Age-specific percentile-based reference curve of serum procalcitonin concentrations in Japanese preterm infants

Procalcitonin (PCT) levels are elevated early after birth in newborn infants; however, the physiological features and reference of serum PCT concentrations have not been fully studied in preterm infants. The aims of the current study were to establish an age-specific percentile-based reference curve of serum PCT concentrations in preterm infants and determine the features. The PCT concentration peaked in infants at 1 day old and decreased thereafter. At 1 day old, serum PCT concentrations in preterm infants <34 weeks’ gestational age were higher than those in late preterm infants between 34 and 36 weeks’ gestational age or term infants ≥37 weeks’ gestational age. Although the 50-percentile value in late preterm and term infants reached the adult normal level (0.1 ng/mL) at 5 days old, it did not in preterm infants. It took 9 weeks for preterm infants to reach it. Serum PCT concentrations at onset in late-onset infected preterm infants were over the 95-percentile value. We showed that the physiological feature in preterm infants was significantly different from that in late preterm infants, even in those <37 weeks’ gestational age. To detect late-onset bacterial infection and sepsis, an age-specific percentile-based reference curve may be useful in preterm infants.


Serum PCT concentrations from birth to 5 days old.
shows the postnatal age-specific percentile-based reference curves of serum PCT concentrations in preterm, late preterm, and term infants from birth to 5 days old. The PCT concentration peaked at 1 day old and decreased thereafter in all infant groups. At 1 day old, the serum PCT concentration in preterm infants was higher compared with that in late preterm or term infants (50-percentile values were 11.1 ng/mL in preterm, 1.2 ng/mL in late preterm, and 2.2 ng/mL in term infants). In contrast, the 50-percentile value of late preterm and term infants reached about 0.1 ng/mL at 5 days old (0.11 ng/mL and 0.12 ng/mL in late preterm and preterm infants, respectively), but not in preterm infants (0.31 ng/mL).
Comparison of serum PCT concentrations among the groups. Serum PCT concentrations were compared among the preterm, late preterm, and term infant groups each day of age (Fig. 2). Serum PCT concentrations in preterm infants were significantly higher than those in late preterm and term infants from birth to 5 days old, except for the values between preterm and term infants at 3 days old. There were no significant differences in serum PCT concentrations between late preterm and term infants from birth to 5 days old.  Late-onset infected cases. Case 1. A Japanese infant weighing 482 g was born vaginally at 22 weeks and 4 days. At 42 days old, dyschromia and hypotension appeared, and the blood test showed that the WBC count was 16,900/μL (stab cells: 28%, segmented cells: 49%), platelet count was 0.3 × 10 4 /μL, and serum CRP value was 13.64 mg/dL. The infant was treated with cefotaxime and vancomycin. Enterobacter cloacae was detected from the blood culture. This case was diagnosed with culture-proven sepsis.

Case 2.
A Japanese infant weighing 1,014 g was born by emergency caesarean section at 27 weeks and 5 days. At 12 days old, the infant was extubated; however, an apnea attack occurred. Chest radiography showed infiltration in the lung field, and the blood test showed that the WBC count was 32,500/μL (neutrophils: 76%) and the serum CRP value was 1.75 mg/dL. The infant was treated with cefmetazole. Pseudomonas aeruginosa was detected from the tracheal tube culture. This patient was diagnosed with ventilator-associated pneumonia.
Case 3. A Japanese infant weighing 1,546 g was born by emergency caesarean section at 30 weeks and 3 days. At 55 days old, an apnea attack occurred. The blood test showed that the WBC count was 9,600/μL (neutrophils: 52%) and serum CRP value was 3.73 mg/dL. The infant was treated with tazobactam/piperacillin and arbekacin. As no bacteria were detected from the blood culture, this patient was diagnosed with clinical sepsis.  Time course of serum PCT concentrations of late-onset infected preterm infants. Figure 4 shows the changes in serum PCT concentrations in late-onset infected preterm infants. All cases clearly showed a higher value at onset than the 95-percentile value, and this returned within the reference range, indicating improvement.

Discussion
In the current report, we established a postnatal age-specific percentile-based reference of serum PCT in newborn infants, and showed that physiological changes in late preterm infants were significantly different from those in preterm infants, but not in term infants. Furthermore, we established a postnatal age-specific percentile-based reference in preterm infants, and determined that the serum PCT concentration reaches adult normal levels (0.1 ng/mL) after 9 weeks old. As the serum PCT concentration in preterm infants is age-dependent, the reference for the serum PCT concentration should be used before 9 weeks of age. Using this reference curve, preterm infants with late-onset sepsis or a bacterial infection were clearly over the 95-percentile value for PCT at the time of onset. Neonatal sepsis or bacterial infection is a major cause of growth outcomes and neurological sequelae, especially in preterm infants 27 . As an early diagnosis and subsequent antibiotic therapy are needed to improve infants' outcomes, neonatologists are eagerly awaiting the clinical application of a new early diagnostic marker because existing markers such as the CRP level have a low sensitivity in preterm infants 28 . Although earlier studies have reported that PCT was a more sensitive marker than existing markers such as the WBC count, CRP level, and interleukin-6 levels in adults and children in the clinical setting 5-10 , the sensitivity and specificity of serum PCT in newborns were low compared to those in adults for diagnosing sepsis (respective sensitivity and specificity: 80% and 94% in adults 29 and 77% and 62% in newborns 30 using a cut-off value of 1 ng/mL). The reason for this low accuracy may be why serum PCT becomes physiologically elevated after birth even in non-infected newborns (e.g., those with respiratory and hemodynamic disorders or asphyxia even with mild symptoms) [31][32][33][34] . Therefore, to use serum PCT in the neonatal clinical setting, a nomogram and reference curve of the serum PCT concentration are needed.
Some studies in western countries have already reported the nomogram and references of the serum PCT concentration within 5 days after birth 15,16,18 . This is the first report of a serum PCT reference curve in a Japanese preterm infant population. Concordant with previous studies 15,16,18 , our results also observed that the serum PCT concentration peaked at 1 day old, decreased thereafter, and returned to adult normal levels until 5 days old in term and late preterm infants, suggesting no ethnic difference. Physiological changes of the serum PCT in late preterm infants should be similar to that in term infants. However, the PCT concentration in preterm infants reached adult or child normal levels after 9 weeks, and the duration was longer than that in late preterm and term infants (5 days old in this study), and in preterm infants studied previously (96-120 hours) 16,18 . Similar to our results, Hahn et al. reported the reference interval of serum PCT from 7-60 days postnatally in preterm infants, and preterm infants ≤ 32 weeks' GA still had age dependency until 60 days old 35 . Preterm infants may take a long time to reach adult normal levels due to 5-to 10-fold higher PCT amounts at peak (1 day old) and more immature endocrine and metabolism systems than term infants (generally a half-life of PCT is around 22-26 hours 1 ). The serum PCT concentration in preterm infants is higher compared with that in term infants, because PCT, a precursor of calcitonin, may be associated with osteogenesis and the metabolism of calcium 2 . Further studies are needed to determine the mechanism of this phenomenon.
Our reference curve only used the upper limit of the 95 percentile. The reason for this was that PCT has a non-detection level or an extremely low level in the blood of healthy conditions except for newborn infants because it is produced from the thyroid and transformed into calcitonin quickly 2 . When adopting this percentile curve in preterm infants to various late-onset infected cases, the serum PCT was over the 95-percentile value at onset in all cases. This postnatal age-specific 95-percentile value or other percentile values such as the 90 percentile or 97.5 percentile may be useful for detecting late-onset sepsis or bacterial infection in preterm infants.
Limitations of this study are as follows: 1) as this was a retrospective study, the sample number was biased at each time point after birth. Ideally, the same infants should be longitudinally sampled at every time point to establish the reference; however, the necessary blood amount for measuring the PCT concentration is relatively too much in newborn infants; therefore, this would be unethical. The usefulness of this 95-percentile postnatal age-specific nomogram in preterm infants was not fully determined because there were only three late-onset infected infants. To draw this conclusion, further studies with more late-onset infected cases should be performed, and the test performance characteristics should be shown using the determined cut-off criteria.
In conclusion, we established a reference for serum PCT concentrations in Japanese preterm, late preterm, and term infants during the early period after birth, and for the first time, we found that GA dependency was different at the boundary of 34 weeks' GA. In addition, the percentile-based curve of the longer period until 12 weeks old was established in preterm infants, and we determined that there is age-dependency until 9 weeks old. An age-specific percentile-based reference curve may be useful for diagnosing late-onset bacterial infection and sepsis in preterm infants.

Methods
Study design and subjects. This retrospective study was conducted in Japanese newborns admitted to our NICU at Kobe University Hospital, Japan between June 2014 and December 2014. To establish an age-specific percentile-based reference curve, newborns were excluded if they met the following conditions: (1) congenital malformations and genetic disorders, including a chromosomal abnormality; (2) clinical signs of infection, including sepsis and treatment with antibiotics, or a positive culture; and (3) a history of surgery because this condition may affect the physiological PCT status. Other infants were enrolled in this study. To analyse changes in serum PCT concentrations in late-onset infection, 3 late-onset infected preterm infants were enrolled. Residual serum samples after completing routine laboratory tests were used for this study. Data collection and the use of human materials for this study were approved by the Ethical Committee of Kobe University Graduate School of Medicine (no. 1688), and written informed consent was obtained from the parents of all enrolled newborn infants. The methods were carried out in accordance with the approved guidelines.
During the study period, 299 newborn patients were admitted to our NICU. Of these, 16 patients were excluded because 4 were non-Japanese, 7 had congenital malformations and genetic disorders, 4 had infections (3 late-onset and 1 early onset), and 1 had undergone surgery. Finally, serum samples (n = 1,267) were obtained from 283 newborn patients at birth to 136 days old at the discretion of the attending neonatologist, and based on their clinical conditions, with multiple serum samples obtained from some patients at different days.
Definition. The onset of infection within 72 hours and after 72 hours of birth was defined as early-and late-onset infection, respectively. Small-, appropriate-, or heavy-for-GA was defined as a birthweight less than 10th percentile, between 10th and 90th percentile, or more than 90th percentile for the same GA in a Japanese population 36 . Serum PCT measurement. Using 30 μL of serum, the serum PCT concentration was measured by electrochemical luminescence immunoassay using the COBAS 8000e analyser (Roche Diagnostics, Basel, Switzerland). Data are shown as ng/mL. Study methods. The enrolled newborns were divided into three infant groups as follows: preterm (< 34 weeks' GA, n = 37), late preterm (34-36 weeks' GA, n = 61), and term infants (≥ 37 weeks' GA, n = 185). First, a postnatal age-specific percentile-based reference curve was established for serum PCT from birth to 5 days old in preterm, late preterm, and term infants. Second, serum PCT concentrations were compared each day among infants within 5 days of age. Third, a postnatal age-specific percentile-based reference curve of serum PCT concentrations was created until the concentrations returned to adult normal levels (0.1 ng/mL) in preterm infants. Fourth, the time course in serum PCT concentrations in late-onset infected infants were analysed using our age-specific percentile-based reference curve for preterm infants.
Statistical analysis. The Kruskal-Wallis test was used to compare multiple independent data sets, and differences between the individual data sets were assessed using the all-pairwise test from the Steel-Dwass method. The percentile was determined by the non-parametric method. Statistically significant differences were considered for p values < 0.05.