This was a prospective observational descriptive study conducted in Department of Neonatology and Pathology of a tertiary care hospital (JSS hospital, Mysuru) in the month of January and February 2016. The PIVs, which were removed and discarded after their clinical utility were used for this study. Inclusion criteria were: a) Peripheral intravenous catheter of 24 g; b) Has been in situ for at least 6 hours; c) No continuous infusion through the PIV for last 6 hours; d) No local infection at the site of PIV insertion.

We conducted a pilot study with 10 samples and noted that 90% of the sample demonstrated visible blood in the catheter after flushing [Table/Fig-1]. Considering the prevalence to be 90%, alpha of 0.05 and confidence levels of 89%, we needed 92 samples. Categorical variables were reported as proportions. Continuous variables were analysed for their distribution and summarized as mean and standard deviation or median and interquartile range as appropriate. Microsoft excel 2013 was used for data tabulation and ananlysis.

The experiment consisted of three steps: Step 1 was to estimate the volume of the PIV catheter that was done at the beginning of the study and only once; Step 2 was to obtain the residual blood from the hub of the catheter by flushing; and Step 3 was counting the RBCs in the flush thus obtained.

Step 1 - The volume of the PIV (24 Gauge) was measured as follows. PIV 24 G was completely filled with saline using a syringe. The plunger of the syringe was operated till the amount of saline left in the syringe was 0.5 ml (injected volume). The catheter tip was positioned into a Pediatric EDTA vacutainer. 0.5 ml of saline was now pushed through the catheter which was followed by 1ml air flush to empty the catheter. The saline collected in the vacutainer was aspirated into another syringe and the volume was noted (final observed volume). The volume of the catheter was calculated as final observed volume minus injected volume (0.5 ml). The experiment was repeated three times and the average volume calculated. We measured the volume of the PIV catheter to be 0.2 ml. Finally, this volume was used to flush the experimental catheters.

Step 2 - Once the catheter was removed from the patient, it was held over an EDTA vacutainer with tip pointing into the vacutainer. 0.2 ml of saline was pushed through the catheter. This was followed by 1 ml air flush to empty the catheter. The total volume obtained in the vacutainer was the volume of the catheter plus the volume of saline flush which was equal to 0.4 ml. This resulted in a dilution factor of 2 of the residual flush. Then, the solution in the vacutainer was labelled with the serial number and transported to the Department of Pathology.

Step 3 - RBC count was performed by standard technique using Neubauer chamber within 30 minutes of receipt of the sample. To obtain the actual value of residual blood in the hub of the PIV catheter, the value was multiplied by dilution factor 2. The values were tabulated in Microsoft excel 2013.

We included 106 catheters, which were to be discarded. One of the catheter was completely contaminated with blood during removal and hence not included for further analysis. Four samples were lost in the transit and seven were analysed after 12 hours of collection. These values were considered unreliable due to hemolysis and hence, were not included for final analysis. Thus, a total of 94 samples were finally analysed.

A total of 56% of PIV catheters were removed from term babies and 44% from preterm babies. The minimum duration of PIV catheter in place was 8 hours and maximum duration was 120 hours before we studied them. The minimum time duration between the last routine flush and study was one hour and maximum was 24 hours. Baseline characteristics are depicted in [Table/Fig-2].

Before microscopic analysis, naked eye observation revealed 84 % of the residual flush to be blood tinged and 15% showed blood clot. On microscopic examination, all 100% of the samples revealed RBCs. [Table/Fig-3] depicts the RBC count of residual flush in our study. There was a weak correlation between the duration of PIV catheter stay in situ and the RBC counts. The Pearson product moment correlation ‘r’ was 0.12 [Table/Fig-4].

[Table/Fig-4]:

Correlation between the duration of catheter in situ and RBC count from the Hub.

Our study found blood contamination of the PIV hub in all (100%) of the PIV catheters. Catheter related infection has been a long standing challenge in neonatal intensive care units. The standard practice of PIV catheter care in our unit is to fasten a Bionector to the hub. The Bionector is cleaned with alcohol swab before each use, medications are administered and this is followed by 1ml normal saline push over 30 seconds. The Bionector is cleaned again with the alcohol swab and left in place. As soon as the indication for having PIV catheter ceases, catheter is removed and discarded as per the hospital policy.

The major determinants of catheter related infection are breech in the sterility during insertion and handling of vascular catheters. The safety devices such as Bionector, which are attached to the vascular catheter, provide a closed system. This prevents the back flow of blood and entry of organisms from outside surroundings when in closed position.

Blood from the vein can flow back into the vascular catheter hub during administration of medications. This blood is flushed back into the vein after administering medications. If there is residual blood left over in the hub, this can facilitate the growth of bacteria. A tiny inoculum of the surrounding commensals can grow in this good culture medium. The consequence of this can range from mild thrombophlebitis to fatal sepsis.

There are several studies in the neonatal literature looking into the magnitude of catheter related infections [5,6] and also different flushing techniques to reduce catheter hub contamination [7,8]. Most of them have been conducted on central venous catheters. There is no clear data on similar issues with peripheral venous catheters. The experimental in vitro studies have used external indicators such as proteins or bacterial inoculum to assess the efficacy of different flushing techniques to reduce the catheter hub contamination [9]. In this study, we attempted to quantify the magnitude of hub contamination in peripheral venous catheter in real life scenario using patient’s own RBCs as the indicator of contamination. Our study has shown contamination of the catheter hub in all the cases studied indicating need for impact assessment. Duration of stay did not have influence on the hub contamination.

The concentration of RBCs in the hub after flush noted in our study varied from as low as 170/cumm to as high as 20,80,000/cumm. This can act as a good culture media for the bacteria to grow if inoculated even in small numbers. Our findings warrant revisiting the current flushing technique and exploration of new techniques in real life scenario to minimize PIV catheter hub contamination.

This study did not look into how many of the neonates from whom the catheter was obtained did have either local or systemic infection.

Blood contamination of the small bore peripheral intravenous catheter is universal in spite of regular continuous flushing in neonatal population. We found blood contamination in all the peripheral venous catheters. Majority of them were visible to naked eye. There was no correlation between the duration of stay and magnitude of blood contamination. We are planning to check for microbial contamination in such contaminated catheters in our future study.