Risk Factors of Central Line-Associated Bloodstream Infection (CLABSI): A Prospective Study From a Paediatric Intensive Care Unit in South India

Background Central line-associated bloodstream infection (CLABSI) is the most common hospital-acquired infection. However, studies evaluating the factors associated with the risk of CLABSI in pediatric intensive care units (PICU) were limited in India. Objective The objective of our study is to evaluate the association of factors and the etiology causing CLABSI. Study design This is a hospital-based single-center prospective study conducted in the pediatric intensive care unit (PICU) of our tertiary care hospital spanning one year. Participants Children aged between two months to 15 years admitted in the PICU for more than 48 hours with central venous catheterization were included. Pearson's chi-squared test with Yates' continuity correction and logistic regression with odds ratio were calculated by R statistical software (R Foundation for Statistical Computing, Vienna, Austria) and a p-value less than 0.05 was considered statistically significant. Results Our analysis showed that factors such as young age (2-12 months), high pediatric risk of mortality (PRISM III) score (> 15), leukocytosis, neutrophilia, anemia, change of central venous catheter, duration of catheterization (>7 days), exposure to blood products, use of steroids, inotropes, and prophylactic antibiotics were significantly associated with increased risk of CLABSIs with an odds ratio of 4.53, 4.54, 2.91, 4.56, 4.76, 3.74, 2.49, 2.41, 7.22, 6.77 and 5.16 respectively (p<0.05). Further, factors such as older age (>12 months) and low PRISM III score (≤ 15) significantly reduce the risk of CLABSIs by 83.64% and 69.14% respectively (p<0.05). Conclusion In conclusion, our results revealed that factors such as young age, high PRISM III score, leukocytosis, neutrophilia, anemia, change of central venous catheter, duration of catheterization (> 7 days), exposure to blood products during the hospital stay, use of steroids, inotropes, and prophylactic antibiotics were identified as risk factors for CLABSI.


Materials And Methods
This is a prospective, single-center study conducted for a period of one year (January 2021 to February 2022) in the pediatric intensive care unit (PICU) of our hospital. The PICU receives patients from the emergency department, outpatient department, referrals from other hospitals, and postoperative patients requiring PICU care. The study was approved by our institutional ethical review board (Reference ID: ECASM-AIMS-2021-346). A minimum sample size of 54 was calculated with an 80% power and a 95% confidence interval based on the previous study [9]. Our study included 118 patients aged two months to 15 years admitted to our PICU in whom a central venous catheter was inserted for more than 48 hours. Patients with any previous infections and central venous catheter lines inserted outside our PICU were excluded. Patients' demographics, relevant clinical parameters, central venous catheter-related factors, pediatric risk of mortality (PRISM III) scores, bacterial etiology, and outcomes were collected in a predesigned standardized form. A CLABSI was defined as a primary bloodstream infection in a patient who had 48 hours of central venous catheterization and is not related to infection at other sites [1,10,11,13]. The PRISM III score was scored within 24 hours of hospital admission. Blood samples collected for blood culture were inoculated into brain heart infusion broth and incubated at 37°C for 24 hours in our hospital microbiological laboratory. Then subcultures were performed on blood agar (BA) along with gram films and antimicrobial susceptibility testing was performed by BACTEC (Becton Dickinson Diagnostic Systems, Franklin Lakes, NJ) [11,16]. Blood culture was considered negative if no growth was identified on the final subculture after seven days. Blood culture was considered positive if any growths were identified based on morphology and biochemical reactions during the seven days. Patients were managed as per institutional PICU protocols and antibiotics were administered based on antimicrobial susceptibility.

Statistical analysis
Statistical analysis was performed using R statistical software for Windows, v. 4.1.0 (R Foundation for Statistical Computing, Vienna, Austria) [17][18][19]. Continuous variables were depicted as mean ± standard deviation (SD) and categorical variables were depicted as frequency and percentages. Pearson's chi-squared test with Yates' continuity correction and logistic regression with odds ratio were calculated by the "epitools" package [19]. A p-value of less than 0.05 was considered statistically significant.

Results
Out of the 518 patients admitted to the PICU, our study included 118 patients with central venous catheterization (CVC). Among the 118 patients with CVC included in our study, 27% (32) of patients had CLABSI with an observed incidence of 61.78 CLABSIs per 1000 PICU admissions and 30.42 CLABSIs per 1000 central venous catheter days. Table 1 shows the characteristics of patients included in our study such as age, gender, diagnostic categorization, nutritional status, PRISM III score, hematological factors, catheterrelated factors, exposures to medications, and outcomes. Patients with CLABSI under the age of 12 months were significantly higher than the patients without CLABSI (p = 0.0011). The mean (± SD) age of patients included in our study was 57.75 ± 40.28 months in the CLABSI groups and 60.15 ± 31.90 months in the non-CLABSI groups. Among the patients with CLABSI, no statistically significant difference was observed in gender (p = 0.3374) as shown in Table 1. Higher PRISM III scores of >15 (21.87%) were significantly associated with patients with CLABSI as compared to patients without CLABSI (p = 0.0262) as shown in Table  1. Four (12.5%) patients with CLABSI died during their PICU stay. Further, we observed a significant association between CLABSI and factors such as leukocytosis, neutrophilia, anemia, use of steroids, inotropes, and prophylactic antibiotics as shown in Table 1. However, we observed no significant association between CLABSI and factors such as nutritional status, diagnostic categorization (medical and surgical), other hematological factors (such as lymphocytosis and thrombocytosis), catheter-related factors, exposure to blood products, total parenteral nutrition and overall mortality in the analysis of Pearson's chi-squared test with Yates' continuity correction as statistical significance.     Table 3 shows the microbial etiology in patients with CLABSI. We observed that 34.3% (11) of the samples tested were identified as gram-negative organisms, 21.8% (7) were identified as gram-positive organisms, 6.3% (2) were due to Candida albicans and 37.5% (12) were identified as polymicrobial growth based on blood culture. Of the five Klebsiella pneumonia isolates, three were multidrug-resistant and two were sensitive to carbapenem and extended-spectrum beta-lactamase (ESBL). Out of the four Acinetobacter baumani isolates, two were multidrug-resistant and two were sensitive to carbapenem and ESBL. Both Burkholderia cepacia and Salmonella were sensitive to carbapenem and ESBL. Among the four coagulasenegative staphylococci isolates, three were methicillin-resistant and one was methicillin-sensitive, whereas both isolates of Staphylococcus aureus were methicillin-resistant. Among the Candida albicans, one was fluconazole-resistant and one was fluconazole sensitive.   [20]. We also observed a female predominance of CLABSI cases with no statistical significance in contrast to the results reported in previous studies [3,4,9,10,20]. We observed an overall mortality of 12.5% in patients with CLABSI compared to the mortality of 12.9% to 56% reported in patients with CLABSI in previous studies [10][11][12][13]. This may be attributed to the lack of analyses based on the severity of the disease. Among the four deaths in patients with CLABSI, two had a PRISM III score of > 15 and another two had a PRISM III score of ≤ 15, whereas among the 17 deaths in patients without CLABSI, three had a PRISM III score of > 15 and 14 had PRISM III score of ≤ 15. Our analysis showed that factors such as patients in the age group of 2-12 months, high PRISM III score of greater than 15, leukocytosis, neutrophilia, anemia, change of central venous catheter, duration of central venous catheterization for more than seven days, exposure to blood products during the hospital stay, use of steroids, inotropes and prophylactic antibiotics were significantly associated with increased risk of central line-associated bloodstream infections (CLABSIs) as shown in Table 2. The reported factors such as age, change of central venous catheter, and duration of catheterization were consistent with the previous reports [10][11][12][13]. In our study, the central venous lines in 11 patients were changed after seven days of catheterization; this was found to be significantly associated with CLABSI. However, further studies addressing the change of central venous catheters before and after seven days were needed to validate these findings. Further, a significant association of higher PRISM III scores indicates an association with the increased severity of the disease. Exposure to blood products during the hospital stay, exposure to steroids, and exposure to previous antibiotics were found to be significantly associated with the risk of CLABSI and our results were consistent with the previous reports [12]. However, the significant association between CLABSI and the exposure to vasoactive agents (inotropic agents) reported in our study was contrary to the previous reports [12]. Moreover, factors such as the use of steroids, inotropes, and prophylactic antibiotics were significantly associated with CLABSI when adjusted for the PRISM III scores (severity of disease). However, exposure to blood products was not significantly associated with CLABSI when adjusted for PRISM III scores. Further, factors such as patients in the age group of greater than 12 months and patients with low PRISM III score of ≤ 15 were significantly associated with reduced risk of central line-associated bloodstream infections (CLABSIs) by 88.44% and 69.14% respectively as shown in Table 2. The observed major site of insertion was the internal jugular vein in our study which is consistent with the previous studies [21,22] and contradictory to those reported by Tomar et al [11].
Overall, we observed that Klebsiella pneumonia was the most common organism causing CLABSI in our study which is consistent with those results previously reported in India [10-13, 15, 20]. Further, we observed that the majority of CLABSIs were caused by gram-negative organisms in our study which are consistent with those reported in the previous studies [10][11][12][13][14][15]20]. However, some studies reported that gram-positive organisms were the most common etiology of CLABSI, which is contradictory to the results reported in our study [1,23,24]. Regarding antimicrobial susceptibility, we observed that overall 15.62% were multidrug-resistant bacteria with 18.75% susceptible to carbapenem and extended-spectrum betalactamase which is consistent with rates reported in previous studies [11,13]. We also observed that 50% of Staphylococcus aureus and 75% of coagulase-negative staphylococcus were methicillin-resistant and 50% of candida species were fluconazole-resistant, which is consistent with those reported in the previous studies [10,13].
Overall, the results observed in our study underscore the need for continuous surveillance and robust multidisciplinary prevention strategies such as providing bundle care approaches and antibiotic stewardship programs [25,26]. The bundle of care approach is an evolving multidisciplinary approach that focuses on prevention strategies such as primary prevention (neonate, staff, caretaker, environment, devices), secondary prevention (detection, screening, and epidemiological surveillance), tertiary prevention (antibiotic prescription, stewardships, outbreak controls) and implementation of the bundle care practices (by audit, feedback, awareness, organizing implementation team meetings, education, training, licensure standards) to promote and sustain the implementation of these practices in the hospitals [25,26]. Further, providing education to the health care workers regarding the antibiotic stewardship practices such as promoting appropriate selection of treatment regimens and prevention of unnecessary exposure to antibiotics might reduce the emergence of antibiotic resistance [27]. We also suggest that close adherence to the bundle of care practices regarding central venous catheter insertion and maintenance and education to healthcare workers about the maintenance of catheters might help in tackling the increased prevalence of CLABSIs [25,26]. Further, we also suggest that antibiotic optimization policy and guidance protocols based on regional data on antimicrobial susceptibility need to be regularly updated to combat the increased incidence of antimicrobial resistance [25][26][27].
Moreover, there are some limitations to be considered when interpreting our results such as small sample size, short duration of study, and lack of 28-day mortality rates and data regarding the care provided during central venous catheter insertion, access, and maintenance. However, our study provided a prospective and comprehensive analysis of factors associated with CLABSI and the profile of organisms in the pediatric intensive care unit in southern India when compared to previous studies. Further, prospective studies with large sample sizes and longer duration, and involving multiple centers are needed to validate these findings and generalize these results for clinical implementation.

Conclusions
In conclusion, our results revealed that factors such as patients in young age, high PRISM III scores, leukocytosis, neutrophilia, anemia, change of central venous catheter, duration of central venous catheterization for more than seven days, use of steroids, inotropes, and prophylactic antibiotics significantly increases the risk of CLABSI. However, factors such as being older than 12 months of age and having lower PRISM III scores play a protective role in significantly reducing the risk of CLABSIs. Moreover, long-term multicentric and interventional studies with large sample sizes were needed to confirm our findings. Further, robust analysis with additional factors based on the severity of the diseases was needed.

Appendices
The methods used for the statistical analysis in R language using R-statistical software for windows version 4.1.0 is shown in this appendix section. # To perform logistic regression analysis and odds ratio with 95% confidence intervals > logitmodel <-glm(clabsi~factors, family = 'binomial', data = data) > summary(logitmodel) # output for logistic regression analysis.

Disclosures
Human subjects: Consent was obtained or waived by all participants in this study. Amrita Institute of Medical Sciences, Kochi issued approval ECASM-AIMS-2021-346. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.