Hypothermia in Young Infants

Pediatricians encounter and care for young infants with hypothermia in a variety of clinical settings after their birth hospitalization. Most cases of hypothermia in early infancy, defined here as 90 days old or less, will be attributed to benign causes, including immature thermoregulation or environmental exposures; however, hypothermia can also be a sign of a serious infection or other significant pathology. Recently, numerous studies have been published focusing on hypothermia in young infants. This article will summarize the available literature on infants with hypothermia, acknowledge current limitations, and discuss directions for future research.Hypothermia occurs when the body loses more heat than it can generate or absorb and can affect the function of all organ systems. Young infants, particularly those with low birth weight or prematurity, are at increased risk for hypothermia. Factors that render young infants at risk for hypothermia include having a large surface area compared with body weight, a low percentage of body fat for insulation and heat production, and thin skin, all of which contribute to the body easily losing heat. Systemic inflammation, which can occur in the setting of infection, can also result in hypothermia. Why hypothermia instead of fever sometimes accompanies infection is not well understood, but one theory posits it occurs when there is an altered inflammatory cascade or an atypical immune response.While fever in infants is commonly accepted as a temperature of 38.0 °C or higher, there is no universally accepted definition of hypothermia in infants. Two frequently referenced definitions include (1) a temperature less than 36.0 °C, which is one of the criteria for diagnosing systemic inflammatory response syndrome among children from term birth through adolescence, according to the 2005 International Pediatric Sepsis Consensus Conference, and (2) a temperature less than 36.5 °C, which the World Health Organization uses to define hypothermia in neonates less than 28 days old. In practice, a recent survey from 2023 of pediatric emergency medicine clinicians found that most respondents used a temperature cutoff of 36.0 °C or less for identifying hypothermia.Several challenges exist in interpreting the literature on hypothermia in early infancy. First, there has been heterogeneity in study definitions of hypothermia, as well as in early infancy and serious or invasive infection. Second, several studies have used billing codes to identify participants with hypothermia, which is less precise and likely less accurate than using documented temperatures. Third, many of the studies have been limited by small sample sizes. Compared with fever, hypothermia in early infancy is relatively uncommon, making it more difficult to conduct large-scale studies of this population. That said, sepsis accompanied by hypothermia vs fever has been associated with higher mortality. Lastly, multiple studies have only included infants who had blood cultures collected, thereby limiting the generalizability of the findings to infants who are evaluated for bacteremia. Even with these limitations, the results of this research have contributed to a better understanding of cohort characteristics and outcomes.Hypothermia may be the first sign of serious bacterial infection (SBI) in an infant. The prevalence of SBI, which includes urinary tract infection, bacteremia, and meningitis, among young infants with hypothermia presenting to the emergency department is estimated to be 2% to 8%. Urinary tract infection is the most common SBI. SBI should be suspected in older infants who present with hypothermia. Older infants are more resistant to environmental hypothermia than younger infants due to their relatively high body weight and percentage of body fat and mature systems for thermoregulation; therefore, when hypothermia does occur in older infants, infection is more likely to be the culprit. SBI should also be suspected in infants with hypothermia who have the following history, physical examination, and laboratory findings: a history of complex chronic condition or recurrent temperature instability, ill appearance on examination, abnormal platelet count or elevated absolute neutrophil count on complete blood cell count, or an abnormal urinalysis finding (pyuria, presence of leukocyte esterase, or presence of nitrites). The importance of biomarkers such as C-reactive protein and procalcitonin levels for risk stratification of infants with hypothermia who may have SBI is unclear.Hypothermia can also be associated with neonatal herpes simplex virus (HSV) infection. The American Academy of Pediatrics’ clinical practice guideline for infants with fever recommends HSV testing in infants 8 to 21 days old with hypothermia. The prevalence of HSV infection in young infants presenting with hypothermia is estimated to be 0.2% to 1.3%. Additionally, other (non-SBI and non-HSV) infections are described in infants with hypothermia including viral bronchiolitis, viral meningitis, and pneumonia.Although infection should be front of mind when evaluating infants with hypothermia, infants may become hypothermic from noninfectious etiologies as well. Young infants are at risk for temperature dysregulation from benign physiologic and environmental causes. In addition, hypothermia in early infancy may be caused by inadequate caloric intake due to conditions such as pyloric stenosis or congenital bowel disease or by excess caloric expenditure due to conditions such as congenital cardiac disease. Other diagnoses to consider include genetic abnormalities such as Prader-Willi syndrome, inborn errors of metabolism, and endocrine abnormalities. Clinicians should maintain a broad differential diagnosis when considering etiologies of hypothermia.Currently, there is no standardized, evidence-based approach to evaluating the underlying cause of hypothermia in infancy. As a result, there is significant variation among clinicians in their diagnostic workup and decision-making. Given the broad list of causes of hypothermia as described above, clinicians must take a detailed history and perform a thorough physical examination when evaluating these patients.Some pediatric medical centers include the treatment of well-appearing infants with hypothermia in their clinical care pathway designed for infants with fever. However, because these pathways are based on literature focused on infants with fever, their use in infants with hypothermia may lead to unnecessary invasive testing, hospitalization, and antimicrobial exposure. Nonetheless, if there is reason to suspect infection in this scenario based on the infant’s history or physical examination findings, strongly consider performing HSV testing and starting treatment empirically with acyclovir and antibiotics.On the opposite end of the evaluation workup spectrum, some clinicians do not perform any diagnostic testing on well-appearing infants with hypothermia. It is unclear what information guides clinicians to this decision, but it may be personal experience of these infants generally faring well. A recent study reported significant variation in the care of well-appearing young infants with hypothermia, with no difference in the frequency of missed cases of bacteremia, bacterial meningitis, or HSV by the care they received. Another study described a cohort of neonates 28 days old or less with incidental hypothermia detected at an outpatient visit. Upon laboratory testing, none were found to have SBI or HSV infection. Together, these studies suggest that a full SBI evaluation may not be necessary for all well-appearing infants with hypothermia.An ill-appearing infant with hypothermia warrants a full evaluation for serious infection, including blood, urine, and cerebral spinal fluid bacterial cultures, as well as HSV polymerase chain reaction assays or cultures from blood, cerebral spinal fluid, and surface swabs of the mouth, nasopharynx, conjunctivae, and anus. These infants should be given empirical acyclovir and antibiotics while awaiting test results.Significant pathology aside from infection should remain in consideration, too. Additional testing for noninfectious causes should be tailored based on suggestive findings in the patient’s medical history and on physical examination.Young infants with hypothermia are a unique patient population with their own risk factors for serious infection and other pathology. Future efforts should focus on reaching a consensus on the standard definition of hypothermia in young infants, including temperature and age range limits based on risk for pathology. Having standard definitions would reduce heterogeneity among studies, allowing for comparisons across studies. Research is also needed to determine the utility of biomarkers for SBI risk stratification, followed by the derivation and validation of clinical decision tools to differentiate between low and high risk for SBI in infants with hypothermia.Comment: To avoid unnecessary testing, the decision to evaluate a well-appearing infant with hypothermia should start with confidence that the infant really is hypothermic. After all, most of us can recall a time when an overly chilly triage room lowered an infant’s peripheral temperature, causing us undue alarm. While electronic-thermometer, rectal temperature is considered most reflective of core temperature in term infants less than 3 months old, rectal temperatures require the patient to be unbundled and are not recommended in the setting of prematurity, immunodeficiency, rectal anomaly, or bleeding disorder. During the COVID-19 pandemic, many pediatric practices invested in noncontact infrared thermometers because they are quick, limit the spread of infection between patients, can be used with fully clothed infants, and are very sensitive for detecting fever. That said, if the lens is dirty, readings can be falsely low, and if the thermometer is in sunlight or near radiant heat, readings may be falsely high. The accuracy of infrared thermometers for detecting hypothermia in infants also varies by manufacturer. As such, if hypothermia is detected using an infrared thermometer in a well-appearing, term infant, it may be worth double-checking the reading with a rectal temperature before starting an SBI workup.Linda Y. Fu, MD, MSAssociate Editor, In Brief