549999
research-article2014
PENXXX10.1177/0148607114549999Journal of Parenteral and Enteral NutritionDibb et al
Brief Communication
Central Venous Catheter Salvage in Home Parenteral Nutrition Catheter-Related Bloodstream Infections: Long-Term Safety and Efficacy Data
Journal of Parenteral and Enteral Nutrition Volume XX Number X Month 201X 1–7 © 2014 American Society for Parenteral and Enteral Nutrition DOI: 10.1177/0148607114549999 jpen.sagepub.com hosted at online.sagepub.com
Martyn J. Dibb, MB, ChB, MD, MRCP1; Arun Abraham, MB, BS, MRCS1; Paul R. Chadwick, MB, ChB, MD, MRCP, FRC(Path)2; Jon L. Shaffer, MB, ChB, MD, FRCP1; Antje Teubner, MD, MRCS1; Gordon L. Carlson, MB, ChB (Hons), MD, FRCS1; and Simon Lal, MB, ChB, PhD, FRCP1
Abstract Background: Catheter-related bloodstream infections (CRBSIs) are a serious complication in the provision of home parenteral nutrition (HPN). Antibiotic salvage of central venous catheters (CVCs) in CRBSI is recommended; however, this is based on limited reports. We assessed the efficacy of antibiotic salvage of CRBSIs in HPN patients. Materials and Methods: All confirmed CRBSIs occurring in patients receiving HPN in a national intestinal failure unit (IFU), between 1993 and 2011, were analyzed. A standardized protocol involving antibiotic and urokinase CVC locks and systemic antibiotics was used. Results: In total, 588 patients were identified with a total of 2134 HPN years, and 297 CRBSIs occurred in 137 patients (65 single and 72 multiple CRBSIs). The overall rate of CRBSI in all patients was 0.38 per 1000 catheter days. Most (87.9%) infections were attributable to a single microorganism. In total, 72.5% (180/248) of CRBSIs were salvaged when attempted (coagulase-negative staphylococcus, 79.8% [103/129], Staphylococcus aureus, 56.7% [17/30]; polymicrobial infections, 67.7% [21/30]; and miscellaneous, 66.1% [39/59]). CVC salvage was not attempted in 49 episodes because of life-threatening sepsis (n = 18), fungal infection (n = 7), catheter problems (n = 20), and CVC tunnel infection (n = 4). Overall, the CVC was removed in 33.7% (100/297) of cases. There were 5 deaths in patients admitted to the IFU for management of the CRBSI (2 severe sepsis at presentation, 3 metastatic infection). Conclusions: This is the largest reported series of catheter salvage in CRBSIs and demonstrates successful catheter salvage in most cases when using a standardized protocol. (JPEN J Parenter Enteral Nutr. XXXX;xx:xx-xx)
Keywords parenteral nutrition; nutrition; venous access; nutrition support teams; nutrition support practice; outcomes research/quality; sepsis; research and diseases
Clinical Relevancy Statement Antibiotic salvage of catheter-related bloodstream infections in home parenteral nutrition (HPN) patients is important in preserving venous access. The rate of successful salvage with different organisms is an important finding, and the safety of the protocol is also clinically relevant for clinicians supervising patients receiving HPN.
Introduction Home parenteral nutrition (HPN) delivered by a dedicated central venous catheter (CVC) can be a life-saving therapy for patients with intestinal failure (IF). Aseptic care of CVCs for patients with IF is mandatory, and strict adherence to catheter care protocols has been demonstrated to reduce the incidence of catheter-related bloodstream infection (CRBSI) and prolong CVC viability.1-4 CRBSI is a serious and common complication of long-term HPN, and CRBSI rates are a useful surrogate marker of the overall quality of HPN care. CRBSI rates have been variably reported
as 0.35–2.27 episodes/1000 catheter days.5-9 A number of risk factors for the development of infections have been identified, including type and characteristics of the infusion device, education and training in CVC care, nature of underlying disease and intestinal anatomy, opiate dependency, and the use of the CVC for obtaining blood samples and administration of medication.9-11 Coagulase-negative staphylococci (CNS) have been reported to be responsible for approximately 50% of episodes of CRBSI in a recent systematic review, with gram-negative and polymicrobial infections accounting for the majority of the remainder.10 From the 1Intestinal Failure Unit and 2Microbiology Department, Salford Royal NHS Foundation Trust, Salford, England. Financial disclosure: None declared. Received for publication May 28, 2014; accepted for publication August 3, 2014. Corresponding Author: Martyn J. Dibb, MB, ChB, MD, MRCP, Department of Gastroenterology, Royal Liverpool University Hospital, Prescott Street, Liverpool L7 8XP, UK. Email: [email protected]
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
2
Journal of Parenteral and Enteral Nutrition XX(X)
Table 1. Guidelines for the Diagnosis of CRBSI.1 Diagnosis of CRBSI is best achieved (a) by quantitative or semi-quantitative culture of the catheter (when the CVC is removed or exchanged over a guide wire); (b) by paired quantitative blood cultures or paired qualitative blood cultures from a peripheral vein and from the catheter, with continuously monitoring of the differential time to positivity (if the catheter is left in place) (Grade A). CRBSI, catheter-related bloodstream infection; CVC, central venous catheter.
Since the maintenance of safe central venous access is of crucial importance to the patient with IF and since repeated CVC removal and reinsertion can compromise venous access, international guidelines advocate attempted salvage of an infected tunneled CVC wherever possible.1 Small studies have demonstrated that CVC salvage can be obtained in a high percentage of CVCs with coagulase-negative staphylococcal infection using standardized antibiotic protocols.12-14 However, the long-term effectiveness of antibiotic salvage protocols outcome in CRBSI attributable to other organisms has not been adequately addressed. The aim of the present study was therefore to determine the efficacy and safety of antibiotic CVC salvage for CRBSI in patients requiring long-term HPN.
Materials and Methods Between January 1993 and December 2011, all confirmed CRBSIs occurring in patients receiving PN managed at a national intestinal failure unit (IFU) were entered into a prospectively maintained database. All venous access devices used for HPN were included, although the standard device in use at our institution is a single-lumen tunneled CVC (Broviac). Diagnosis of a CRBSI was based on quantitative and qualitative assessment of central and peripheral blood cultures and pour plates as recommended by European guidelines (Table 1).1 An initial standardized treatment protocol (Figure 1) involving antibiotic and urokinase CVC locks and systemic antibiotic administration was used. Blood cultures and pour plates were taken both from peripheral venous blood and from the CVC in patients with a suspected CRBSI. The CVC was then not used until a CRBSI was excluded. Blood cultures and pour plates were repeated immediately in those patients in whom initial pour plates were negative but who had a positive central venous blood culture to confirm the diagnosis (Figure 1). Patients with septic shock, mechanical catheter complications, or fungal CVC infections were excluded from attempted antibiotic CVC salvage and the CVC was removed in these cases. Patients were initially treated with vancomycin and urokinase CVC locks, as well as systemic vancomycin, until microbiological antibiotic sensitivities were available. Patients received daily medical review, and if there was no clinical improvement
after 48 hours, the catheter was removed. This was defined as ongoing symptoms, signs of infection, or laboratory evidence of treatment failure (worsening inflammatory markers, renal impairment, or neutrophilia). The standard treatment length was 14 days, during which time the CVC was not used for intravenous (IV) feeding. Antibiotic choice and duration was adapted if a resistance pattern was identified. Repeated sets of peripheral and central blood cultures and pour plates were taken 48 hours after completion of antimicrobial therapy at day 16 to evaluate the efficacy of antibiotic salvage. Echocardiography was arranged in all patients with severe sepsis and those in whom catheter salvage failed. Failed salvage was defined as recurrent positive blood or pour plate culture identified from a catheter lumen within 30 days of the end of the initial treatment course. Data collected included patient demographics, mechanism and diagnosis of IF, length of small bowel remaining and the presence of a stoma, laterality of CVC, time since CVC insertion, CRBSI microorganism(s) isolated, antibiotic therapy administered, duration of antibiotic treatment, and the number of CRBSIs occurring per patient. Statistical analysis was performed by using the SPSS software package (version 19.0; SPSS, Inc, an IBM Company, Chicago, IL). Student t tests and 1-way analysis of variance (ANOVA) with Bonferroni comparisons were used for parametric data, and the Mann-Whitney U test was used for nonparametric data. Cox regression was performed for time to first infection. Statistical significance was taken as P < .05.
Results In total, 588 patients were identified during the study period, encompassing a total of 2134 HPN years. A total of 297 CRBSIs occurred in 137 patients (65 single CRBSI, 72 multiple CRBSIs); the remaining 451 patients receiving HPN during this period had no CRBSIs. Single-lumen tunneled CVCs were present in 292 of 297 (98.3%) episodes (Broviac, 272/297; Hickmann, 20/297), as well as in a small number of other devices (Port-a-cath, 3/297; dual-lumen Hickman line, 1/297; Groshon, 1/297). The overall rate of infection in all patients was 0.38 per 1000 catheter days (0.14 per HPN year). The median (range) number of CVC days prior to developing an infection was 403 (5–6128). Patients with single or multiple episodes of CRBSI received longer mean durations of HPN than did patients without a CRBSI (Figure 2). Patients with a bowel length of less than 100 cm were more likely to have multiple CVC infections (P = .003), but univariate analysis of the time to first CVC infection revealed that mechanism of IF, underlying diagnosis, nutrition outcome, CVC thrombosis, presence/absence of a stoma, and the presence of remaining colon did not significantly affect the incidence of CRBSI. In total, 219 of 297 (73.7%) patients with a CRBSI had pyrexia and/or rigors, and 57 of 297 (19.2%) had nonspecific
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
Dibb et al
3
Figure 1. Management of suspected catheter-related bloodstream infection (CRBSI). AB, antibiotics; CBC, central blood culture; CFU, colony-forming units; CPP, central pour plate; D/W, discuss with; ECHO, echocardiogram; IFU, intestinal failure unit; PBC, peripheral blood culture; PN, parenteral nutrition; PPP, peripheral pour plate; UTI, urinary tract infection.
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
4
Journal of Parenteral and Enteral Nutrition XX(X)
Figure 2. The effect of time on parenteral nutrition (PN) against number of catheter-related bloodstream infections (CRBSIs). Histogram demonstrating the median and range of time receiving PN against patients who had no, single, or multiple CRBSIs.
symptoms, such as headache and malaise. The remaining patients had chest/neck pain (11/297 [3.7%]) or were asymptomatic (9/297 [3%]), or symptom data were not recorded (3/297 [1%]). Microbiological confirmation was available in 278 of 297 (93.6%) cases; the diagnosis was confirmed in 255 of 278 (91.7%) by the presence of a positive CVC blood culture, while in 15 of 278 (5.4%) cases, the diagnosis was confirmed by either a central pour plate >1000 colony-forming units (CFU) or a central pour plate/peripheral pour plate ratio of ≥4:1. In the remaining 8 of 278 (2.88%) cases, diagnosis of CRBSI was made on a positive peripheral blood culture and with high clinical suspicion, with laboratory and radiological exclusion of other possible causes. A single microorganism, most commonly CNS (50.5% cases), caused 87.9% of all infections (Table 2). Fungi were isolated in 11 (3.6%) cases, either as a single organism (7/11 [63.6%]) or as part of a polymicrobial infection (4/11 [36.4%]). CVC salvage was not attempted in 49 episodes of CRBSI because of life-threatening sepsis (n = 18), sole fungal CVC infections (n = 7), mechanical catheter problems (eg, coexisting CVC fracture; n = 20), or infection of the catheter tunnel (n = 4). Patients in whom salvage was initially attempted received a median of 14 days of antibiotic therapy (range, 2–25). In 68.1% (171/248) of CRBSI episodes, a standard 14-day course of antibiotic therapy was completed. The reasons for not completing a full course of antibiotics varied but included persistent pyrexia or sepsis, identification of mechanical line complication, identification of a fungal or resistant organism, and CVC thrombosis. Excluding those episodes in whom attempted catheter salvage was deemed inappropriate, 72.5% (180/248) of CRBSIs were salvaged, including coagulase-negative staphylococcus, 79.8% (103/129); Staphylococcus aureus, 56.7%
(17/30 [methicillin-resistant Staphylococcus aureus (MRSA), 4/9; methicillin-sensitive Staphylococcus aureus (MSSA), 13/21]); polymicrobial infections, 67.7% (21/30); and miscellaneous, 66.1% (39/59) (Figure 3). The catheter was removed in 33.7% (100/297) of cases. CVC reinfection within 30 days occurred in 4.4% (11/248) of cases. There were 4 cases of infective endocarditis complicating CRBSI (3 cases of coagulase-negative staphylococcus, 1 case of MSSA). In all cases of endocarditis, the CVC was removed. There were 5 deaths in patients admitted for management of the CRBSI (3 cases of CNS, 1 Enterobacter, and 1 multiple organisms). Of the patients who died, antibiotic salvage was not attempted in 2 of 5 due to the presence of septic shock. The remaining 3 patients who died had varying lengths of antibiotic salvage (7, 14, and 14 days, respectively) but had developed a metastatic focus of infection at other sites via hematogenous spread, with 1 patient having infective endocarditis and 2 patients having bronchopneumonia.
Conclusion The present study describes the largest reported series to date of CVC salvage of CRBSIs occurring in patients requiring long-term HPN; successful salvage rates of over 70% can be safely achieved by following a standardized initial protocol that is adapted based on microbiological organism, antimicrobial susceptibility results, and clinical response. Furthermore, we demonstrate for the first time in a large series of patients that CVCs infected with organisms not traditionally salvaged, such as S aureus, can also be safely salvaged in a significant proportion of cases. The successful salvage rates reported in this study are vital for patients with type 3 IF, since CVC salvage preserves precious venous access by obviating the need for repeated CVC removal and reinsertion. The effectiveness of CVC salvage in this study relied on a focused and stepwise protocol for managing patients with suspected CVC infections. Once a diagnosis of CRBSI has been suspected and then established, it is essential that prompt treatment be initiated, where appropriate. It is notable that only 5 patients (1.7%) died in the present series, which is a relatively low mortality rate.15 This likely reflects the practice used in this study that highlighted the importance of recognizing conditions where salvage should not be attempted; CVCs were removed immediately and salvage not attempted in unstable patients meeting standard criteria for severe sepsis.16 Equally, early identification of metastatic infections in patients with CRBSIs is paramount, and removal of the infected CVC would be suggested in these situations to remove the primary source of infection. The present study is also the largest to demonstrate longterm safety of salvage if CVCs are infected with organisms other than coagulase-negative staphylococci. Current international guidelines advocate CVC removal, rather than attempted salvage, in CRBSIs due to S aureus unless the patient has a
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
Dibb et al
5
Table 2. Organisms Identified in Home Parenteral Nutrition Catheter-Related Bloodstream Infections. Organism Gram-positive bacteria CNS MSSA MRSA Gram-positive bacillus Bacillus sp. Enterococcus faecalis Micrococcus sp. Propionibacterium sp. Enterococcus sp. Streptococcus sp. Lactococcus sp. Staphylococcus warneri Gram-negative bacteria Klebsiella sp. Enterobacter sp. Escherichia coli Acinetobacter sp. Serratia liquefaciens Pseudomonas sp. Other organisms Multiple organisms Candida sp. Yeast Missing data Total
Frequency
%
Valid %
Cumulative %
150 29 11 7 5 5 3 2 2 2 1 1
50.5 9.8 3.7 2.4 1.7 1.7 1.0 0.7 0.7 0.7 0.3 0.3
51.5 10.0 3.8 2.4 1.7 1.7 1.0 0.7 0.7 0.7 0.3 0.3
50.5 60.3 64.0 66.3 68.0 69.7 70.7 71.4 72.1 72.7 73.1 73.4
11 8 6 2 2 1
3.7 2.7 2.0 0.7 0.7 0.3
3.8 2.7 2.1 0.7 0.7 0.3
77.1 79.8 81.8 82.5 83.2 83.5
36 5 2 6 297
12.1 1.7 0.7 2.0 100
12.4 1.7 0.7 0.0 100
95.6 97.3 98.0 100.0 100
CNS, central nervous system; MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus.
contraindication such as lack of alternative venous access, significant bleeding diathesis, and/or if quality-of-life issues take priority over the need for reinsertion of a new catheter at another site; furthermore, in the setting of relative contraindications to catheter removal, 28 days of antibiotics would normally be recommended for S aureus CRBSIs.17 Notably, successful CVC salvage was achieved in this study in a significant proportion of CVCs infected by S aureus and other bacteria following 14 days of targeted antibiotic therapy. Clinicians and patients must be vigilant for subtle symptoms or laboratory abnormalities that may represent an early CRBSI. More than one-fourth of patients with proven CRBSIs in this study were apyrexial and around one-fifth had nonspecific symptoms. We have previously demonstrated that patients with CRBSIs may present with hyperbilirubinemia, and it is therefore important to adopt a low index of suspicion for CRBSIs in patients receiving long-term HPN who may present with atypical features.12 Once suspected, accurate diagnosis currently relies on the combined quantitative and qualitative assessment of central and peripheral blood cultures and pour plates. In this study, a CRBSI was confirmed with >1000 CFU in the central pour plate or by a ≥4-fold higher colony count in quantitative central vs peripheral pour plates. European and North American guidelines underscore the greater specificity
of quantitative over qualitative methods and advocate the combined use of both to diagnose CRBSIs, which was the method used in this study.1,17 New diagnostic approaches, including the use of real-time polymerase chain reaction, may improve diagnostic sensitivity and reduce time to diagnosis, but utility in the HPN population has not yet been established.18 Clearly, it is better to avoid a CRBSI rather than to have to salvage an infected catheter. Incidence rates of CRBSI in patients requiring HPN have been previously variably reported between 0.35 and 2.27 per 1000 catheter days.5-9 Evidence has shown that nursing, patient, and carer adherence to a strict aseptic catheter care protocol, supervised by a multidisciplinary nutrition team, is paramount in minimizing CVC-related septic episodes4,7 and, undoubtedly, the stringent training program and catheter care protocol (Table 3) offered at our center accounts for the very low CRBSI rate of 0.38 per 1000 catheter days described in the present large series of patients. It is, however, notable that only a small proportion (12%) of patients receiving HPN were responsible for the majority (75%) of the total number of CRBSIs in this study. A number of patient-related factors, such as the presence of a stoma and/or opiate use, have previously been associated with a higher risk of CVC infection.19 In this study, as reported in other series, shorter small bowel length was associated with an increased risk of CRBSI.20,21 This may relate to
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
6
Journal of Parenteral and Enteral Nutrition XX(X) carers, can minimize the occurrence of this life-threatening complication to very low levels. When a CVC is infected, CVC salvage can be achieved in a high proportion of cases by following an initial standardized protocol that is then tailored to the patient’s clinical condition and microbiological sensitivities. Constant vigilance is required for the development of severe sepsis or metastatic infection that, despite treatment, can be associated with a poor outcome. We demonstrate for the first time in HPN patients that catheter salvage can be achieved not only in coagulase-negative staphylococcal infections but also for other bacterial species.
References
Figure 3. Catheter-related bloodstream infection (CRBSI) salvage rate by microorganism and case frequency. The percentages of cases successfully salvaged by microorganism is indicated. CNS, central nervous system; MRSA, methicillinresistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus. Table 3. Areas Covered by the Salford Home Parenteral Nutrition Catheter Maintenance Protocol. Hand washing Trolley washing Preparing aseptic pack Line lock Changing the line connectors Exit site dressing Additives Running fluids through a line Connecting to fluids Infusion pumps Disconnecting the line Dealing with problems
patients with a shorter bowel having higher stoma effluent output, leading to increased fluid/electrolytes and requiring increased nights of PN and catheter hub access, which in turn has been associated with an increased risk of CRBSIs.22 In summary, CRBSI remains a significant complication of HPN delivery, which continues to have an associated mortality. The present study demonstrates that robust aseptic catheter care protocols, with dedicated training of nurses, patients, and
1. Pittiruti M, Hamilton H, Biffi R, MacFie J, Pertkiewicz M. ESPEN guidelines on parenteral nutrition: central venous catheters (access, care, diagnosis and therapy of complications). Clin Nutr. 2009;28:365-377. 2. Nehme AE. Nutritional support of the hospitalized patient: the team concept. JAMA. 1980;243:1906-1908. 3. Goldstein M, Braitman LE, Levine GM. The medical and financial costs associated with termination of a nutrition support nurse. JPEN J Parenter Enteral Nutr. 2000;24:323-327. 4. Sutton CD, Garcea G, Pollard C, Berry DP, Dennison AR. The introduction of a nutrition clinical nurse specialist results in a reduction in the rate of catheter sepsis. Clin Nutr. 2005;24:220-223. 5. Lloyd DAJ, Vega R, Bassett P, Forbes A, Gabe SM. Survival and dependence on home parenteral nutrition: experience over a 25-year period in a UK referral centre. Aliment Pharmacol Ther. 2006;24:1231-1240. 6. Vantini I, Benini L, Bonfante F, et al. Survival rate and prognostic factors in patients with intestinal failure. Dig Liver Dis. 2004;36:46-55. 7. Crispin A, Thul P, Arnold D, Schild S, Weimann A. Central venous catheter complications during home parenteral nutrition: a prospective pilot study of 481 patients with more than 30,000 catheter days. Onkologie. 2008;31:605-609. 8. Marra AR, Opilla M, Edmond MB, Kirby DF. Epidemiology of bloodstream infections in patients receiving long-term total parenteral nutrition. J Clin Gastroenterol. 2007;41:19-28. 9. Buchman AL, Opilla M, Kwasny M, Diamantidis TG, Okamoto R. Risk factors for the development of catheter-related bloodstream infections in patients receiving home parenteral nutrition. JPEN J Parenter Enteral Nutr. 2013;38(6):744-749. 10. Dreesen M, Foulon V, Spriet I, et al. Epidemiology of catheter-related infections in adult patients receiving home parenteral nutrition: a systematic review. Clin Nutr. 2013;32:16-26. 11. Richards DM, Scott NA, Shaffer JL, Irving M. Opiate and sedative dependence predicts a poor outcome for patients receiving home parenteral nutrition. JPEN J Parenter Enteral Nutr. 1997;21:336-338. 12. Clare A, Teubner A, Shaffer JL. What information should lead to a suspicion of catheter sepsis in HPN? Clin Nutr. 2008;27:552-556. 13. Santarpia L, Pasanisi F, Alfonsi L, et al. Prevention and treatment of implanted central venous catheter (CVC)–related sepsis: a report after six years of home parenteral nutrition (HPN). Clin Nutr. 2002;21: 207-211. 14. Santarpia L, Alfonsi L, Tiseo D, et al. Central venous catheter infections and antibiotic therapy during long-term home parenteral nutrition: an 11-year follow-up study. JPEN J Parenter Enteral Nutr. 2010;34:254-262. 15. Messing B, Crenn P, Beau P, Boutron-Ruault MC, Rambaud JC, Matuchansky C. Long-term survival and parenteral nutrition dependence in adult patients with the short bowel syndrome. Gastroenterology. 1999;117:1043-1050. 16. Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock. Crit Care Med. 2013;41:580-637.
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
Dibb et al
7
17. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 update by the Infectious Diseases Society of America. Clin Infect Dis. 2009;49:1-45. 18. Dark P, Dunn G, Chadwick P, et al. The clinical diagnostic accuracy of rapid detection of healthcare-associated bloodstream infection in intensive care using multipathogen real-time PCR technology. BMJ Open. 2011;1:e00018. 19. Dibb M, Teubner A, Theis V, Shaffer J, Lal S. Review article: the management of long-term parenteral nutrition. Aliment Pharmacol Ther. 2013;37:587-603.
20. Terra RM, Plopper C, Waitzberg DL, et al. Remaining small bowel length: association with catheter sepsis in patients receiving home total parenteral nutrition: evidence of bacterial translocation. World J Surg. 2000;24:1537-1541. 21. Reimund J-M, Aroundel Y, Finck G, Zimmermann F, Duclos B, Baumann R. Catheter-related infection in patients on home parenteral nutrition: results of a prospective survey. Clin Nutr. 2002;21:33-38. 22. Bozzetti F, Mariani L, Bertinet DB, et al. Central venous catheter complications in 447 patients on home parenteral nutrition: an analysis of over 100,000 catheter days. Clin Nutr. 2002;21:475-85.
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
research-article2014
PENXXX10.1177/0148607114549999Journal of Parenteral and Enteral NutritionDibb et al
Brief Communication
Central Venous Catheter Salvage in Home Parenteral Nutrition Catheter-Related Bloodstream Infections: Long-Term Safety and Efficacy Data
Journal of Parenteral and Enteral Nutrition Volume XX Number X Month 201X 1–7 © 2014 American Society for Parenteral and Enteral Nutrition DOI: 10.1177/0148607114549999 jpen.sagepub.com hosted at online.sagepub.com
Martyn J. Dibb, MB, ChB, MD, MRCP1; Arun Abraham, MB, BS, MRCS1; Paul R. Chadwick, MB, ChB, MD, MRCP, FRC(Path)2; Jon L. Shaffer, MB, ChB, MD, FRCP1; Antje Teubner, MD, MRCS1; Gordon L. Carlson, MB, ChB (Hons), MD, FRCS1; and Simon Lal, MB, ChB, PhD, FRCP1
Abstract Background: Catheter-related bloodstream infections (CRBSIs) are a serious complication in the provision of home parenteral nutrition (HPN). Antibiotic salvage of central venous catheters (CVCs) in CRBSI is recommended; however, this is based on limited reports. We assessed the efficacy of antibiotic salvage of CRBSIs in HPN patients. Materials and Methods: All confirmed CRBSIs occurring in patients receiving HPN in a national intestinal failure unit (IFU), between 1993 and 2011, were analyzed. A standardized protocol involving antibiotic and urokinase CVC locks and systemic antibiotics was used. Results: In total, 588 patients were identified with a total of 2134 HPN years, and 297 CRBSIs occurred in 137 patients (65 single and 72 multiple CRBSIs). The overall rate of CRBSI in all patients was 0.38 per 1000 catheter days. Most (87.9%) infections were attributable to a single microorganism. In total, 72.5% (180/248) of CRBSIs were salvaged when attempted (coagulase-negative staphylococcus, 79.8% [103/129], Staphylococcus aureus, 56.7% [17/30]; polymicrobial infections, 67.7% [21/30]; and miscellaneous, 66.1% [39/59]). CVC salvage was not attempted in 49 episodes because of life-threatening sepsis (n = 18), fungal infection (n = 7), catheter problems (n = 20), and CVC tunnel infection (n = 4). Overall, the CVC was removed in 33.7% (100/297) of cases. There were 5 deaths in patients admitted to the IFU for management of the CRBSI (2 severe sepsis at presentation, 3 metastatic infection). Conclusions: This is the largest reported series of catheter salvage in CRBSIs and demonstrates successful catheter salvage in most cases when using a standardized protocol. (JPEN J Parenter Enteral Nutr. XXXX;xx:xx-xx)
Keywords parenteral nutrition; nutrition; venous access; nutrition support teams; nutrition support practice; outcomes research/quality; sepsis; research and diseases
Clinical Relevancy Statement Antibiotic salvage of catheter-related bloodstream infections in home parenteral nutrition (HPN) patients is important in preserving venous access. The rate of successful salvage with different organisms is an important finding, and the safety of the protocol is also clinically relevant for clinicians supervising patients receiving HPN.
Introduction Home parenteral nutrition (HPN) delivered by a dedicated central venous catheter (CVC) can be a life-saving therapy for patients with intestinal failure (IF). Aseptic care of CVCs for patients with IF is mandatory, and strict adherence to catheter care protocols has been demonstrated to reduce the incidence of catheter-related bloodstream infection (CRBSI) and prolong CVC viability.1-4 CRBSI is a serious and common complication of long-term HPN, and CRBSI rates are a useful surrogate marker of the overall quality of HPN care. CRBSI rates have been variably reported
as 0.35–2.27 episodes/1000 catheter days.5-9 A number of risk factors for the development of infections have been identified, including type and characteristics of the infusion device, education and training in CVC care, nature of underlying disease and intestinal anatomy, opiate dependency, and the use of the CVC for obtaining blood samples and administration of medication.9-11 Coagulase-negative staphylococci (CNS) have been reported to be responsible for approximately 50% of episodes of CRBSI in a recent systematic review, with gram-negative and polymicrobial infections accounting for the majority of the remainder.10 From the 1Intestinal Failure Unit and 2Microbiology Department, Salford Royal NHS Foundation Trust, Salford, England. Financial disclosure: None declared. Received for publication May 28, 2014; accepted for publication August 3, 2014. Corresponding Author: Martyn J. Dibb, MB, ChB, MD, MRCP, Department of Gastroenterology, Royal Liverpool University Hospital, Prescott Street, Liverpool L7 8XP, UK. Email: [email protected]
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
2
Journal of Parenteral and Enteral Nutrition XX(X)
Table 1. Guidelines for the Diagnosis of CRBSI.1 Diagnosis of CRBSI is best achieved (a) by quantitative or semi-quantitative culture of the catheter (when the CVC is removed or exchanged over a guide wire); (b) by paired quantitative blood cultures or paired qualitative blood cultures from a peripheral vein and from the catheter, with continuously monitoring of the differential time to positivity (if the catheter is left in place) (Grade A). CRBSI, catheter-related bloodstream infection; CVC, central venous catheter.
Since the maintenance of safe central venous access is of crucial importance to the patient with IF and since repeated CVC removal and reinsertion can compromise venous access, international guidelines advocate attempted salvage of an infected tunneled CVC wherever possible.1 Small studies have demonstrated that CVC salvage can be obtained in a high percentage of CVCs with coagulase-negative staphylococcal infection using standardized antibiotic protocols.12-14 However, the long-term effectiveness of antibiotic salvage protocols outcome in CRBSI attributable to other organisms has not been adequately addressed. The aim of the present study was therefore to determine the efficacy and safety of antibiotic CVC salvage for CRBSI in patients requiring long-term HPN.
Materials and Methods Between January 1993 and December 2011, all confirmed CRBSIs occurring in patients receiving PN managed at a national intestinal failure unit (IFU) were entered into a prospectively maintained database. All venous access devices used for HPN were included, although the standard device in use at our institution is a single-lumen tunneled CVC (Broviac). Diagnosis of a CRBSI was based on quantitative and qualitative assessment of central and peripheral blood cultures and pour plates as recommended by European guidelines (Table 1).1 An initial standardized treatment protocol (Figure 1) involving antibiotic and urokinase CVC locks and systemic antibiotic administration was used. Blood cultures and pour plates were taken both from peripheral venous blood and from the CVC in patients with a suspected CRBSI. The CVC was then not used until a CRBSI was excluded. Blood cultures and pour plates were repeated immediately in those patients in whom initial pour plates were negative but who had a positive central venous blood culture to confirm the diagnosis (Figure 1). Patients with septic shock, mechanical catheter complications, or fungal CVC infections were excluded from attempted antibiotic CVC salvage and the CVC was removed in these cases. Patients were initially treated with vancomycin and urokinase CVC locks, as well as systemic vancomycin, until microbiological antibiotic sensitivities were available. Patients received daily medical review, and if there was no clinical improvement
after 48 hours, the catheter was removed. This was defined as ongoing symptoms, signs of infection, or laboratory evidence of treatment failure (worsening inflammatory markers, renal impairment, or neutrophilia). The standard treatment length was 14 days, during which time the CVC was not used for intravenous (IV) feeding. Antibiotic choice and duration was adapted if a resistance pattern was identified. Repeated sets of peripheral and central blood cultures and pour plates were taken 48 hours after completion of antimicrobial therapy at day 16 to evaluate the efficacy of antibiotic salvage. Echocardiography was arranged in all patients with severe sepsis and those in whom catheter salvage failed. Failed salvage was defined as recurrent positive blood or pour plate culture identified from a catheter lumen within 30 days of the end of the initial treatment course. Data collected included patient demographics, mechanism and diagnosis of IF, length of small bowel remaining and the presence of a stoma, laterality of CVC, time since CVC insertion, CRBSI microorganism(s) isolated, antibiotic therapy administered, duration of antibiotic treatment, and the number of CRBSIs occurring per patient. Statistical analysis was performed by using the SPSS software package (version 19.0; SPSS, Inc, an IBM Company, Chicago, IL). Student t tests and 1-way analysis of variance (ANOVA) with Bonferroni comparisons were used for parametric data, and the Mann-Whitney U test was used for nonparametric data. Cox regression was performed for time to first infection. Statistical significance was taken as P < .05.
Results In total, 588 patients were identified during the study period, encompassing a total of 2134 HPN years. A total of 297 CRBSIs occurred in 137 patients (65 single CRBSI, 72 multiple CRBSIs); the remaining 451 patients receiving HPN during this period had no CRBSIs. Single-lumen tunneled CVCs were present in 292 of 297 (98.3%) episodes (Broviac, 272/297; Hickmann, 20/297), as well as in a small number of other devices (Port-a-cath, 3/297; dual-lumen Hickman line, 1/297; Groshon, 1/297). The overall rate of infection in all patients was 0.38 per 1000 catheter days (0.14 per HPN year). The median (range) number of CVC days prior to developing an infection was 403 (5–6128). Patients with single or multiple episodes of CRBSI received longer mean durations of HPN than did patients without a CRBSI (Figure 2). Patients with a bowel length of less than 100 cm were more likely to have multiple CVC infections (P = .003), but univariate analysis of the time to first CVC infection revealed that mechanism of IF, underlying diagnosis, nutrition outcome, CVC thrombosis, presence/absence of a stoma, and the presence of remaining colon did not significantly affect the incidence of CRBSI. In total, 219 of 297 (73.7%) patients with a CRBSI had pyrexia and/or rigors, and 57 of 297 (19.2%) had nonspecific
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
Dibb et al
3
Figure 1. Management of suspected catheter-related bloodstream infection (CRBSI). AB, antibiotics; CBC, central blood culture; CFU, colony-forming units; CPP, central pour plate; D/W, discuss with; ECHO, echocardiogram; IFU, intestinal failure unit; PBC, peripheral blood culture; PN, parenteral nutrition; PPP, peripheral pour plate; UTI, urinary tract infection.
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
4
Journal of Parenteral and Enteral Nutrition XX(X)
Figure 2. The effect of time on parenteral nutrition (PN) against number of catheter-related bloodstream infections (CRBSIs). Histogram demonstrating the median and range of time receiving PN against patients who had no, single, or multiple CRBSIs.
symptoms, such as headache and malaise. The remaining patients had chest/neck pain (11/297 [3.7%]) or were asymptomatic (9/297 [3%]), or symptom data were not recorded (3/297 [1%]). Microbiological confirmation was available in 278 of 297 (93.6%) cases; the diagnosis was confirmed in 255 of 278 (91.7%) by the presence of a positive CVC blood culture, while in 15 of 278 (5.4%) cases, the diagnosis was confirmed by either a central pour plate >1000 colony-forming units (CFU) or a central pour plate/peripheral pour plate ratio of ≥4:1. In the remaining 8 of 278 (2.88%) cases, diagnosis of CRBSI was made on a positive peripheral blood culture and with high clinical suspicion, with laboratory and radiological exclusion of other possible causes. A single microorganism, most commonly CNS (50.5% cases), caused 87.9% of all infections (Table 2). Fungi were isolated in 11 (3.6%) cases, either as a single organism (7/11 [63.6%]) or as part of a polymicrobial infection (4/11 [36.4%]). CVC salvage was not attempted in 49 episodes of CRBSI because of life-threatening sepsis (n = 18), sole fungal CVC infections (n = 7), mechanical catheter problems (eg, coexisting CVC fracture; n = 20), or infection of the catheter tunnel (n = 4). Patients in whom salvage was initially attempted received a median of 14 days of antibiotic therapy (range, 2–25). In 68.1% (171/248) of CRBSI episodes, a standard 14-day course of antibiotic therapy was completed. The reasons for not completing a full course of antibiotics varied but included persistent pyrexia or sepsis, identification of mechanical line complication, identification of a fungal or resistant organism, and CVC thrombosis. Excluding those episodes in whom attempted catheter salvage was deemed inappropriate, 72.5% (180/248) of CRBSIs were salvaged, including coagulase-negative staphylococcus, 79.8% (103/129); Staphylococcus aureus, 56.7%
(17/30 [methicillin-resistant Staphylococcus aureus (MRSA), 4/9; methicillin-sensitive Staphylococcus aureus (MSSA), 13/21]); polymicrobial infections, 67.7% (21/30); and miscellaneous, 66.1% (39/59) (Figure 3). The catheter was removed in 33.7% (100/297) of cases. CVC reinfection within 30 days occurred in 4.4% (11/248) of cases. There were 4 cases of infective endocarditis complicating CRBSI (3 cases of coagulase-negative staphylococcus, 1 case of MSSA). In all cases of endocarditis, the CVC was removed. There were 5 deaths in patients admitted for management of the CRBSI (3 cases of CNS, 1 Enterobacter, and 1 multiple organisms). Of the patients who died, antibiotic salvage was not attempted in 2 of 5 due to the presence of septic shock. The remaining 3 patients who died had varying lengths of antibiotic salvage (7, 14, and 14 days, respectively) but had developed a metastatic focus of infection at other sites via hematogenous spread, with 1 patient having infective endocarditis and 2 patients having bronchopneumonia.
Conclusion The present study describes the largest reported series to date of CVC salvage of CRBSIs occurring in patients requiring long-term HPN; successful salvage rates of over 70% can be safely achieved by following a standardized initial protocol that is adapted based on microbiological organism, antimicrobial susceptibility results, and clinical response. Furthermore, we demonstrate for the first time in a large series of patients that CVCs infected with organisms not traditionally salvaged, such as S aureus, can also be safely salvaged in a significant proportion of cases. The successful salvage rates reported in this study are vital for patients with type 3 IF, since CVC salvage preserves precious venous access by obviating the need for repeated CVC removal and reinsertion. The effectiveness of CVC salvage in this study relied on a focused and stepwise protocol for managing patients with suspected CVC infections. Once a diagnosis of CRBSI has been suspected and then established, it is essential that prompt treatment be initiated, where appropriate. It is notable that only 5 patients (1.7%) died in the present series, which is a relatively low mortality rate.15 This likely reflects the practice used in this study that highlighted the importance of recognizing conditions where salvage should not be attempted; CVCs were removed immediately and salvage not attempted in unstable patients meeting standard criteria for severe sepsis.16 Equally, early identification of metastatic infections in patients with CRBSIs is paramount, and removal of the infected CVC would be suggested in these situations to remove the primary source of infection. The present study is also the largest to demonstrate longterm safety of salvage if CVCs are infected with organisms other than coagulase-negative staphylococci. Current international guidelines advocate CVC removal, rather than attempted salvage, in CRBSIs due to S aureus unless the patient has a
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
Dibb et al
5
Table 2. Organisms Identified in Home Parenteral Nutrition Catheter-Related Bloodstream Infections. Organism Gram-positive bacteria CNS MSSA MRSA Gram-positive bacillus Bacillus sp. Enterococcus faecalis Micrococcus sp. Propionibacterium sp. Enterococcus sp. Streptococcus sp. Lactococcus sp. Staphylococcus warneri Gram-negative bacteria Klebsiella sp. Enterobacter sp. Escherichia coli Acinetobacter sp. Serratia liquefaciens Pseudomonas sp. Other organisms Multiple organisms Candida sp. Yeast Missing data Total
Frequency
%
Valid %
Cumulative %
150 29 11 7 5 5 3 2 2 2 1 1
50.5 9.8 3.7 2.4 1.7 1.7 1.0 0.7 0.7 0.7 0.3 0.3
51.5 10.0 3.8 2.4 1.7 1.7 1.0 0.7 0.7 0.7 0.3 0.3
50.5 60.3 64.0 66.3 68.0 69.7 70.7 71.4 72.1 72.7 73.1 73.4
11 8 6 2 2 1
3.7 2.7 2.0 0.7 0.7 0.3
3.8 2.7 2.1 0.7 0.7 0.3
77.1 79.8 81.8 82.5 83.2 83.5
36 5 2 6 297
12.1 1.7 0.7 2.0 100
12.4 1.7 0.7 0.0 100
95.6 97.3 98.0 100.0 100
CNS, central nervous system; MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus.
contraindication such as lack of alternative venous access, significant bleeding diathesis, and/or if quality-of-life issues take priority over the need for reinsertion of a new catheter at another site; furthermore, in the setting of relative contraindications to catheter removal, 28 days of antibiotics would normally be recommended for S aureus CRBSIs.17 Notably, successful CVC salvage was achieved in this study in a significant proportion of CVCs infected by S aureus and other bacteria following 14 days of targeted antibiotic therapy. Clinicians and patients must be vigilant for subtle symptoms or laboratory abnormalities that may represent an early CRBSI. More than one-fourth of patients with proven CRBSIs in this study were apyrexial and around one-fifth had nonspecific symptoms. We have previously demonstrated that patients with CRBSIs may present with hyperbilirubinemia, and it is therefore important to adopt a low index of suspicion for CRBSIs in patients receiving long-term HPN who may present with atypical features.12 Once suspected, accurate diagnosis currently relies on the combined quantitative and qualitative assessment of central and peripheral blood cultures and pour plates. In this study, a CRBSI was confirmed with >1000 CFU in the central pour plate or by a ≥4-fold higher colony count in quantitative central vs peripheral pour plates. European and North American guidelines underscore the greater specificity
of quantitative over qualitative methods and advocate the combined use of both to diagnose CRBSIs, which was the method used in this study.1,17 New diagnostic approaches, including the use of real-time polymerase chain reaction, may improve diagnostic sensitivity and reduce time to diagnosis, but utility in the HPN population has not yet been established.18 Clearly, it is better to avoid a CRBSI rather than to have to salvage an infected catheter. Incidence rates of CRBSI in patients requiring HPN have been previously variably reported between 0.35 and 2.27 per 1000 catheter days.5-9 Evidence has shown that nursing, patient, and carer adherence to a strict aseptic catheter care protocol, supervised by a multidisciplinary nutrition team, is paramount in minimizing CVC-related septic episodes4,7 and, undoubtedly, the stringent training program and catheter care protocol (Table 3) offered at our center accounts for the very low CRBSI rate of 0.38 per 1000 catheter days described in the present large series of patients. It is, however, notable that only a small proportion (12%) of patients receiving HPN were responsible for the majority (75%) of the total number of CRBSIs in this study. A number of patient-related factors, such as the presence of a stoma and/or opiate use, have previously been associated with a higher risk of CVC infection.19 In this study, as reported in other series, shorter small bowel length was associated with an increased risk of CRBSI.20,21 This may relate to
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
6
Journal of Parenteral and Enteral Nutrition XX(X) carers, can minimize the occurrence of this life-threatening complication to very low levels. When a CVC is infected, CVC salvage can be achieved in a high proportion of cases by following an initial standardized protocol that is then tailored to the patient’s clinical condition and microbiological sensitivities. Constant vigilance is required for the development of severe sepsis or metastatic infection that, despite treatment, can be associated with a poor outcome. We demonstrate for the first time in HPN patients that catheter salvage can be achieved not only in coagulase-negative staphylococcal infections but also for other bacterial species.
References
Figure 3. Catheter-related bloodstream infection (CRBSI) salvage rate by microorganism and case frequency. The percentages of cases successfully salvaged by microorganism is indicated. CNS, central nervous system; MRSA, methicillinresistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus. Table 3. Areas Covered by the Salford Home Parenteral Nutrition Catheter Maintenance Protocol. Hand washing Trolley washing Preparing aseptic pack Line lock Changing the line connectors Exit site dressing Additives Running fluids through a line Connecting to fluids Infusion pumps Disconnecting the line Dealing with problems
patients with a shorter bowel having higher stoma effluent output, leading to increased fluid/electrolytes and requiring increased nights of PN and catheter hub access, which in turn has been associated with an increased risk of CRBSIs.22 In summary, CRBSI remains a significant complication of HPN delivery, which continues to have an associated mortality. The present study demonstrates that robust aseptic catheter care protocols, with dedicated training of nurses, patients, and
1. Pittiruti M, Hamilton H, Biffi R, MacFie J, Pertkiewicz M. ESPEN guidelines on parenteral nutrition: central venous catheters (access, care, diagnosis and therapy of complications). Clin Nutr. 2009;28:365-377. 2. Nehme AE. Nutritional support of the hospitalized patient: the team concept. JAMA. 1980;243:1906-1908. 3. Goldstein M, Braitman LE, Levine GM. The medical and financial costs associated with termination of a nutrition support nurse. JPEN J Parenter Enteral Nutr. 2000;24:323-327. 4. Sutton CD, Garcea G, Pollard C, Berry DP, Dennison AR. The introduction of a nutrition clinical nurse specialist results in a reduction in the rate of catheter sepsis. Clin Nutr. 2005;24:220-223. 5. Lloyd DAJ, Vega R, Bassett P, Forbes A, Gabe SM. Survival and dependence on home parenteral nutrition: experience over a 25-year period in a UK referral centre. Aliment Pharmacol Ther. 2006;24:1231-1240. 6. Vantini I, Benini L, Bonfante F, et al. Survival rate and prognostic factors in patients with intestinal failure. Dig Liver Dis. 2004;36:46-55. 7. Crispin A, Thul P, Arnold D, Schild S, Weimann A. Central venous catheter complications during home parenteral nutrition: a prospective pilot study of 481 patients with more than 30,000 catheter days. Onkologie. 2008;31:605-609. 8. Marra AR, Opilla M, Edmond MB, Kirby DF. Epidemiology of bloodstream infections in patients receiving long-term total parenteral nutrition. J Clin Gastroenterol. 2007;41:19-28. 9. Buchman AL, Opilla M, Kwasny M, Diamantidis TG, Okamoto R. Risk factors for the development of catheter-related bloodstream infections in patients receiving home parenteral nutrition. JPEN J Parenter Enteral Nutr. 2013;38(6):744-749. 10. Dreesen M, Foulon V, Spriet I, et al. Epidemiology of catheter-related infections in adult patients receiving home parenteral nutrition: a systematic review. Clin Nutr. 2013;32:16-26. 11. Richards DM, Scott NA, Shaffer JL, Irving M. Opiate and sedative dependence predicts a poor outcome for patients receiving home parenteral nutrition. JPEN J Parenter Enteral Nutr. 1997;21:336-338. 12. Clare A, Teubner A, Shaffer JL. What information should lead to a suspicion of catheter sepsis in HPN? Clin Nutr. 2008;27:552-556. 13. Santarpia L, Pasanisi F, Alfonsi L, et al. Prevention and treatment of implanted central venous catheter (CVC)–related sepsis: a report after six years of home parenteral nutrition (HPN). Clin Nutr. 2002;21: 207-211. 14. Santarpia L, Alfonsi L, Tiseo D, et al. Central venous catheter infections and antibiotic therapy during long-term home parenteral nutrition: an 11-year follow-up study. JPEN J Parenter Enteral Nutr. 2010;34:254-262. 15. Messing B, Crenn P, Beau P, Boutron-Ruault MC, Rambaud JC, Matuchansky C. Long-term survival and parenteral nutrition dependence in adult patients with the short bowel syndrome. Gastroenterology. 1999;117:1043-1050. 16. Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock. Crit Care Med. 2013;41:580-637.
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
Dibb et al
7
17. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 update by the Infectious Diseases Society of America. Clin Infect Dis. 2009;49:1-45. 18. Dark P, Dunn G, Chadwick P, et al. The clinical diagnostic accuracy of rapid detection of healthcare-associated bloodstream infection in intensive care using multipathogen real-time PCR technology. BMJ Open. 2011;1:e00018. 19. Dibb M, Teubner A, Theis V, Shaffer J, Lal S. Review article: the management of long-term parenteral nutrition. Aliment Pharmacol Ther. 2013;37:587-603.
20. Terra RM, Plopper C, Waitzberg DL, et al. Remaining small bowel length: association with catheter sepsis in patients receiving home total parenteral nutrition: evidence of bacterial translocation. World J Surg. 2000;24:1537-1541. 21. Reimund J-M, Aroundel Y, Finck G, Zimmermann F, Duclos B, Baumann R. Catheter-related infection in patients on home parenteral nutrition: results of a prospective survey. Clin Nutr. 2002;21:33-38. 22. Bozzetti F, Mariani L, Bertinet DB, et al. Central venous catheter complications in 447 patients on home parenteral nutrition: an analysis of over 100,000 catheter days. Clin Nutr. 2002;21:475-85.
Downloaded from pen.sagepub.com at University of British Columbia Library on November 17, 2015
