall risk bias of this study was assessed as low by NOS, the different diagnostic criteria for hypothermia among studies made it possible that there was some information bias in this study. At the same time, because the influencing factor meta-analysis methodology only extracted the OR values for analysis, this study team was unable to explore the potential interactions among the influence factors, indicating that the final results of the influence factor analysis may have some confounding bias. Third, the number of included studies was limited, which may lead to imprecise results of the meta-analysis. The reason for this is the special population of VLBW/ELBW infants, which makes it difficult to obtain more clinical cohort data for research and analysis. Furthermore, due to the particularity of the population, relevant research focused on clinical nursing quality improvement projects, and the discussion and quantification of influencing factors were not involved.

Conclusion

Through literature generalization, we determined a total of 16 factors influencing hypothermia in VLBW/ELBW infants. Of these, four influencing factors (race, age, socio-economic status, and spontaneous labor) were included in only one study, and so the meta-analysis was not performed on them. The remaining 12 factors were included in the meta-analysis, and the results showed that failure to keep warm in time, neonatal resuscitation, gestational age, maternal combined complications, Cesarean section and 5-min Apgar score were contributing factors for hypothermia in VLBW/ELBW infants.

Supplemental Information