Catheter-Related Bloodstream Infections in Children with Short Bowel Syndrome: A Single Center Retrospective Study

Background: Pediatric short bowel syndrome (SBS) is a devastating clinical entity that is usually managed with parenteral nutrition (PN). Catheter-related bloodstream infection (CRBSI) is a serious complication of long-term PN. The aim of this study was to evaluate the incidence and risk factors of CRBSI in SBS children. Methods: Nineteen pediatric patients with SBS were retrospectively recruited from Shanghai Children’s Hospital between August 2015 and December 2019. Clinical data, including demographics, aetiology of SBS, site and type of catheters, duration of PN, comorbidities, and microbiological data regarding CRBSI, were collected and reviewed to assess the CRBSI incidence and associated risk factors. Results: A total of 57 catheter insertions for PN were administered in 19 SBS children. CRBSI was identied in 11 cases among 6 patients during a total catheter days of 1907. The rate of CRBSI was 5.8 per 1000 catheter days. The common manifestations of the SBS children with CRBSI were fever (6/11, 54.5%), lethargy (5/11, 45.4%), stagnation of weight (7/11, 63.6%), and increment of enterostomy uid (7/11, 63.6%). Potential risk factors for CRBSI included an absence of ileocecal valve (p=0.046), and long duration of PN (p=0.019). The most common microorganism isolated from the blood cultures was Klebsiella pneumoniae (6/11, 54.55%). Conclusions: The rate of CRBSI was 5.8 per 1000 catheter days, and klebsiella pneumoniae was the most common isolated pathogen of CRBSI in the studied SBS children. An absence of ileocecal valve and long duration of PN were potential risk factors of CRBSI. potential risk factors of CRBSI. However, the results need to be further conrmed by a multi-centre study with a large cohort.


Introduction
Short bowel syndrome (SBS) is a severe malabsorptive condition resulting from a reduced length of the small intestine [1]. Pediatric SBS is de ned as a need for parental nutrition (PN) more than 42 days after intestinal resection or a bowel length of less than 25% as expected [2,3]. The most common etiologies are necrotizing enterocolitis (NEC) and congenital abnormalities in pediatric SBS [4]. The children with SBS experience diarrhea, malabsorption, growth stunting, and malnutrition as a result of loss of intestinal surface area. The intestinal function is inadequate to maintain the minimum necessary for the absorption of macronutrients, uid, and electrolytes, therefore, PN is required to keep up health and/or growth in children with SBS [2]. Intravascular catheters are integral to deliver the nutrition to the patients with SBS. However, the use of catheters is often complicated by mechanical or infectious complications, which can be life-threatening to the patients [5]. Catheter-related bloodstream infection (CRBSI) is a serious complication of long-term PN, which can lead to loss of vascular access, advancing intestinal failure-associated liver disease, sepsis, and death [6,7]. Children with SBS are at high risk of CRBSI. Previous studies have reported that the incidence of CRBSI among children with SBS and intestinal failure ranges from 0.9 to 11.5 per 1000 catheter days [8][9][10][11]. The incidence of CRBSI in children are associated with impaired gut permeability, altered gut microbiota, immune system dysfunction, comorbidities, the presence of an indwelling central venous catheter (CVC), and prolonged PN [12,13].
In this study, we conducted a single-center retrospectively study to evaluate the incidence and risk factors of CRBSI in SBS children with PN.

Study population
Nineteen children with SBS from Shanghai Children's Hospital, China, between August 2015 and December 2019, were retrospectively enrolled to the study cohort. The inclusion criteria were pediatric SBS patients due to a primary intestinal disease who were received PN for more than 42 days. Patients without a suspected CRBSI were excluded. This study was in compliance with the Helsinki Declaration and was approved by the Ethical Review Board of Shanghai Children's Hospital.

Data collection
Clinical data, including demographics, aetiology of SBS, site and type of catheters, duration of PN, comorbidities, laboratory ndings, and microbiological data regarding CRBSI, were collected and reviewed to assess the CRBSI incidence and associated risk factors. CRBSI was identi ed by positive blood culture drawn from any site in a symptomatic SBS patient receiving PN with suspected bloodstream infection [5].

Statistical analysis
Statistical analysis was performed by SPSS 20.0 software. Descriptive analyses were performed to catalog the characteristics of the patients and catheters as number with percentage. Quantitative data were summarized as medians and interquartile range (IQR 25th-75th), and compared using the t test or χ 2 test. A p value of <0.05 was considered to be statistically signi cance.

Results
As shown in Table 1  The results of the univariate analysis showed that there were no signi cant differences in catheter site, catheter type, or the catheter repair or placement in past 7 days between children with CRBSI and those without CRBSI (Table 2). We found that an absence of ileocecal valve (p = 0.046), and long-term duration of PN prior to CRBSI (p = 0.019) was positively associated with an increase risk of CRBSI, respectively. Furthermore, use of antibiotics, use of antacids at least 1 day that month of catheterization, percentage of calories from carbohydrate (CHO), fat, or protein was not associated with CRBSI (Table 2). However, multivariate analysis did not determine an absence of ileocecal valve, and long-term duration of PN as independent risk factors for CRBSI in our study cohort (Table 3).   Table 4, a total of 11 microorganisms grew in the harvested blood cultures, including 2 grampositive bacterial species (18.2%), 1 gram-negative bacterial species in 6 cases (54.5%), 1 fungal species in 2 cases (18.2%), and 1 case (9.1%) of co-infection with Staphylococcus aureus and Klebsiella pneumoniae. The most common microorganism cultured was Klebsiella pneumoniae (6/11, 54.55%). The catheters in patients with CRBSI were removed, and the patients were treated with antimicrobial therapy to clear the microorganisms.

Discussion
Children with SBS suffer from strikingly high rates of morbidity and mortality, due in part to their susceptibility to life-threatening microbial infections [4,14]. SBS children are at high risk for CRBSI and associated complications for their initial surgical management, malabsorption, immune system dysfunction, intestinal microecological changes, and their need for PN [12,13,15]. In this study, the rate of CRBSI in SBS children with long-term PN was 5.8 per 1000 catheter days, which was comparable with several previous studies [9,11,16]. The rate of CRBSI was 4.6 per 1000 catheter days in a study enrolled 16 children with intestinal failure for a total follow up of 233 months [16]. In another study from China, the incidence of CRBSI was 5.85 per 1000 catheter days in a group of infants with intestinal failure with PICC [9].
The most common symptoms at the time of diagnosis of CRBSI in children are fever, chills, and lethargy [16][17][18]. In the current study, not all SBS children had fever during CRBSI, which was different from previous study [17]. Some patients presented increasing enterostomy uid or stagnation of weight without fever when CRBSI happened. Therefore, early CRBSI should not be evaluated simply by fever. The change of enterostomy uid and stagnation of weight may be identi ed as indicators of CRBSI with a large cohort.
Potential risk factors for CRBSI in the current study included an absence of ileocecal valve, and long-term duration of PN. However, multivariate analysis did not determine an absence of ileocecal valve, or longterm duration of PN as independent risk factors for CRBSI. A previous study showed that use of doublelumen tunneled CVCs, jugular placement of CVC, higher doses of lipid emulsion, and use of antacids were