INTRODUCTION
Fever is a regulated elevation in body temperature caused by an upward resetting of the hypothalamic thermoregulatory set point[1]. In childhood, it is one of the most common clinical manifestations of illness, a major source of parental concern, and a leading reason for seeking medical attention[2,3]. Although most febrile illnesses in children are self-limiting and do not require specific treatment, they place a substantial burden on healthcare services. In a Dutch cohort study, 31.1% of pediatric out-of-hours primary care contacts were fever-related[4]. Febrile illness was also the second most common presenting problem among medical attendances at a United Kingdom pediatric emergency department[5].
In children, fever is most commonly caused by infection, particularly self-limiting viral illnesses, whereas serious bacterial infections occur less frequently[1-3]. Although viral infections account for the majority of febrile episodes, bacterial infections such as pneumonia, urinary tract infections, acute otitis media, sinusitis, and tonsillitis remain important diagnostic considerations because of their potential severity[1,2,6]. Although infection is by far the predominant cause of fever in childhood, noninfectious conditions-including inflammatory, immunologic, neoplastic, metabolic, hereditary, and toxic disorders-should also be considered in the appropriate clinical context[7].
The febrile response represents a tightly regulated host defense mechanism triggered by infectious or noninfectious inflammatory stimuli. Accumulating evidence suggests that fever enhances several innate and adaptive immune functions and contributes to host defense against invading pathogens[8]. Recognition of these stimuli induces the release of endogenous pyrogenic cytokines, which stimulate prostaglandin E2 (PGE2) synthesis, leading to elevation of the hypothalamic thermoregulatory set point and activation of heat-conserving and heat-producing mechanisms, including peripheral vasoconstriction, shivering, and thermogenesis[9-11]. Antipyretic agents reduce fever primarily by inhibiting PGE2 synthesis, thereby restoring the hypothalamic thermoregulatory set point[11]. Figure 1 illustrates the principal mechanisms involved in fever generation and the primary site of action of antipyretic therapy.
Figure 1
Pathophysiological pathway of fever and the primary target of antipyretic therapy.
Despite its established role as a physiological host defense mechanism, fever remains a major source of parental anxiety. It is frequently associated with misconceptions about febrile seizures, dehydration, brain damage, coma, and death—a phenomenon commonly referred to as fever phobia[12]. These concerns often contribute to unnecessary healthcare utilization and inappropriate fever management. However, current evidence indicates that fever itself is generally well tolerated in otherwise healthy children and rarely causes harm[3]. Unlike fever, hyperthermia is an unregulated increase in core body temperature resulting from failure of thermoregulation rather than an elevation of the hypothalamic thermoregulatory set point. It results from excessive heat production, increased heat absorption, impaired heat dissipation, or a combination of these mechanisms. If not promptly recognized and treated, hyperthermia may rapidly lead to tissue injury and life-threatening complications[13].
Over the past decades, a growing body of evidence has substantially advanced the understanding of fever and its management in children. Nevertheless, translating this evidence into everyday clinical practice remains challenging because of considerable heterogeneity across study populations, age groups, clinical settings, diagnostic criteria, outcome measures, and methodological approaches[14]. These factors complicate the development of evidence-based recommendations and may contribute to variations in clinical practice. To promote consistent, evidence-based care, numerous clinical practice guidelines (CPGs) have been developed. These guidelines aim to standardize the assessment and management of febrile children, reduce unwarranted variations in clinical practice, promote the rational use of antipyretic therapy, and address persistent misconceptions and excessive parental concern regarding fever.
This systematic review aimed to compare practical recommendations from national and international pediatric fever CPGs and identify areas of consensus and divergence in fever assessment and management.
MATERIALS AND METHODS
A systematic review of CPGs addressing the assessment and management of fever in children was conducted. This review was reported in accordance with the PRISMA 2020 statement. The review protocol was not prospectively registered with PROSPERO. Nevertheless, the eligibility criteria, search strategy, data extraction procedures, and comparative analysis methods were predefined before study selection and applied consistently throughout the review. Because the objective of this review was to compare the content of guideline recommendations, no formal appraisal of methodological or reporting quality using instruments such as the Appraisal of Guidelines for Research and Evaluation II (AGREE II) or Reporting Items for Practice Guidelines in Healthcare (RIGHT) was performed.
Eligibility criteria
CPGs addressing the assessment and management of fever in children were eligible for inclusion. Eligible documents had to be developed by national or international professional societies, governmental organizations, or recognized guideline development groups and contain explicit clinical recommendations for fever assessment or management. Only the most recent version of each guideline published or updated between January 1, 2010, and June 12, 2026, was included. No language restrictions were applied. Guidelines published in languages other than English were assessed in the original language or translated into English as needed.
Documents focusing exclusively on specific clinical conditions or populations (e.g., febrile neutropenia, fever without source, young febrile infants, or pediatric oncology), emergency department protocols, institutional pathways, review articles, editorials, commentaries, consensus statements without clinical recommendations, duplicate publications, and superseded guideline versions were excluded.
Search strategy
A systematic literature search was conducted to identify CPGs that met the predefined eligibility criteria. Electronic searches were conducted in MEDLINE (via PubMed), the Cochrane Library, Scopus, and Web of Science. Search strategies combined terms related to fever, children, and CPGs, using both free-text terms and, where applicable, controlled vocabulary (Medical Subject Headings in MEDLINE). In Scopus and Web of Science, the term “recommendation” was additionally incorporated into the search strategy to improve specificity and reduce retrieval of non-guideline publications. The complete search strategies for all databases are provided in Supplementary Table 1.
Additional targeted searches were conducted in guideline repositories and the websites of relevant organizations to identify potentially eligible CPGs not captured through database searches. The reference lists of the included guidelines and relevant review articles were screened for additional eligible documents. Authors, experts, professional societies, and other stakeholders were not contacted to identify additional guidelines or obtain supplementary information. No restrictions were applied regarding country or region of origin during the search process.
All records identified through database and supplementary searches were imported into Rayyan (Rayyan Systems Inc., Cambridge, MA, United States). Duplicate records were identified and removed before title and abstract screening.
Study selection and data extraction
Titles and abstracts identified through the literature search were independently screened for eligibility by two reviewers (Momčilo T Pavlović and Karolina F Berenji). Potentially eligible records underwent full-text review independently by both reviewers according to the predefined eligibility criteria, and any disagreements regarding study eligibility were resolved through discussion and consensus. When multiple versions of the same guideline were available, only the most recent version was included. For each included guideline, one reviewer (Momčilo T Pavlović) extracted data using a standardized data collection form, and a second reviewer (Karolina F Berenji) independently verified the extracted information for accuracy and completeness. Extracted data included publication year, country or region of origin, developing organization, target population, and key methodological characteristics. In addition, recommendations relevant to the predefined domains of fever assessment and management were extracted, including temperature assessment, antipyretic therapy, physical methods, and caregiver education. Any discrepancies were resolved through discussion and consensus.
Comparative analysis
Recommendations extracted from the included CPGs were grouped into four predefined clinical domains: Temperature assessment, antipyretic therapy, physical methods, and caregiver education. Within each domain, comparable recommendation items were identified and harmonized to enable direct comparison across guidelines. Recommendations addressing the same clinical question but expressed using different wording were considered equivalent when their clinical meaning was judged to be consistent. Each recommendation item was assessed independently. Consequently, a single guideline could contribute to more than one recommendation category (e.g., different temperature measurement methods recommended for different age groups or clinical settings). For each recommendation item, the level of agreement among guidelines was determined according to the majority recommendation and calculated as the proportion of included CPGs supporting that recommendation relative to the total number of included CPGs. Conditional recommendations were retained as a separate category in the data extraction but were considered to support the corresponding recommendation category when determining the majority position for each recommendation item. Agreement was categorized as excellent (≥ 80%), partial (60%-79%), or low (< 60%). Only explicit recommendations contained within the included guideline documents were extracted. Recommendations available only through referenced external documents were not extracted and were classified as “not reported”. Guidelines in which a recommendation was classified as “not reported” remained in the denominator but were not counted as supporting the majority recommendation. Comparative findings were synthesized descriptively and presented in tables and figures.
RESULTS
Guideline selection
The literature search identified 762 records through database and supplementary searches. After removal of 190 duplicate records, 572 records remained for title and abstract screening. Following full-text assessment of 41 potentially eligible documents, 11 CPGs met the predefined eligibility criteria and were included in the comparative analysis. The study selection process is presented in Figure 2, and reasons for exclusion after full-text assessment are summarized in Supplementary Table 2.
Figure 2
PRISMA 2020 flow diagram illustrating the study selection process for the systematic review.
Characteristics of included guidelines
The 11 included CPGs represented recommendations developed by organizations from 10 countries or regions and by the World Health Organization. The included guidelines were developed by the American Academy of Pediatrics (AAP)[15], a Brazilian expert panel[16], a Chinese expert guideline group[17], the Dutch College of General Practitioners (NHG)[18], the German Society of Pediatrics and Adolescent Medicine (DGKJ)[19], the Hungarian Professional College of Pediatrics[20], the Italian Pediatric Society (SIP)[21], the National Institute for Health and Care Excellence (NICE)[22], a Polish multidisciplinary expert panel[23], a Sub-Saharan African (SSA) expert group[24], and the World Health Organization (WHO)[25]. Target populations ranged from children younger than 5 years to broader pediatric populations, and guideline development methods varied from expert consensus–based recommendations to formal evidence-based approaches using systematic reviews and structured recommendation frameworks. Detailed characteristics of the included guidelines are presented in Supplementary Table 3.
Summary of recommendations
A detailed comparison of recommendations across all included guidelines is presented in Supplementary Table 4, and the level of agreement across recommendation items is summarized in Figure 3.
Figure 3
Level of agreement across key recommendation items in pediatric fever clinical practice guidelines.
Temperature assessment
Recommendations regarding temperature assessment showed the greatest variability across guidelines. Only six guidelines explicitly defined fever, using different temperature thresholds and reference measurement sites. The Chinese guideline defined fever according to the measurement site (≥ 37.6 °C for axillary or oral measurements and ≥ 38.0 °C for rectal or tympanic measurements). The SSA guideline defined fever as an axillary temperature ≥ 37.5 °C, whereas the NHG and Poland guidelines used a rectal temperature ≥ 38.0 °C. Hungary defined fever as a core temperature > 38.0 °C, while the DGKJ guideline used a core temperature ≥ 38.5 °C, with a lower threshold of ≥ 38.0 °C for infants younger than 3 months. Preferred measurement sites also differed substantially. Rectal measurement was recommended by the DGKJ, Hungarian, NHG, and Polish guidelines and conditionally recommended by the Chinese guideline, whereas the NICE, SIP, and SSA guidelines explicitly discouraged its routine use. Axillary temperature measurement was recommended by the Chinese, Hungarian, NICE, Polish, SIP, and SSA guidelines but was discouraged by the DGKJ and NHG guidelines. Tympanic thermometry was recommended by most guidelines, although several restricted its use to children older than 4 weeks (NICE, SIP, and SSA) or 6 months (Hungary, NHG, and Poland). As shown in Figure 3, agreement was low for rectal (45%) and axillary (55%) measurement and partial for tympanic measurement (73%). Forehead thermometers were generally not recommended, except by the DGKJ guideline, which approved their use. Overall, agreement within this domain ranged from low for preferred measurement sites to excellent (91%) for the recommendation that the height of fever alone should not be used as an indicator of serious bacterial infection; only the WHO guideline did not provide an explicit recommendation.
Antipyretic therapy
Compared with temperature assessment, recommendations regarding antipyretic therapy were considerably more consistent. All guidelines recommended antipyretics primarily to relieve discomfort rather than to normalize body temperature (100% agreement). Only the Chinese and WHO guidelines recommended initiating antipyretic treatment at a specified body temperature threshold ≥ 38.2 °C and ≥ 39.0 °C, respectively. The Hungarian and SSA guidelines permitted use of a temperature threshold under specific clinical circumstances, whereas the remaining seven guidelines advised against using a temperature threshold (64% agreement). Paracetamol and ibuprofen were recommended by all guidelines (100% agreement for each), although recommended doses, minimum ages, and maximum daily doses varied. Routine combined antipyretic therapy was discouraged by seven guidelines (64% agreement), whereas recommendations regarding alternating antipyretic therapy showed low agreement (55%). The AAP, Chinese, DGKJ, Hungarian, NICE, and Polish guidelines conditionally permitted alternating paracetamol and ibuprofen under specific clinical circumstances, whereas four guidelines discouraged this practice and the WHO guideline did not provide an explicit recommendation. High agreement was observed regarding recommendations against the use of antipyretics for the prevention of febrile seizures, with 10 of the 11 included guidelines advising against this practice (91% agreement). The WHO guideline did not provide an explicit recommendation. Routine prophylactic administration of antipyretics around vaccination was discouraged by seven of the 11 guidelines (64% agreement). The AAP guideline provided a conditional recommendation, whereas the NICE and WHO guidelines did not explicitly address this practice. The DGKJ guideline discouraged routine prophylactic use but noted an exception for meningococcal B vaccination, referring clinicians to recommendations issued by the German Standing Committee on Vaccination (STIKO).
Physical methods and caregiver education
Recommendations regarding physical methods and caregiver education showed contrasting patterns. Most guidelines discouraged routine tepid sponging (64% agreement), whereas recommendations regarding underdressing showed low agreement (55%). Several guidelines provided limited or no recommendations on physical methods. In contrast, all included guidelines recommended providing caregivers with advice on home management of febrile children and on when to seek medical attention, resulting in 100% agreement for both recommendations.
DISCUSSION
This systematic review compared recommendations from national and international pediatric fever CPGs and demonstrated that, despite important differences in several areas, substantial international consensus exists regarding the core principles of fever management in children. Some of the observed variation may reflect differences in healthcare systems, resource availability, clinical practice, and regional epidemiology across countries. For example, the SSA guideline includes malaria-specific considerations that do not apply to most other settings, illustrating how local disease patterns may shape guideline recommendations. Consequently, recommendations developed in one healthcare setting may not be directly applicable across all geographical, cultural, or socioeconomic contexts, partly explaining the heterogeneity observed among the included CPGs.
Previous studies have examined pediatric fever guidelines from different perspectives. Green et al[26] systematically compared practical recommendations across international pediatric fever guidelines and evaluated the evidence supporting individual recommendations. In contrast, Chiappini et al[27] focused on the methodological quality of pediatric fever guidelines using the AGREE II instrument, while Zhang et al[28] additionally assessed reporting quality using the RIGHT instrument. The present review complements these studies by providing an updated comparison of practical recommendations across a partly different and more contemporary set of pediatric fever CPGs, including several recently published or substantially revised guidelines that were not evaluated previously. Consequently, it provides a comprehensive overview of current recommendations for pediatric fever management.
The included CPGs demonstrated considerable variation in fever definitions and recommended methods of temperature measurement. For example, the SSA guideline defines fever as an axillary temperature ≥ 37.5 °C, whereas the Chinese guideline uses measurement site–specific thresholds (≥ 37.6 °C for axillary or oral measurements and ≥ 38.0 °C for rectal or tympanic measurements). The NHG and Polish guidelines use the same rectal threshold, whereas the DGKJ guideline defines fever as a core temperature ≥ 38.5 °C (≥ 38.0 °C in infants younger than 3 months), with core temperature typically assessed by rectal or tympanic measurement. Recommendations regarding the preferred measurement site also differed substantially. The DGKJ, Hungarian, NHG, and Polish guidelines recommended rectal thermometry, particularly in younger children, whereas the NICE, SIP, and SSA guidelines discouraged its routine use and preferred axillary measurement. Tympanic thermometry was accepted by most guidelines, although several restricted its use according to age. In addition, some guidelines recommended more than one measurement site without providing corresponding site-specific fever thresholds, potentially complicating interpretation of body temperature.
Some of these differences may simply reflect alternative methodological approaches rather than clinically meaningful disagreement. However, conversion scales (“offsets”) have been proposed to estimate equivalent temperatures across measurement sites but have not been adequately validated in children[29]. Furthermore, because the relationship between temperatures obtained at different anatomical sites is inherently variable, no mathematical formula has been shown to reliably convert measurements from one site to another[30]. Consequently, variation in recommended measurement sites, together with differences in fever thresholds, may contribute to inconsistent interpretation of body temperature. Such variation may have implications for clinical decision-making, particularly in situations where accurate temperature measurement is important[31], and may also contribute to parental confusion regarding the definition and interpretation of childhood fever[32]. The observed inconsistency extends beyond CPGs. Hernandez et al[33] found that most commercially available thermometers in the United Kingdom provided temperature measurement advice that differed from NICE recommendations, and almost none explicitly referred to current evidence-based guidance.
Although the included guidelines differed substantially in fever definitions and recommended methods of temperature assessment, these differences had relatively little influence on recommendations regarding antipyretic therapy. All CPGs recommended administering antipyretics primarily to relieve the child’s distress or discomfort. However, some guidelines also incorporated body temperature thresholds under specific clinical circumstances, although the thresholds and indications differed. The Chinese guideline recommended antipyretic therapy at an axillary temperature ≥ 38.2 °C, the WHO guideline for uncomfortable or distressed children with a rectal temperature ≥ 39.0 °C, whereas the Hungarian guideline recommended antipyretics only when body temperature exceeded 40 °C in children with increased metabolic demand or significant clinical compromise. Although consistently recommended, distress is an inherently subjective clinical outcome. Some children tolerate temperatures approaching 40.0 °C without apparent discomfort, whereas others become irritable at much lower temperatures[34]. Recent evidence has further highlighted the lack of a standardized definition and assessment of fever-related discomfort in children, with most studies relying primarily on caregiver- or patient-reported subjective assessments rather than validated objective measures[35]. Although most included CPGs recommend basing antipyretic therapy primarily on the child’s distress or discomfort, the absence of a standardized definition and validated assessment of these clinical manifestations may contribute to variability in the interpretation and implementation of this recommendation in clinical practice.
Despite this broad agreement regarding the indication for treatment, important differences remained in recommendations regarding alternating antipyretic therapy and paracetamol dosing. Both paracetamol and ibuprofen were considered effective antipyretic agents. Routine combined antipyretic therapy was generally discouraged, whereas several guidelines allowed alternating antipyretic therapy only for children with persistent or recurrent distress or when necessary to avoid exceeding the maximum recommended dose of a single antipyretic. This approach is consistent with previous reports suggesting that routine combined or alternating therapy may increase the risk of dosing errors and adverse events, including acute kidney injury[36]. In contrast, evidence from acute pain management indicates that combined administration of paracetamol and ibuprofen provides superior analgesia and a longer duration of pain relief than either agent alone[37]. This cautious approach is further supported by a Cochrane systematic review, which concluded that although combined or alternating therapy may reduce body temperature more effectively than monotherapy, evidence that it provides greater relief of child discomfort remains inconclusive[38]. In addition, the recommended maximum daily dose of paracetamol ranged from 40 mg/kg/day to 90 mg/kg/day across the included guidelines, illustrating substantial variation despite a common therapeutic goal. This variability is clinically relevant because paracetamol overdose is the leading cause of drug-induced acute liver failure in children[39]. It has been suggested that children receiving more than 75 mg/kg/day of paracetamol for more than two consecutive days may be at increased risk of hepatic toxicity[40]. Selecting an appropriate dose for children who are overweight or obese may be even more challenging because neither age- nor weight-based dosing fully accounts for differences in body composition and their effects on paracetamol pharmacokinetics[41].
High agreement was observed regarding recommendations against the use of antipyretics for the prevention of febrile seizures. This consensus is supported by a recent systematic review, which found very limited evidence that antipyretics may reduce seizure recurrence within the same febrile episode and no evidence that they prevent febrile seizures during subsequent febrile illnesses[42]. Despite these consistent recommendations, misconceptions regarding the preventive role of antipyretics remain common. In a Japanese survey, only 24.7% of experienced pediatricians reported advising against the use of antipyretics for febrile seizure prevention[43], while a study from Turkey found that 84% of parents believed that untreated fever would lead to febrile seizures[44]. Similar evidence–practice gaps have also been reported in Switzerland, where 54% of surveyed pediatricians reported frequently lowering the temperature threshold for initiating antipyretic therapy in children with a history of febrile seizures[45].
A high level of agreement was also observed regarding avoidance of routine prophylactic antipyretic administration around vaccination. This consensus is supported by evidence indicating that, although prophylactic paracetamol reduces the frequency of post-vaccination fever, it is also associated with lower antibody responses to several vaccine antigens[46]. A subsequent systematic review confirmed these findings while emphasizing that antibody concentrations generally remained above accepted seroprotective thresholds and that the clinical significance of the observed reduction remains uncertain[47]. Given this uncertainty, most guidelines reserve antipyretics for the treatment of post-vaccination fever or discomfort rather than for routine prophylaxis. The recommendation in the DGKJ guideline reflects STIKO guidance, which recommends prophylactic paracetamol administered at the time of or shortly after routine meningococcal B vaccination in young children to reduce expected post-vaccination fever and pain associated with the highly reactogenic 4CMenB vaccine[48]. This recommendation represents a vaccine-specific strategy rather than a general endorsement of prophylactic antipyretic use.
Despite the broad consensus among contemporary CPGs, translating these recommendations into everyday clinical practice remains challenging. Surveys conducted in different countries have consistently shown that misconceptions regarding childhood fever, inappropriate antipyretic use, and poor adherence to evidence-based recommendations remain common among both caregivers and healthcare professionals[49-51]. In a recent survey of Australian emergency department clinicians, fewer than 10% reported practices consistent with international guideline recommendations, while almost half still endorsed combined antipyretic therapy and most administered antipyretics primarily to reduce body temperature rather than to relieve child discomfort[52]. Similarly, a recent national survey of Italian pediatricians demonstrated that more than one-third still prescribed antipyretics based on body temperature rather than on the child’s level of discomfort, and nearly one-fifth continued to recommend alternating paracetamol and ibuprofen despite guideline recommendations[53]. These findings suggest that publication of evidence-based guidelines alone is insufficient to change clinical practice.
Recommendations regarding physical methods showed variable agreement across the included guidelines. Most CPGs discouraged the routine use of tepid sponging and advised against underdressing. Unlike antipyretic drugs, which lower the hypothalamic thermoregulatory set point, physical cooling removes heat without affecting the underlying set point. Consequently, external cooling may trigger vasoconstriction and shivering, increasing metabolic rate and child discomfort despite a transient reduction in body temperature[34]. Consistent with these recommendations, a Cochrane systematic review found that although tepid sponging may produce a more rapid reduction in body temperature, it is also associated with increased discomfort, shivering, and goose pimples, without clear evidence of clinically meaningful benefit[54].
Caregiver education was consistently emphasized across all included CPGs as a fundamental component of childhood fever management. Caring for a febrile child is often emotionally challenging for parents, making access to clear, consistent, and evidence-based information particularly important. Earlier studies consistently identified healthcare professionals as parents’ principal source of information regarding childhood fever, although the relative contribution of family members, friends, and printed educational materials varied across countries[55,56]. Recent evidence suggests that parental information-seeking behavior has evolved substantially. While healthcare professionals remain the primary source of advice, parents increasingly seek information through online resources, online social networks, and official health service websites[32,57]. However, access to professional advice alone does not necessarily translate into better parental knowledge or reduced fever phobia. In a large nationwide Australian survey, parents who identified healthcare professionals as their primary source of information did not demonstrate better knowledge or fewer misconceptions about childhood fever[58]. These findings suggest that information provided by healthcare professionals should be aligned with current evidence-based guideline recommendations.
This review has several limitations. Although recommendations were extracted using a predefined data collection form, comparison of practical recommendations inevitably involved a degree of interpretative judgment because some recommendations were presented narratively rather than as explicit statements. To minimize this limitation, classification of recommendations was based exclusively on the original guideline text, without reliance on secondary sources, and any uncertainties were resolved by consensus among the reviewers. In addition, the review protocol was not prospectively registered with PROSPERO, representing a limitation of the review. This review focused on general pediatric fever CPGs and did not include guidelines developed for specific diseases, clinical conditions, or specialized patient populations. Guidelines published before the predefined study period or after the literature search was completed were also excluded. Finally, agreement across guidelines reflects the consistency of recommendations rather than their validity or the quality of the underlying evidence. Consequently, a high level of agreement should not be interpreted as indicating stronger evidence or higher methodological quality.
