Elevated procalcitonin levels in febrile children with exanthema subitum undergoing clinically indicated PCT testing: a retrospective cohort study

Abstract

Background

Human herpesvirus 6 (HHV-6), the causative agent of exanthema subitum (ES), is a major viral cause of acute encephalopathy in Japan. Serum procalcitonin (PCT) is widely used as a biomarker of severe bacterial infections and has also been proposed as an early predictor of encephalopathy. However, elevated PCT levels are occasionally observed in ES without bacterial co-infection or encephalopathy, potentially complicating interpretation of PCT levels in febrile children. The association between primary HHV-6 infection and PCT elevation remains unclear. We therefore investigated clinical and laboratory factors associated with serum PCT levels in young children with primary HHV-6 infection.

Methods

We conducted a retrospective cohort study of 188 febrile children aged < 60 months who underwent serum PCT measurement between April 2021 and May 2024. Thirty children with clinically diagnosed ES and laboratory-confirmed primary HHV-6 infection were compared with 53 children with other virologically confirmed febrile illnesses. In exploratory analyses, multivariable logistic regression identified factors associated with ES, and multivariable linear regression explored factors associated with serum PCT levels in the ES group.

Results

Median serum PCT levels were significantly higher in the ES group than in controls (0.28 [interquartile range (IQR) 0.10–0.61] vs. 0.10 [IQR 0.10–0.31] ng/mL; P = 0.018). The PCT/C-reactive protein ratio was also higher in ES (P = 0.018). ES patients had lower white blood cell, neutrophil, and platelet counts, and higher aspartate aminotransferase and lactate dehydrogenase levels. In exploratory multivariable analysis, younger age (odds ratio [OR] 0.92; 95% confidence interval [CI] 0.87–0.98; P = 0.006), lower platelet count (OR 0.88; 95% CI 0.82–0.94; P < 0.001), and higher PCT level (OR 1.74; 95% CI 1.02–2.95; P = 0.042) were associated with ES. Among patients with ES, lower lymphocyte count and absence of febrile seizures were associated with higher PCT levels. All ES patients had a self-limited clinical course without encephalopathy.

Conclusions

Among febrile children undergoing clinically indicated PCT testing, primary HHV-6 infection was associated with modest PCT elevation. These elevations may reflect HHV-6-associated hematologic alterations rather than bacterial co-infection or disease severity, highlighting the need for cautious interpretation of PCT in ES.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-026-13825-2.

Introduction

Human herpesvirus 6 (HHV-6) is the causative agent of exanthema subitum (ES), a typical self-limited illness characterized by high fever for several days, followed by the sudden appearance of a transient rash [1]. HHV-6 is also recognized as one of the major viral causes of acute encephalopathy in Japan [2]. Serum procalcitonin (PCT) has been established as a biomarker of severe bacterial infections. However, PCT has recently been investigated as a potential early predictive marker of acute encephalopathy [3, 4]. Notably, HHV-6 accounts for a substantial proportion of cases of acute encephalopathy. Nevertheless, elevated serum PCT levels are sometimes observed in clinical practice, even in children with ES who show no evidence of bacterial co-infection or acute encephalopathy. Such elevations may complicate the interpretation of PCT as a biomarker of bacterial infection and as a potential predictive marker of acute encephalopathy [5]. To our knowledge, few studies have systematically evaluated virus-specific differences in serum PCT responses among febrile children. Therefore, this study aimed to investigate the association between primary HHV-6 infection and PCT levels. We hypothesized that primary HHV-6 infection contributed to the unexpected elevation in serum PCT levels.

Materials and methods

Study population

This retrospective single-center cohort study was conducted at Hagi Civil Hospital (Hagi, Japan), a secondary-care community hospital providing pediatric emergency and inpatient services in western Japan. We included febrile children younger than 60 months with clinically diagnosed ES who underwent clinically indicated serum PCT measurement at Hagi Civil Hospital between April 2021 and May 2024. The clinical information and laboratory data of patients with ES were collected from medical records.

ES was diagnosed clinically based on the following criteria: body temperature of greater than 37.5 °C for at least 24 h followed by a maculopapular rash appearing around the time of defervescence and resolving within 1–4 days [6]. Primary HHV-6 infection was defined by the detection of serum HHV-6 deoxyribonucleic acid (DNA) using real-time polymerase chain reaction (PCR) in patients who were seronegative for anti-HHV-6 immunoglobulin G (IgG) antibodies at the initial hospital visit [7].

Immunocompromised children were excluded from this study. Febrile children aged < 60 months with virologically confirmed viral infections served as controls. All had negative blood cultures. Viral infections in the control group were diagnosed based on a combination of clinical presentation and virological testing, in accordance with routine clinical practice. Virological confirmation was obtained using rapid antigen tests or PCR assays, as clinically indicated. The confirmed viruses were respiratory syncytial virus (n = 22), severe acute respiratory syndrome coronavirus 2 (n = 13), adenovirus (n = 9), human metapneumovirus (n = 5), Epstein-Barr virus (n = 2), and influenza A virus (n = 2). Epstein–Barr virus testing was not performed routinely and was reserved for patients with clinical findings suggestive of infectious mononucleosis, including atypical lymphocytosis and liver function abnormalities. Routine screening for HHV-6 infection was not performed in the control group because none exhibited clinical features suggestive of primary HHV-6 infection, such as sudden defervescence followed by the characteristic rash of ES.

Blood cultures were obtained from all patients in this study and yielded no bacterial growth. Urine cultures, when obtained at the discretion of the attending physician, were also negative for bacterial growth. Imaging studies, including chest radiography and abdominal ultrasonography, revealed no identifiable bacterial focus. Antibacterial agents were either not administered or discontinued early, and all patients showed a favorable clinical course without clinical deterioration suggestive of bacterial infection. In addition, laboratory findings, including C-reactive protein (CRP) levels and white blood cell (WBC) counts, were not consistent with bacterial infection and improved during follow-up. Based on this comprehensive assessment, patients were classified as having viral infection without evidence of concomitant severe bacterial infection. A flowchart of the study is shown in Fig. 1.

Fig. 1.

Flowchart of the study process. Abbreviations: PCT, procalcitonin; ES, exanthema subitum; HHV-6, human herpesvirus 6; PCR, polymerase chain reaction; IgG, immunoglobulin G

This study was approved by the Ethical Review Board of the Hagi Civil Hospital (approval No. H2024001). The requirement for written informed consent was waived given the retrospective design. The study participation was based on an opt-out procedure.

Criteria for procalcitonin measurement

Serum PCT measurements were not performed systematically in all febrile children or all patients with ES. Serum PCT was routinely measured in patients presenting with clinical features suggestive of sepsis in accordance with published guidelines [8]. PCT testing was performed in patients with any of the following: (1) abnormal age-adjusted vital signs meeting the pediatric systemic inflammatory response syndrome criteria [9]; (2) poor general appearance (e.g., lethargy, decreased responsiveness); or (3) clinical evidence of impaired peripheral perfusion (e.g., delayed capillary refill > 2 s, cool or mottled extremities, weak peripheral pulses).

Serum PCT levels were measured using an immunochromatographic assay (RapidChip^® PCT; SEKISUI MEDICAL Co., Tokyo, Japan) according to the manufacturer’s instructions. This assay has a lower limit of quantification of 0.2 ng/mL, with an intra-assay coefficient of variation of 15%. Samples were analyzed immediately after blood collection.

Serological tests and real-time PCR assay for HHV-6

Acute-phase serum samples were obtained from all patients and immediately stored at − 30 °C until serological analyses of anti-HHV-6 IgG antibodies and real-time PCR for HHV-6 DNA. Anti-HHV-6 IgG antibody titers were determined using an indirect immunofluorescence assay. To distinguish between primary HHV-6 infection and viral reactivation, anti-HHV-6 IgG antibody titers were measured in acute-phase serum at the initial visit using an HHV-6 IgG enzyme-linked immunosorbent assay (ELISA) kit (Abnova, Taipei, Taiwan). A sample was considered positive if its index value was greater than 1.10, negative if it was < 0.90, and indeterminate if between 0.90 and 1.10. However, the ELISA kit did not distinguish between HHV-6 A and HHV-6B variants, and cross-reactivity with HHV-7 was observed.

Laboratory confirmation of HHV-6 infection was performed using real-time PCR of preserved serum samples from all patients with clinically diagnosed ES. The lower limit of detection of HHV-6 DNA was approximately 100 copies/mL.

Statistical analysis

When serum PCT concentrations were below the assay’s lower limit of quantification (0.2 ng/mL), values were replaced with 0.1 ng/mL (one-half the lower limit of quantification) for statistical analysis [10]. Based on previous reports [3, 4], we additionally evaluated the PCT/CRP ratio as an exploratory index to characterize the relationship between serum PCT elevation and CRP response. Differences between patients with primary HHV-6 and those with other viral infections were compared using the Mann–Whitney U test. Comparisons between three or more groups were performed using the Kruskal–Wallis test, followed by the Bonferroni-corrected Mann–Whitney U test. Categorical variables were compared using the chi-square test. Correlations among parameters were calculated using Spearman’s rank correlation coefficients. Statistical significance was set at P < 0.05.

The clinical parameters that distinguished the ES group from the control group were explored using multivariable logistic regression analysis with stepwise forward variable selection. All clinical and laboratory parameters listed in Table 1 were considered candidate explanatory variables. Model discrimination was evaluated using the area under the receiver operating characteristic curve (AUC). Multicollinearity among candidate variables was assessed using variance inflation factors (VIFs). To assess model stability, internal validation was performed using five-fold cross-validation. The full dataset was randomly divided into five subsets. In each iteration, four subsets were used to develop the regression model and the remaining subset was used for validation. This process was repeated five times so that each subset served once as the validation dataset. Model performance was evaluated using the AUC, and mean AUC values were calculated for the training and validation datasets. Factors associated with serum PCT level in ES patients were explored using multiple regression analysis with stepwise variable selection. Serum PCT, CRP, PCT/CRP ratio, ALT, lymphocyte count, CK, WBC count, and serum iron levels were logarithmically transformed because of their highly skewed distributions. Statistical analyses were performed using StatFlex version 6.0 (Artech Co., Osaka, Japan) and JMP Pro version 16.1.0 (SAS Institute Inc., Cary, NC, USA).

Table 1.

Comparison of backgrounds, clinical symptoms, and laboratory findings between ES and control groups

Variable	ES group (n = 30)	Controls (n = 53)	P value
Characteristics
Age, months, median (IQR)	14 (12–20)	19 (5–29)	0.294
Male sex, n (%)	17 (57%)	24 (45%)	0.319
Clinical symptoms
Duration of fever at sampling, days, median (IQR)	3 (2–5)	2 (1–5)	0.436
Febrile convulsion, n (%)	7 (23%)	9 (17%)	0.481
Convulsive seizure duration (min)	3 (1–5)	20 (5–35)	0.019
Time from seizure onset to sampling (hours)	1 (1–2)	1 (0.5–1)	0.176
Laboratory findings, median (IQR)
WBC (×103/µL)	6.10 (4.80–9.08)	8.20 (5.85–11.55)	0.022
Neutrophil count (×103/µL)	2.38 (1.42–4.30)	3.58 (2.07–5.61)	0.021
Lymphocyte count (×103/µL)	2.65 (1.78–3.21)	2.80 (1.85–4.37)	0.341
Hemoglobin (g/dL)	11.9 (11.3–12.6)	11.6 (10.8–12.4)	0.068
Platelet count (×104/µL)	22.6 (16.7–30.1)	30.6 (23.5–40.2)	0.001
AST (U/L)	45 (35–61)	36 (32–43)	0.009
ALT (U/L)	17 (12–20)	15 (13–23)	0.779
LDH (U/L)	351 (282–391)	304 (249–338)	0.004
CK (U/L)	92 (74–141) *	89 (63–115)	0.315
Creatinine (mg/dL)	0.27 (0.23–0.30)	0.24 (0.22–0.29)	0.521
CRP (mg/dL)	0.40 (0.30–1.76)	0.86 (0.17–2.36)	0.483
PCT (ng/mL)	0.28 (0.10–0.61)	0.10 (0.10–0.31)	0.018
PCT ≥ 0.2 ng/mL (LLOQ), n (%)	21 (70%)	21 (40%)	0.011
PCT /CRP ratio	0.69 (0.27–1.55)	0.24 (0.08–0.75)	0.018
Serum iron (µg/dL)	15 (13–21) †	17 (12–29) ‡	0.443
Ferritin (ng/mL)	80 (29–105) §	54 (26–111) ||	0.493

^* n = 29, ^† n = 18, ^‡ n = 42, ^§ n = 12, ^|| n = 17

Abbreviations: AST, aspartate aminotransferase; ALT, alanine aminotransferase; CK, creatine kinase; CRP, C-reactive protein;

ES, exanthema subitum; IQR, interquartile range; LDH, lactate dehydrogenase; LLOQ, lower limit of quantification;

PCT, procalcitonin; WBC, white blood cell

Results

A total of 188 febrile children aged < 60 months underwent serum PCT measurement during the study period (Fig. 1). Of these, 42 (22%) were clinically diagnosed with ES. Twelve patients were excluded as their acute-phase serum samples tested negative for HHV-6 DNA and/or anti-HHV-6 IgG antibodies. Finally, 30 patients exhibiting the typical clinical features of ES were confirmed to have primary HHV-6 infection.

The patient characteristics, clinical symptoms, and laboratory findings are presented in Table 1. No significant differences in age or sex distribution were found between the 30 patients with primary HHV-6 infection and 53 control patients. The duration of fever before sampling was comparable between the groups. Although the frequency of febrile seizures was comparable between the groups, the duration of convulsive seizures was significantly shorter in HHV-6–infected patients than in controls (median, 3 vs. 20 min; P = 0.019). No cases of epilepsy or acute encephalopathy were observed in the control group.

The patients with primary HHV-6 infection had significantly lower WBC (median 6.10 vs. 8.20 × 10³/µL; P = 0.022), neutrophil (2.38 vs. 3.58 × 10³/µL; P = 0.021), and platelet counts (22.6 vs. 30.6 × 10⁴/µL; P = 0.001), whereas aspartate aminotransferase (AST) (45 vs. 36 U/L; P = 0.009) and lactate dehydrogenase (LDH) levels (351 vs. 304 U/L; P = 0.004) were significantly higher than those in the control group. Other laboratory parameters, including hemoglobin, alanine aminotransferase, creatine kinase, creatinine, CRP, serum iron, and ferritin levels, showed no significant differences between the groups. However, because ferritin and serum iron measurements were available only in a limited number of patients, these findings should be interpreted cautiously.

Multivariable logistic regression analysis was performed to identify parameters associated with the ES group. ES status was set as the dependent variable. The laboratory parameters that were significant in the univariate analyses and relevant clinical findings were considered candidate explanatory variables in the stepwise regression model building. The serum PCT levels (median [interquartile range (IQR)]: 0.28 [0.1–0.61] vs. 0.1 [0.1–0.31] ng/mL; P = 0.018) and the PCT/CRP ratio (median [IQR]: 0.69 [0.27–1.55] vs. 0.24 [0.08–0.74]; P = 0.018) were significantly higher in the ES group than those in the control group (Fig. 2A and B).

Fig. 2.

Serum PCT levels and PCT/CRP ratio in the ES and control groups. Box-and-whisker plots showing serum PCT levels (A) and serum PCT/CRP ratios (B) in the ES and control groups. Horizontal lines within the boxes indicate median values; the boxes represent the interquartile range; the lower and upper whisker limits indicate the 2.5th and 97.5th percentile, respectively; and dots represent individual observed values. Serum PCT values below the lower limit of quantification (0.2 ng/mL) were substituted with 0.1 ng/mL for statistical analysis and are shown as open circles, whereas directly measured values are shown as filled circles. Serum PCT levels and PCT/CRP ratios were significantly higher in the ES group than in the control group. Abbreviations: PCT, procalcitonin; CRP, C-reactive protein; ES, exanthema subitum

In an exploratory analysis stratified by viral etiology, serum PCT levels were compared across individual viral infections. Viral subgroups with extremely small sample sizes were excluded from this analysis. Although patients with ES tended to show higher PCT levels than those with other viral infections, no statistically significant differences were observed among viral groups (Kruskal–Wallis test, P = 0.071; Supplementary Figure S1). In exploratory multivariable logistic regression analysis, lower platelet counts (odds ratio [OR] 0.88; 95% confidence interval [CI] 0.82–0.94; P < 0.001), younger age (OR 0.92; 95% CI 0.87–0.98; P = 0.006) and higher serum PCT level (OR 1.74; 95% CI 1.02–2.95; P = 0.042) were associated with the ES group (Table 2). The multivariable logistic regression model showed acceptable discriminative ability (AUC = 0.80). Five-fold cross-validation suggested stable model performance, with comparable mean AUC values in the training and validation datasets (0.818 and 0.820, respectively). No substantial multicollinearity was observed (maximum VIF = 1.19).

Table 2.

Multivariable logistic regression analysis of factors distinguishing ES and control groups

Variable	β coefficient	SE (β)	z value	P value	OR (95% CI)
Platelet count [×104/µL]	–0.134	0.036	–3.70	< 0.001	0.88 (0.82–0.94)
Age [months]	–0.085	0.031	–2.78	0.006	0.92 (0.87–0.98)
Serum PCT [ng/mL]	0.552	0.271	2.04	0.042	1.74 (1.02–2.95)

Model fit indices: AIC = 89.11; AUC = 0.802

Abbreviations: CI, confidence interval; ES, exanthema subitum; OR, odds ratio; PCT, procalcitonin; SE, standard error

Multiple regression analysis was performed to identify clinical and laboratory factors associated with serum PCT levels in patients with ES (n = 30). The regression model for PCT level demonstrated a significant overall fit (R² = 0.368, adjusted R² = 0.267; Table 3). In exploratory multivariable regression analysis, lower lymphocyte counts and the absence of febrile seizures were associated with higher serum PCT levels in the ES group (Table 3). The final model demonstrated moderate explanatory performance (R = 0.61) without problematic multicollinearity (maximum VIF = 1.52). Internal validation using five-fold cross-validation was not performed because the sample size was too small to provide stable estimates.

Table 3.

Multivariable regression analysis of factors associated with serum PCT levels in the ES group

Variable	β	SE (β)	Standardized β	t-value	P value
Intercept	2.410	1.447	—	—	—
Age [months]	0.030	0.026	0.051	0.312	0.757
Sex	0.110	0.352	0.190	1.140	0.266
Lymph [/µL]	–0.125	0.040	–0.609	–3.124	0.004
Seizure	–1.271	0.484	–0.501	–2.627	0.015

Overall fit: R = 0.610, R² = 0.368, adjusted R² = 0.267

Abbreviations: ES, exanthema subitum; Lymph, lymphocyte count; PCT, procalcitonin; SE, standard error. Sex was coded as female = 1 and male = 0 (reference category), and febrile seizure was coded as present = 1 and absent = 0 (reference category)

A negative association between PCT and lymphocyte count was observed, with a Spearman rank correlation coefficient of -0.404 (P = 0.027), as shown in Fig. 3.

Fig. 3.

Relationship between serum PCT levels and lymphocyte count in the ES group. The solid line represents the linear regression line. Spearman’s rank correlation coefficient is shown (rS = -0.404, P = 0.027). Abbreviations: PCT, procalcitonin; ES, exanthema subitum

Discussion

A recent systematic review and additional studies have shown that elevated serum PCT levels can distinguish serious bacterial infections from other causes of acute fever, such as viral infection [11–13]. PCT production is inhibited by cytokines such as interferon-γ, which is preferentially released during viral infection [14]. Our data demonstrated that both serum PCT levels and the PCT/CRP ratio were significantly higher in patients with ES than in those with other viral infections. Although we did not measure these cytokines, a recent study showed significantly higher interferon-γ expression during the febrile phase in patients with primary HHV-6 infection than in those without the infection [15]. Collectively, these findings suggest that the PCT elevation observed in patients with ES may be influenced by additional mechanisms beyond cytokine-mediated regulation. The observed elevation of the PCT/CRP ratio may reflect relative dissociation between serum PCT and CRP responses during primary HHV-6 infection. However, the underlying biological mechanisms and potential clinical utility of this parameter remain uncertain, and the PCT/CRP ratio should therefore be interpreted as an exploratory rather than an established diagnostic marker.

In this study, the ES group showed lower WBC, neutrophil, and platelet counts, whereas the AST and LDH levels were significantly higher than those in patients with other viral infections. These findings are consistent with previous reports on the acute phase of HHV-6 infection [16]. Although PCT elevation is not specific to ES, our data indicate that elevated PCT levels occurred significantly more frequently in patients with ES than in those with other viral infections. Furthermore, multivariable analysis demonstrated that PCT remained associated with ES in exploratory multivariable analysis after adjustment for AST, LDH, WBC, and neutrophil counts. In addition, PCT levels were inversely correlated with lymphocyte counts within the ES group. In many infectious diseases, including coronavirus disease 2019, elevated PCT and lymphopenia have been associated with severe systemic inflammation and poor clinical outcomes [17, 18]. However, despite the observed inverse correlation between PCT and lymphocyte counts, all patients with ES in our study experienced a self-limited clinical course without sequelae. These findings suggest that the inverse correlation observed in ES does not necessarily reflect disease severity. These findings should be interpreted cautiously because the present analyses were exploratory and based on a limited sample size. Therefore, the observed associations between serum PCT levels, lymphocyte counts, and seizures should be considered hypothesis-generating rather than confirmatory. Although the pathological significance of the inverse association remains unclear, it may indicate transient lymphocyte alterations due to the lymphotropic nature of HHV-6 rather than disease severity [16, 19].

Previous studies have suggested that convulsive seizures themselves may increase serum PCT levels independently of bacterial infection [20]. Based on this evidence, we hypothesized that ES patients with febrile seizures would exhibit higher PCT levels than those without seizures. However, our findings did not support this assumption. In fact, an inverse correlation was observed between serum PCT levels and the presence of febrile seizures in the ES group. One possible explanation is the timing of blood sampling. In our cohort, blood samples were obtained shortly after seizure recognition—within 2 h in 86% of ES patients with febrile seizures. Given that serum PCT levels generally begin to rise within 3–4 h after inflammatory stimulation and peak at 6–12 h [21], this early sampling may have been insufficient to capture seizure-related PCT elevation in ES patients with febrile seizures. Additionally, the pathophysiological changes associated with primary HHV-6 infection itself may have contributed more substantially to PCT elevation than seizure activity, thereby attenuating any independent effect of convulsive episodes. Importantly, because PCT was measured at a single time point in this study, the independent effect of febrile seizures on PCT levels in ES cannot be accurately determined. Serial PCT measurements beginning immediately after seizure onset are required to clarify the relationship between febrile seizures and serum PCT levels. This approach may improve the reliability of PCT-based assessments in predicting the development of acute encephalopathy [4].

Despite its contributions, this study had several limitations. First, this was a retrospective, single-center analysis with a relatively small sample size. Because the number of patients in each viral subgroup was limited, virus-specific trends in PCT levels could not be adequately evaluated. In addition, analyses stratified by conventional PCT cutoffs could not be reliably performed. Larger studies focusing on individual viral pathogens are warranted to further clarify virus-specific PCT responses.

Second, HHV-6 testing was not routinely performed in the control group. Although these patients lacked clinical features suggestive of primary HHV-6 infection and had alternative viral etiologies that were clinically and virologically evident, the possibility of undetected or asymptomatic HHV-6 infection cannot be completely excluded. In addition, enterovirus and certain other viral infections were not systematically excluded in all patients because virological testing was performed according to routine clinical practice. Furthermore, the ELISA assay used in this study could not distinguish HHV-6 A from HHV-6B, and potential serological cross-reactivity with HHV-7 may have contributed to diagnostic misclassification in some cases. However, because all included patients fulfilled the typical clinical criteria for ES and showed detectable HHV-6 DNA in serum during the acute phase, major diagnostic misclassification was considered less likely, although it cannot be completely excluded.

Third, serum PCT measurements were performed based on clinical indications rather than systematically in all patients with ES. In routine clinical practice, PCT testing was primarily performed in children with suspected systemic illness or sepsis. Therefore, the present cohort may preferentially represent patients with relatively more severe clinical presentations, introducing potential selection bias and limiting the generalizability of our findings.

Fourth, a substantial proportion of PCT measurements (30% in the ES group and 60% in the control group) fell below the assay’s lower limit of quantification. Therefore, these values were substituted with half of the limit of quantification (0.1 ng/mL). This practice may bias the estimates of central tendency and dispersion, particularly when a high proportion of values are censored [22]. In addition, the point-of-care immunochromatographic assay used in this study had relatively limited analytical sensitivity compared with fully automated laboratory-based assays. The relatively high lower limit of quantification and limited analytical precision may have attenuated subtle intergroup differences in serum PCT levels. Consequently, our results regarding absolute PCT levels and the magnitude of between-group differences should be interpreted with caution. In addition, analyses involving ferritin and serum iron levels were limited by the relatively small number of available measurements.

Finally, patients with primary HHV-6 infection who did not develop rash were not systematically evaluated. However, previous epidemiological studies from Japan have shown that characteristic rash occurs in approximately 80% of children with primary HHV-6B infection [7]. Therefore, the present cohort likely reflects a substantial proportion of the typical clinical presentations encountered in Japanese pediatric practice, although the findings may not fully represent the entire clinical spectrum of primary HHV-6 infection.

Conclusions

Among febrile children undergoing clinically indicated PCT testing, PCT elevation occurred more frequently in ES than in the control viral infections included in this study. Serum PCT elevation in ES patients remained associated with ES after adjustment for conventional inflammatory markers associated with bacterial infection, including CRP levels, WBC counts, and neutrophil counts. The inverse association between PCT and lymphocyte counts observed in the ES group may reflect transient lymphocyte injury caused by HHV-6, rather than disease severity. Recognition of this phenomenon may facilitate interpretation of modest PCT elevations in children with suspected ES. However, serum PCT levels should always be interpreted in conjunction with clinical findings, microbiological evaluation, inflammatory markers, imaging studies when appropriate, and the overall clinical course. These results provide additional insights into the pathophysiology of HHV-6 infection and the interpretation of PCT levels in pediatric viral diseases.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ALT

Alanine aminotransferase

AST

Aspartate aminotransferase

CI

Confidence interval

CRP

C-reactive protein

DNA

Deoxyribonucleic acid

ELISA

Enzyme-linked immunosorbent assay

ES

Exanthema subitum

HHV-6

Human herpesvirus 6

IgG

Immunoglobulin G

IQR

Interquartile range

LDH

Lactate dehydrogenase

OR

Odds ratio

PCR

Polymerase chain reaction

PCT

Procalcitonin

RSV

Respiratory syncytial virus

SE

Standard error

WBC

White blood cell

Author contributions

HI conceived the study. HI, YT, and YN collected data. HI and KI performed statistical analyses. HI drafted the manuscript. All authors critically revised and approved the final version.

Funding

This work was supported by a research grant from the Yamaguchi Medical Association. The funder had no role in study design, data collection, analysis, interpretation, manuscript preparation, or decision to submit.

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethical Review Board of Hagi Civil Hospital (approval No. H2024001) and was conducted in accordance with the principles of the 2013 revision of the Declaration of Helsinki. The requirement for written informed consent was waived due to the retrospective design and the use of an opt-out procedure.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.