a b s t r a c t

Objectives: hypothermic infants are at risk for serious bacterial and herpes simplex virus infections, but there are no Evidence-based guidelines for managing these patients. We sought to characterize variations and trends in care for these infants in the emergency department (ED).

Methods: We conducted a retrospective cross-sectional study of infants under 90 days old presenting to 32 pedi- atric EDs from 2009 through 2019 with an International Classification of Diseases diagnosis code for hypother- mia. We characterized variation in diagnostic testing, antimicrobial treatment, and disposition of children in three age groups (<=30 days, 31-60 days, and 61-90 days old) and analyzed care trends.

Results: Of 7828 ED encounters meeting inclusion criteria, most (81%) were <= 30 days of age. Infants in the 0-30 days old age group, compared to 61-90 days old age group, had a higher proportion of blood (75% vs. 68%), urine (72% vs. 64%), and cerebrospinal fluid (CSF; 35% vs. 22%) cultures obtained (p < 0.01) and greater antimicrobial use (81% vs. 68%; p < 0.01) in the ED. From 2009 to 2019, C-reactive protein (CRP), and procalcitonin usage steadily increased, from 25% to 40% and 0% to 30% respectively, while antibiotic use (83% to 77%), CSF testing (53% to 44%), and chest radiography (47% to 34%) decreased. Considerable interhospital variation was noted in testing and treatment, including CSF testing (14-70%), inflammatory markers (CRP and procalcitonin; 8-88%), and antibiotics (56-92%).

Conclusion: Substantial hospital-level variation exists for managing hypothermic infants in the ED. Long-term trends are notable for changing practice over time, particularly with increased use of inflammatory markers. Prospective studies are needed to risk stratify and optimize care for this population.

(C) 2022

1. Introduction

Abbreviations: ED, emergency department; SBI, serious bacterial infections; UTI, urinary tract infections; PHIS, Pediatric Health Information System; CHA, Children’s Hospital Association; ICD, International Classification of Diseases; CBC, complete blood count; ESR, erythrocyte sedimentation rate; CRP, c-reactive protein; CSF, cerebrospinal fluid; HSV, herpes simplex virus; RSV, respiratory syncytial virus; CXR, chest X-rays; ICU, intensive care unit; CCC, complex chronic conditions; AAP, American Academy of Pediatrics; CPG, clinical practice guideline.

* Corresponding author at: Department of Emergency Medicine, New York Presbyterian Hospital/Weill Cornell Medical Center, 525 East 68th Street, Box 179, Room M-130, New York, NY 10065, United States of America.

E-mail addresses: [email protected] (Y.H.J. Lo), [email protected] (S. Ramgopal), [email protected] (A.N. Hashikawa), [email protected] (J.A. Cranford), [email protected] (A.J. Rogers).

Hypothermia may be a presenting sign of serious infection in young infants presenting to the Emergency Department (ED). Recent retro- spective studies, mainly from single-center sources, suggest that hypo- thermic young infants are at risk for serious bacterial infections (SBI; defined as urinary tract infection [UTI], bacteremia, and/or Bacterial meningitis), and herpes simplex infection [1-11]. The host response of hypothermia to infection is poorly understood but may be related to cy- tokine system abnormalities [12-19] resulting in end-organ dysfunction with impaired thermoregulatory function [20-24].

Data on the ED evaluation and management of hypothermic young infants (<=90 days old) are limited. In the absence of robust research or

https://doi.org/10.1016/j.ajem.2022.07.058

0735-6757/(C) 2022

practice care guidelines, prior single-center work has suggested that there may be variation in the performance of diagnostic testing and treatment for hypothermic infants evaluated in the ED [1-4]. This type of unwarranted variation, which refers to practice pattern differences that are not explained by illness or other medical needs, is an important cause of low-value care [25]. Low-value care in turn can result in over- utilization of potentially harmful treatments and underuse in warrantED treatments. To date no study has used a nationally representative data- base to report on multicenter ED practice patterns for the evaluation and management of hypothermic infants.

Demonstration of practice pattern variation and trends among ED clinicians evaluating hypothermic infants will highlight opportunities to improve and standardize care and identify areas where knowledge gaps exist. To address this, we aimed to use a nationally representative database to describe hospital-level variation and trends in the care of young hypothermic infants evaluated in ED.

1. Methods
   1. Data source

We performed a cross-sectional study using the Pediatric Health In- formation System (PHIS), an administrative database managed by the Children’s Hospital Association (CHA; Lenexa, KS) that contains inpa- tient, ambulatory surgery, observation, and ED encounter-level data from geographically diverse, tertiary-care pediatric hospitals in the United States. PHIS Data quality and reliability are assured through a joint effort between the CHA and participating hospitals [26]. Participat- ing hospitals provide discharge/Encounter data for external bench- marking, including demographics, diagnoses, and procedures. Data are de-identified at submission and are subjected to reliability and validity checks before being included in the database [27]. Data from 32 hospi- tals were included for the current study after excluding hospitals due to incomplete data or data quality issues. This study was considered ex- empt from review by our Institutional Review Board.

• 1. Study population

We included all infants <=90 days old presenting to the ED from Jan- uary 1, 2009, to December 31, 2019, with any International Classifica- tion of Diseases (ICD) revision 9 or 10 diagnosis code for hypothermia, on the basis that the use of any diagnosis code has previously been shown to best identify febrile infants (Supplementary Table 1) [28]. Subjects with missing data, patients born at the hospital during that encounter, or transferred from a non-PHIS facility were excluded.

• 1. Outcomes measures

Our outcomes of interest included diagnostic testing, antimicrobial treatments, and disposition. Diagnostic testing included complete blood count , chemistries, inflammatory markers (C-reactive protein [CRP], and procalcitonin), urinalysis, blood, urine, and cerebrospinal fluid cultures, viral studies, including Herpes simplex virus , respiratory syncytial virus , varicella, influenza tests, and chest x- rays (CXR). Antimicrobial treatments included systemic antibiotics and the antivirals acyclovir and oseltamivir. Tests and antimicrobials in the ED were queried using associated billing codes, with any testing or treat- ment occurring on hospital days 0 or 1 considered performed in the ED (Supplementary Table 2). Disposition included admission (defined as in- tensive care unit [ICU], acute care or Observation status admission) [29], utilization of the ICU or neonatal ICU, and in-hospital mortality.

• 1. Data acquisition and analysis

Demographic data obtained on each patient included sex, ethnicity, and race. We identified patients with complex chronic conditions

(CCC) based on ICD-9 or 10 codes during the index encounter, with con- ditions expected to last longer than 12 months with multi-organ or se- vere single organ involvement requiring pediatric specialty care or hospitalization [30,31].

We stratified infants into three age groups: <=30 days, 31-60 days, and 61-90 days, then performed chi-squared analyses to evaluate inter- group differences in testing, treatment, and disposition. To reflect the recent American Academy of Pediatrics (AAP) clinical practice guideline (CPG) for febrile infants [32], we analyzed the cohort using 0-7, 8-21, 22-28, 29-60, and 61-90 day age bands. Overall testing and treatment rates for each calendar year were plotted and analyzed with chi-squared testing for statistical significance between calendar years. Following a similar approach used to evaluate variation in care for febrile young in- fants [10], we also assessed overall hospital-level variation in testing, antimicrobial, and hospitalization rates and ranked them into quintiles. The sum of ranks was obtained for each hospital, and an overall compar- ative utilization rank was assessed in tertiles: high, moderate, or low utilizers. An alpha of 0.05 was used for all analyses, and all statistical tests were two-tailed. Analyses were conducted using SPSS (version 27.0, IBM Corp., 2020).

1. Results
   1. Patient population

We initially identified 11,371 encounters with infants <=90 days old during our study period using ICD-9 and 10 Diagnosis codes for hypo- thermia. After applying exclusions, 7828 patients were included (Fig. 1). Most patients were <= 30 days old (81.0%), with the majority of Non-Hispanic ethnicity (79.1%) and white race (59.3%). A CCC was identified in 22.4% (Table 1).

• 1. Overall practice pattern and trends

Among all patients, the most frequently obtained tests were CBC (85.2%), blood culture (74.5%), urine culture (71.0%), and urinalysis (70.0%). Inflammatory markers were obtained in 35.9% of patients. CSF testing was pursued in 50.9% of patients. The most common viral test used was HSV (30.3%), followed by influenza (10.7%). CXR were obtained in 39.2% of patients. Antibiotics were administered in 79.5% of patients and acyclovir in 40.0% (Table 2).

Infants 0-90 days old, seen in the ED,

with any hypothermia ICD codes, from 1/1/2009-12/31/2019

n = 11,371

Excluded (n = 1,978):

Born inside hospital (11) Transferred to the ED (1,967)

Excluded (n = 1,565):

Encounters from hospitals without full clinical and billing data during study period Hospital (KC) which generates

additional separate encounter for admissions

Main Cohort

n = 7,828

Fig. 1. Subject Inclusion Flowchart.

Table 1 Patient demographics by age group distributions.
	Total (n = 7828) n (%)	0-30 days (n = 6344) n (%)	31-60 days (n = 1006) n (%)	61-90 days (n = 478) n (%)	P value
Sex Male	4124 (52.7)	3314 (52.3)	553 (55.0)	257 (53.8)	0.23
Ethnicity Hispanic	863 (11.0)	668 (10.8)	126 (12.5)	49 (10.3)	0.57
Non-Hispanic	6190 (79.1)	5030 (79.3)	780 (7.5)	380 (79.5)
Unknown Race White	775 (9.9) 4643 (59.3)	626 (9.9) 3867 (61.0)	100 (9.9) 552 (54.9)	49 (10.3) 224 (46.9)	<0.001
Black	1642 (21.0)	1179 (18.6)	294 (29.2)	169 (35.4)
Asian	359 (4.6)	313 (4.9)	26 (2.6)	20 (4.2)
Other	1184 (15.1)	985 (15.5)	134 (13.3)	65 (13.6)
Complex Chronic Condition	1753 (22.4)	1124 (17.7)	371 (36.9)	258 (54.0)	<0.001

Infants <=30 days old had statistically significantly higher rates of cul- tures from all sites (blood, urine, and CSF), HSV testing, antibiotic and antiviral administration. Conversely, older infants had higher RSV and influenza testing rates, CXR studies, ICU admission, and mortality. When our cohort was subdivided and analyzed in 0-7, 8-21, 22-28, 29-60, and 61-90 days old groups, results were similar with respect to clinical testing and antibiotic administration to the primary analysis (Supplementary Table 3). From 2009 through 2019, there were statisti- cally significant increases in the proportions of CRP (25.0% to 39.6%, p < 0.001) and procalcitonin (0% to 30.0%, p < 0.001) obtained, and statisti- cally significant decreases in antibiotic use (82.4% to 77.0%, p < 0.001), CSF testing (53.7% to 44.5%, p < 0.001) and CXR (46.3% to 33.9%, p <

0.001) (Fig. 2).

• 1. Hospital-level variation

Within the 32 included hospitals in our cohort, ranges of testing were as follows: Urine tests in 61-90%, CBC or blood culture in 4-95%, CSF tests in 14-70%, inflammatory markers (CRP and procalcitonin) in

8-88%, and antibiotics in 56-92%. We identified a similar range of vari- ation when patients were categorized based on age. When ranked by overall utilization of diagnostic tests and antibiotic treatments, 15 hos- pitals (46.9%) remained within the same tertile in both age groups. Of the 12 hospitals in the lowest utilization tertile for the <=30 days old group, 1 (11.1%) were in a higher utilization tertile for the 31-90 days group. Conversely, 8 (80.0%) hospitals in the highest utilization tertile for the <=30 days old group were in a lower utilization tertile for the 31-90 days old group. Moreover, similar utilization variations were noted when the younger age group was further stratified into 0-7, 8-21, and 22-28 days old (Fig. 3).

1. Discussion

We used a multicenter pediatric dataset to evaluate both variation and longitudinal trends in the ED management of hypothermic young infants. We identified wide practice variation for diagnostic testing, par- ticularly with performance of CSF studies, acquisition of inflammatory markers, and antibiotic administration. Furthermore, an evaluation of

Table 2

Testing, treatment, and outcome of hypothermic infants by age.

	Total (n = 7828) n (%)	0-30 days (n = 6344) n (%)	31-60 days (n = 1006) n (%)	61-90 days (n = 478) n (%)	P value
Diagnostic Tests
CBC	6669 (85.2)	5397 (85.1)	861 (85.6)	411 (86.0)	0.720
Chemistries	4715 (60.2)	3750 (59.1)	645 (64.1)	320 (66.9)	<0.001
C-reactive Protein	2579 (32.9)	2092 (33.0)	312 (31.0)	175 (36.6)	0.100
Procalcitonin	630 (8.0)	513 (8.1)	82 (8.2)	35 (7.3)	0.832
Urinalysis	5478 (70.0)	4421 (69.7)	728 (72.4)	329 (68.8)	0.193
Blood culture	5830 (74.5)	4755 (75.0)	751 (74.7)	324 (67.8)	0.001
Urine culture	5561 (71.0)	4535 (71.5)	722 (71.8)	304 (63.6)	0.001
CSF culture	2617 (33.4)	2224 (35.1)	288 (28.6)	105 (22.0)	<0.001
Any CSF testing1	3984 (50.9)	3369 (53.1)	457 (45.4)	158 (33.1)	<0.001
Herpes Simplex Vius2	2375 (30.3)	2127 (33.5)	180 (18.2)	65 (13.6)	<0.001
Varicella	73 (0.9)	65 (1.0)	7 (0.7)	1 (0.2)	0.142
Respiratory Syncytial Virus	1205 (15.4)	850 (13.4)	248 (24.7)	107 (22.4)	<0.001
Influenza	835 (10.7)	600 (9.5)	160 (15.9)	75 (15.7)	<0.001
CXR	3065 (39.2)	2148 (33.9)	608 (60.4)	309 (64.6)	<0.001
Treatments Antibiotics	6221 (79.5)	5110 (80.5)	785 (78.0)	326 (68.2)	<0.001
Acyclovir	3135 (40.0)	2823 (44.5)	237 (23.6)	75 (15.7)	<0.001
Oseltamivir	43 (0.5)	24 (0.4)	10 (1.0)	9 (1.9)	<0.001
Disposition Overall admissions	7061 (90.2)	5722 (90.2)	909 (90.4)	430 (90.0)	0.970
ICU3	3623 (43.3)	2735 (43.1)	600 (59.6)	288 (60.3)	<0.001
Mortality	136 (1.7)	74 (1.2)	33 (3.3)	29 (6.1)	<0.001

Abbreviations: CBC, complete blood count; CSF, cerebral spinal fluid; CXR, chest x-ray; ICU, intensive care unit.

¹ Based on lumbar puncture procedure codes.

² PCR, serology, cultures.

³ Neonatal ICU and Pediatric ICU.

Fig. 2. Testing and Treatment Trends. Annual frequency of diagnostic tests and antibiotic medications from 2009 through 2019 across 32 hospitals in the cohort.

Changes in management demonstrated an increasing use of inflamma- tory markers over time, in the absence of any published evidence on the optimal ED management of hypothermic infants. Our findings sug- gest that hypothermic young infants presenting to the ED receive incon- sistent care and highlight the need for robust research in this patient population to establish evidence-based guidelines.

Association of hypothermia and SBI among hypothermic infants have been previously reported in single-centered studies [1-4], with one PHIS study using a similar cohort evaluating outcomes of hypothermic infants showing an SBI prevalence of 8% [5]. However, this prior work did not an- alyze ED care trends and variations in depth. We observed considerable hospital-level variation in antimicrobial administration, inflammatory markers use, and CSF studies in hypothermic infants presenting to the ED, a finding that was also notable when infants were stratified into two age groups. Furthermore, some hospitals with a high utilization pat- tern in one age group were low utilizers in another and vice versa. Single-center studies have also reported similar diagnostic variations. Perry et al. demonstrated that only a quarter of hypothermic infants at one pediatric academic institution underwent SBI evaluation, and of those patients, only 59% had CSF studies done [4]. In another PHIS- based study on the variation of care in febrile infants <90 days old, CSF testing rates varied from 26% to 77% [11]. While robust research and quality efforts have been made in standardizing the management of fe- brile infants, best practice guidelines remain lacking in infants with hy- pothermia, magnifying the dearth of evidence to help clinicians stratify hypothermic infants into high and low-risk categories.

In 2021, AAP published the CPG on the evaluation and management of young febrile infants, synthesizing decades of research into three age- based risk stratification algorithms: 8-21 days, 22-28 days, and 29-60 days old [32]. This was supported by multiple extensive national pro- spective studies. However, the relationship between age and SBI in hy- pothermic infants has not been fully investigated. Single-centered studies on SBI and hypothermic infants are limited by their retrospec- tive nature, small number of infants, varying cohort selection criteria, and different temperature thresholds and therefore are difficult to gen- eralize. One multicenter retrospective PHIS study did not find statisti- cally significant changes in SBI rates with age [5]. Furthermore, other studies have suggested an association between SBI and older age

among hypothermic infants [3,4]. Therefore, the applicability of the new AAP CPG for febrile infants to hypothermic infants is one which requires greater exploration.

Biomarkers for systemic inflammation have emerged as key screen- ing tools for SBI in febrile infants. The role of inflammatory markers, such as absolute Neutrophil count, CRP, and procalcitonin, are estab- lished in the risk stratification of well-appearing febrile infants [32-36]. When combined with other clinical and laboratory findings, inflammatory markers have high sensitivity in identifying SBI in febrile infants [34]. However, the role of inflammatory markers in hypothermic infants is not clear, a point that may be particularly relevant as some in- vestigators have postulated that hypothermia as a response to sepsis may represent immune dysfunction [37-39]. Studies of adult hypother- mic patients in the setting of Systemic Infection have suggested that hy- pothermia may be representative of an altered anti-inflammatory regulatory response [39], a result of systemic inflammatory activation to bacterial pyrogens [40], or due to endothelial dysfunction during sep- sis [41]. To date, there are no published investigations evaluating the role of inflammatory markers in hypothermic young infants. As such, an increase in testing for inflammatory markers among young hypo- thermic infants may represent an attempt at risk stratification in the absence of robust evidence to support their safe implementation.

Past studies have attempted to identify clinical risk factors for SBI in hypothermic infants, which can assist in efforts to prognosticate disease course, severity, and direct resource allocation. Single-center retrospec- tive studies have suggested prematurity, apnea, poor feeding, lethargy, failure to thrive, and ED physician characterization of patients as “ill-ap- pearing” as associated risk factors for SBI [1,2,4]. In addition, most single-center studies, and the recent PHIS study, suggest that hypother- mic patients with underlying complex chronic conditions are at in- creased risk for SBI. Our analysis identified a higher rate of CCC in the older population. Since the designation of CCC occurs at the encounter level via ICD codes affiliated with that visit, older infants may have more chronic conditions already diagnosed when presenting to the ED with hypothermia. Moreover, when older infants present with hypo- thermia, clinicians may have a higher suspicion for underlying infection, as they are more resistant to environmental hypothermia due to more extensive fat storage and metabolic reserves. These factors may also

Y.H.J. Lo, S. Ramgopal, A.N. Hashikawa et al.

American Journal of Emergency Medicine 60 (2022) 121–127

125

Fig. 3. Hospital-specific laboratory testing and antibiotic administration utilization for the overall cohort, infants <=30 days old and 31-90 days old. <= 30 days old group further stratified into 0-7, 8-21, and 22-28 days old. Individual hospital utilization ranks for urine, blood, CSF tests, and antibiotic rates are indicated by shades of black, with high utilizers as darker boxes. Diagnostic tests and antibiotic use were summed and ranked into tertiles based on utilization rate.

Abbreviations: CBC, complete blood count; CSF, cerebral spinal fluid.

account for the higher CXR, ICU admission, and mortality rates in older infants.

• 1. Limitations

Our findings are subject to limitations. First, although ICD-9 and 10 codes are accurate in certain conditions [28,42], they have not been val- idated in hypothermic infants. Including infants with any diagnosis code for hypothermia may have subjected the cohort to misclassification bias, with potential over inclusion of patients (e.g., clinicians diagnosing hypothermia based on the complaint of subjective cold temperature at home without documented hypothermia) and under inclusion (e.g., patients with numerous concurrent medical pathologies, so hypo- thermia was not included in diagnosis list). Second, PHIS lacks granular physiologic and clinical data, such as vital signs confirming hypother- mia, to examine thresholds for testing and treatment across hospitals. Further work is required to identify and validate diagnosis codes that best correlate with measured hypothermic temperature, as has been done for fever [28], and to evaluate differences in care based on re- corded temperature. Third, PHIS batches procalcitonin into the same billing code with calcitonin and thyrocalcitonin. However, given the context and timing of the performance of this test, we suspect that this test accurately represents procalcitonin. Fourth, some hospitals bill cultures via a generic unspecified code, which contributes to under inclusion and may account for the discrepancy between lumbar punc- ture performed and CSF culture obtained. Lastly, because our cohort was limited only to patients from pediatric hospitals, our results may not be generalizable to patients presenting in other settings.

1. Conclusion

Our multicenter cross-sectional study identified substantial practice pattern variation in evaluating and managing hypothermic young in- fants presenting to the emergency department. We identified an in- creasing trend in the use of inflammatory markers for evaluating hypothermic infants for serious infections and differences in diagnostic testing among hospitals. Given the current lack of evidence for applying febrile infant guidelines to evaluate for serious infections in hypother- mic infant population, prospective, multicenter studies of hypothermic infants are needed to address this knowledge gap, standardize care, and improve patient-centered outcomes.

Funding source

S.R. is sponsored by PEDSnet Scholars Training Program (Depart- ment of Pediatrics, Ann and Robert H Lurie Children’s Hospital of Chi- cago). This source had no role in the study design, the collection, analysis and interpretation of data, in the writing of the report, and in the decision to submit the article for publication. All authors have no financial relationships relevant to this article to disclose.

Prior presentations

• Pediatric Academic Societies Annual Scientific Meeting (Virtual). Highlighted e-Poster with Live Q&A. May 2021
• Society for Academic Emergency Medicine Annual Scientific Meeting (Virtual). Lightning Oral Abstract. May 2021

Contributor’s statement

Dr. Lo conceptualized and designed the study, refined data collection elements, queried the data, reviewed data analyses, drafted the initial manuscript, and edited the manuscript.

Dr. Cranford conducted the data analyses, critically reviewed and revised the manuscript.

Dr. Ramgopal refined the study design, supervised data collection and analyses, and critically reviewed and revised the manuscript. Dr. Hashikawa refined the study design, supervised data collection and analyses, and critically reviewed and revised the manuscript. Dr. Rogers refined the study design, supervised data collection and analyses, and critically reviewed and revised the manuscript.

All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Credit authorship contribution statement

Yu Hsiang J. Lo: Writing – review & editing, Writing – original draft, Resources, Methodology, Investigation, Data curation, Conceptualiza- tion. Sriram Ramgopal: Writing – review & editing, Supervision, Meth- odology, Investigation, Conceptualization. Andrew N. Hashikawa: Writing – review & editing, Supervision, Methodology, Investigation, Conceptualization. James A. Cranford: Writing – review & editing, Visu- alization, Software, Investigation, Formal analysis. Alexander J. Rogers: Writing – review & editing, Supervision, Methodology, Investigation, Conceptualization.

Declaration of Competing Interest

None.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2022.07.058.

References

1. Kasmire KE, Vega C, Bennett NJ, Laurich VM. Hypothermia: a sign of sepsis in young infants in the emergency department. Pediatr Emerg Care. 2021;37:e124-8. https:// doi.org/10.1097/PEC.0000000000001539.
2. Wood JK, Halvorson EE, Auriemma JR, et al. Clinical characteristics and health out- comes of neonates reporting to the emergency department with hypothermia. Hosp Pediatr. 2018;8(8):458-64. https://doi.org/10.1542/hpeds.2017-0176.
3. Ramgopal S, Walker LW, Vitale MA, Nowalk AJ. Factors associated with serious bac- terial infections in infants <=60 days with hypothermia in the emergency department. Am J Emerg Med. 2019;37(6):1139-43. https://doi.org/10.1016/j.ajem.2019.04.015.
4. Perry MC, Yaeger SK, Noorbakhsh K, Cruz AT, Hickey RW. Hypothermia in young in- fants: frequency and yield of sepsis workup. Pediatr Emerg Care. 2021;37(8): e449-55. https://doi.org/10.1097/PEC.0000000000001674.
5. Ramgopal S, Noorbakhsh KA, Pruitt CM, Aronson PL, Alpern ER, Hickey RW. Out- comes of young infants with hypothermia evaluated in the emergency department. J Pediatr. 2020;221:132-137.e2. https://doi.org/10.1016/j.jpeds.2020.03.002.
6. Cruz AT, Freedman SB, Kulik DM, et al. Herpes simplex virus infection in infants un- dergoing meningitis evaluation. Pediatrics. 2018;141(2):e20171688. https://doi.org/ 10.1542/peds.2017-1688.
7. Caviness AC, Demmler GJ, Selwyn BJ. Clinical and Laboratory features of neonatal herpes simplex virus infection: a case-control study. Pediatr Infect Dis J. 2008;27 (5):425-30. https://doi.org/10.1097/INF.0b013e3181646d95.
8. Watt K, Waddle E, Jhaveri R. Changing epidemiology of serious bacterial infections in febrile infants without localizing signs. PLoS One. 2010;5:e12448. https://doi.org/10. 1371/journal.pone.0012448.
9. Greenhow TL, Hung Y-Y, Herz AM, Losada E, Pantell RH. The changing epidemiology of serious bacterial infections in young infants. Pediatr Infect Dis J. 2014;33:595-9. https://doi.org/10.1097/INF.0000000000000225.
10. Mahajan P, Kuppermann N, Mejias A, et al. Association of RNA biosignatures with bacterial infections in febrile infants aged 60 days or younger. JAMA. 2016;316: 846-57. https://doi.org/10.1001/jama.2016.9207.
11. Aronson PL, Thurm C, Alpern ER, et al. Variation in care of the febrile young infant

<90 days in US pediatric emergency departments. Pediatrics. 2014;134:667-77. https://doi.org/10.1542/peds.2014-1382.

1. Leijh PC, van den Barselaar MT, van Zwet TL, Dubbeldeman-Rempt I, van Furth R. Ki- netics of phagocytosis of Staphylococcus aureus and Escherichia coli by human granulocytes. Immunology. 1979;37:453-65. https://doi.org/10.1542/peds.2014- 1382.
2. van Oss CJ, Absolom DR, Moore LL, Park BH, Humbert JR. Effect of temperature on the chemotaxis, phagocytic engulfment, digestion and O2 consumption of human poly- morphonuclear leukocytes. J Reticuloendothel Soc. 1980;27:561-5.
3. Wenisch C, Narzt E, Sessler DI, et al. Mild intraoperative hypothermia reduces pro- duction of reactive oxygen intermediates by polymorphonuclear leukocytes. Anesth Analg. 1996;82:810-6. https://doi.org/10.1097/00000539-199604000-00023.
4. Le Deist F, Menasche P, Kucharski C, Bel A, Piwnica A, Bloch G. Hypothermia during cardiopulmonary bypass delays but does not prevent neutrophil-endothelial cell ad- hesion. A clinical study. Circulation. 1995;92:II354-8. https://doi.org/10.1161/01.cir. 92.9.354.
5. Akriotis V, Biggar WD. The effects of hypothermia on neutrophil function in vitro. J Leukoc Biol. 1985;37:51-61. https://doi.org/10.1002/jlb.37.1.51.
6. Sung Y, Akriotis V, Barker C, Calderwood S, Biggar WD. Susceptibility of human and porcine neutrophils to hypothermia in vitro. Pediatr Res. 1985;19:1044-7. https:// doi.org/10.1203/00006450-198510000-00020.
7. Biggar WD, Bohn DJ, Kent G, Barker C, Hamilton G. Neutrophil migration in vitro and in vivo during hypothermia. Infect Immun. 1984;46:857-9. https://doi.org/10.1128/ iai.46.3.857-859.1984.
8. Biggar WD, Bohn D, Kent G. Neutrophil circulation and release from bone marrow during hypothermia. Infect Immun. 1983;40:708-12. https://doi.org/10.1128/iai. 40.2.708-712.1983.
9. El-Radhi AS, Jawad MH, Mansor N, Ibrahim M, Jamil II. Infection in neonatal hypo- thermia. Arch Dis Child. 1983;58:143-5. https://doi.org/10.1136/adc.58.2.143.
10. Knobel R, Holditch-Davis D. Thermoregulation and heat loss prevention after birth and during neonatal intensive-care unit stabilization of extremely low-birthweight infants. J Obstet Gynecol Neonatal Nurs. 2007;36:280-7. https://doi.org/10.1111/j. 1552-6909.2007.00149.x.
11. Lunze K, Hamer DH. Thermal protection of the newborn in resource-limited envi- ronments. J Perinatol. 2012;32:317-24. https://doi.org/10.1038/jp.2012.11.
12. Hackman PS. Recognizing and understanding the cold-stressed term infant. Neona- tal Netw. 2001;20:35-41. https://doi.org/10.1891/0730-0832.20.8.35.
13. Blackburn ST. Maternal, fetal, and neonatal physiology: A clinical perspective. 3rd ed.. Saunders; 2007; 716-7.
14. Gray M. Value based healthcare. BMJ. 2017;356:j437. https://doi.org/10.1136/bmj. j437.
15. Mongelluzzo J, Mohamad Z, Ten Have TR, Shah SS. Corticosteroids and mortality in children with bacterial meningitis. JAMA. 2008;299:2048-55. https://doi.org/10. 1001/jama.299.17.2048.
16. Fletcher DM. Achieving data quality. How data from a pediatric health information system earns the trust of its users. J AHIMA. 2004;75:22-6.
17. Aronson PL, Williams DJ, Thurm C, et al. Accuracy of diagnosis codes to identify fe- brile young infants using administrative data. J Hosp Med. 2015;10:787-93. https://doi.org/10.1002/jhm.2441.
18. Macy ML, Hall M, Shah SS, et al. Pediatric observation status: are we overlooking a growing population in children’s hospitals? J Hosp Med. 2012;7:530-6. https://doi. org/10.1002/jhm.1923.
19. Feudtner C, Christakis DA, Connell FA. Pediatric deaths attributable to complex chronic conditions: a population-based study of Washington state, 1980-1997. Pediatrics. 2000;106:205-9.
20. Feudtner C, Feinstein JA, Zhong W, Hall M, Dai D. Pediatric complex chronic condi- tions classification system version 2: updated for ICD-10 and complex medical tech- nology dependence and transplantation. BMC Pediatr. 2014;14:199. https://doi.org/ 10.1186/1471-2431-14-199.
21. Pantell RH, Roberts KB, Adams WG, et al. Evaluation and management of well- appearing febrile infants 8 to 60 days old. Pediatrics. 2021;148:e2021052228. https://doi.org/10.1542/peds.2021-052228.
22. Gomez B, Mintegi S, Bressan S, et al. Validation of the “Step-by-Step” approach in the management of young febrile infants. Pediatrics. 2016;138:e20154381. https://doi. org/10.1542/peds.2015-4381.
23. Kuppermann N, Dayan PS, Levine DA, et al. A clinical prediction rule to identify fe- brile infants 60 days and younger at low risk for serious bacterial infections. JAMA Pediatr. 2019;173:342-51. https://doi.org/10.1001/jamapediatrics.2018.5501.
24. Mintegi S, Bressan S, Gomez B, et al. Accuracy of a sequential approach to identify young febrile infants at low risk for invasive bacterial infection. Emerg Med J. 2014;31:e19-24. https://doi.org/10.1136/emermed-2013-202449.
25. Gomez B, Diaz H, Carro A, Benito J, Mintegi S. Performance of blood biomarkers to rule out invasive bacterial infection in febrile infants under 21 days old. Arch Dis Child. 2019;104:547-51. https://doi.org/10.1136/archdischild-2018-315397.
26. Marik PE, Zaloga GP. Hypothermia and cytokines in septic shock. Norasept II Study Investigators. North American study of the safety and efficacy of murine monoclonal antibody to tumor necrosis factor for the treatment of septic shock. Intensive Care Med. 2000;26:716-21. https://doi.org/10.1007/s001340051237.
27. Arons MM, Wheeler AP, Bernard GR, et al. Effects of ibuprofen on the physiology and survival of hypothermic sepsis. Ibuprofen in Sepsis Study Group. Crit Care Med. 1999;27:699-707. https://doi.org/10.1097/00003246-199904000-00020.
28. Drewry AM, Fuller BM, Skrupky LP, Hotchkiss RS. The presence of hypothermia within 24 hours of sepsis diagnosis predicts persistent lymphopenia. Crit Care Med. 2015;43:1165-9. https://doi.org/10.1097/CCM.0000000000000940.
29. Garami A, Steiner AA, Romanovsky AA. Fever and hypothermia in systemic inflam- mation. Handb Clin Neurol. 2018;157:565-97. https://doi.org/10.1186/s13054- 016-1510-3.
30. Wiewel MA, Harmon MB, van Vught LA, et al. Risk factors, host response and out- come of hypothermic sepsis. Crit Care. 2016;20:328. https://doi.org/10.1542/peds. 2010-2064.
31. Tieder JS, Hall M, Auger KA, et al. Accuracy of administrative billing codes to detect urinary tract infection hospitalizations. Pediatrics. 2011;128:323-30. https://doi.org/ 10.1542/peds.2010-2064.