Investigation of a cluster of ventriculostomy associated gram-negative bacillary meningitis cases in a neurosurgical intensive care unit.
Kennedy C, McNeely D, Houston T, Wallace C, McCarthy MK, Conly J.
APIC 1997 Conference, New Orleans, June 8-12, 1997 (poster presentation)
ABSTRACT
Although postoperative infections of cerebrospinal fluid (CSF) associated with external ventriculostomy drainage systems occur with a frequency of approximately 10% and are not infrequently caused by gram-negative bacilli (GNB), there have been few reported clusters. During a 3 month period, June-Sept 1996, 9 patients were identified with meningitis in the early postoperative period. Of these patients, 8 were in the neurosurgical intensive care unit, with 1 in a medical intensive care unit. Six had GNB identified from ventriculostomy samples, as a single isolate in 3 and polymicrobial in 3 other cases. A total of 23 different isolates were identified with 8 (35%) identified as either Klebsiella-Enterobacter or Pseudomonas. The remainder of the isolates were staphylococci or streptococci. The Enterobacter isolates had dissimilar antibiograms. All patients had an external ventriculostomy drainage system in place. Potential risk factors identified included concomitant GNB endotracheal colonization, tracheostomy and ventriculostomy procedures performed on the same day, multiple and/or bedside ventriculostomy tubing changes, embolization or coiling procedures performed, and a leaking ventriculostomy site. The isolation of dissimilar strains between these patients and the colonization at other body sites did not support a common source outbreak. Recommendations made included the adherence to strict aseptic technique including a review of sampling procedures, the use of sterile gloves during calibration of the line, discontinuation of the practice of bedside tubing changes, consistent use of a closed endotracheal suctioning system, increased use of the Camino drainage system and ventriculostomy changes to be done prior to tracheotomy procedures. In the following 6 months, there has been only one episode of ventriculostomy associated GNB meningitis.
METHODS
Description of the Cluster:
- 9 ICU patients with external ventriculostomy systems in place with 1-7 episodes of ventriculitis
- 4 month period (June-Sept 1996)
- 5 patients had 1 infection ; 4 had 2-7 episodes
- 20 episodes of infection ( 23 isolates )
- No GNB CSF isolates had been identified within the previous 6 months
Venticulostomy Infection Episodes 1996
CLINICAL and MICROBIOLOGIC INVESTIGATIONS
- Medical record review of identified cases
- Prospective surveillance to identify new cases
- Review of current policies and procedures
- Observation of present practices
- Review of microbiology reports
- Comparison of extended antibiograms of a similar isolates
- Formation of multi-disciplinary team to address issues
RESULTS
- Patient Characteristics:
- 3 female, 6 male
- Age range 33-73 years
- Diagnosis: hemorrhage (5), tuberculous meningitis (1),
tumor (1), cyst (1), closed head trauma (1)
- 36 venticulostomy insertions/ changes (1-10 times)
- Catheter days: mean 6.6 +/- 1.8 (SD) range 1-10
- CSF Infections:1 (5 patients), 2 (1 patient), 3 (1 patient), 4 (1 patient), 7 (1 patient)
MICROBIOLOGIC
- 4/20 episodes (20%) were polymicrobial
- 6/9 patients (66%) had a GNB identified in CSF
- 10/29 isolates (34%) were gram-bacilli [GNB]
- Klebsiella pneumoniae (3), Enterobacter cloacae (3), Pseudomonas aeruginosa (2) and Stenotrophomonas maltophilia (1) (antibiograms were dissimilar)
- 6 episodes (30%) the identical GNB was also identified in an endotrachial specimen within the same week
- 2 episodes (10%) the identical GNB was also identified at the ventriculostomy exit site within the same week
INTRODUCTION
External intracranial pressure (ICP) monitoring systems consist of a
catheter or screw (bolt), sensor, pressure tubing, transducer and a recording device. They are used in the management of cerebral hypertension after a neurological episode.
Intraventricular catheter:
- inserted via burr hole, through meninges and into lateral ventricle
- risk of ventriculitis is 10-17% (2,3)
- monitor intracranial pressure (ICP) intracranial haemorrhage or head injury
- drain cerebrospinal fluid (CSF) to reduce ICP
- collection of CSF specimens
- administer intrathecal antibiotics or contrast media
Point of Practice Observations:
- Sterile gloves were not consistently used when calibrating or obtaining CSF from the system
- Ventriculostomy catheters were changed q6-7 days
- A few ventriculostomy changes had been done in the ICU
- Sampling was done from both the proximal and distal ports
- A tracheostomy was done prior to a ventriculostomy change in the OR
- The system was entered daily for the routine collection of CSF for culturing
- In-line suction catheters were not used consistently
SUMMARY and CONCLUSIONS
- Only 2 ventriculostomy associated infections were identified during the next 6 mos (1 associated with GNB)
- Implementation of the recommendations appeared to be associated with a significant decrease in ventriculitis
- It is uncertain which of the recommendations was responsible for the reduction but those directed towards reducing cross-contamination between ETT secretions and the ventriculostomy system appear to be significant
- Surveillance of ICP system infections continues and documentation of bedside insertions has been implemented
RISK FACTORS FOR INFECTION
- intracranial hemorrhage, open head injury,
- CSF leak, craniotomy
- manipulation, replacement or irrigation of the system
- concurrent systemic infections e.g. pneumonia
- insertion technique (risk- tunneling catheter)
- insertion location (OR, ER or ICU)
- duration of monitoring and catheter days
- ? catheter care (dressing or stopcocks)
- ? prophylactic antibiotic or steroid use
OBSERVATIONS
- 9- ICP drainage
- 3- ICU bedside changes
- 3- ventriculostomy/tracheotomy procedures
- 2- intra-thecal therapy
- 3- embolization
- 1- implant (Omaya)
- 1- acute cholecystitis post-op
RECOMMENDATIONS
- Aseptic technique and sterile gloves to be used while manipulating an ICP system
- In-line suction catheters to be used
- Assess necessity of routine changes of ventriculostomy catheter and sets
- It is preferable to change ventriculostomy catheters in the OR
- Ventriculostomy changes should precede a tracheotomy
- A closed monitoring system should be used if drainage is not required
- CSF samples should be collected from a port rather than a stopcock.
- CSF samples should be collected from a distal rather than proximal port
- The need for routine daily CSF sampling should be reviewed.
- Policies should indicate the person responsible for obtaining CSF specimens
MICROBIOLOGY
Patient 1
June 13 ventriculostomy
June 14 embolization, CSF: Klebsiella pneumonia, group A haemolytic Streptococcus
June 15 sputum: Klebsiella pneumonia, group A haemolytic Streptococcus
Patient 2
June 18 ventriculostomy
June 20 CSF Klebsiella pneumonia, Enterobacter cloacae
June 27 sputum Klebsiella pneumonia
Patient 3
June 25 ventriculostomy
June 27 embolization
June 30 CSF Streptococcus mitis
Patient 4
July 18 ventriculostomy
July 19 CSF Pseudomonas aeruginosa
July 13 sputum Pseudomonas aeruginosa
Patient 5
June 24 ventriculostomy removed/ VP shunt inserted
July 7 CSF Staphylococcus aureus
June 14 sputum Staphylococcus aureus
Patient 6
July 19 ventriculostomy and tracheotomy
July 19 CSF Pseudomonas aeruginosa
July 22 sputum Pseudomonas aeruginosa
Patient 7
July 19 ventriculostomy
July 20 CSF Klebsiella pneumonia
July 20 sputum Klebsiella pneumonia
Patient 8
August 7 ventriculostomy (bedside)
August 12 CSF Enterobacter cloacae
August 15 exit site(leaking) Enterobacter cloacae
Patient 9
July 24 Omaya removed, VP shunt
July 24 CSF Staphylococcus aureus
REFERENCES
- Aucoin PJ et al. Intracranial pressure monitors: epidemiological study of risk factors and infections. Am J Med 1986;80:369-76.
- Brader MK et al. Ventriculostomy and intracranial pressure monitoring: in search of a 0% infection rate. Heart & Lung 1995;24:166-72.
- Holloway KL et al. Ventriculostomy infections: the effect of monitoring duration and catheter exchange in 584 patients. J Neurosurg 1996;85:419-24.
- Paramore CG et al. Relative risks of ventriculostomy infection and morbidity. Acta Neurochir (Wien) 1994;127:79-84.
- Hickman KM et al. Intracranial pressure monitoring: review of risk factors associated with infection. Heart & Lung 1990;19:84-91.
- Schultz M et al. Bacterial ventriculitis and duration of ventriculostomy catheter insertion. J Neurosci Nurs 1993;25:158-164.
- Franges EZ et al. Infections related to intracranial pressure monitoring. J Neurosci Nurs 1988;20:94-103.
