Bacteriological Profile and Antimicrobial Susceptibility Pattern of Cerebrospinal Fluid Shunt Infections in Infants and Children.

BACKGROUND: Despite the advent of modern neurosurgical techniques, new antibiotics, and modern imaging techniques, infection after ventriculoperitoneal (VP) shunt insertion and/or ventriculostomy is still a serious issue. AIM: The aim of this work was to study bacteriological profile and antimicrobial susceptibility pattern of cerebrospinal fluid shunt infections in infants and children.
MATERIALS AND METHODS: A total of 90 patients under the age of 12 years undergoing cerebrospinal fluid shunt procedures were included. The CSF shunt fluid, external ventricular drain (EVD), shunt tube tip specimens were processed as per standard microbiological techniques. The organisms isolated were subjected to antimicrobial susceptibility using the Kirby-Bauer disk diffusion method.
RESULTS: Of 20 infected patients 10 (50%) were culture positive. Most common organisms isolated were Acinetobacter baumannii 03 (30%) followed by Enterococcus faecalis 2 (20%). Two isolates of A. baumannii and one isolate of Klebsiella pneumoniae showed carbapenem resistance, which were sensitive to colistin. All Gram-positive isolates were sensitive to vancomycin and linezolid. Reinfection was found only in one (8.33%) patient. In 12 (60%) infected cases with shunt failure, revision of shunt was done. The shunt related mortality in this study was 1.11%.
CONCLUSION: Most common organisms isolated were A. baumannii followed by E. faecalis. Carbapenem resistance was noted in these isolates which were sensitive to colistin. All Gram-positive isolates were sensitive to vancomycin and linezolid. Copyright:

INTRODUCTION

Cerebrospinal fluid (CSF) shunt has been the treatment of choice for hydrocephalus since the invention of the shunt valve by John Holter in 1959. CSF shunting procedures provide a rapid means of normalizing intracranial pressure and can prevent neuronal damage as well as other detrimental sequelae.[1]

Despite the advent of modern neurosurgical techniques, new antibiotics, and modern imaging techniques, infection after ventriculoperitoneal (VP) shunt insertion and/or ventriculostomy is still a serious issue.[2] Shunt infection rates range from 20 to 40%.[1]

The main causative agents of shunt infections are Gram-positive like Staphylococcus epidermidis (52.8 to 88.9%), S. aureus (12 to 40%) also Streptococcus Group B and Enterococcus species. Gram-negative bacilli (9 to 22%) like Enterobacter species, Klebsiella pneumoniae, and Pseudomonas aeruginosa.[345] Carbapenem-resistant isolates and MRSA are increasing across the globe and making it difficult to treat shunt infections. Although some infections have been managed successfully with antimicrobial therapy alone. Inadequate or inappropriate treatment can cause reinfection of shunt. Hence, early diagnosis and effective treatment of shunt infections are compulsory to reduce morbidity and mortality.[26] This study was undertaken to study causative pathogens of CSF shunt infections and their antimicrobial susceptibility pattern.

MATERIALS AND METHODS

This prospective study was conducted for 1 year in the Department of Microbiology of a tertiary care teaching hospital.

Patients under age of 12 years undergoing CSF shunt procedures.

Patient giving consent of study.

Children on corticosteroid, immunosuppressive, and antimetabolite therapy.

Children with hematological disorder.

Methodology

A total of 90 patients with VP shunt [Figures 1 and 2] were included in the study. Demographic findings, predisposing medical conditions, clinical sign and symptoms, age at insertion of shunt, shunt infections, other laboratory parameters, antibiotics used for perioperative prophylaxis, treatment modality including type and duration of antibiotics, follow-up, and final outcomes were noted.

Figure 1

X-ray showing ventricular end of VP shunt

Figure 2

X-ray showing abdominal end of VP shunt

Criteria used for the CSF shunt infections (CDC/NHSN)[7]

The United States Centers for Disease Control (CDC) and the National Healthcare Safety Network (NHSN) has described healthcare-associated ventriculitis or meningitis as follows.

patients who meet at least one of the following criteria:

An organism cultured from the CSF

At least two of the following signs or symptoms with no other recognized cause in patients aged >1 year of age: fever >38°C or headache, meningeal signs, or cranial nerve signs, or at least two of the following signs or symptoms with no other recognized cause in patients aged ≤1 year of age: fever >38°C or hypothermia <36°C, apnoea, bradycardia, or irritability.

At least one of the following: increased CSF white blood cell count, elevated CSF protein, and decreased CSF glucose; organisms seen on a CSF Gram stain; organisms cultured from the blood; positive nonculture diagnostic test from the CSF, blood, or urine; diagnostic single-antibody titer (immunoglobulin M) or fourfold increase in paired sera (immunoglobulin G) for organism.

Specimen collection[8,9]

The CSF shunt fluid, EVD, shunt tube tip specimens were collected by the clinician by aseptic precautions in sterile container. The specimens were immediately transported to the laboratory and processed as per standard microbiological techniques. Do not refrigerate.

Processing of specimen

Gross findings like volume, appearance of CSF, that is, clear, bloody, cloudy, xanthochromic was noted. If <1 mL volume specimen was vortexed. If >1 mL volume, the specimen was centrifuged at 2500–3000 rpm for 15–20 min. Supernatant was removed carefully with a sterile pipette and 0.5 mL of the sediment was mixed well.

The specimen/sediment was inoculated in Trypticase soy broth (TSB) and thioglycollate broth. A wet mount of specimen/sediment was prepared and looked for pus cells, RBC, microorganisms. Gram stain was also prepared.

The specimen was inoculated onto Sheep Blood agar (SBA), Sheep chocolate agar (SCA), and MacConkey agar (MA). SBA and SCA were kept for incubated in candle jar at 37°C for 48 h. MA was incubated aerobically at 37°C for 24 h. Subculture of the TSB was done after 4 h and 24 h on MA, SBA and SCA.

Thioglycollate broth was examined for turbidity every day for seven days. If broth was turbid then subcultured done on SBA, SCA, and MA. If no turbidity in broth was observed then broth was discarded after seven days of incubation. Organisms were identified using Gram stain characteristics, colony characteristics, and biochemical reactions as per standard microbiological techniques.

Antimicrobial susceptibility testing[10]

From the primary culture plate bacterial suspension was prepared in trypticase soya broth till turbidity matched with 0.5McFarland standard. The organisms isolated were subjected to antimicrobial susceptibility using the Kirby–Bauer disc diffusion method on Muller Hinton agar and Muller Hinton Blood agar as per standard microbiological techniques. Results obtained were interpreted as per Clinical and Laboratory Standards Institute (CLSI) guidelines 2017.[10]

Antibiotic disks used for testing antimicrobial susceptibility test for Gram-positive organisms were as follows: penicillin (P) 10 μg, erythromycin (E) 15 μg, clindamycin (Cd) 2 μg, linezolid (LZ) 30 μg, cefoxitin (CX) 30 μg, vancomycin (VA) 0.016–256 μg/mL, chloramphenicol (C) 30 μg, ampicillin (AMP) 10 μg, and trimethoprim/sulfamethoxazole (COT) 23.75/1.25 μg.

Antibiotic disks used for testing antimicrobial susceptibility test for Gram-negative organisms were as follows: ampicillin (AMP) 10 μg, amoxicillin-clavulanic acid (AMC) 20/10 μg, ceftriaxone (CTR) 30 μg, ceftazidime (CAZ) 30 μg, ciprofloxacin (CIP) 5 μg, cefepime (CPM) 30 μg, imipenem (IMI) 10 μg, meropenem (MERO) 10 μg, piperacillin-tazobactam (PTZ) 100/10 μg, colistin strip (COLI) 0.016–256, trimethoprim/sulfamethoxazole (COT) 23.75/1.25 μg, polymyxin B (300 units), ampicillin-sulbactam (A/S) 10/10 μg, chloramphenicol (C) 30 μg, and tigecycline (TG) 15 μg.

The standard strains of S. aureus ATCC 25923, Escherichia coli ATCC 25922 and P. aeruginosa ATCC 27853 were used for quality control.

RESULTS

In this prospective study, a total of 90 patients who had undergone CSF shunt were analyzed. Maximum patients belonged to age group 0 to 1 month 29 (32.23) followed by >1 month to ≤6 months 17 (18.89%). Maximum patients 72 (80%) were under the age of 2 years. Of the total 90 cases 53 (59%) were males and 37 (41%) were females.

Of the 90 cases, maximum cases were of congenital malformations (MMC and aqueductal stenosis) 61 (67.78%), followed by tuberculous meningitis 24 (26.6%) [Table 1]. So, congenital malformations were the commonest etiologies for shunt surgery in this study.

Table 1

Distribution of cases as per etiology of shunt surgery (n = 90)

Etiology		No. of cases (n = 90)	Percentage %
Congenital malformations	Meningomyelocele (MMC)	52	57.78
	Aqueductal stenosis	09	10
Tuberculous meningitis		24	26.67
Hemorrhage (subarachnoid and intraventricular)		02	2.22
Tumor (craniopharyngioma)		01	1.11
Head Injury		01	1.11
Colpocephaly		01	1.11

Of the 20 infected patients most, the common age group was of >1 year to ≤2 years 08 (40%) followed by 0 to ≤1 month 5 (25%) and > 1 month to ≤6 months 04 (20%) [Table 2]. Mean 16.5 months and the standard deviation 15.9 months. All infected patients were younger than 2 years old 18 (90%) except two patients who were of age 3 years and 5 years. Shunt infections were observed more in males 15 (75%) than females 5 (25%).

Table 2

Age-wise distribution of shunt infection cases (n = 20)

Age	No. of cases (n = 20)	Percentage (%)
0 to ≤1 month	05	25
>1 month to ≤6 months	04	20
>6 month to ≤1 year	01	05
>1 year to ≤2 years	08	40
>2 year to ≤5 years	02	10
>5 year to ≤12 years	0	0

The most common underlying condition amongst all the infected cases were congenital malformations (MMC+ aqueductal stenosis) in 13 (65%) cases followed by tuberculous meningitis 6 (30%) cases and head injury 1 (5%) case.

In five infected patients, cell count was increased (>5/μL) and mean 10.82/μL. CSF Shunt Fluid Protein was increased (>45 mg/dL) with mean of 27.17 mg/dL and fluid sugar was decreased (<45 mg/dL) with mean of 40 mg/dL.

Pus cells and organisms were seen on CSF shunt fluid Gram stain 5 (25%) of the infected cases. Of 20 infected patients 10 (50%) were culture positive. According to Overturf’s 11 diagnostic criteria for VP shunt infections, 10 culture-positive cases were definite shunt infections and other 10 culture-negative were cases of probable shunt infections.[16]

Most common organisms isolated were A. baumannii 3 (30%) followed by E. faecalis 2 (20%) followed by one isolate each of E. aerogenes, P. aeruginosa, and Methicillin-resistant S. aureus (MRSA) [Table 3].

Table 3

Bacteriological spectrum of organisms isolated in shunt infection cases

Isolates	No. of cases (n = 10)	Percentage
Gram-negative bacilli (n = 6)
Acinetobacter baumannii	03	30
Klebsiella pneumoniae	01	10
Pseudomonas aeruginosa	01	10
*Enterobacter aerogenes	01	10
Gram-positive cocci (n = 4)
Enterococcus faecalis	02	20
Streptococcus Grp B	01	10
Methicillin-resistant Staphylococcus aureus (MRSA)	01	10

*For this patient revision of the shunt was done and after 4 months, he got infected with Proteus mirabilis which was sensitive only to imipenem and meropenem and was treated with carbapenems. This was the only case of reinfection in this study

Among Acinetobacter baumannii two were resistant to baseline drugs and carbapenems, which were sensitive to colistin and tigecycline: One isolate of Klebsiella pneumoniae also showed carbapenem resistance which was sensitive to colistin [Tables 4 and 5].

Table 4

Susceptibility of gram-negative organisms in shunt infection cases (n = 6)

Isolates	Baseline antibiotics
	AMC	A/s	CTR/CAZ	COT	C
A. baumannii (n = 3) *(CRE 02)	0	0	01	01	0
K. pneumoniae (n = 1) (CRE 01)	0	01	0	0	0
E. aerogenes (n = 1)	NA	NA	01	01	0
*P. aeruginosa (n = 1)	NA	NA	01	NA	0

NA = not applicable

*For Pseudomonas aeruginosa baseline antibiotics are CAZ, MERO, IMI, PTZ, and C

Table 5

Susceptibility of gram-negative organisms in shunt infection cases (n = 6)

Isolates	Higher antibiotics
	PTZ	IMI	MERO	COLI	TG
A. baumannii (n = 3) *(CRE 02)	02	01	01	03	02
K. pneumoniae (n = 1) (CRE 01)	0	0	0	01	0
E. aerogenes (n = 1)	01	01	01	01	01
P. aeruginosa (n = 1)	01	01	01	01	NA

NA = not applicable, CRE = carbapenem-resistant isolate

All Gram-positive isolates were sensitive to vancomycin and linezolid [Table 6]. The patient infected with MRSA isolate succumbed to infection.

Table 6

Susceptibility of Gram-positive organisms in shunt infection cases (n = 4)

Antibiotics	P	AMP	VA	LZ	C	COT	CTR/CX
E. faecalis (n = 2)	02	02	02	02	0	NA	NA
MRSA (n = 1)	0	NA	01	01	01	01	0 (CX)
Strepto. Group B (n = 1)	0	NA	01	01	01	0	01 (CTR)

In seven patients, conservative management was done. Two patients underwent shunt exteriorization followed by revision of shunt. Other 10 patients were managed by complete shunt replacement. So, in 12 (60%) patients, complete revision of shunt was done. One patient with MRSA infection succumbed to the infection.

DISCUSSION

Hydrocephalus is a common neurosurgical disease that develops via a variety of etiologies, including congenital anomaly, intracranial hemorrhage, infections, and tumor.[11] Although CSF shunts contributed to the significant improvement of the management and outcome of hydrocephalus, the shunt has several complications, including infections, which is a major threat to pediatric patients.[12] Many factors have been reported to be associated with increased risk of infections including the age of patient, etiologies of hydrocephalus, type of shunt implanted, surgeons experience, presence of previous shunt infection, patient skin colonization and surgical technique used.[4] So, special consideration should be paid concerning the patients preoperative condition. The incidence of infection following VP shunt placement is reported to be approximately 20%–40%.[11]

Shunt complications can be divided into three general categories: mechanical failure (improper insertion of shunt, partial shunt blockage, fracture or displacement of the shunt portions or shunt movement from the initial place), dysfunctional shunts and infections.

This study showed that infections occurred in 20 (22.22%) of the 90 patients who underwent VP shunt insertion. This result is similar to the studies by Bokhary et al.,[13] Braga et al.[14] However, studies by Mancao et al.,[15] McGirt et al.,[16] Lee et al.[17] showed lower infection rate.

Of 20 infections only 10 (11.11%) cases were culture positive which is statistically significant (P = 0.000). According to Overturf’s 11 diagnostic criteria for VP shunt infection 10 were definite shunt infections and other 10 were cases of probable shunt infections.[18] Pan et al.[19] and Wang et al.[2] also reported definite shunt infection rate of 16.21% and 9.3%, respectively.

In this study, 10 organisms were isolated of which 6 (60%) were Gram-negative and 4 (40%) were Gram-positive. A. baumannii 03 (30%) was the most common isolated organism. Bisno et al.[4] also showed 20% Acinetobacter species in shunt infections. It should not be forgotten that the cause of nosocomial shunt infections are Gram-negative organisms. In our study, the rate of Gram-negative microorganisms was fairly higher. It can be speculated that, simultaneous infections in other parts of the body which were caused by the same Gram-negative microorganisms may be responsible for that higher incidence.

Bokhary et al.[13] also showed more Gram-negative isolates (81.3%). In other studies, Gram-positive microorganisms are found to predominate in the shunt infections where Coagulase negative Staphylococcus are common cause of shunt infections as they are the skin commensal and infants resident skin flora keeps on changing. In the study by Mancao et al.[15] showed 79.5% of Gram-positive and 10.2% Gram-negative organisms. Acinetobacter baumannii 3 (50%) was the most common isolated Gram-negative organism of which two were resistant to baseline drugs and carbapenems. These carbapenem-resistant isolates were sensitive to colistin and tigecycline. Other Gram-negative isolates were one isolate each of E. aerogenes, P. aeruginosa, and K. pneumoniae. Klebsiella pneumoniae also showed carbapenem resistance which was sensitive to colistin.

Acinetobacter baumannii infections are a recognized problem in healthcare, causing VP shunt infections and ventriculitis. Such infections are serious intracranial infection that can lead to serious complication and death. Treatment of infections caused by A. baumannii becomes difficult because of its inclination to develop pandrug resistance to the universally used antibiotic. In the study by Demoz et al.[20] reported baseline and carbapenem-resistant A. baumannii from shunt fluid which was only sensitive to ampicillin-sulbactam and colistin, similar to our study. Enterococcus faecalis 02 (50%) was the commonest Gram-positive isolate followed by one isolate each of Methicillin-resistant Staph. aureus (MRSA) and Streptococcus Group B. All Gram-positive isolates were sensitive to vancomycin and linezolid.

All infected patients with Gram-negative isolates were managed by ceftriaxone (50–100 mg/kg) and carbapenems (40 mg/kg), carbapenem-resistant isolates were treated with colistin and tigecycline. Patients infected with Gram-positive isolates were treated with ceftriaxone and vancomycin (10–15 mg/kg). However, MRSA-infected patient was treated vigorously with vancomycin and linezolid, but developed sepsis and respiratory distress and expired after 45 days of shunt placement. Patients with tuberculous meningitis were treated with antitubercular therapy along with shunt surgery. Two patients underwent shunt exteriorization however EVD specimen grew E. aerogenes and E. faecalis. So, they were managed by revision of shunt.

Reinfection was found only in one (8.33%) patient. This patient had first infection with E. aerogenes and was managed with revision of shunt. After 4 months, patient developed repeat shunt infection with P. mirabilis which was managed with carbapenems. Repeat shunt infection of 10.41% was seen in the study by Agarwal et al.[21] (10.41%). Yilmaz et al.[6] reported repeat shunt infection 25.86%.

CONCLUSION

Most common organisms isolated were A. baumannii followed by E. faecalis. Carbapenem resistance was noted in these isolates which were sensitive to colistin. All Gram-positive isolates were sensitive to vancomycin and linezolid. Diagnosis and management of carbapenem-resistant Gram-negative isolates and MRSA isolates is essential as shunt infections with these resistant organisms can lead to mortality and morbidity.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.