Procalcitonin Levels in Febrile Infants After Recent Immunization

METHODS

This study was part of a larger study investigating the performance of procalcitonin as a discriminator of SBI in this age group.⁷ The complete study methods are detailed in the original publication. We conducted a prospective cohort study of infants ≤90 days of age who presented to the emergency department with a documented temperature of ≥38.0°C. Infants were excluded if they had an underlying chronic illness or antibiotic use within the previous 48 hours.

Patients were classified as having an SBI if they had a blood, urine, or cerebrospinal culture that grew a pathogenic organism or had a chest radiograph that was interpreted by an attending radiologist who diagnosed pneumonia. Subjects were divided into 3 groups. The first group included all patients with an SBI regardless of their recent immunization status (SBI group). The second group included those patients without SBI who had received immunizations within the previous 48 hours (RI group). The third group included those patients without SBI who had not received immunizations in the previous 48 hours (NRI group).

At the time of enrollment, the attending physician responsible for the care of the patient completed a questionnaire to assess the appearance of the infant on a 5-point scale. Procalcitonin was measured on stored samples by using an immunometric assay (Brahms [Hennigsdorf, Germany] procalcitonin-sensitive Kryptor kit).

Statistical analysis was performed by using SPSS 14.0 (SPSS Inc, Chicago, IL). For continuous variables, independent-samples t tests and the nonparametric Wilcoxon rank-sum test were used; for categorical data, Fisher's exact test or χ² analysis was used. Analyses were performed between the SBI and RI groups, the SBI and NRI groups, and RI and NRI groups. The 95% confidence intervals (CIs) for proportions were calculated by using Stata 6 (Stata Corp, College Station, TX).

The study was approved by the Children's Hospital Boston institutional review board and was compliant with the Health Insurance Portability and Accountability Act of 1996.

RESULTS

During the study period, 874 infants ≤90 days old with a temperature of ≥38°C were evaluated in the emergency department. Fifty-two patients met the exclusion criteria. Of the remaining 822 patients, consent was obtained for 501 (61%) of them. Of those patients, we were able to measure the procalcitonin level from an available blood specimen in 271 patients (54% of consented eligible patients).

Of the 271 patients comprising the study group, there were 44 (16%) patients in the SBI group, 35 (13%) patients in the RI group, and 192 (71%) patients in the NRI group. Of the 44 SBIs, 33 were urinary tract infections (UTIs), 2 were UTIs with bacteremia, 4 were bacteremia, and 5 were pneumonia. Demographics of the infants in these 3 groups are presented in Table 1. As predicted by the timing of immunizations, the mean age of patients in the RI group was significantly higher than in the SBI and NRI groups. The patients were significantly more well-appearing in the RI group in comparison to those in the SBI group but not in comparison to those in the NRI group.

The white blood cell (WBC) and procalcitonin data for the 3 groups are presented in Table 1. WBC values were significantly higher in the SBI and RI groups when compared with the NRI group. There was no difference in the mean WBC count between the RI and SBI groups. Median procalcitonin levels were significantly higher in the SBI group when compared with both the RI and NRI groups. The procalcitonin levels in the RI group were also higher than in the NRI group. Figure 1 depicts the distribution of procalcitonin values in the SBI, RI and NRI groups.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

Comparison of the study subgroups. The boxes represent the limits of the 25th and 75th percentiles and are divided with a line at the median. The whiskers represent 1.5 times the IQR.

Two of 37 patients with recent immunization had an SBI. The first patient was a 77-day-old girl with a temperature of 40.0°C who had received immunizations 2 days before presentation. Her urine culture grew >100000 colony-forming units/mL Escherichia coli. Her procalcitonin level was 0.36 ng/mL. The second patient was a 71-day-old boy who had received immunizations earlier that day and presented with a temperature of 38.4°C. His urine culture grew 30000 colony-forming units/mL E coli. His procalcitonin level was 0.211 ng/mL.

Using our previously published cut point of 0.12 ng/mL, the sensitivity of procalcitonin for SBI in the entire cohort was 96% (95% CI: 83%–99%), the specificity was 23% (95% CI: 18%–29%), and the negative predictive value (NPV) was 96% (95% CI: 86%–99%). Using the same cut point for the 37 patients who had received immunizations in the past 48 hours (35 in the RI group and 2 with recent immunization in the SBI group), the sensitivity of procalcitonin for SBI was 100% (2 of 2), and the specificity for non-SBI was 8.6% (3 of 35). This calculation was limited by the small sample size of 2 patients with SBI and recent immunizations.

DISCUSSION

Fever is the most frequently reported “serious” and “nonserious” adverse event after immunization according to the Vaccine Adverse Event Reporting System.⁸ This presents a unique problem in those infants <3 months of age who are at higher risk of having an SBI. There is no consensus on whether to treat these patients differently than those with fever and no recent history of immunization.

The effect of vaccination on procalcitonin levels has not been well studied. A Medline database search revealed only 1 article about procalcitonin levels after vaccination: in a letter to the editor of the Pediatric Infectious Disease Journal,⁹ Korczowski reported on a cohort of 17 patients who presented to a pediatric emergency department in Poland with vaccination-associated adverse events. Sixteen of these patients had temperatures of >38.5°C. The mean age of the infants was 20 weeks (range: 6–37 weeks). The mean serum procalcitonin level in these patients was 0.8 ± 0.9 ng/mL (range: 0.1–3.6 ng/mL). There were no SBIs. Korczowski concluded that procalcitonin levels can be modestly elevated after vaccination in the absence of infection.

Ours was a prospective cohort study that included 271 infants ≤90 days old with documented temperatures of ≥38°C in a pediatric emergency department. We previously reported the utility of procalcitonin level for predicting SBI in those patients who had not received immunizations in the previous 48 hours. In that cohort, a cut point of 0.12 ng/mL had a sensitivity of 95% (95% CI: 83%–99%), specificity of 26% (95% CI: 20%–32%), NPV of 96% (95% CI: 85%–99%), and negative likelihood ratio of 0.19 (95% CI: 0.05–0.74); all cases of bacteremia were identified accurately with the cut-point value.

In the current study, we further subcategorized the patients as those with SBI, those without SBI who had recently received immunizations, and those without SBI who had not recently received immunizations. The median procalcitonin level was significantly higher in the group of infants who had received immunizations compared with those who had not, but it was significantly lower than in the SBI group. On the basis of this, it seems that immunization leads to an increase in serum procalcitonin values, but it can still be used to identify a group at low risk for SBI regardless of immunization status. The WBC count was also elevated in the RI group when compared with non-SBI patients without recent immunization but was similar to those with SBI. Therefore, WBC count, a commonly used marker for SBI, is less useful for patients who have recently received vaccinations. When using our previously published cut point of 0.12 ng/mL, the sensitivity of procalcitonin for SBI in the entire cohort was 96% (95% CI: 83%–99%), specificity was 23% (95% CI: 18%–29%), and NPV was 96% (95% CI: 86%–99%). Overall, the addition of vaccinated infants into the analysis of the entire cohort led to a similar sensitivity and NPV but decreased the specificity of procalcitonin level as a marker of SBI. This change is because of the vast majority of immunized patients having a procalcitonin value that was greater than our cut point of 0.12 ng/mL. However, procalcitonin levels can continue to be used to identify patients at low risk for SBI, and by using the same cut point, both patients in our cohort with SBIs who had recently received immunizations were identified.

Our study has several limitations. The study was conducted at an academic pediatric emergency department and, therefore, may not represent the larger population of patients after immunization. In addition, even among those febrile infants who were recently vaccinated and presented with fever, there may be a bias of who had any laboratory evaluation. Furthermore, we did not collect information on the type of vaccinations received and, thus, cannot provide further details on the specific effects of individual vaccines on procalcitonin values. Finally, our definitions of SBI were intended to be conservative to include all possible SBIs, because we were hoping to identify a low-risk group for SBI. We acknowledge that some of the patients with low-colony-count UTIs and radiographic pneumonia may not have represented true bacterial infections. Our initial study provides further analysis on definite and possible SBIs to account for this difficulty in definitions.

CONCLUSIONS

Among young febrile infants with recent immunization, procalcitonin levels are increased compared with patients with fever and no identified bacterial infection. Despite this increase, procalcitonin levels can still reliably discriminate infants with SBI.