A study of cerebrospinal fluid C-reactive protein in bacterial meningitis among paediatric age group in a tertiary care hospital

INTRODUCTION

Meningitis is defined as an inflammatory condition involving the membranes (meninges) covering the brain and spinal cord. It can have infectious causes, such as bacteria, mycobacteria, viruses, fungi, or parasites, or be associated with autoimmunity, cancer, or reactions to medication.¹ Meningitis continues to be a formidable illness with high morbidity and mortality in India. Gram positive cocci and gram-negative bacilli have been incriminated as bacterial aetiological agents of pyogenic meningitis in various studies.^2,3 Bacterial meningitis is a life-threatening illness. Early recognition and appropriate antibiotic treatment is crucial to reduce morbidity and mortality.

The diagnosis of bacterial meningitis poses the biggest challenge as the clinical and biochemical picture is often masked because of prior improper antibiotic usage. This becomes even more difficult in a population who are first catered usually by private practitioners and receive antibiotic course before confirming diagnosis. They are then referred to tertiary care hospital in the due course. In such a scenario, the isolation of the organisms from blood or CSF becomes very difficult.⁴ In a developing country like India, facilities to appropriately isolate blood- or CSF-borne organisms are scarce and even if available culture reports are usually time consuming. Moreover, CSF cultures for pyogenic organisms are positive in only 30–60% of cases, according to various researchers.⁵ The lack of a single test to diagnose the aetiology of meningitis promptly and accurately, makes it more important to find a quick and reliable method for early bedside diagnosis.

CSF-CRP is a test which meets all this criterion, unlike CSF cytology and biochemistry does not require a lot of knowledge and skills and equipment’s to interpret the results.⁶ Hence, the present study was undertaken to evaluate the diagnostic significance of CRP in CSF as an early indicator in the differentiation of bacterial from non-bacterial meningitis.

METHODS

After obtaining institutional ethical committee approval, this hospital based observational study was conducted in the paediatric ward of tertiary care hospital in central India over a period of 1 year from 01 of January 2021 to 31 December 2021. A total 123 children aged between 1 month to 12 years, in whom meningitis was suspected during the study period with any of the following clinical criteria suggestive of CNS infection.

Signs of meningeal irritation like neck rigidity, Kernig sign positive, and Brudzinski sign positive.

Signs of raised Intra cranial tension like vomiting, headache, cranial nerve involvement, bulging anterior fontanelle, and Cushing’s triad. Signs of encephalopathy, like convulsions, lethargy, irritability, coma, altered sensorium, behavioural disturbances, and any other neurological deficit.

Diagnosed case of TB meningitis, children with known immune function disorder, known case of intracranial malformation, patients who were discharged against medical advice, who were referred to higher centre and failure to obtain CSF sample even after 3 attempts were excluded from the study.

Assent of the child was also obtained verbally if the child is more than 7 years of age as per ICMR guidelines. Simultaneously socio-demographic data (age and sex), clinical history, physical examination of each patient was done, and management initiated based on provisional diagnosis. Lumbar puncture performed as per protocol after obtaining informed consent for the same. Under strict aseptic precaution, patient in knee chest position/flexion position and care was taken not to obstruct airway. Lumbar puncture performed at L3- L4 after properly painting and draping the prescribed area. Appropriately sized lumbar puncture needle was used to do the procedure (1.5 for <1 year, 2.5 for 1 year to middle childhood, 3.5 for older children). If LP was unsuccessful at the L3- L4 level, one level above was tried, and a maximum of three attempts were done. If CSF is not obtained even after 3 attempts, then the patient was excluded from study. CSF sample not more than 3 ml collected in 4 sterile containers and 1 ml sent for protein and sugar, 0.5 ml each sent for CSF cytology and CSF CRP, 1 ml sent for CSF culture sensitivity and gram staining.

C-reactive protein was measured in CSF by latex agglutination method using RHELEX-CRP® manufactured by Coral clinical systems company based on Uttarakhand. CRP levels more than 6 mg/dl is considered positive in this qualitative study. CSF culture and Gram stain, cytology and biochemical analysis were used as confirmatory markers. CSF- CRP was compared against these established parameters. Final diagnosis of the patients was considered based on the below stated working diagnosis-

Pyogenic meningitis

Among the clinically suspected cases of meningitis, irrespective of other parameters, CSF culture positivity marks the diagnosis of bacterial meningitis. Since cerebrospinal fluid culture for pyogenic organisms is positive, only in 30-60% of cases according to various Indian studies.^10,11

Culture negative cases were considered as pyogenic meningitis if the other CSF parameters cellularity – leukocytes >10 with neutrophilic predominance >20%, protein >40 mg/dl and sugar <40 mg/dl or less than 2/3^rd of parallel blood sugar was found. CSF antigen detection, PCR and MRI was not considered as the lack of infrastructure and also due economical infeasibility.⁷

Viral encephalitis

CSF with features of lymphocytic pleocytosis (5 to 500 cells/ml), mildly elevated protein content (30 to 65 mg/dl) and normal glucose content. Normal CSF examination does not eliminate the possibility of viral encephalitis.⁸

Complex febrile convulsion

A seizure occurring in association with a febrile illness (temp >38.3-degree Celsius), in the absence of central nervous system infection or acute electrolyte imbalance in children older than 1 month of age upto 5 years without prior afebrile seizures. It is considered complex if it is focal, prolonged (lasting for more than 15 minutes or multiple (occurrence of more than one seizure during the febrile illness).⁹

Dengue encephalopathy

Acute encephalopathy is the most commonly reported neurological disorder associated with dengue virus infection. Dengue encephalopathy involves a diminished level of consciousness that can be precipitated or caused by several factors, including prolonged shock, anoxia, cerebral oedema, metabolic disturbances (e.g. hyponatraemia), systemic or cerebral haemorrhages, acute liver failure, or renal failure. CSF analyses, including measurements of protein, glucose, and cell count, are usually normal.¹⁰

Acute encephalopathy

Generalized disorder of cerebral function (cognitive disorders, convulsions, and mood/personality disorders) that is acute onset with normal CSF findings, with radiological evidence of encephalopathy, non-specific. It can be autoimmune, idiopathic, or metabolic.¹¹

Statistical analysis

The collected data were analysed using statistical package for the social sciences (SPSS) software version 20. For Descriptive statistics standard deviation was used whereas inferential statistics chi square test was used. P value less than 0.05 was taken as significant level.

RESULTS

Total admissions in the ward from the period of January 2021 to December 2021, who were belonging to age 1 month to 12 years were 3086. Out of which, a total of 123 children were clinically suspected of bacterial meningitis were included in the study. Among them 104 patients were diagnosed as a case of bacterial meningitis. Rest were diagnosed as viral encephalitis-7, complex febrile convulsions-4, dengue encephalopathy-4, acute encephalopathy-3, and inborn error of metabolism-1. The majority of children (46.15%) were from the age group of 1 month to 1 year with mean age of patients was 28.30±1.23 months, ranging from 1month to 12 years. The male to female ratio was 1.08:1 (Table 1).

Fever (93.26%) and altered sensorium (92.30%), were the commonest presenting features and neck rigidity (58.65%) was the commonest meningeal sign. Rest of the data are enumerated in the Table 2.

Developmental milestones were delayed in 5 (4.80%) cases, out of which 4 had a history of NICU stay in their earlier life. Immunization based on national immunization schedule was partially received by 13 (12.50%) and 10 (9.61%) children did not receive any vaccines at all. The neurologic assessment showed abnormal mental status (34.61%), abnormal cranial nerve examination findings (10.57%) and abnormal motor function (13.46%) (Table 3).

Among the total 123 patients, who were suspected of meningitis, final diagnosis of bacterial meningitis was done based of CSF examination report, among which 58 (55.77%) of the cases had a positive culture report. Among those who had negative reports (46 cases), diagnosis was established based on CSF protein, CSF sugar and CSF microscopy reports. Among the culture negative cases, routine microscopy was found to be normal in 3 cases, all of which had received some antibiotics before CSF examination. CSF Analysis are shown in Table 4. The mean protein levels in children were 74.22 mg% and the mean CSF sugar in the study group was 58.2 mg%.

Among the organisms grown in CSF culture pneumococcus was the most commonly isolated one accounting for 53.48% of the cases as depicted in Figure 2. The growth in the CSF revealed that ceftriaxone was sensitive in maximum i.e., 38 of the cases (65.51%), followed by cefotaxime 30 (51.72%) and amikacin (39.65%).

Among the total 123 patients, 104 (84.55%) diagnosed with bacterial meningitis and remaining 19 (15.44%) cases diagnosed other than meningitis. It was observed that CRP in CSF was positive in maximum number of children with bacterial meningitis i.e., 86 (82.69%) but it was negative in 18 (17.30%) children. In children diagnosed other than meningitis, CRP was positive in 3 (15.78%) cases and negative in 16 (84.21%). The mean CRP level was 15.36±0.67 mg/l in children with meningitis.

Among the culture positive patients, 53 (91.37%) out of 58, showed positive CSF CRP levels. Comparing cytology reports with CSF CRP positivity, 62 (81.57%) out of 76 abnormal cytology reports were found to be positive. In contrasting abnormal CSF biochemical reports with CSF CRP positivity, 66 (80.48%) out of 82 were found to be positive as show in Table 5.

Among 90 patients who were discharged, 84 (80.76%) had no new neurological deficit, but 6 of the discharged patients had neurological deficit. This included 3 patients which had facial nerve palsy, 3 with mild motor deficits. The institutional mortality of the study was 13.46%. (Figure 3).

Among 104 bacterial meningitis cases, 86 children (82.69%) had CSF CRP positivity (sensitivity 83%) and CSF culture was negative for 16 children among 19 children who tested to be CSF CRP negative (specificity 84.21%). Out of the total 89 children who show CSF CRP positivity, 86 meningitis patients had some organism grown in their CSF culture (positive predictive value 96.62%) and CSF culture was negative for 16 non meningitis children among total 34 children who tested to be CSF CRP negative (negative predictive value 47.05%).

DISCUSSION

The institutional prevalence of bacterial meningitis in our tertiary care hospital was 3.37% which was low compared to with study conducted by Banu et al (2021) (7.5%).¹² On contrary, a higher prevalence of bacterial meningitis was reported in Indian studies conducted in Kolkata by Joardar et al (2012) (34.66%) and in Tamil Nadu by Minz et al (2008) (18.56%).^13,14 The lower incidence may be attributed to an improved vaccination coverage, and better social environment achieved over years. Among the 104 cases of bacterial meningitis, majority i.e., 46.15% were belonging to the age group of 1 month to 1 year, making it the most vulnerable population to be affected by bacterial meningitis. The mean age of patients was 28.30±1.23 months, ranging from 1 month to 12 years with near equitable gender division (male: 51.92%). These findings are comparable with the previous studies.^15,16

Fever was a most common presenting complaints for the cases with bacterial meningitis followed by altered sensorium. The most common meningeal sign was neck rigidity. A similar observation was also recorded by other authors.^17,18 Developmental milestones were delayed in 5 (4.80%) cases, out of which 4 had a history of NICU stay in their earlier life. Immunization based on national immunization schedule was partially received by 13 (12.50%) and 10 (9.61%) children did not receive any vaccines at all. The neurologic assessment show that the mental status was abnormal in 34.61% of children. 14 out of the 104 patients required assisted breathing on admission in view of GCS score less than 8/15. Cranial nerve involvement was seen in 11 patients, of which maximum had lateral rectus palsy (VIth nerve). Tone and power abnormality were observed in 14 patients (13.46 %).

Among 104 bacterial meningitis cases, 58 (55.76%) had a positive culture report. Among the culture negative cases, routine microscopy was found to be normal in 3 cases, all of which had received some antibiotics before CSF examination. Although 73.07% of the cases had cells in the CSF, mostly polymorphs (67.30%) and 6 (5.76%) cases had predominantly lymphocytes, the rest of the 28 cases did not show any significant cells. The reason for this observation may be due to inadvertent delay in CSF processing which may cause desiccation of cells, previous antibiotic administration, variable skill factor of the pathologist examining the CSF.¹⁷ In present study, CSF in the bacterial meningitis group was most commonly clear. It was turbid or frankly purulent in only 11.2% of cases. The mean protein levels in children were 138.4 mg% and mean CSF sugar level was 32.2 mg%. These findings are in accordance with the study conducted by Bansal et al and Kalpana et al.^15,17

Among the organisms grown in CSF culture pneumococcus (Streptococcus pneumoniae) was the most commonly isolated one accounting for 46.55% of the 58 CSF culture positive cases. This can be attributed to the lack of immunization coverage of pneumococcal vaccine among socioeconomically weak families that are being catered by the centre. The similar observations were made by other studies done elsewhere, where the most common isolates were streptococcus pneumonia.^19,20

Among the total 123 patients, 104 (84.55%) diagnosed with bacterial meningitis and remaining 19 (15.44%) cases had a diagnosis other than meningitis. It was observed that CRP in CSF was positive in maximum number of children with bacterial meningitis i.e., 86 (82.69%) but it was negative in 18 (17.30%) children. In children diagnosed other than meningitis, CRP was positive in 3 (15.78%) cases and negative in 16 (84.21%). These findings are in hands with previous studies by Bansal et al, and Hansson et al.^15,21

Out of 104 children 58 were CSF culture positive of them 53 (91.37%) were CSF-CRP positive and 5 (8.62%) cases were CSF-CRP negative. Whereas 46 children were CSF culture negative, out of them 33 (71.73%) were CSF-CRP positive and 13 (28.26%) were CSF-CRP negative. CSF cytology was abnormal in 76 children and normal in 28 children. Among 76 children, 62 (81.57%) were CSF-CRP positive and remaining 14 (18.42%) were CSF-CRP negative. While among 28 children, 24 (85.71%) were CSF-CRP positive and 4 (14.28%) were CSF-CRP negative. However, CSF biochemical study was abnormal in 85 cases and normal in 19 cases. Among the abnormal cases, 76 (89.41%) were CSF-CRP positive and 9 (10.58%) were CSF-CRP negative. Whereas out of normal cases, 10 (52.63%) were CSF-CRP positive and 9 (47.36%) were CSF-CRP negative. The current study suggests that a higher number of false negatives value of CSF-CRP were observed in lesser age group that too less than 3 years of age. Our results correlated with other studies.^17,22 The institute mortality of patients with bacterial meningitis was 13.46% which is comparable with the previous studies.^23,24

CRP in CSF was positive in 86 (82.69%) of the CSF samples with culture positivity which was statistically significant (p<0.01) in detecting cases of bacterial meningitis. The sensitivity was 83% and specificity was 84.21%. It had a positive predictive value of 96.62% and a negative predictive value of 47.05%. So, though CSF CRP is reasonably sensitive in picking up cases of pyrogenic meningitis; it can miss up to 17% of the cases. But it is highly specific for bacterial meningitis. So, if CSF CRP is positive, it virtually makes a diagnosis of bacterial meningitis irrespective of other corroborative evidence whereas a negative test need not necessarily rule out pyogenic meningitis. A similar finding was also reported in various other studies.^5,19,22

The level of CRP in CSF is variable and usually less than that of plasma. It depends on the amount of CRP released into CSF from plasma by ultrafiltration and secretion by inflamed meninges. Among all these mechanisms, CSF leakage or passive diffusion from the blood appears more plausible rather than active secretion.²⁵ In current study, the CSF-CRP was higher in the bacterial meningitis group, indicating some amount of active secretion in CSF. CSF-CRP also had a good correlation with other CSF parameters. Currently, the gold standard for the diagnosis of meningitis is CSF culture. But it has its own drawbacks such as the low positivity rate and chances of contamination. The CSF picture can also be altered by administrating antibiotics, delayed analysis, and in partially treated meningitis. CSF-CRP can come as a handy investigation in such settings whose level is not affected when it is analysed early in the course of illness.

Limitations

This study was limited by a small sample size and lack of quantitative estimation of CSF-CRP. A larger sample size and quantitative estimation of CSF-CRP would have given a better insight. No serial CSF-CRP levels were tested to see the response to the treatment, but this would have required repeated LPs, which is not ethically justifiable. We have also not compared our results with other markers such as CSF procalcitonin and IL-6.

CONCLUSION

In the present study, CRP levels in CSF offers a moderate sensitivity and specificity as well as high positive predictive value. This indicates that CSF CRP is a better marker in differentiating bacterial meningitis. Also, it can be concluded that, in addition to current conventional diagnostic methods (biochemistry, cultures and smears), CSF-CRP is a simple, rapid, and accurate approach for the laboratory diagnosis of bacterial meningitis. It can be used as an initial test for the diagnosis of bacterial meningitis till other confirmatory test reports are awaited. Thus, while concluding that CSF CRP levels can serve as a useful screening test for bacterial meningitis, a further study with a larger population is required to validate our results. Measuring CRP levels before and after antibiotics would also give a better interpretation.

Recommendations

CSF CRP can be used as a bedside diagnostic tool for the rapid diagnosis of bacterial meningitis in children. It should be made available in every paediatric ICU. It is economical, quick and have high positive predictive value with good sensitivity. Current guidelines for empirical antibiotics for the treatment of bacterial meningitis can be used with good coverage for the common pathogens found in the community.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. Zunt JR, Kassebaum NJ, Blake N, Glennie L, Wright C, Nichols E, et al. Global, regional, and national burden of meningitis, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2018;17(12):1061-82.
2. 2. Smith ES. Purulent meningitis in infants and children. J Pediatr. 1954;45(4):425-36.
3. 3. Coonrod JD, Rytel Michael W. Determination of aetiology of bacterial meningitis by counterimmunoelectrophoresis. The Lancet. 1972;299(7761):1154-7.
4. 4. Tacon CL, Flower O. Diagnosis and Management of Bacterial Meningitis in the Paediatric Population: A Review. Emerg Med Int. 2012;1-8.
5. 5. Kumar L, Chitlangiya S, Ayyagari A. The current status of pyogenic meningitis in children. Indian Pediatr. 1980;17(5):438-44.
6. 6. Malla KK, Malla T, Rao KS, Basnet S, Shah R. Is Cerebrospinal Fluid C-reactive Protein a Better Tool than Blood C-Reactive Protein in Laboratory Diagnosis of Meningitis in Children? Sultan Qaboos Univ Med J. 2013;13(1):93-9.
7. 7. Weingberg G, Thompson-Stone R. Bacterial Infections of the Nervous System. In: Swaiman K, Ashwal S, Ferriero D, Schor N, Finkel R, Gropman A, et al, editors. Swaiman’s Pediatric Neurology, Principles and Practices. 6th ed. Elsevier. 2017;883-94.
8. 8. Bonthius DJ, Bale JF. Viral infections of the nervous system. In: Swaiman K, Ashwal S, Ferriero D, Schor N, Finkel R, Gropman A, et al., editors. Swaiman’s Pediatric Neurology. 6th edition. Elsevier. 2017;895-906.
9. 9. Seinfeld S, Shinnar S. Febrile Seizures. In: Swaiman K, Ashwal S, Ferriero D, Schor N, Finkel R, Gropman A, et al., editors. Swaiman’s Pediatric Neurology. 6th edition. Elsevier. 2017;519-23.
10. 10. Carod-Artal FJ, Wichmann O, Farrar J, Gascón J. Neurological complications of dengue virus infection. Lancet Neurol. 2013;12(9):906-19.
11. 11. Li GH, Ning ZJ, Liu YM, Li XH. Neurological Manifestations of Dengue Infection. Front Cell Infect Microbiol. 2017;7.
12. 12. Sumayyah Banu S, Sudha Madhuri D, Shashikala Reddy P, Author C. Aetiology of Meningitis in Paediatric Age Group at a Tertiary Care Centre in Hyderabad, India-A Cross Sectional Study. Acta Scientific Microbiol. 2021;4.
13. 13. Joardar S, Joardar GK, Mandal PK, Mani S. Meningitis in Children: A Study in Medical College &amp; Hospital, Kolkata. Bangladesh J Child Health. 2012;36(1):20-5.
14. 14. Minz S, Balraj V, Lalitha MK, Murali N, Cherian T, Manoharan G, et al. Incidence of Haemophilus influenzae type b meningitis in India. Indian J Med Res. 2008;128(1):57-64.
15. 15. Bansal K, Bhatt D, Dadhich SK, Kariya D. Study of Estimation of Cerebrospinal Fluid C-Reactive Protein in Diagnosis of Acute Meningitis. Pediatric Rev Int J Pediatric Res. 2021;8(4):175-81.
16. 16. Goenka S, Singhal A, Matlani M, Kasana D. Cerebrospinal Fluid as an Alternative Tool to Blood C-reactive Protein in Laboratory Diagnosis of Bacterial Meningitis in Children: a Comparative Analysis. Int J Contemp Med Res. 2021;8(3).
17. 17. Kalpana S, Priyadharishini D. Cerebrospinal fluid c-reactive protein - a point of care test in the diagnosis of bacterial meningitis. Indian J Child Health (Bhopal). 2018;05(03):170-3.
18. 18. Subashini B, Adhikari D, Verghese V, Jeyaseelan V, Veeraraghavan B, Prakash J. CNS infections in children: Experience from a tertiary care center. Vol. 9, Journal of Global Infectious Diseases. Medknow Publications. 2017;35-6.
19. 19. Natani BS, Goyal P, Agarwal A, Bhatia S, Kumar M. Study of CSF C-reactive protein in meningitis to differentiate bacterial meningitis from aseptic meningitis in children between 1 month and 12 years of age. Int J Contemp Pediatrics. 2017;4(3):943.
20. 20. Robbins JB, Schneerson R, Gotschlich EC. Surveillance for Bacterial Meningitis by Means of Polymerase Chain Reaction. Clin Infect Dis. 2005;40(1):26-7.
21. 21. Hansson LO, Lindquist L, Linné T, Sego E. Quantitation of C-reactive protein in cerebrospinal fluid and serum by zone immunoelectrophoresis assay (ZIA). J Immunol Methods. 1987;100(1-2):191-5.
22. 22. Singh N, Arora S, Kahlon PS. Cerebrospinal Fluid C-Reactive Protein in Meningitis. Indian Pediatr. 1995;32:687-8.
23. 23. Hussain IHMI, Sofiah A, Ong LC, Choo KE, Musa MN, Teh KH, et al. Haemophilus influenzae meningitis in Malaysia. Pediatr Infect Dis J. 1998;17(Supplement):S189-90.
24. 24. Basri R, Zueter AR, Mohamed Z, Alam MK, Norsa’adah B, Hasan SA, et al. Burden of bacterial meningitis: a retrospective review on laboratory parameters and factors associated with death in meningitis, Kelantan Malaysia. Nagoya J Med Sci. 2015;77(1-2):59-68.
25. 25. Sindic CJM, Collet-Cassart D, Deprl~ A, Laterre EC, Masson PL. C-reactive protein in serum and cerebrospinal fluid in various neurological disorders Apparent Local Consumption during Bacterial Meningitis. J Neurol Sci. 1984;63.