- Email: [email protected]
Central venous catheter repair is not associated with an increased risk of central line infection or colonization in intestinal failure pediatric patients
Journal of Pediatric Surgery 51 (2016) 395–397
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Central venous catheter repair is not associated with an increased risk of central line infection or colonization in intestinal failure pediatric patients☆ Claire McNiven a, Noah Switzer a,b, Melisssa Wood a, Rabin Persad a,c, Marie Hancock d, Sarah Forgie a,e, Bryan J. Dicken a,f,⁎ a
Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB T6G 2B7, Canada Department of Surgery, University of Alberta, Edmonton, AB T6G 2B7, Canada c Division of Pediatric Gastroenterology, University of Alberta, Edmonton, AB T6G 2B7, Canada d Alberta Health Services, Home TPN Program University of Alberta, Edmonton, Alberta, Canada e Division of Pediatric Infectious Disease, University of Alberta, Edmonton, AB T6G 2B7, Canada f Division of Pediatric General Surgery, University of Alberta, Edmonton, AB T6G 2B7, Canada b
a r t i c l e
i n f o
Article history: Received 19 March 2015 Received in revised form 31 July 2015 Accepted 2 August 2015 Key words: Central venous catheter Bloodstream infection Line repair Short bowel syndrome Pediatric
a b s t r a c t Purpose: The intestinal failure (IF) population is dependent upon central venous catheters (CVC) to maintain minimal energy requirements for growth. Central venous catheter infections (CVCI) are frequent and an independent predictor of intestinal failure associated liver disease. A common complication in children with longterm CVC is the risk of line breakage. Given the often-limited usable vascular access sites in this population, it has been the standard of practice to perform repair of the broken line. Although widely practiced, it is unknown if this practice is associated with increased line colonization rates and subsequent line loss. Methods: A retrospective review of our institutional IF population over the past 8 years (2006–2014) was performed. Utilizing a prospectively constructed database, all pediatric patients (n = 13, ages 0–17 years) with CVC dependency enrolled in the Children's Intestinal Rehabilitation Program with IF were included who underwent a repair and/or replacement procedure of their line. The control replacement group was CVCs that were replaced without being repaired (36), the experimental repair group was CVCs that were repaired (8). The primary outcome of interest was the mean number of days in each group from the intervention (replacement or repair) to line infection/colonization. Mann–Whitney tests for significance were performed with p-values b 0.05 being the threshold value for significance. Results: There were no catheter repair associated CVCI. The mean number of days from the replacement or repair of a CVC to its removal owing to infection/colonization was 210.0 and 162.8 days respectively. There was no statistically significant difference between these groups in time to removal owing to line infection (p = 0.55). Conclusion: Repair of central venous catheters in the pediatric population with intestinal failure does not lead to an increased rate of central venous catheter infection and should be performed when possible. © 2016 Elsevier Inc. All rights reserved.
Intestinal failure (IF) is the dependence upon parenteral nutrition to maintain minimal energy requirements for growth and development. Short bowel syndrome (SBS) is a form of intestinal failure and can be defined as residual small bowel length b25% predicted by gestational age or by the dependence on total parental nutrition (TPN) for more than 42 days [1,2]. Common etiologies of SBS include gastroschisis, necrotizing enterocolitis, midgut volvulus, extensive aganglionosis, and intestinal atresia [3,4]. The overall incidence rate of SBS is 1200 per 100,000 live births [5]. An unrecoverable condition 50 years ago, owing to ☆ The authors have no conflicts of interests to disclose. ⁎ Corresponding author at: Stollery Children's Hospital, University of Alberta Hospital, 2C3.44 WMC, 8440-112 Street, Edmonton, Alberta T6G 2B7, Canada. E-mail address: [email protected] (B.J. Dicken). http://dx.doi.org/10.1016/j.jpedsurg.2015.08.003 0022-3468/© 2016 Elsevier Inc. All rights reserved.
advancements in nutritional delivery, 80–94% of patients now survive owing to reliance on central venous catheters (CVC) for the delivery of total parenteral nutrition (TPN) [6]. Central venous catheter infections (CVCI) are a feared complication of long term parenteral access and contribute substantially to the overall morbidity and mortality of SBS patients. The reported mean CVCI rates in pediatric patients with SBS/IF ranges from 2.8 per 1000 catheter days to 13.6 per 1000 catheter days [7]. A common complication in children with long-term CVL is the risk of line breakage, with the theoretical risk of increased catheter colonization and infection, with resultant loss of the line. However, given the often limited available vascular access sites in the complex SBS/IF population, it has been the standard of practice across North America to perform repair of the broken line under sterile technique with the aim of
396
C. McNiven et al. / Journal of Pediatric Surgery 51 (2016) 395–397
preserving line access. Although widely practiced, it is unknown if this practice is associated with increased line colonization rates and subsequent line loss. New infection control standards at the University of Alberta are aimed at eliminating the practice of fixing potentially reparable lines for fear of increased infection rates, with limited supporting data from the literature. The purpose of this study is to identify if repair compared to replacement of central venous lines in the pediatric short bowel population leads to an increased risk of central venous catheter infection.
1. Methods 1.1. Study design This study is a retrospective review of prospectively collected data on intestinal failure population enrolled in the Children's Intestinal Rehabilitation Program (CHIRP) in Edmonton, Alberta, Canada from the years 2006–2014. The criterion for enrollment into CHIRP is short gut defined as residual small bowel length b25% predicted by gestational age. The study protocol included a replacement group (group 1) where lines were replaced immediately owing to irreparable mechanical break, or refractory sepsis. The repair group (group 2) had lines with mechanical breaks repaired where feasible. Central line care was undertaken by a nurse practitioner, and adhered to institutional protocols in both the repair and replacement groups. CVC infection control protocols have evolved over the study period to include infection control nursing, sterile dressing changes, chlorhexidine patches at exit sites, and alcohol wipes before and after each use. The institution's ethics board granted approval for this study [Re: Pro00048831].
1.2. Data recorded The primary outcome of interest was mean number of lines removed for infection within 30 days of repair. The definition of CVCI was previously described by Lundgren et al., as bacteremia in a patient with an indwelling CVC with a pathogen not related to an infection elsewhere. CVCIs within 30 days of a repair procedure, for the purpose of this study, are attributed to the repair procedure itself [8]. Secondary outcomes collected included mean line longevity, gestational age, gender, etiology of short gut, residual bowel length and type of microorganism.
1.3. Repair of CVC line technique Criteria for repair included any long-tem indwelling CVC with a mechanical fracture amenable to repair. Repairable catheter breaks require at least 3 cm of undamaged external catheter length from skin exit site to allow for repair and prevent retraction of line under the skin and at least 2.5 cm of undamaged length proximal to bifurcation or trifurcation of catheter. Repairs were carried out as previously described [8]. In brief, a glued repair was applied (Bard Access Systems, Salt Lake City, UT), following cutting of the catheter with sterile scissors approximately 1.5 cm proximal to the break, an extension bridge was applied with adhesive, and a plastic sheath slide into place to bridge the repair site. In addition, the site was further splinted with cut tongue depressors and secured to the infant for a period of 24 hours before use.
Table 1 Basic demographics of pediatric intestinal failure patients included in the study. Characteristic Gender Male Female Age at line insertion ± SD (range) Diagnosis Gastroschisis Necrotizing enterocolitis Volvulus Tufting enteropathy Other Residual bowel length ± SD (range)
Repair
Replacement
83.3% 16.7% 1.9 ± 1.9 years (0.4–6.1)
76.9% 23.1% 1.1 ± 1.1 years (0.0–5.0)
33.3% (2/6) 33.3% (2/6)
30.7% (4/13) 38.5% (5/13)
16.7% (1/6) 0.0% (0/6) 16.7% (1/6) 45.8 ± 17.3 cm (17.0–75.0)
15.4% (2/13) 7.7% (1/13) 7.7% (1/13) 43.4 ± 21.2 cm (17.0–65.0)
2. Results Thirteen patients were included in the study, 10 males (76.9%) and 3 females (23.1%). The average gestational age of these patients was 32.5 weeks and the mean birth weight 2010.5 grams. Etiologies of intestinal failure by group are presented in Table 1. All 13 patients had at least one CVC replaced and 6/13 (46.1%) patients had a CVC repaired. The mean number of lines per patient was 3.4. A total of 44 CVC lines were either replaced (36) or repaired (8). Of the replaced CVCs, 15 were replaced owing to irreparable mechanical break and 15 were replaced owing to line infection, despite appropriate antibiotic therapy. Six CVCs were replaced for other unspecified reasons. The repair group had 8 CVCs repaired owing to mechanical break, and then later replaced owing to irreparable mechanical break, line infection, or other unspecified reasons. Of the 8 repaired lines, none became infected within the 30 day postprocedure time period. The mean time from replacement to line infection (210.0 days) was not significantly different (p = 0.55) from the mean time from repair to line infection (162.8 days). The longevity of the CVCs was also analyzed, where longevity represents the average number of days from line placement to line replacement (owing to mechanical break, line infection and other unspecified removal events). The mean longevity of CVCs in the replacement group was 154.2 days, versus a mean longevity of 552.25 days for the repair group. The longevity of a repair group CVCs was obtained from the sum of days prerepair and postrepair to line replacement. The difference in mean longevity between the replacement and repair groups was statistically significant (p b 0.0001) (Table 2). The difference between the days to replacement of a CVC owing to irreparable mechanical break in the replacement group (114.4 days) and the days to repair of a CVC (333.5 days) was statistically significant (p = 0.002) (Fig. 1). A total of 19 lines were ultimately removed owing to infection (15/36, 41.7%) and repair (4/8, 50%). Organisms identified from patient cultures included Enterococcus faecalis (31.6%), Candida albicans (21.0%), Klebsiella oxytoca (21.0%), Klebsiella pneumoniae (10.5%), and Escherichia coli (10.5%). Other organisms, including Staphylococcus aureus (5.3%) and coagulase-negative staphylococci (5.3%), were isolated less frequently. 31.6% of the cultures contained more than one type of organism.
1.4. Statistical analysis Descriptive summaries of time to line replacement and repair, gestational age, gender, etiology of short gut, and residual bowel length were conducted in our series. Mean and standard deviations were used when appropriate. Mann–Whitney tests for significance were performed with p-values b 0.05 considered significant.
Table 2 Comparison of longevity of lines replaced and lines repaired (values in days). Replacement group Mechanical break Other
Repairgroup 114.4 114.0
Mechanical break Other
P value 341.7 74
0.002
C. McNiven et al. / Journal of Pediatric Surgery 51 (2016) 395–397
Mean Replace Mean Repair
397
114.4d 341.7d
333.5d
Line replacement Line repair d=days, 1cm = 60d Fig. 1. Time to mechanical break comparing CVC replacement and CVC repair.
3. Discussion This review was conducted to determine if the current practice of repairing broken CVCs was causing more infections in CVCs. Of the 8 CVCs that were repaired, none were replaced owing to line infection attributable to the repair itself, defined as infection within 30 days of the repair procedure. None of the CVCIs were attributed to the repair procedure, as defined by Lundgren et al. [8]. To determine if repaired CVCs were more susceptible to infection, days to line infection postrepair were compared to days to line infection in replaced CVCs. The difference between the mean time to line infection in the replacement group (210.0 days) and the repair group (162.8 days) was found to be not statistically significant (p = 0.55). This data suggests that repaired CVCs are equal to replaced CVCs with respect to risk of infection. In contrast to this study, Lundgren et al. reported an increase in CVCI rates after CVC repair in pediatric patients. However, this study included a heterogeneous cohort of immunocompromised, oncology and SBS patients, with SBS patients making up less than 20% of the retrospective cohort [8]. Interestingly, within their subgroup of pediatric patients with SBS they found no difference between the incidence of CVCI before and after CVC repair, a similar finding to this review of our pediatric SBS population. Repairing CVCs increases line longevity and preserves the use of limited vascular access sites. In this review the mean longevity of a repaired CVC was 552.25 days in comparison to only 154.2 days for a replaced CVC. Increasing the longevity of a CVC increases the longevity of the vascular access site, thereby reducing the number of vascular access sites compromised by CVC replacement. Pediatric IF patients with no usable vascular access sites ultimately need to be transplanted, so preservation of these sites is crucial [9]. Furthermore, we noted the time from first repair to irreparable mechanical break (341.7 days) was equal/exceeded the longevity of the primary CVC (333.5 days), and exceeded the life of a primary replaced CVC (114.4 days). Additional advantages of CVC repair include decreased exposure to general anesthetic, especially in light of recent concerns of developmental outcomes from multiple anesthetics [10,11]. This also equates to decreased technical complications of re-siting a CVC, risk of CVC dislodgement, cost, and repeated fluoroscopy exposure [12] must be considered. A systematic review of the current literature on central venous line repairs and infection rates was conducted to explore available literature to help guide this project. An electronic search of MEDLINE, Embase, Scopus, Web of Science, Dare, Clinical Evidence, TRIP, Health Technology Database, Conference abstracts, and the Cochrane Library databases was completed. Search terms included: “central venous line” or “catheter” AND “short gut” or “short bowel”. All human studies limited to English language and published before December of 2014 were included. A total of 1384 articles were identified using our search criteria for screening. After removing duplicates and limiting to pediatric studies, 128 articles were included for title screening, abstract or full text assessment. However once assessed using explicit criteria, only 7 papers were found that reported the incidence of line breakage with only 2 mentioning CVCI rates after repair or replacement. Further, all 7 studies had small patient populations, ranging from a total of 6 to 38 patients. This
highlights the current lack of data and evidence based recommendations in this area. Several limitations of this study exist. The first is the study design itself, being a retrospective review. However, based on these findings, it might not be ethical to perform a prospectively collected randomized controlled trial on this population. Furthermore, we report on a small population of pediatric IF patients with CVCs. However, although our patient population was small, it is congruent with similar studies in like subpopulations [8,13,14]. Another limitation of this review is the statistically significant difference between the mean time to mechanical break in the replacement group (114.4 days) and the mean time to repair in the repair group (333.5 days)(p = 0.001). It is possible that the lines that were repaired were already more resistant to infection because they were optimally located or better cared for. However, the replacement group also accounts for broken lines that could not be repaired owing to the location of the break in the CVC. Further analysis of specific data on the eligibility for repair may change this difference. 4. Conclusion Repaired CVCs were not associated with increased rate of line infection compared to replaced lines in the pediatric short bowel population. Based on this study, we recommend that primary repair, when possible, of CVCs in the SBS population should continue to be the standard of care. References [1] Wales PW. Surgical therapy for short bowel syndrome. Pediatr Surg Int 2004;20(9):647–57. [2] Dicken BJ, Sergi C, Rescoria FJ, et al. Medical management of motility disorders in patients with intestinal failure: a focus on necrotizing enterocolitis, gastroschisis, and intestinal atresia. J Pediatr Surg 2011;46(8):1618–30. [3] Sabharwal G, Strouse PJ, Islam S, et al. Congenital short-gut syndrome. Pediatr Radiol 2004;34(5):424–7. [4] Wales PW, Jancelewicz T, Romao RL, et al. Delayed primary serial transverse enteroplasty as a novel management strategy for infants with congenital ultrashort bowel syndrome. J Pediatr Surg 2013;48(5):993–9. [5] Bishop WP. Short Bowel Syndrome, in Pediatric Practice, Gastroenterology. New York: McGraw-Hill Medical; 2010. [6] Sigalet DL. Short bowel syndrome in infants and children: an overview. Semin Pediatr Surg 2001;10(2):49–55. [7] Wales PW, Kosar C, Carricato M, et al. Ethanol lock therapy to reduce the incidence of catheter-related bloodstream infections in home parenteral nutrition patients with intestinal failure: preliminary experience. J Pediatr Surg 2011;46(5):951–6. [8] Lundgren IS, Zhou C, Malone FR, et al. Central venous catheter repair is associated with an increased risk of bacteremia and central line-associated bloodstream infection in pediatric patients. Pediatr Infect Dis J 2012;31(4):337–40. [9] Kaufman SS, Atkinson JB, Bianchi A, et al. Indications for pediatric intestinal transplantation: a position paper of the American Society of Transplantation. Pediatr Transplant 2001;5(2):80–7. [10] Ing C, DiMaggio C, Whitehouse A, et al. Long-term differences in language and cognitive function after childhood exposure to anesthesia. Pediatrics 2012;130(3):e476–85. [11] Flick RP, Katusic SK, Colligan RC, et al. Cognitive and behavioral outcomes after early exposure to anesthesia and surgery. Pediatrics 2011;128(5):e1053–61. [12] Brenner D, Elliston C, Hall E, et al. Estimated risks of radiation-induced fatal cancer from pediatric CT. Am J Roentgenol 2001;176(2):289–96. [13] Gordon S, Gardiner S. Central Line Infections in Repaired Catheters: A Retrospective Review. J Assoc Vasc Access 2013;18(3):164–6. [14] Small M, Major GAD, Gabe SM. PMO-065 Repair of fractured central venous catheters for parenteral nutrition: techniques and risk of infection. Gut 2012;61(Suppl. 2) page A99.
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Central venous catheter repair is not associated with an increased risk of central line infection or colonization in intestinal failure pediatric patients☆ Claire McNiven a, Noah Switzer a,b, Melisssa Wood a, Rabin Persad a,c, Marie Hancock d, Sarah Forgie a,e, Bryan J. Dicken a,f,⁎ a
Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB T6G 2B7, Canada Department of Surgery, University of Alberta, Edmonton, AB T6G 2B7, Canada c Division of Pediatric Gastroenterology, University of Alberta, Edmonton, AB T6G 2B7, Canada d Alberta Health Services, Home TPN Program University of Alberta, Edmonton, Alberta, Canada e Division of Pediatric Infectious Disease, University of Alberta, Edmonton, AB T6G 2B7, Canada f Division of Pediatric General Surgery, University of Alberta, Edmonton, AB T6G 2B7, Canada b
a r t i c l e
i n f o
Article history: Received 19 March 2015 Received in revised form 31 July 2015 Accepted 2 August 2015 Key words: Central venous catheter Bloodstream infection Line repair Short bowel syndrome Pediatric
a b s t r a c t Purpose: The intestinal failure (IF) population is dependent upon central venous catheters (CVC) to maintain minimal energy requirements for growth. Central venous catheter infections (CVCI) are frequent and an independent predictor of intestinal failure associated liver disease. A common complication in children with longterm CVC is the risk of line breakage. Given the often-limited usable vascular access sites in this population, it has been the standard of practice to perform repair of the broken line. Although widely practiced, it is unknown if this practice is associated with increased line colonization rates and subsequent line loss. Methods: A retrospective review of our institutional IF population over the past 8 years (2006–2014) was performed. Utilizing a prospectively constructed database, all pediatric patients (n = 13, ages 0–17 years) with CVC dependency enrolled in the Children's Intestinal Rehabilitation Program with IF were included who underwent a repair and/or replacement procedure of their line. The control replacement group was CVCs that were replaced without being repaired (36), the experimental repair group was CVCs that were repaired (8). The primary outcome of interest was the mean number of days in each group from the intervention (replacement or repair) to line infection/colonization. Mann–Whitney tests for significance were performed with p-values b 0.05 being the threshold value for significance. Results: There were no catheter repair associated CVCI. The mean number of days from the replacement or repair of a CVC to its removal owing to infection/colonization was 210.0 and 162.8 days respectively. There was no statistically significant difference between these groups in time to removal owing to line infection (p = 0.55). Conclusion: Repair of central venous catheters in the pediatric population with intestinal failure does not lead to an increased rate of central venous catheter infection and should be performed when possible. © 2016 Elsevier Inc. All rights reserved.
Intestinal failure (IF) is the dependence upon parenteral nutrition to maintain minimal energy requirements for growth and development. Short bowel syndrome (SBS) is a form of intestinal failure and can be defined as residual small bowel length b25% predicted by gestational age or by the dependence on total parental nutrition (TPN) for more than 42 days [1,2]. Common etiologies of SBS include gastroschisis, necrotizing enterocolitis, midgut volvulus, extensive aganglionosis, and intestinal atresia [3,4]. The overall incidence rate of SBS is 1200 per 100,000 live births [5]. An unrecoverable condition 50 years ago, owing to ☆ The authors have no conflicts of interests to disclose. ⁎ Corresponding author at: Stollery Children's Hospital, University of Alberta Hospital, 2C3.44 WMC, 8440-112 Street, Edmonton, Alberta T6G 2B7, Canada. E-mail address: [email protected] (B.J. Dicken). http://dx.doi.org/10.1016/j.jpedsurg.2015.08.003 0022-3468/© 2016 Elsevier Inc. All rights reserved.
advancements in nutritional delivery, 80–94% of patients now survive owing to reliance on central venous catheters (CVC) for the delivery of total parenteral nutrition (TPN) [6]. Central venous catheter infections (CVCI) are a feared complication of long term parenteral access and contribute substantially to the overall morbidity and mortality of SBS patients. The reported mean CVCI rates in pediatric patients with SBS/IF ranges from 2.8 per 1000 catheter days to 13.6 per 1000 catheter days [7]. A common complication in children with long-term CVL is the risk of line breakage, with the theoretical risk of increased catheter colonization and infection, with resultant loss of the line. However, given the often limited available vascular access sites in the complex SBS/IF population, it has been the standard of practice across North America to perform repair of the broken line under sterile technique with the aim of
396
C. McNiven et al. / Journal of Pediatric Surgery 51 (2016) 395–397
preserving line access. Although widely practiced, it is unknown if this practice is associated with increased line colonization rates and subsequent line loss. New infection control standards at the University of Alberta are aimed at eliminating the practice of fixing potentially reparable lines for fear of increased infection rates, with limited supporting data from the literature. The purpose of this study is to identify if repair compared to replacement of central venous lines in the pediatric short bowel population leads to an increased risk of central venous catheter infection.
1. Methods 1.1. Study design This study is a retrospective review of prospectively collected data on intestinal failure population enrolled in the Children's Intestinal Rehabilitation Program (CHIRP) in Edmonton, Alberta, Canada from the years 2006–2014. The criterion for enrollment into CHIRP is short gut defined as residual small bowel length b25% predicted by gestational age. The study protocol included a replacement group (group 1) where lines were replaced immediately owing to irreparable mechanical break, or refractory sepsis. The repair group (group 2) had lines with mechanical breaks repaired where feasible. Central line care was undertaken by a nurse practitioner, and adhered to institutional protocols in both the repair and replacement groups. CVC infection control protocols have evolved over the study period to include infection control nursing, sterile dressing changes, chlorhexidine patches at exit sites, and alcohol wipes before and after each use. The institution's ethics board granted approval for this study [Re: Pro00048831].
1.2. Data recorded The primary outcome of interest was mean number of lines removed for infection within 30 days of repair. The definition of CVCI was previously described by Lundgren et al., as bacteremia in a patient with an indwelling CVC with a pathogen not related to an infection elsewhere. CVCIs within 30 days of a repair procedure, for the purpose of this study, are attributed to the repair procedure itself [8]. Secondary outcomes collected included mean line longevity, gestational age, gender, etiology of short gut, residual bowel length and type of microorganism.
1.3. Repair of CVC line technique Criteria for repair included any long-tem indwelling CVC with a mechanical fracture amenable to repair. Repairable catheter breaks require at least 3 cm of undamaged external catheter length from skin exit site to allow for repair and prevent retraction of line under the skin and at least 2.5 cm of undamaged length proximal to bifurcation or trifurcation of catheter. Repairs were carried out as previously described [8]. In brief, a glued repair was applied (Bard Access Systems, Salt Lake City, UT), following cutting of the catheter with sterile scissors approximately 1.5 cm proximal to the break, an extension bridge was applied with adhesive, and a plastic sheath slide into place to bridge the repair site. In addition, the site was further splinted with cut tongue depressors and secured to the infant for a period of 24 hours before use.
Table 1 Basic demographics of pediatric intestinal failure patients included in the study. Characteristic Gender Male Female Age at line insertion ± SD (range) Diagnosis Gastroschisis Necrotizing enterocolitis Volvulus Tufting enteropathy Other Residual bowel length ± SD (range)
Repair
Replacement
83.3% 16.7% 1.9 ± 1.9 years (0.4–6.1)
76.9% 23.1% 1.1 ± 1.1 years (0.0–5.0)
33.3% (2/6) 33.3% (2/6)
30.7% (4/13) 38.5% (5/13)
16.7% (1/6) 0.0% (0/6) 16.7% (1/6) 45.8 ± 17.3 cm (17.0–75.0)
15.4% (2/13) 7.7% (1/13) 7.7% (1/13) 43.4 ± 21.2 cm (17.0–65.0)
2. Results Thirteen patients were included in the study, 10 males (76.9%) and 3 females (23.1%). The average gestational age of these patients was 32.5 weeks and the mean birth weight 2010.5 grams. Etiologies of intestinal failure by group are presented in Table 1. All 13 patients had at least one CVC replaced and 6/13 (46.1%) patients had a CVC repaired. The mean number of lines per patient was 3.4. A total of 44 CVC lines were either replaced (36) or repaired (8). Of the replaced CVCs, 15 were replaced owing to irreparable mechanical break and 15 were replaced owing to line infection, despite appropriate antibiotic therapy. Six CVCs were replaced for other unspecified reasons. The repair group had 8 CVCs repaired owing to mechanical break, and then later replaced owing to irreparable mechanical break, line infection, or other unspecified reasons. Of the 8 repaired lines, none became infected within the 30 day postprocedure time period. The mean time from replacement to line infection (210.0 days) was not significantly different (p = 0.55) from the mean time from repair to line infection (162.8 days). The longevity of the CVCs was also analyzed, where longevity represents the average number of days from line placement to line replacement (owing to mechanical break, line infection and other unspecified removal events). The mean longevity of CVCs in the replacement group was 154.2 days, versus a mean longevity of 552.25 days for the repair group. The longevity of a repair group CVCs was obtained from the sum of days prerepair and postrepair to line replacement. The difference in mean longevity between the replacement and repair groups was statistically significant (p b 0.0001) (Table 2). The difference between the days to replacement of a CVC owing to irreparable mechanical break in the replacement group (114.4 days) and the days to repair of a CVC (333.5 days) was statistically significant (p = 0.002) (Fig. 1). A total of 19 lines were ultimately removed owing to infection (15/36, 41.7%) and repair (4/8, 50%). Organisms identified from patient cultures included Enterococcus faecalis (31.6%), Candida albicans (21.0%), Klebsiella oxytoca (21.0%), Klebsiella pneumoniae (10.5%), and Escherichia coli (10.5%). Other organisms, including Staphylococcus aureus (5.3%) and coagulase-negative staphylococci (5.3%), were isolated less frequently. 31.6% of the cultures contained more than one type of organism.
1.4. Statistical analysis Descriptive summaries of time to line replacement and repair, gestational age, gender, etiology of short gut, and residual bowel length were conducted in our series. Mean and standard deviations were used when appropriate. Mann–Whitney tests for significance were performed with p-values b 0.05 considered significant.
Table 2 Comparison of longevity of lines replaced and lines repaired (values in days). Replacement group Mechanical break Other
Repairgroup 114.4 114.0
Mechanical break Other
P value 341.7 74
0.002
C. McNiven et al. / Journal of Pediatric Surgery 51 (2016) 395–397
Mean Replace Mean Repair
397
114.4d 341.7d
333.5d
Line replacement Line repair d=days, 1cm = 60d Fig. 1. Time to mechanical break comparing CVC replacement and CVC repair.
3. Discussion This review was conducted to determine if the current practice of repairing broken CVCs was causing more infections in CVCs. Of the 8 CVCs that were repaired, none were replaced owing to line infection attributable to the repair itself, defined as infection within 30 days of the repair procedure. None of the CVCIs were attributed to the repair procedure, as defined by Lundgren et al. [8]. To determine if repaired CVCs were more susceptible to infection, days to line infection postrepair were compared to days to line infection in replaced CVCs. The difference between the mean time to line infection in the replacement group (210.0 days) and the repair group (162.8 days) was found to be not statistically significant (p = 0.55). This data suggests that repaired CVCs are equal to replaced CVCs with respect to risk of infection. In contrast to this study, Lundgren et al. reported an increase in CVCI rates after CVC repair in pediatric patients. However, this study included a heterogeneous cohort of immunocompromised, oncology and SBS patients, with SBS patients making up less than 20% of the retrospective cohort [8]. Interestingly, within their subgroup of pediatric patients with SBS they found no difference between the incidence of CVCI before and after CVC repair, a similar finding to this review of our pediatric SBS population. Repairing CVCs increases line longevity and preserves the use of limited vascular access sites. In this review the mean longevity of a repaired CVC was 552.25 days in comparison to only 154.2 days for a replaced CVC. Increasing the longevity of a CVC increases the longevity of the vascular access site, thereby reducing the number of vascular access sites compromised by CVC replacement. Pediatric IF patients with no usable vascular access sites ultimately need to be transplanted, so preservation of these sites is crucial [9]. Furthermore, we noted the time from first repair to irreparable mechanical break (341.7 days) was equal/exceeded the longevity of the primary CVC (333.5 days), and exceeded the life of a primary replaced CVC (114.4 days). Additional advantages of CVC repair include decreased exposure to general anesthetic, especially in light of recent concerns of developmental outcomes from multiple anesthetics [10,11]. This also equates to decreased technical complications of re-siting a CVC, risk of CVC dislodgement, cost, and repeated fluoroscopy exposure [12] must be considered. A systematic review of the current literature on central venous line repairs and infection rates was conducted to explore available literature to help guide this project. An electronic search of MEDLINE, Embase, Scopus, Web of Science, Dare, Clinical Evidence, TRIP, Health Technology Database, Conference abstracts, and the Cochrane Library databases was completed. Search terms included: “central venous line” or “catheter” AND “short gut” or “short bowel”. All human studies limited to English language and published before December of 2014 were included. A total of 1384 articles were identified using our search criteria for screening. After removing duplicates and limiting to pediatric studies, 128 articles were included for title screening, abstract or full text assessment. However once assessed using explicit criteria, only 7 papers were found that reported the incidence of line breakage with only 2 mentioning CVCI rates after repair or replacement. Further, all 7 studies had small patient populations, ranging from a total of 6 to 38 patients. This
highlights the current lack of data and evidence based recommendations in this area. Several limitations of this study exist. The first is the study design itself, being a retrospective review. However, based on these findings, it might not be ethical to perform a prospectively collected randomized controlled trial on this population. Furthermore, we report on a small population of pediatric IF patients with CVCs. However, although our patient population was small, it is congruent with similar studies in like subpopulations [8,13,14]. Another limitation of this review is the statistically significant difference between the mean time to mechanical break in the replacement group (114.4 days) and the mean time to repair in the repair group (333.5 days)(p = 0.001). It is possible that the lines that were repaired were already more resistant to infection because they were optimally located or better cared for. However, the replacement group also accounts for broken lines that could not be repaired owing to the location of the break in the CVC. Further analysis of specific data on the eligibility for repair may change this difference. 4. Conclusion Repaired CVCs were not associated with increased rate of line infection compared to replaced lines in the pediatric short bowel population. Based on this study, we recommend that primary repair, when possible, of CVCs in the SBS population should continue to be the standard of care. References [1] Wales PW. Surgical therapy for short bowel syndrome. Pediatr Surg Int 2004;20(9):647–57. [2] Dicken BJ, Sergi C, Rescoria FJ, et al. Medical management of motility disorders in patients with intestinal failure: a focus on necrotizing enterocolitis, gastroschisis, and intestinal atresia. J Pediatr Surg 2011;46(8):1618–30. [3] Sabharwal G, Strouse PJ, Islam S, et al. Congenital short-gut syndrome. Pediatr Radiol 2004;34(5):424–7. [4] Wales PW, Jancelewicz T, Romao RL, et al. Delayed primary serial transverse enteroplasty as a novel management strategy for infants with congenital ultrashort bowel syndrome. J Pediatr Surg 2013;48(5):993–9. [5] Bishop WP. Short Bowel Syndrome, in Pediatric Practice, Gastroenterology. New York: McGraw-Hill Medical; 2010. [6] Sigalet DL. Short bowel syndrome in infants and children: an overview. Semin Pediatr Surg 2001;10(2):49–55. [7] Wales PW, Kosar C, Carricato M, et al. Ethanol lock therapy to reduce the incidence of catheter-related bloodstream infections in home parenteral nutrition patients with intestinal failure: preliminary experience. J Pediatr Surg 2011;46(5):951–6. [8] Lundgren IS, Zhou C, Malone FR, et al. Central venous catheter repair is associated with an increased risk of bacteremia and central line-associated bloodstream infection in pediatric patients. Pediatr Infect Dis J 2012;31(4):337–40. [9] Kaufman SS, Atkinson JB, Bianchi A, et al. Indications for pediatric intestinal transplantation: a position paper of the American Society of Transplantation. Pediatr Transplant 2001;5(2):80–7. [10] Ing C, DiMaggio C, Whitehouse A, et al. Long-term differences in language and cognitive function after childhood exposure to anesthesia. Pediatrics 2012;130(3):e476–85. [11] Flick RP, Katusic SK, Colligan RC, et al. Cognitive and behavioral outcomes after early exposure to anesthesia and surgery. Pediatrics 2011;128(5):e1053–61. [12] Brenner D, Elliston C, Hall E, et al. Estimated risks of radiation-induced fatal cancer from pediatric CT. Am J Roentgenol 2001;176(2):289–96. [13] Gordon S, Gardiner S. Central Line Infections in Repaired Catheters: A Retrospective Review. J Assoc Vasc Access 2013;18(3):164–6. [14] Small M, Major GAD, Gabe SM. PMO-065 Repair of fractured central venous catheters for parenteral nutrition: techniques and risk of infection. Gut 2012;61(Suppl. 2) page A99.