Background & Aims: Fever is a frequent reason for pediatric visits and often accompanies other clinical signs that facilitate diagnosis (1). Fever without localized symptoms in children <36 months remains a diagnostic challenge (1). Etiology and required evaluations differ by age; unimmunized or incompletely immunized young children are at greatest risk for occult bacteremia due to pneumococcus (1). Rapid recognition of serious bacterial infection (SBI) is essential to enable timely treatment and prevent complications, yet empirical parenteral antibiotics expose children to adverse effects, allergy risk, and higher costs (3).
Procalcitonin (PCT), the precursor of calcitonin hormone, is one of the most reliable markers of sepsis (4). Under normal conditions PCT is produced by thyroid C cells and converted to calcitonin before entering circulation (5–7), with levels <0.10 ng/ml in healthy individuals (8). During bacterial infection, non-thyroidal tissues produce PCT throughout the body (9), whereas cytokines released in viral infections suppress its level (10). PCT is increasingly used as a biomarker for bacterial infection, early diagnosis, and monitoring of antimicrobial therapy (2). Its role has been evaluated in respiratory, neonatal, hemato-oncologic, hospital-acquired, and surgical infections (3).
Given the difficulty of diagnosing bacterial causes of fever without focus and the expanding evidence for PCT as a biomarker, this study evaluated the diagnostic value of procalcitonin in young children presenting with fever without localized symptoms.
Methods: Children aged 3–36 months admitted with fever without localizing signs were prospectively enrolled after informed parental consent. Fever was confirmed and a detailed physical examination performed. Children who had received antibiotics in the previous 48 hours were excluded.
If no clinical focus was identified, laboratory tests including CBC with differential, ESR, CRP, urinalysis (U/A), urine culture (U/C), and blood culture (B/C by BACTEC method) were obtained. A blood sample for quantitative PCT measurement was collected; laboratory staff were blinded to clinical diagnosis. A checklist recorded demographic data, age group, gender, COVID-19 contact history, and vaccination status. Patients were followed throughout hospitalization.
Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of PCT were calculated in relation to culture results.
Results: Forty-two patients met inclusion criteria. Mean age was 16.01 ± 9.30 months; 24 (57.14%) were male and 18 (42.86%) female. Seventeen children (40.5%) had a positive PCT test and 25 (59.5%) a negative test. Blood culture was positive in 3 patients (7.14%). COVID-19 PCR was negative in 40 cases (95.24%). The most frequent final diagnosis was pyelonephritis (4 cases, 36.36% of culture-proven infections).
Among the 17 PCT-positive patients, 9 had leukocytosis (>15,000), 10 had elevated ESR (>30), and 8 had both; all but one were CRP positive. Diagnoses included 4 cases of pyelonephritis, 2 sepsis, 1 bacterial meningitis, and 1 shigellosis.
Among the 25 PCT-negative patients, 5 had leukocytosis, 2 had high ESR, and 1 had both; 8 were CRP positive. Diagnoses included 2 aseptic meningitis, 2 COVID-19, and 1 bacteremia. The positive blood culture with Staphylococcus lugdunensis showed PCT of only 0.1 ng/ml; clinical evaluation suggested colonization rather than true infection, and the patient improved without antibiotics.
Overall, the PPV of PCT for diagnosing bacterial infection was 47.05%, whereas the NPV was 96%, demonstrating strong ability to rule out serious bacterial infection.
Conclusion: This study confirms the high NPV of PCT in detecting bacterial infection in children with fever without focus. Similar findings have been reported in meta-analysis by Chia-Hung et al., where PCT outperformed CRP for diagnosing serious bacterial infection (11). Galto et al. evaluated 99 children aged 7 days to 36 months with fever >38 °C and found PCT to have the best sensitivity (93%) and NPV (96%), though a modest PPV (38%) (12). Carnino et al. showed PCT rises significantly in bacterial but not fungal infections (9). Gomes et al. demonstrated superior performance of PCT over CRP for identifying invasive bacterial infection in young infants with fever without source (13).
Our findings of low PPV (47.05%) but high NPV (96%) are consistent with these studies and emphasize PCT’s utility as a rule-out test. The small number of positive blood cultures limited precise estimation of PPV and may explain occasional discordant results, such as the Staphylococcus lugdunensis case with low PCT. Nonetheless, the ability of PCT to reliably exclude serious bacterial infection can reduce unnecessary antibiotic exposure, hospitalization costs, and adverse effects.
Fever without localized sign in young children remains a complex diagnostic problem, particularly in settings such as Iran where pneumococcal vaccination is not routine and clinical signs are unreliable predictors of SBI. While early intravenous antibiotics are often recommended to avoid complications, inappropriate use carries risks.
This study supports incorporation of PCT measurement, alongside clinical and routine laboratory criteria, in the initial evaluation of febrile young children without focus. High negative predictive value makes PCT a valuable adjunct to guide early decision-making and avoid unnecessary antibiotic administration.