Klebsiella oxytoca outbreak in an intensive care unit: a probable link to common insulin vial use.
The ICU of our university hospital is a 13-bed medical/surgical unit. All ICU patients have daily reviews by infectious diseases and clinical microbiology specialists. Klebisella oxytoca was isolated from blood cultures of four patients who had developed fever within two consecutive days. All patients responded well to antibiotic therapy and two cases were accepted as breakthrough bacteraemia.
The infection control team conducted an investigation to identify the source of the outbreak. The medical records of four patients were thoroughly reviewed. Demographic data, location of the patients in the ICU, nursing care, invasive procedures, preparation and administration of intravenous fluids and drugs were evaluated.
Evaluation of patient localisation in the ward did not show a local clustering. Bacteraemic patients were given medical care by different nursing staff. Evaluation of nursing and invasive procedures, preparation and administration of intravenous fluids and drugs did not reveal any source for the outbreak. There was no shared equipment, parenteral or enteral nutrition between patients. All parenteral fluids and drugs including opioids were separately prepared for each patient, except for insulin which was drawn from common stock vials. Samples taken from multi-dose vial insulin solutions were sterile. To eliminate the probability of a pseudo-outbreak, antiseptic solutions used for decontamination of the skin were cultured but also were sterile. Parenteral fluids were not available for culture as they had already been discarded.
As the source of the outbreak could not be determined, a case-controlled study was conducted. A case was defined as a patient having fever and a positive blood culture for Klebisella oxytoca during the period of outbreak. A control was defined as a patient followed up in the ICU during the outbreak period with no positive blood culture for Klebisella oxytoca. The outbreak period included the day preceding and one week following the day of first culture positivity for Klebisella oxytoca. Epi Info[TM] software (version 6.04) was used in statistical analysis. Fisher's exact test was used for comparison between the groups.
A detailed analysis showed that the sole difference between case and control patients was insulin infusion (Table 1). All case patients compared to only one control patient had insulin infusion (4/4, 1/7; P=0.01). Detailed re-investigation revealed that insulin was injected to the saline solution of each patient by a syringe filled with insulin from a common vial. The insulin vials did not contain any preservatives. Before each use, the stopper of the vial was cleansed with 70% alcohol. The vial used for insulin infusion during the outbreak was already discarded at the time of investigation; a sample from a current insulin vial was cultured and was sterile. Similar to the other reports of Klebisella oxytoca outbreaks due to contaminated parenteral fluids, we thought that a shared insulin vial may have been contaminated with Klebisella oxytoca due to poor aseptic techniques applied during cleansing of the stopper of the vial (4,5). Nevertheless, the statistical power of our study was compromised by the small sample size of the case-controlled group since the outbreak was swiftly controlled.
The molecular typing of the isolates was performed by pulsed field gel electrophoresis, a method frequently used in epidemiological analysis with a high discrimination power in most of gram negative bacteria including Klebsiella spp (6). The pulsed field gel electrophoresis analysis revealed that the outbreak was monoclonal, supporting the hypothesis of common insulin vial contamination (Figure 1).
In conclusion, Klebsiella oxytoca may lead to sudden and rapidly disseminating outbreaks in intensive care units. Contamination of parenteral fluids due to inappropriate aseptic techniques during preparation may be the most likely cause of the outbreak. Rapid and detailed investigation and identification of the source is crucial for the control of the outbreak.
Caption: Figure 1: PFGE analysis of the bacteria isolated from the patients. PFGE=pulse field gel electrophoresis.
(1.) French GL. Antimicrobial resistance in hospital flora and nosocomial infections. In: Mayhall CG (ed). Hospital Epidemiology and Infection Control, 3rd ed. Philadelphia: Lippincott Williams and Wilkins 2004; p. 1613-1636.
(2.) Worku B. Klebsiella oxytoca outbreak at the Ethio-Swedish Children's Hospital (ESCH). Ethiop Med J 1997; 35:177-183.
(3.) Morgan ME, Hart CA, Cooke RW. Klebsiella infection in a neonatal intensive care unit: role of bacteriological surveillance. J Hosp Infect 1984; 5:377-385.
(4.) Sardan YC, Zarakolu E Altun B, Yildirim A, Yildirim G, Hascelik G et al. A cluster of nosocomial Klebsiella oxytoca bloodstream infections in a university hospital. Infect Control Hosp Epidemiol 2004; 25:878-882.
(5.) Toldos CM, Ortiz G, Camara M, Segovia M. Application of pulsed-field gel electrophoresis in an outbreak of infection due to Klebsiella oxytoca. J Med Microbiol 1997; 46:889-890.
(6.) Stratton CW, Greene J. Role of the microbiology in hospital epidemiology and infection control. In: Mayhall CG (ed). Hospital Epidemiology and Infection Control, 3rd ed. Philadelphia: Lippincott Williams and Wilkins 2004; p.1809-1825.
Y. Aybar Bilir
Table 1 Characteristics of the case and control patients Characteristics Cases, n=4 Control, n=7 Age, mean (SD) 53 ([+ or -] 25.86) 57 ([+ or -] 28.54) Male 3 3 APACHE H score (SD) 22 ([+ or -] 22.25) 21 ([+ or -] 21.43) Diagnosis at admission Postoperative follow-up 1 1 Respiratory insufficiency 2 2 Sepsis 1 2 Neurological diseases 2 Central venous catheter 3 5 Total parenteral nutrition 1 0 Intubation 4 5 Intravenous administration 4 4 of antibiotics Insulin infusion 4 1 Transfusion of blood 3 3 products Characteristics Total, n=11 Pvalue * Age, mean (SD) 55 ([+ or -] 33) 0.82 Male 6 0.54 APACHE H score (SD) 22 ([+ or -] 8.88) 0.88 Diagnosis at admission Postoperative follow-up 2 Respiratory insufficiency 4 Sepsis 3 Neurological diseases 2 Central venous catheter 7 -- Total parenteral nutrition 1 0.36 Intubation 9 0.49 Intravenous administration 8 0.23 of antibiotics Insulin infusion 5 0.01 * Transfusion of blood 6 0.54 products * A P value of <0.05 was considered statistically significant. SD=standard deviation, APACHE=acute physiology and chronic health evaluation.
|Printer friendly Cite/link Email Feedback|
|Author:||Mete, B.; Bilir, Y. Aybar; Aygun, G.; Yilmaz, M.; Urkmez, S.; Dikmen, Y.; Ozturk, R.|
|Publication:||Anaesthesia and Intensive Care|
|Article Type:||Viewpoint essay|
|Date:||Mar 1, 2013|
|Previous Article:||An unusual cause of respiratory distress in the neurosurgical intensive care unit.|
|Next Article:||An alternative method of transverse in-plane ultrasound-guided thoracic paravertebral block.|