Infections and colonisation of central venous catheters in patients admitted to intensive care units at two tertiary care hospitals Demographic and clinical characteristics of patients with CRBSI

Introduction: Central venous catheters (CVC) are frequently used in modern health care systems. However, CVCs are likely to get colonised with microorganisms, resulting in catheter-related blood stream infections (CRBSI). This study investigated microorganisms causing CRBSI and CVC colonisation, their antibiotic resistance, and factors associated with CRBSI in Methods: The study included 300 adult patients in ICUs with a CVC in-situ for >48 hours. Blood taken through the CVC, peripheral blood and CVC tips were cultured. Microorganisms were identified and tested for antibiotic susceptibility. Demographic factors were recorded. Results: Seventeen patients (13.1%) developed CRBSI. The CRBSI rate was 13.3 per 1,000 catheter days in CNTH while it was 10 (5.9%) cases and 5.2 per 1,000 catheter days in Apeksha hospital. In CNTH, CVC colonisation was detected in 35 (26.9%). Coagulase negative staphylococci (CoNS) were the leading cause of both CRBSI (41%) and colonisation (58%). In Apeksha hospital, Klebsiella sp. were the predominant pathogens that caused CRBSI. Eighty-nine CVCs (52.3%) were colonised by CoNS (58%). In CNTH, CRBSI was detected more in males (14/17) (p<0.05). The majority of microorganisms that caused CRBSI and CVC colonisation in both hospitals showed resistance to commonly used antibiotics. Conclusion: The CRBSI rates and the incidences were higher in CNTH. CoNS was the most common cause of CRBSI in CNTH and Gram negative bacteria in Apeksha hospital. CVC colonisation with CoNS was common in both hospitals. Antibiotic resistance was high among bacteria causing CRBSI and colonisation.


Introduction
In the past few decades, the use of intravascular catheters, including central venous catheters (CVC) have become common practices in health care settings. These CVCs are used to administer chemotherapeutic drugs, blood products, nutrients, and antibiotics as well as for hemodialysis procedures. 1 CVCs can be a predisposing factor of catheter-related bloodstream infections (CRBSI) which lead to hospital acquired bacteremia and sepsis. CRBSIs are associated with high morbidity and mortality rates as well as increased healthcare costs. 2 Microorganisms which are normal flora of the skin such as Staphylococcus aureus and Staphylococcus epidermidis are common aetiological agents of CRBSI. 1,3 However, some Gramnegative bacteria, such as Escherichia coli and Klebsiella species have also been reported, especially in patients with malignancies. 3,4 Studies have found some risk factors associated with CRBSI such as old age, prolonged catheterisation, co-existing diseases such as head and neck cancer, immunosuppression, neutropaenia, diabetes, malnutrition, and chronic use of steroids. 4 Early diagnosis and proper treatment plans are the key to successful management of CRBSI. Information regarding the causative organisms and their antimicrobial susceptibility patterns in the local health care settings are important for timely initiation of appropriate empirical therapy. Identification of risk factors help not only to determine the most effective patient management plan in the case of development of CRBSI, but also to recognise better prevention methods. There is a lack of available information regarding CRBSI, its etiological agents and their susceptibility patterns in south asian healthcare facilities. Studies conducted on CRBSI will be most useful for developing and updating treatment and infection prevention guidelines in this context, especially in the ICU setting.
The current study was conducted to investigate the incidence of CRBSI and CVC colonisation with causative microorganisms and their antibiotic susceptibility, along with factors associated with the development of CRBSI in ICU patients at two tertiary care hospitals, CNTH, Ragama and Apeksha hospital, Maharagama, Sri Lanka.

Study population
This descriptive cross-sectional study was conducted over a 10-month period from November 2018 to August 2019 at CNTH and Apeksha hospital, Maharagama, Sri Lanka. Both male and female adult (>18 years) patients admitted to the ICUs who had a central venous catheter (CVC) in-situ for more than 48 hours were included in this study.

Definitions and calculations
CRBSI was identified according to the following standard definitions. 2 1. A shorter time to positive culture (>2 hours earlier) with the same organism in a central line blood sample than a peripheral blood sample drawn at the same time. 2. When both positive peripheral blood cultures (one or more) and catheter tip culture isolate the same microbial species Catheter colonisation was determined by the presence of a significant growth of microorganisms using a semi-quantitative culture ≥15 colony forming units (CFU)] using a 4cm segment from the tip of the catheter or a positive culture from blood drawn from the catheter and a negative peripheral blood culture. 2 The occurrence of CRBSI was calculated as incidence per 1000 catheter-days. 5

Patients' demographics and clinical data
Clinical data including duration of the catheterisation, type of catheter, catheter insertion site, underlying diseases, type of malignancy (haematological or nonhaematological), neutrophil status (in cancer patients), days of ICU stay and demographic data including gender and age were collected from each patient and analysed to identify the associated factors for developing CRBSI.

Microbiological analysis
Peripheral blood and line blood from CVC were processed according to standard microbiological methods 6 and CVC tips were cultured according to Maki's roll over plate method. 7 The isolated microorganisms were identified using biochemical tests and commercially available kits (API ® Staph and API ® 20E-Analytical Profile Index System; bioMérieux, Paris, France).
Antibiotic susceptibility testing was carried out by disk diffusion method according to the Clinical and Laboratory Standard Institute (CLSI -2016) methodology. 8

Statistical analysis
Data were analysed by a computer-based programme, statistical package for social sciences (SPSS V26). All p values were based on 2-tailed tests and a p value of less than 0.05 was considered significant (p<0.05).

Results
Three hundred (300) patients, 130 from CNTH and 170 from Apeksha hospital were included in the study.

Demographic and clinical characteristics of patients with CRBSI
In CNTH the mean age of patients was 48.3 (±16.9) years. Fourteen of 67 males and 3 of 63 females developed CRBSI which was statistically significant (P=0.006). The most frequent comorbid conditions included diabetes in 02 (11.7%) patients. The jugular vein (n=15; 88.2%) was the most common site of CVC insertion. Local complications were seen in 2 (11.7%) patients. Mean duration to develop CRBSI was 6.5 days after catheterisation (Table 1).
Patients in Apeksha hospital developed CRBSI after a mean duration of 18.9 days following CVC insertion. The patient's age, gender, presence of diabetes mellitus, type of malignancy, neutropenia, location, or the type of catheter did not have a statistically significant association with development of CRBSI (Table 1).

Causative organisms and antibiotic susceptibility
In CNTH, coagulase negative Staphylococcus spp (CoNS) was the leading cause of both CRBSI (n=7; 41%) and CVC colonisation (n=20;58%). Pseudomonas spp. (n=2; 12%), Acinetobacter spp. (n=2;12%), diphtheroids (n=2; 12%) were isolated from patients with CRBSI followed by other Gram-negative organisms and Candida spp. The second leading cause of CVC colonisation was Pseudomonas spp. (n=6; 17%) followed by other organisms as shown in Table 2. Two specimens revealed polymicrobial colonisation, Staphylococcus aureus with Pseudomonas spp. and Acinetobacter spp. with CoNS. The majority of isolates from patients with CRBSI showed resistance to several antibiotics.  Cloxacillin and clindamycin resistance rates in CoNS isolated from the CRBSI patients in CNTH were 71% and 57% respectively (Table 3). Among the CoNS that caused CVC colonisation in Apeksha hospital, 83% showed resistance to erythromycin and 66% to clindamycin. All Acinetobacter strains (n=5) isolated in CNTH were resistant to cefoperazone-sulbactam (Table  4). In addition, Pseudomonas spp. isolated from the patients in CNTH showed higher resistance to the antimicrobial agents than isolates from the patients in Apeksha hospital.   Table 6.
Among CoNS isolates, no resistance was detected to linezolid and vancomycin.

Discussion
CVCs are significant in the care of critically ill patients. However, the use of CVCs is associated with severe complications, especially CRBSI. It is one of the most serious hospital acquired infections in the ICU setting, with increased rates of morbidity and mortality. 16 Colonisation of the CVC tip is considered to be a pre-requisite for CRBSI. 2   9 reported the incidence of CRBSI to be 14.59/1000 catheter days, which was higher than that of this study. Meanwhile, almost a similar CRBSI rate (13.64/1000 catheter days) was reported by Rao in 2018. 10 The mean age of patients with CRBSI in CNTH and Apeksha hospitals were 48.3 and 47.5 years respectively, which is lower than in the study done by Rao 10 where the mean age was 51.25 years. The study group in CNTH showed CRBSI is more common in males (14/17) than in females (3/17). However, there was no significant gender difference in other studies. 9,10 Hospital acquired infections are a significant problem, particularly in patients with malignancies.
In the study group at Apeksha hospital, the CRBSI rate was 5.9 with an incidence of 5.2 per 1000 catheter days. Agrawal et al. 13 reported a rate of 2.1 per 1000 CVC days which was less than the present study in patients at medical and surgical oncology departments.
When comparing the two hospitals, the percentage of CRBSI cases at CNTH was higher than at Apeksha hospital (P=0.020). This may be due to several reasons. Cancer patients, who are immunocompromised are treated more often with antibiotics to prevent infections than other patients. It is also possible that infection control procedures at Apeksha Hospital contributed to the lower CRBSI rate. Oberoi et al. 15 suggested this irregularity of the incidence of CRBSI in various studies could be due to factors such as site of catheterisation, insertion techniques used, type(s) of catheter used, catheter care techniques and diagnostic criteria used to diagnose CRBSI.
CoNS were the most common organisms causing CRBSI in CNTH (41.17%). This was consistent with the findings of Patil et al. 16 in a study conducted on patients admitted to the medical intensive care unit, where CoNS accounted for 65% of CRBSI. In contrast to the above findings, Kumari et al. and Malek et al. found that the majority of organisms responsible for CRBSIs in medical/surgical ICUs at a tertiary care hospital and medical/coronary and surgical ICUs at a private hospital respectively were Gram-negative bacteria. 11,17 In addition to the bacterial pathogens, Candida spp also contributed to 6% (n= 1) of the CRBSI cases in the CNTH study population. This is consistent with the findings of Rao. 10 Gram negative bacilli were the predominant microorganisms that cause CRBSI in the study population at Apeksha hospital. However, CVC colonisation was caused mainly by Gram positive cocci. This finding is consistent with the findings of Gupta et al. 18 which stated that the major cause of CRBSI was Pseudomonas aeruginosa while the most common coloniser was Staphylococcus aureus. As Staphylococcus spp. are skin commensals, they can cause a high proportion of catheter colonisation but in this susceptible group of patients, Gram negative bacilli seem to be the more important invasive pathogens.
During the last two decades, an epidemiological shift has taken place in CRBSI among cancer patients. Overall, CVCs became a less frequent source of blood stream infections (BSI) and microorganisms that cause BSI and CRBSI have moved from Gram positive bacteria toward Gram-negative bacteria. 3 Organisms responsible for CRBSI and colonisation in both hospitals showed high resistance rates to a number of commonly used antibiotics which is consistent with other similar studies. 9,10 In this study, all the isolated Acinetobacter strains from CNTH showed a very high rate of multidrug resistance which is a major cause for worry as Acinetobacter species are very frequently isolated from patients in Sri Lankan ICUs. 14 Antibiotic susceptibility patterns of bacteria show a significant variation in different geographical areas depending on many factors. 18 Data from this study show the gravity of the problem in Sri Lankan hospitals, especially due to the high resistance rates of Gram negative bacilli to available broad-spectrum life-saving antibiotics.
The presence of diabetes was found to be a significant factor in some studies 9,12 but in this study, there was no statistically significant association in both hospitals. Neutropenia is also reported as an important risk factor associated with CRBSI in cancer patients 13 , because when the absolute neutrophil count is <500/µL, immune control of the microbial flora, including pathogens, become insufficient. 19 The current study however did not find any statistically significant correlation between neutropenia and CRBSI in cancer patients. This result is consistent with those of Cairo et al. 2011 20 who reported that neutropenia was mainly seen in patients with non-CRBSI.
The majority of CRBSI are preventable. Implementation of educational programmes and the adhesion to infection prevention practices as hand washing, cleaning the insertion site with chlorhexidine, using maximum barrier precautions when inserting CVC and avoiding the femoral insertion in adult patients if possible, can be used to minimise CRBSI. 21 The hospital infection control manual of the Sri Lanka College of Microbiologists 22, states that CVC care should include use of 2%-4% chlorhexidine for hand washing and insertion site cleaning, adherence to strict aseptic precautions during the insertion, observation of insertion site for evidence of infection daily, and removal of CVC as soon as no longer required. However, in the present study, information on adherence to these guidelines was not obtained.

Conclusion
CRBSI rate is higher in CNTH compared to Apeksha hospital. CoNS are the main cause of CRBSI in CNTH and Gram negative bacilli in Apeksha hospital. CVC colonisation with CoNS is common in both hospitals. Males are significantly associated with CRBSI in the CNTH study population. Antibiotic resistance is extremely high among bacteria causing colonisation and CRBSI, which necessitates urgent action to curtail the problem.