Evidence-based Protocol Implementation to Prevent External Ventricular Drain Infection – 5-year Experience

Abstract Description

Presentation Upload :

Authors (including presenting author) :

Boo MS, Taw BBT, Lui WM

Affiliation :

Department of Neurosurgery, Queen Mary Hospital

Introduction :

External ventricular drain (EVD) is commonly seen in neurosurgical intensive care unit. It is used to treat hydrocephalus by allowing cerebral spinal fluid (CSF) drainage and to monitor intracranial pressure in patient with intracranial pathology. EVD infection has always been a problem. Although the EVD infection rate in our Department being about 1% in the previous years, EVD infection and its associated clinical ventriculitis are causes of significant morbidity and mortality. We advocate the principal: “one case is too many” in our practice. Therefore, our Department has implemented an evidenced-based protocol on the prevention of EVD infection since 1 January 2014. Our target being a zero EVD infection rate.

Objectives :

(1) To target a zero EVD infection by implementing the evidence-based protocol. (2) To review the result after 5-year implementation.

Methodology :

A clinical study protocol from the United States was adopted as reference and was modified to fit our clinical setting. The protocol includes 3 parts: EVD insertion, dressing and manipulation. In the protocol, 2% Chlorhexidine in alcohol is used as disinfectant and a dressing with Chlorhexidine gluconate patch is applied to the EVD exit site. Both plain EVD and antibiotic-coated EVD were used. Other than the peri-operative antibiotics given, no routine prophylactic antibiotics were prescribed to patients with EVDs during the course of CSF drainage.

Result & Outcome :

Data has been reviewed from 1 January 2014 to 31 December 2018. A total of 631 EVDs in 554 patients, with 5546 catheter-days have been analyzed. The duration of EVD placement ranged from 1 to 34 days. Primary diagnoses including intracerebral haemorrhage, subarachnoid haemorrhage, traumatic brain injury and brain tumour. Patients with pre-existing central nervous system (CNS) infection as well as those with cranial-facial and cranial-nasal pathology which is known to have association with CNS infection were excluded from the analysis. One case of ventriculitis has been identified.