Indian J Pediatr DOI 10.1007/s12098-014-1409-4

ORIGINAL ARTICLE

Meningitis Related Ventriculitis - Experience from a Tertiary Care Centre in Northern India Rakesh Kumar & Pratibha Singhi & Parag Dekate & Meenu Singh & Sunit Singhi

Received: 8 July 2013 / Accepted: 6 March 2014 # Dr. K C Chaudhuri Foundation 2014

Abstract Objective To assess clinical profile and short term outcome of patients with meningitis related ventriculitis. Methods Case records of 47 patients admitted with a diagnosis of ventriculitis (complicating meningitis) over 3 y were analysed retrospectively. Results The mean age of patients was 4.96 mo (20 d – 42 mo). The common symptoms were fever (27/47), enlarging head (23/47) and seizures (12/47) with signs of raised intracranial pressure (29/47). Ventricular cerebro-spinal fluid (CSF) showed mean cells - 237 cells/μL, sugar - 36 mg/dL and protein - 568 mg/dL. Gram stain and/or culture were positive in 5 patients. Ultrasonography of the head (done in 32 of 47 patients) showed hydrocephalus in 29 and internal echoes, debris or septations in ventricles in 5 patients. Computed tomography (CT) of brain (done in 42 of 47 patients) showed hydrocephalus in 36, abscesses in 7, ependymal enhancement in 4, infarcts in 4 and subdural empyema in 3 patients. MRI brain (done in 13 patients) showed hydrocephalus in all, 6 had cysts and 3 had ependymal enhancement. Thirty seven patients were started on ceftriaxone combined with vancomycin/ cloxacillin with/without amikacin, and 10 patients received vancomycin and meropenem. Mean duration of antibiotics treatment was 4.17 wk (range; 0.8–12 wk). External ventricular drainage (EVD) was done in 22/47 patients. Cure rate (improvement in clinical and CSF parameters) in patients with external ventricular drainage was 59 % as against 50 % in patients who received antibiotics alone. At least 19 patients required ventriculoperitoneal (VP) shunt for unresolved significant hydrocephalus by discharge. There were 10 deaths, R. Kumar : P. Singhi (*) : P. Dekate : M. Singh : S. Singhi Department of Pediatrics, Advanced Pediatrics Centre, Post Graduate Institute of Medical Education and Research, Chandigarh 160012, India e-mail: [email protected]

four of them occurred within first 2 wk of admission, one in 3rd wk and 5 within few months of discharge from hospital. Conclusions Ventriculitis associated with meningitis is a difficult to treat infection with significant mortality and morbidity. It requires prolonged administration of antibiotics. External ventricular drainage may improve cure rate and hasten clearing of CSF infection. Keywords Meningitis . Ventriculitis . Children . Hydrocephalus

Introduction Ventriculitis is a well known complication of pyomeningitis commonly seen in infants. It leads to hydrocephalus acutely and significant neurological morbidity even after adequate treatment. There are no guidelines or recommendations available for its diagnosis and treatment. Ventriculitis requires a prolonged course of antibiotics with good cerebro-spinal fluid (CSF) penetration. The optimal duration of antibiotics required to treat ventriculitis is not known. This retrospective study was done to look for clinical profile of patients with ventriculitis and their short term outcome.

Material and Methods Retrospective analysis of medical records of all children with the diagnosis of meningitis related ventriculitis was done over 3 y i.e., from January 2008 through December 2010. The information was recorded in a predesigned Performa. Follow up information was collected from the follow up clinic files. For those who did not come for follow up, telephonic follow up was done whereever possible.

Indian J Pediatr

All the patients with ventricular CSF features suggestive of ventriculitis [i.e., CSF sugar less than 25 mg/dL or CSF/blood glucose ratio of 0.45 or less, CSF protein >50 mg/dL, and CSF white blood cell count (WBC’s)>10/mm3 with predominance of polymorphs] and clinical features of ventriculitis/ meningitis with or without positive CSF cultures were included for analysis. Intravenous antibiotics were mainstay of treatment in all patients. Some patients also underwent CSF drainage procedures like external ventricular drainage (EVD), Ommaya chamber placement and CSF shunt. Statistical analysis was performed using the Statistical Package for Social Sciences Version 19.0 for Windows.

Results A total of 47 patients who had pyogenic meningitis with ventriculitis were analysed. Majority of the patients were below 6 mo of age at the time of presentation (36/47). Male to female ratio was 39:8. General characteristics and clinical features of the patients are given in Tables 1 and 2, respectively. Most common symptoms at presentation were rapidly enlarging head (23/47), fever (27/47) and seizures (12/47). Around one-third of patients (17/47) had recognized neurodevelopmental delay at the time of presentation. Most common clinical signs were features of raised intracranial pressure (29/47) i.e., decreased level of consciousness, vomiting or bulging anterior fontanel. More than half (28/47) of the patients had proven pyomeningitis prior to referral to the authors’ centre. Many (18/47) of patients however also had history suggestive of neonatal sepsis and prolonged hospitalization in the neonatal period and after. Average time interval from the diagnosis of meningitis to ventriculitis (as calculated from first Lumbar puncture to first ventricular tap suggesting ventriculitis) was 17.5 d (range of 1–46 d). Almost two-third (32/47) of the patients were diagnosed to have ventriculitis within 3 wk of diagnosis of meningitis. Initial ventricular CSF showed a mean of 237 cells/μL, 36 mg/dL of sugar and 568 mg/dL of protein. Comparison of CSF characteristics in lumbar and Table 1 General characteristics of patients (n=47) Characteristics

Value (range)

Mean age of presentation (months) Male: Female ratio Mean duration of symptoms (days) Total hospital stay (days) Total duration of antibiotics (weeks) Time to be afebrile (days)

4.96 (20 d – 42 mo) 39:8 14.29(1 – 90 d) 31.72 (5 – 90 d) 4.17 (0.8 – 12 wk) 6.19 (1 – 56 d)

ventricular CSF in 22 patients (where lumbar puncture was done within 24 h of ventricular tap) is shown in Fig. 1. Five patients had positive Gram stain or cultures of CSF sometime during their treatment. Two patients had Gram negative bacilli on stain and Enterobacter and Pseudomonas on culture. One patient had Gram positive cocci on smear and culture showed it to be methicillin resistant Staphylococcus aureus (MRSA). Two other organisms grown in CSF culture in two different patients were coagulase negative staphylococcus (CONS) and yeast. All these patients had EVD in situ when CSF sample was taken for microbiology. Ultrasonography (USG) of head was done in 32 of 47 patients. Hydrocephalus was seen in 29 cases. Changes specific to ventriculitis, such as internal echoes/debris/septations in ventricles were seen in 5, cyst formation and encephalomalacia in 4, and abscesses in 2 cases. Computed tomography (CT) of brain was done in 42 of 47 patients. Communicating hydrocephalus was seen in 24 patients. Noncommunicating hydrocephalus was seen in 12 patients. Other findings seen were abscesses (7), ependymal enhancement (4), infarcts (4), intraventricular hemorrhage (1), basal exudates (3) and subdural empyema (3). MRI brain was performed in 13patients. Communicating hydrocephalus was seen in 7 and non-communicating hydrocephalus in 6 patients. Other findings were presence of cysts (5), ependymal enhancement (3), infarcts (1), encephalomalacia (1) and abscesses (1). Ten patients had already received antibiotics (for > 1wk) before referral to the authors’ centre and were therefore directly started on vancomycin and meropenem. Remaining 37 patients were started on any of the ceftriaxone based combinations i.e., ceftriaxone plus cloxacillin plus amikacin or ceftriaxone plus vancomycin. Ten patients received only antibiotics (no CSF drainage done). Mean duration of antibiotic treatment given was 4.17 wk (range; 0.8 –12 wk). Five patients, who died, had received minimum duration of antibiotics (0.8 to 2.7 wk). The comparison of CSF parameters among patients who received treatment for 4 wk and those who received treatment for longer duration is shown in Fig. 2. EVD was done in 22 out of 47 patients. Primary CSF shunt (without EVD/Ommaya insertion before VP shunt) was done in 15 patients. Ommaya chamber was also placed in 4 of these 37 patients (who had undergone EVD or shunt surgery) for a brief period. Four more patients required VP shunt after EVD/Ommaya was removed and hydrocephalus persisted. In follow up 2 more patients underwent VP shunt at 2 wk and 4 wk after discharge, respectively. So, out of 42 patients who survived at discharge, 21 patients (50 %) finally required VP shunt because of persisting significant hydrocephalus even after treating ventriculitis adequately. Documented improvement (in clinical/ventricular CSF) in patients who underwent EVD was 59 % (13/22) as against 50 % (5/10) in patients who received antibiotics alone. In primary VP shunt group improvement was seen in 66 % (10/15) (Table 3).

Indian J Pediatr Table 2 Clinical presentation of patients (n=47)

HC Head circumference; GCS Glasgow coma score

Symptoms

Signs

Increasing head size Fever Vomiting Seizures Altered consciousness

23 27 7 12 8

Excessive cry Poor feeding Irritability Rapid breathing Cough

5 3 6 5 2

Lethargy

5

Poor vision

1

Fourteen patients (parents/caregivers) left against medical advice (LAMA) or were discharged on caregiver’s request (DOR). Five (5/47) patients died in the hospital. Remaining 28 patients were discharged as improved. Of the 5 deaths in hospital, four occurred within first 2 wk. They were not diagnosed to have meningitis before admission. All were below 6 mo of age and had features of severe sepsis inform of shock/pneumonia/respiratory failure requiring respiratory support. Two of them had associated major congenital malformations. Fourteen patients from DOR/LAMA group never came for follow up. Of these (6 could be contacted on phone) 3 had died and 3 survived (status as of 3 y after discharge) with some degree of developmental delay along with seizures. Of the 28 patients who were discharged as improved, follow up of variable duration (0.5 to 30 mo, average 10.7 mo) was available in 14 patients, out of which 2 died. Of the remaining 12, 7 improved without any sequelae and 5 had neurological deficits like developmental delay (5/5), defective hearing (1/5), defective vision (1/5), right hemiparesis with focal seizures (1/5) and refractory seizures in one. So, out of 47 patients outcome/follow up was available for 25 patients (53 %). Of these 25, 7 patients (28 %) improved without any neurological disability, 8(32 %) improved but had significant neurodisability and 10 (40 %) died.

Fig. 1 Comparison of ventricular and lumbar CSF (n=22)

Macrocephaly (HC>2 SD) Bulging Ant. Fontanel Setting sun sign Focal neurological deficits Low GCS at admission (≤13)

16 36 9 11 26

Discussion Ventriculitis following meningitis is a serious complication more commonly seen in infants. The true incidence of ventriculitis is not known as there are no prospective studies that document the frequency of ventriculitis in children with meningitis. Following gram negative neonatal meningitis, the reported incidence of ventriculitis ranges from 52 to 94 % [1, 2]. In an autopsy series of neonatal meningitis, ventriculitis was seen on histology in all the cases [3]. Mir reported 17 neonates having ventriculitis following meningitis over a 4 year period from a neonatal intensive care unit in a tertiary care hospital at Toronto, Canada, but total cases of pyogenic meningitis in the same period was not reported [4]. In a study from India ventriculitis was seen in 14 % cases on ultrasonography of head in 50 infants with pyogenic meningitis [5]. Many cases of neonatal meningitis leading to ventriculitis are misdiagnosed as congenital hydrocephalus when they present later in infancy with hydrocephalus. In a study by Udani et al. 13 of 72 infants with “congenital hydrocephalus” admitted for CSF shunt surgery had associated active meningitis/ventriculitis which was unrecognized [6]. They concluded that unrecognized active meningitis/ventriculitis is an important cause of infantile hydrocephalus in India. There is neither any consensus definition nor any diagnostic and treatment guideline for ventriculitis. Most of the

700 600 500 400 Ventricular CSF 300

Lumbar CSF

200 100 0

Cells (per µL)

Sugar (mg/dL) Proteins (mg/dL)

Indian J Pediatr

500

Fig. 2 CSF parameters as per the duration of treatment in 31 patients

450 400 350 300 250

< 4 wk (n=14)

200

≥ 4 wk (n=17)

150 100 50 0 Cells (per µL)

literature available is on ventriculitis following CSF shunt infection. Different researchers have used different definitions for ventriculitis [7, 8]. Most have taken culture positivity of the ventricular CSF as the essential diagnostic criteria along with cytological and biochemical parameters that suggest infection and a clinical profile of meningitis/ventriculitis. However, culture positivity is only occasionally seen (as in meningitis) in many developing countries because of prior antibiotic administration. Most of the index patients (37/47) had received antibiotics before coming to the hospital. So, the authors took positive cytological and biochemical criteria as diagnostic for ventriculitis (detailed in methodology). Time taken for ventriculitis to develop (if at all) in an infant with meningitis (from first LP suggesting meningitis to first positive ventricular CSF) could be an important consideration as to when should one start looking for features of ventriculitis clinically. The authors tried to analyse the present data for this information and estimated it to be 17.5 d on average. It is worth mentioning here that ventricular tap was not done in any patient before referral to the centre. However, there were 3 patients referred with suspected diagnosis of ventriculitis. This suggests that meningitis related ventriculitis is not commonly suspected and hence not diagnosed/treated at primary or secondary health care level. Table 3 Discharge status of patients with respect to intervention received (n=47) Status

EVD group

Only antibiotics group

Primary VP shunt group

Total

Improved Death Unknown (DOR/LAMA) Total

13 3 6 22

5 1 4 10

10 1 4 15

28 5 14 47

DOR Discharged on request; LAMA Left against medical advice

Sugar (mg/dL)

Proteins (mg/dL)

Ultrasonographic evalution for complications of meninigitis in infants has shown ventriculitis in around 10 to 20 % of meningitis cases in various studies [8–12]. In the index patients, hydrocephalus was seen in more than 90 % cases. However, specific features of ventriculitis in ultrasonography [13] like internal echoes, debris or septations in the ventricles were seen in only 15 % cases. Characteristic CT findings are: ventricular debris, hydrocephalus, ependymal enhancement and meningeal enhancement [12, 14]. Characteristic MRI findings of ventriculitis are intraventricular debris and pus, abnormal periventricular and subependymal signal intensity and enhancement of the ventricular lining [15, 16]. In the index patients, specific findings of ependymal enhancement in CT brain were seen in 10 % and ependymal enhancement on MRI brain in 23 %. Although, neuroimaging is important for evaluation of suspected ventriculitis, the gold standard for diagnosis is ventricular CSF analysis. In the present study, it was observed that there was no significant difference in any of the cytological or biochemical parameters in lumbar and ventricular CSF as also reported by Sommer et al. [17]. It may suggest that in the given scenario of post meningitis hydrocephalus and lumbar CSF showing meningitis, one need not do ventricular tap to prove ventriculitis. However, in infants with depressed sensorium and features of raised ICP, ventricular tap can be therapeutic as well as diagnostic. Ventriculitis requires prolonged antibiotic therapy that covers gram negative rods and staphylococcus species [18]. Treatment guidelines are available only for shunt related ventriculitis/meningitis [19]. Whether these can be extrapolated for meningitis related ventriculitis is debatable as both entities are different with respect to patient age and causative organisms. In the authors’ centre, 3rd generation cephalosporins with aminoglycoside and cloxacillin are empirically started. First line antibiotics are continued for atleast 2–3 wk and changed to vancomycin and meropenem if there is no

Indian J Pediatr

CSF clearance after this duration. However, if there is any other deterioration indicating uncontrolled sepsis, antibiotics are to be changed early. Whereever cultures are positive, antibiotics are changed according to sensitivity reports. If at any stage patient develops features of acute raised ICP, a therapeutic ventricular tap or external ventricular drainage is done. Total duration of treatment is usually 4–6 wk. In the present study it was seen that there was no significant difference in CSF parameters in patients who received antibiotics for 4 wk and those who received for more than 4 wk. It indicates that prolonging antibiotic duration after 4 wk does not make much difference in improving ventricular CSF parameters. Intraventricular antibiotics are recommended for shunt related ventriculitis in patients who do not respond to systemic antibiotics or in whom shunt cannot be removed [19]. There are some old studies in small number of patients that report benefit of using intraventricular aminoglycosides and penicillins [20–22]. The authors did not use intra-ventricular antibiotics in any of the index patients as one need to have CSF drug concentrations to titrate dose of intraventricular antibiotics. Drainage of infected CSF from the ventricles is again recommended for shunt related ventricultis [19]. This can be done either by exteriorizing the infected shunt or putting an EVD. Again this strategy has not been well studied for meningitis related ventriculitis. Most of the patients with meningitis related ventriculitis have large hydrocephalus that warrants CSF drainage particularly when there are features of raised ICP. Different options available for CSF drainage are repeated ventricular taps, Ommaya chamber placement with repeated CSF tap and EVD. There are no studies comparing the relative efficacy of these three strategies in patients with meningitis related ventriculitis. In the present cohort, EVD was done when there were signs of acute raised ICP with hydrocephalus and VP shunt was not possible due to persisting ventriculitis. EVD once done was continued till ventricular CSF showed ventriculitis and CSF from external drain was examined daily for resolution of infection. Once ventriculitis improved, EVD was clamped for 24 – 48 h with close monitoring for acute raised ICP. If features of raised ICP recurred, VP shunt was performed and if there were no features of raised ICP, EVD was removed. Average duration of EVD was 9.1 d (range 3–19 d). Most common complication was infection (5 patients). Only other complication noted was proximal drain block (2 patients) requiring repeat EVD procedure. Being a retrospective study, there are limitations of incomplete follow up of patients and no pre-defined criteria or protocol for various interventions. So, it was difficult to correctly analyze the effect of various interventions on the outcome. However, this is first study looking into management issues of meningitis related ventriculitis.

Conclusions Despite limitations of the present study following suggestions may still be drawn: &

& & & & &

Diagnosis of meningitis related ventriculitis is frequently missed or delayed. One needs to have high index of suspicion (especially below 6 mo of age) to diagnose it early. USG head and a ventricular tap can confirm diagnosis and help monitoring response to treatment. CSF drainage may be helpful in decreasing acute raised ICP and clearing inflammatory material thereby improving the outcome. At least 4 wk of appropriate antibiotics should be given in these patients. After resolution of ventriculitis, majority of the patients continue to have hydrocephalus requiring VP shunt. Prolonged follow up is important especially for neurological assessment, as significant numbers of patients develop various neuro-developmental morbidities.

Contributions RK: Idea and plan of study, data collection, data analysis, writing of the manuscript; PS: Plan of study, manuscript writing; PG: Data collection, data analysis, literature review; MS: Writing of manuscript, analysis of data; SS: Critical review and editing of the manuscript. PS will act as guarantor for this paper. Conflict of Interest None. Role of Funding Source None.

References 1. Lee EL, Robinson MJ, Thong ML, Puthucheary SD, Ong TH, Ng KK. Intraventricular chemotherapy in neonatal meningitis. J Pediatr. 1977;91:991–5. 2. McCracken GH Jr, Mize SG, Threlkeld RN. Gentamycin therapy in gram negative bacillary meningitis of infancy: Report of second neonatal meningitis cooperative study group. Lancet. 1980;1:787– 91. 3. Gilles FH, James JL, Berenberg W. Neonatal meningitis: the ventricle as bacterial reservoir. Arch Neurol. 1977;34:560–2. 4. Mir NA. Ventriculitis complicating pyogenic meningitis in neonates. Indian J Pediatr. 1985;52:395–400. 5. Arrumugham R, Katariya S, Singhi P, Singhi S, Suri S, Walia BN. Sonography in pyogenic meningitis. Indian Pediatr. 1994;31:1329– 36. 6. Udani V, Udani S, Merani R, Bavdekar M. Unrecognised ventriculitis/meningitis presenting as hydrocephalus in infancy. Indian Pediatr. 2003;40:870–3. 7. Lyke KE, Obasanjo OO, Williams MA, O’ Brein M, Chotani R, Perl TM. Ventriculitis complicating use of intraventricular catheters in adult neurosurgical patients. Clin Infect Dis. 2001;33:2028–33.

Indian J Pediatr 8. Yikilmaz A, Taylor GA. Sonographic findings in bacterial meningitis in neonates and young infants. Pediatr Radiol. 2008;38:129–37. 9. Edwards MK, Brown DL, Chua GT. Complicated infantile meningitis: Evaluation by real-time sonography. AJNR Am J Neuroradiol. 1982;3:431–4. 10. Raju VS, Rao MN, Rao VS. Cranial sonography in pyogenic meningitis in neonates and infants. J Trop Pediatr. 1995;41:68–73. 11. Soni JP, Gupta BD, Soni M, Gupta M, Dabi DR, Nemal KR. Cranial ultrasonic assessment of infants with acute bacterial meningitis. Indian Pediatr. 1994;31:1337–43. 12. Mahajan R, Lodha A, Anand R, Patwari AK, Anand VK, Garg DP. Cranial sonography in bacterial meningitis. Indian Pediatr. 1995;32: 989–93. 13. Lozier AP, Sciacca RR, Romagnoli MF, Connolly ES Jr. Ventriculostomy-related infections: A critical review of the literature. Neurosurgery. 2002;51:170–82. 14. Vachon L, Mikity V. Computed tomography and ultrasound in purulent ventriculitis. J Ultrasound Med. 1987;6:269–71. 15. Fukui MB, Williams RL, Mudigonda S. CT and MR imaging features of pyogenic ventriculitis. AJNR Am J Neuroradiol. 2001;22:1510–6.

16. Fujikawa A, Tsuchiya K, Honya K, Nitatori T. Comparison of MRI sequences to detect ventriculitis. AJR Am J Roentgenol. 2006;187: 1048–53. 17. Sommer JB, Gaul C, Heckmann J, Neundörfer B, Erbguth FJ. Does lumbar cerebrospinal fluid reflect ventricular cerebrospinal fluid? A prospective study in patients with external ventricular drainage. Eur Neurol. 2002;47:224–32. 18. Agrawal A, Cincu R, Timothy J. Current concepts and approach to ventriculitis. Infect Dis Clin Pract. 2008;16: 100–4. 19. Tunkel AR, Hartman BJ, Kaplan SL, Kaufman BA, Roos KL, Scheld WM, et al. Practice guidelines for the management of bacterial meningitis. Clin Infect Dis. 2004;39:1267–84. 20. Pickering LK, Ericsson CD, Ruiz-Palacios G, Blevins J, Miner ME. Intraventricular and parenteral gentamicin therapy for ventriculitis in children. Am J Dis Child. 1978;132:480–3. 21. Lorber J, Kalhan SC, Mahgrefte B. Treatment of ventriculitis with gentamicin and cloxacillin in infants born with spina bifida. Arch Dis Child. 1970;45:178–85. 22. Stark G. Treatment of ventriculitis in hydrocephalic infants: Intrathecal and intraventricular use of the new penicillins. Dev Med Child Neurol. 1968;15:36–44.

Meningitis related ventriculitis--experience from a tertiary care centre in northern India.

To assess clinical profile and short term outcome of patients with meningitis related ventriculitis...
537KB Sizes 0 Downloads 4 Views