Diagnostic Utility of Cerebrospinal Fluid Procalcitonin in Neonatal Meningitis

We aimed to study the diagnostic utility of cerebrospinal uid (CSF) procalcitonin in neonates with meningitis. All the neonates with sepsis who qualied for lumbar puncture were prospectively evaluated. The neonates were classied as Meningitis and No meningitis group based on predened criteria. CSF procalcitonin was estimated in these neonates along with cytological and biochemical parameters. A total of 113 neonates were included in the study with 29 in meningitis group and 84 in no meningitis group. The median procalcitonin levels was higher in babies with meningitis as compared to those without meningitis [0.194 (0.034 - 0.534) in meningitis group vs 0.012 (0.012-0.012) ng/ml in no meningitis group, p< 0.001]. The area under curve (AUC) for CSF procalcitonin was 0.867 (0.77 -0.95) and at a cut off level of 0.120 ng/ml CSF procalcitonin had a sensitivity of 83%, specicity of 84% and positive and negative predictive likelihood ratios of 5.35 and 0.20 respectively for the diagnosis of meningitis. Conclusion: CSF procalcitonin has a good diagnostic accuracy similar to other parameters in the diagnosis of neonatal meningitis and can be considered as an additional diagnostic marker particularly, when CSF culture is negative and cytochemical analysis is inconclusive.


Introduction
Neonatal meningitis is a potentially life threatening serious infection in neonates. The reported incidence of this condition is much higher at 0.8-6.1 per 1000 live births, with a mortality of 40-58% in developing countries. 1 The survivors of the disease are at risk of numerous adverse short-and long-term outcome. 2 Hence, early detection and appropriate antimicrobial treatment are cornerstone of management so as to reduce the complications associated with this condition. 2 Although, culture of cerebrospinal uid (CSF) is the gold standard for diagnosis but the yield is often low (15-30%) and in majority of the time the decision to treat meningitis is solely based upon the assessment of cellularity and biochemical parameters. 3 Further, studies have shown signi cant overlap between normal and abnormal CSF values in neonates both in terms of cellularity as well as biochemical parameters with preterm neonates having higher cell, sugar and protein values in CSF as compared to older children and adults. 2 Moreover since the incidence of traumatic taps are higher, it may further affect the interpretation of CSF parameters as well in neonatal age group. 4 Hence, neonatal meningitis is often over diagnosed in clinical setting which usually leads to unnecessary exposure of neonates to prolonged duration of antibiotics which in turn increases the risk of adverse effects due to exposure to drugs as well as it also increases the risk of antimicrobial resistance. 5,6 Hence, there is an urgent need for a new in ammatory marker which can accurately diagnose neonatal meningitis in babies with neonatal sepsis and hence, in turn will reduce the rates of above said complications associated with the disease as well as its treatment. One of such marker is procalcitonin which is a precursor of calcitonin and is a reliable marker of infection and the associated in ammation. 2 Several studies in neonatal sepsis on serum procalcitonin levels have found it be useful both for the diagnosis as well as prognosis. 7 Although, studies in adult population have shown that CSF procalcitonin to be a reliable marker for diagnosis of meningitis. 8 However, there is a dearth of literature regarding its role in the diagnosis of neonatal meningitis. Hence this study was planned to evaluate the role of CSF procalcitonin in diagnosis of neonatal meningitis.

Methods
This was a prospective observational study conducted between Jan 2018 to May 2019 in a level 3 tertiary care neonatal unit in South India. The study was approved by the institutional ethics committee, and informed consents were obtained from the parents before inclusion in the study. The study was registered in Clinical Trial registry of India (CTRI), with trial reg no of CTRI/2018/09/015720

Inclusion criteria
All neonates (£28 days of age) with signs and symptoms suggestive of sepsis and who quali ed for lumbar puncture as per the unit protocol were included in the study. 9 Exclusion criteria Neonates who received antibiotics for more than 3 days before obtaining lumbar puncture, sick neonates who are expected to survive for less than 24 hours, major congenital malformations, presence of deepseated focus of infection other than meningitis e.g. abscess, septic arthritis etc and coagulopathy.
Clinical data Detailed base line information were recorded including gestation age (i.e. preterm or term), age at presentation, type of sepsis (early [<3 days of onset of symptoms] or late [³3 days of onset of symptoms]), clinical symptoms and signs.

Laboratory analysis
All the neonates included in the study underwent lumbar puncture under strict aseptic precautions. CSF was analysed immediately for cell count (total, polymorph and neutrophil count) along with sugar, protein, gram stain, culture and procalcitonin levels. Procalcitonin level estimation was done by ELISA method (Cloud clone corp, USA).
In the present study meningitis was de ned as either positive CSF culture, or abnormal CSF analysis with or without clinical manifestation of central nervous system infection). Abnormal CSF analysis in term neonates was de ned as either CSF WBC count >8 / mm 3 or glucose <20 mg/dl or protein >150 mg/dl and in preterm neonates as CSF WBC count ≥10 mm 3 or glucose <24 mg/dl or protein >170 mg/dl and no meningitis as if CSF WBC count < 25/ mm 3 and glucose ≥ 25 mg/dl and protein <170 mg/dl. The enrolled neonates were followed up till discharge or death. 10

Statistical analysis
Patient information was collected in a predesigned proforma. Statistical analysis was done using the SPSS statistical package (version 18.0). Data were expressed as mean ± standard deviation (SD) or median (IQR) for continuous variables and as percentages for categorical variables.
Student's t test (unpaired) was used for analysis of continuous variables. Categorical variables were compared by chi-square test or Fisher's exact test as applicable. The results were considered signi cant at 5% level of signi cance (P < 0.05).
The receiver operating characteristic (ROC) curves were drawn, and the sensitivity, speci city, and areas under ROC curves (AUCs) were compared to evaluate the diagnostic performance of each indicator of neonatal meningitis,

Sample size estimation
Based on the previous study by Reshi et al 11 the sensitivity of procalcitonin to diagnose neonatal meningitis is 92%. Assuming 5% level of signi cance and 90% power and 10% absolute precision, the required sample size was 113 patients to estimate the diagnostic accuracy in meningitis.

Results
The ow of participants into the study is presented in Fig. 1. A total of 221 neonates were assessed for eligibility out of which 108 neonates were excluded for various reasons and 113 neonates were nally included in the study. On the basis of CSF analysis and culture results the neonates were classi ed into 2 groups: Meningitis group consisting of 29 neonates and No meningitis group consisting of 84 neonates (Fig 1). Table 1 shows the baseline characteristics of meningitis and no meningitis group. The baseline characteristics were comparable across both the groups. However, majority of the enrolled neonates had late onset sepsis [LOS n=105 (93%), EOS n= 8 (7%)]. Blood and CSF culture positivity rates among enrolled neonates were 19.4% and 5% respectively. Microbiological distribution shows infection with gram negative and gram positive pathogens to be 9.7% and 5.3 % of the enrolled cases respectively. Table 2 shows the CSF analysis as well as the concentration of CSF procalcitonin and CRP levels in  At a cut off value of 0.120 ng/ml CSF procalcitonin had a sensitivity of 83%, speci city of 84% and positive and negative predictive likelihood ratios of 5.35 and 0.20 respectively

Discussion
The present study has shown that CSF procalcitonin is a reliable biomarker in the diagnosis of neonatal meningitis with AUC of 0.867 (0.77 -0.95). Further, at cut-off limit of 0.120 ng/mL, it has an acceptable diagnostic accuracy for the diagnosis of neonatal meningitis (sensitivity 83%, speci city 84%, LR+ 5.35).
Serum procalcitonin levels at the cut-off limit of 2.0-2.5 ng/ml serves as a reliable biomarker for diagnosis of bacterial sepsis [sensitivity of 0.85, con dence interval (CI) (0.76; 0.90) and speci city of 0.54, CI (0.38; 0.70)] as well as deciding on the duration of antibiotic use in neonates. 7 However, the role of CSF procalcitonin in the diagnosis of neonatal meningitis has not been evaluated rigorously in neonatal population. Studies in adult and pediatric age group population have shown it to be a reliable marker for the diagnosis of meningitis as well as its ability to differentiate between viral and bacterial etiologies of meningitis. 12,13 In the study by Reshi et al it was observed that median CSF PCT levels were signi cantly higher in neonates with meningitis in comparison to those without meningitis. 11 They hypothesized that PCT in CSF is likely derived from the plasma itself as blood-brain barrier gets disrupted due to infectious process. The median PCT values observed in their study were much higher than our study [median PCT in meningitis and non-meningitis group were 0.47 (0.38-0.88) and 0.26 (0.21-0.28) respectively] this could be explained by a higher blood and/or CSF culture positivity rate (56% vs 24%) as well as more robust de nition of bacterial meningitis used in their study as compared to our study (either positive CSF and/or blood culture or positive gram staining). However, the values are in concordance with the study done by Rajial et al with cytochemistry proven meningitis. 14 Further, we found that at a level of 0.12 ng/ml, CSF procalcitonin had a sensitivity of 83% and speci city of 84% with area under curve of 0.86 indicating good diagnostic value of this marker for diagnosis of meningitis. These cut off values are much lower as compared to previous studies. This can be explained by lower CSF culture positivity rate in our study as compared to previous study. Other in ammatory markers have also been evaluated in neonatal meningitis and one of the important marker being CSF CRP. In our study CSF CRP had a lower diagnostic ability as compared to CSF procalcitonin. At a value of 7 mg/dl it had a sensitivity of 42% and speci city of 82% with AUC of 0.55 indicating lower diagnostic e ciency and similar results have been observed in other studies which have evaluated this marker. 15 We also observed CSF procalcitonin had a similar diagnostic e ciency as compared with CSF protein and neutrophil count and was in conjunction to the ndings observed in other studies. (11) Our study is not without limitations. We measured only a single value of CSF procalcitonin and a repeat value after completion of antibiotic treatment would have improved its diagnostic ability. Additional measurement of serum levels of procalcitonin would have helped to identify the severity of sepsis and its relation with CSF values.
In conclusion, CSF procalcitonin has a good diagnostic accuracy similar to other parameters in the diagnosis of neonatal meningitis and can be considered as an additional diagnostic marker particularly, when CSF culture is negative and cytochemical analysis is inconclusive.