IN-USE MICROBIAL CONTAMINATION OF INTRAVENOUS FLUIDS IN THE NEONATAL WARD OF THE WESLEY GUILD HOSPITAL, ILESA

Abstract

In-use microbial contamination of intravenous fluid (IVF) is a known cause of
nosocomial septicaemia in neonates and is associated with increased morbidity and mortality.
From March 1 to June 30, 2011, a prospective study was conducted in the Neonatal Ward
(NNW) of the Wesley Guild Hospital (WGH), Ilesa to determine the prevalence and pattern of
bacterial and fungal isolates, their antimicrobial susceptibility, some associated factors and
outcome of hospitalisation of in-use contamination of IVF. Three hundred consecutive units of
IVF administered on 206 babies admitted to the ward were studied. Information recorded on
each unit of IVF and administration set included the type and number of IVF units and
administration sets used, time of set up of the infusion, duration of use and whether or not
additives were put in the fluids. The baby’s history was taken at first contact to include age,
gestational age, sex, place of delivery and admission diagnosis. From each unit of IVF, two
samples, each of 20 mL were taken for microbiological analysis, the first at the commencement
of infusion therapy (i.e. sample A) and the second by the end of infusion therapy or when the
unit of IVF was to be changed (i.e. sample B). One millilitre of blood each was taken for blood
culture and sensitivity before commencement as well as after IVF discontinuation or whenever
the patient’s condition deteriorated. Other investigations were done as indicated.
Of the three hundred units of IVF used by the babies, twenty (6.7 percent) units of IVF
were found to be contaminated at the commencement of infusion therapy. These 20 (6.7%) IVF
units with contamination at the commencement of infusion as well as the 20 (9.3%) of the 214
IVF administration sets used with these contaminated sample A and the 20 (9.7%) babies who
received these contaminated IVF units, were excluded from further analysis of the corresponding
sample B i.e. only 280 IVF units administered to 186 babies and the 194 IVF administration sets
15

used for them were analysed for contamination in sample B. Forty-five (16.1 percent) of the IVF
units were contaminated by the end of infusion (in-use) in this study. Significantly higher
proportion of units of IVF used for more than 72 hours, i.e. 18 (25.7 percent) of the 70 units of
IVF compared with 27 (12.9 percent) of the 210 units of IVF used for less than 72 hours were
contaminated (p = 0.011). The average duration of use of these units of IVF found contaminated
by the end of infusion (77.9 hours) was also significantly longer than the 57.4 hours for the
uncontaminated units of IVF (p < 0.001). In addition, there was a strong positive correlation
between the colony count and duration of use of IVFs (r = 0.8, p < 0.001). Also, significantly
higher proportion of units of IVF for which the IVF administration sets were used for more than
72 hours (24.2 percent) were contaminated (p = 0.002). The duration of use of IVF
administration set positively correlated with colony count (r = 0.5, p < 0.001). The 111.5 hours
average duration of use of the IVF administration set for the contaminated fluids was also
significantly longer than the 76.8 hours for the uncontaminated intravenous fluid units (p <
0.001). A significant proportion of babies who used two or more units of IVF each during their
admission (29.2 percent) compared with babies who used one unit of IVF each (8.8 percent) had
their IVF contaminated by the end of infusion (p = 0.0010). Introduction of additives in the IVF
was also identified as a significant risk factor for contamination (p = 0.044). On multivariate
analysis, the duration of use of IVF longer than 72 hours (β = 0.884, OR = 0.413, p = 0.010, CI =
0.210 – 0.813) and the duration of use of IVF administration set longer than 72 hours (β = 0.929,
OR = 0.395, p = 0.035, CI = 0.167 – 0.936) were the only factors that had significant independent
contribution to the contamination of the units of intravenous fluid.
The leading micro-organisms found to contaminate IVFs were the Gram-negative
bacteria- Klebsiella pneumoniae and Proteus vulgaris and the Gram-positive bacteria-
Staphylococcus specie (CoNS and Staphylococcus aureus). K. pneumoniae was the most
common isolate, it was cultured from 22 (48.9 percent) of the contaminated 45 units of IVF by
the end of infusion. Also, the fungus Candida albicans was isolated from four (8.9%) of the
contaminated IVF units.
The bacterial isolates from the units of IVF had high resistance rates to the commonly
used antibiotics in our unit; Ampicillin (96.2 percent), Cloxacillin (96.2 percent), Cefuroxime
(65.4 percent), Ceftriaxone (53.9 percent) and Gentamicin (51.3 percent) but they were mostly
sensitive to Ofloxacin (82.1 percent) and Ciprofloxacin (75.6 percent).
The mean duration of hospital stay for the babies whose IVF units were contaminated
was 305.0 hours and this was significantly longer than the 216.0 hours for babies whose IVF
units were not contaminated (p = 0.037). Three of the 31 babies whose units of IVF were
contaminated by the end of infusion developed infusion-associated septicaemia with K.
pneumoniae isolated. One of these three babies died.
The present study showed that IVF contamination is a problem in the NNW of the
Wesley Guild Hospital, Ilesa and there is widespread resistance to the current empirical
antibiotics. The inclusion of Ciprofloxacin and Ofloxacin in the empirical management of
infusion-associated septicaemia may be imperative.

IN-USE MICROBIAL CONTAMINATION OF INTRAVENOUS FLUIDS IN THE NEONATAL WARD OF THE WESLEY GUILD HOSPITAL, ILESA