• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Parenteral Nutrition As a Risk Factor for Central Venous Catheter- Related Infection

Posted on: Sunday, 4 September 2005, 03:00 CDT

ABSTRACT. Background: The role of parenteral nutrition (PN) therapy as an independent risk factor for central venous catheter (CVC)-related infection in nonselected adult patients is not well established. The aim of this study was to evaluate PN as a risk factor for central venous catheterrelated infection in nonselected adult patients in a general university hospital. Methods: Patients using central venous catheters, exposed or nonexposed to PN, were prospectively followed for development of central venous catheter- related infection. Results: One hundred fifty-three patients were studied; 28 developed central venous catheter-related infection. Patients with central venous catheter-related infection presented higher frequency of PN use than patients without infection (60.7 vs 34.4%; p = .010). Multivariate Cox analysis showed that PN (relative risk (RR) = 3.30; 95% confidence interval [CI], 1.30-8.34; p = .012) was the only risk factor for central venous catheter-related infection. Malnutrition (RR = 0.45; 95% CI, 0.15-1.34; p = .152), days of hospitalization before central venous catheter insertion (RR = 1.00; 95% CI, 0.98-1.02; p = .801), and sustained hyperglycemia (RR = 0.49; 95% CI, 0.98-1.21; p = .091) were not significant in the model. Multiple logistic regression revealed that malnutrition (odds ratio [OR] = 8.05; 95% CI, 1.85-35.03; p = .005), central venous catheter indication for surgical-related pathology (OR = 7.26; 95% CI, 2.51-21.04; p < .001), sustained hyperglycemia (OR = 4.34; 95% CI, 1.79-10.52; p = .001), and days of hospitalization before central venous catheter insertion (OR = 1.04; 95% CI, 1.01-1.07; p = .004) were associated with PN use after adjustment for Assessment Score Intervention System score (OR = 0.33; 95% CI, 0.14-0.80; p = .014). Conclusions: PN therapy is an independent risk factor for central venous catheter-related infection in nonselected hospitalized adult patients. (Journal of Parenteral and Enteral Nutrition 29:367-373, 2005)

Parenteral nutrition (PN) is indicated in patients unable to obtain nutrients by the enteral route in order to prevent the adverse effects of malnutrition.1 The benefits of this therapy to improve nutrition status are well recognized.2 In addition, PN plays a significant role in the outcome of surgery in severely malnourished patients, in patients with short-bowel syndrome, and in critically ill patients.1

However, controversies regarding the impact of PN on mortality and overall morbidity still remain,1,3 especially due to the presence of PN-associated complications.4 One of the most serious complications of PN is catheter-related sepsis, which has been reported to occur in 1.3%-26.2% of the catheters used to administer PN.5-9 Such wide range of frequencies could be explained by differences in the design of studies investigating central venous catheter (CVC)-related bloodstream infection (BSI), in the criteria used to define CVC-related infection and in the type of catheter used, or by the inclusion of only selected groups of patients and the catheter management by teams with varying degrees of skill.

PN was first identified as a risk factor for CVC-related infection in the 1970s.10-12 Later, additional studies confirmed this association.13-19 Nevertheless, data confirming PN as an independent risk factor for CVC-related infection according to multivariate analysis are still scarce.13,17 Furthermore, studies often focus receiving PN in patients from intensive care units (ICUs).13,17 Therefore, the role of PN therapy as a risk factor for CVC-related infection in nonselected adult patients is still uncertain.

The aim of this study was to evaluate PN as a risk factor for CVC- related infection in nonselected adult patients from a general university hospital.

MATERIALS AND METHODS

Patients

A concurrent cohort study was carried out with adult patients using CVC with or without exposure to PN. The patients included in the study were hospitalized at the medical-surgical wards and ICU of Hospital de Clinicas de Porto Alegre, a general university hospital in Brazil, from May 2000 to December 2001.

All consecutive patients exposed to PN and whose CVC tips were cultured on removal were included in the study. For each patient receiving PN, 2 other patients with CVC and not exposed to PN during their entire stay in hospital were randomly selected by lot on the same day, 1 from the same ward and 1 from the ICU. When the selected PN patient was from the ICU, the 2 other nonexposed patients were also from the ICU. This was done because one-third of the patients in our hospital who are exposed to PN spend at least part of their hospital stay in the ICU. All CVC tips were cultured on removal in nonexposed patients.

All patients (exposed and nonexposed to PN) were followed from the moment of insertion of the first catheter to the removal of the last catheter required for the assigned therapy (PN or other). CVC change, with or without guidewire, was not counted as inclusion of a new patient. The same research nurses (MGB and LT) assessed all the patients for clinical and laboratory signs of CVC-related infection 3 times a week. In addition to demographic data, the following aspects were evaluated: nutrition status, mortality, severity of illness, presence of known diabetes mellitus and persistent hyperglycemia, insulin use, CVC insertion for surgical-related pathologies, antibiotic therapy, body mass index (BMI), other invasive procedures, any infection not related to CVC, complications related to CVC insertion, and days of hospitalization before CVC insertion. The hospital's ethics committee approved the protocol.

Definitions

"Days at risk per patient" was defined as the interval between the day of insertion of the first CVC (patient inclusion) and the onset of CVC-related infection (outcome) or catheter removal.

CVC-related infection was defined as recommended by the Centers for Disease Control and Prevention (CDC).20 Exit site infection was diagnosed in the presence of tenderness, skin erythema or induration within 2 cm of the catheter exit site, with or without fever, with positive catheter-tip culture or concomitant purulence in the exit site, even in the absence of a positive culture from the catheter tip.

CVC-related BSI was defined as bacteremia/fungemia or sepsis in patients with CVC, with at least 1 positive semiquantitative or quantitative blood culture from a peripheral vein in which the isolated organism was the same as that isolated from the catheter tip, with systemic manifestations of infection (ie, fever, chills, or hypotension) and without any other apparent source of infection.

Microbiological Analysis

CVCs were removed under aseptic conditions and distal catheter segments (tips) were cultured using the technique recommended by Maki.21 Results were expressed as negative (absence of growth or <15 colony-forming units [CFU]) or positive (≥15 CFU). Blood cultures were requested in the presence of systemic manifestations of infection (fever, chills, or hypotension).

PN Therapy

Preparation and administration of PN solutions were performed under aseptic conditions after the protocols of the Nutritional Support Therapy Program of our hospital. Briefly, the aim of PN was to provide 30 kcal/kg/d of nonprotein energy and 9 gN/d. A 2:1 solution (10% amino acids, 50% glucose) was administered through an exclusive IV line. All patients received a lipid solution, administered through a separate IV line. Physicians prescribed PN according to each patient's estimated needs, choosing 1 of 3 standardized solutions: (1) 2400 kcal/d (amino acids = 100 g); (2) 2080 kcaVd (amino acids = 70 g); (3) 2680 kcal/d (amino acids = 120 g). A pharmacist prepared PN solutions in the PN unit with maximal sterile-barrier precautions (sterile gloves, gown and mask, horizontal laminar air flow hood, Automix 3 + 3 Compounder system, Baxter Healthcare Corporation, Clientec Nutrition Division, Deerfield, IL). Heparin (1000 UI/L) was added to all solutions. After visual inspection of all prepared solutions, randomly selected samples ([radical]n +1; n = number of solutions prepared) were cultured daily and stored for later analysis, if necessary. Growth of microorganism was not observed in any of the samples evaluated during this study.

Catheters and Catheter Care

Nonantimicrobial-impregnated, nontunneled, shortduration, radiopaque, single-lumen Vialon CVCs (Intracath, Becton, Dickinson Ind. Cir. Ltda, Minas Gerais, Brazil; 16G, length 30.5 cm) or 2- lumen polyurethane CVCs (Arrow International, Reading, PA; 18G, 7 Fr, length 20 cm) were inserted in the subclavian or internal jugular veins. The femoral vein approach was not used. Central venous hemodialysis catheters and Shilley or Swan-Ganz catheters were excluded from the analysis because none of these catheter types is used for PN at our institution.

CVC insertion was performed by physicians-in-training or by staff physicians (ICU patients) as bedside procedures with maximal sterile barrier precautions (cap, mask, sterile gown, sterile gloves, and large sterile drapes). CVC management followed the recommendations of the hospital's Infection Control Center. CVCs were changed according to the judgment of staff physicians, with a \new percutaneous insertion or using a guidewire.

Skin antisepsis at the insertion site was performed with 2% alcoholic chlorhexidine. Gauze or transparent semipermeable polyurethane dressings (OpSite, Smith & Nephew, England) were used. The dressings were changed once every 7 days or if they appeared damp, dirty, or not properly attached to the skin.20 PN solutions and all devices used for PN administration were used for 24 hours and then were discarded. The infusion sets used for non-PN solutions were changed every 72 hours.

CVCs were removed if no longer needed or in the presence of mechanical CVC-related complications (arterial puncture, hemo- and pneumothorax, pneumoperitoneum, and CVC occlusion) or local or systemic signs of CVC-related infection.

Clinical and Laboratory Evaluation

Subjective global assessment of nutrition status was performed as described by Detski et al,22 and patients were classified as well nourished or malnourished (including moderately malnourished, possibly malnourished, or severely malnourished). BMI (weight [kg]/ height [m^sup 2^]) was calculated if allowed by the patient's clinical status.

Serum albumin level was determined by the bromocresol green technique and plasma glucose levels were determined by a glucose oxidase method. Hyperglycemia was defined when fasting plasma glucose levels were ≥6 mmol/L.23 Bedside capillary blood glucose monitoring was performed every 6 hours in patients receiving PN and as prescribed by physicians-in-training or by staff physicians in the other patients, using commercial strips and meters (Advantage or Advantage Complete; Roche Produtos Quimicos e Farmaceuticos S/A, So Paulo, Brazil). Sustained hyperglycemia was considered when strip values were ≥11 mmol/L for at least 24 hours. Insulin use during the study, at least 24 hours, was defined as regular subcutaneous administration of rapid or intermediate- acting insulin or continuous IV administration of rapid insulin.

Severity of clinical status was graded according to the Assessment Score Intervention System (ASIS)24: (1) postoperative patients requiring routine postoperative observation but not requiring intensive nursing or physician care; (2) physiologically stable patients requiring prophylactic observation without intensive nursing or medical care; (3) physiologically stable patients requiring intensive nursing care and monitoring; (4) physiologically unstable patients requiring intensive nursing or physician care with the need for frequent reassessment and adjustment of therapy; and (5) physiologically unstable patients who are in coma or shock or who require cardiopulmonary resuscitation or intensive medical and nursing care with need for frequent reassessment. Patients with ASIS score from 1 to 3 were classified as steady and patients with ASIS scores 4 and 5 as unsteady. This classification was used to compare patients with or without CVC-related infection and patients exposed and nonexposed to PN.

Invasive procedures beyond CVC insertion were endotracheal intubation; mechanical ventilation; nasogastric intubation; insertion of Swan-Ganz catheter, Schilley catheter, and peripheral venous catheter; thoracic and Kher tubes and intracranial pressure monitoring catheter; ostomies; and insertion of devices for monitoring central venous pressure and for hemodialysis.

Statistical Analysis

Student's t test or Mann-Whitney's U test was used as appropriate to test differences in mean values. Proportions were compared by the χ^sup 2^ test or Fisher's exact test. Cumulative incidence and incidence density of CVC-related infection (outcome) were calculated. Incidence density was calculated by multiplying the total number of outcomes by 1000 days and dividing this result by the total number of days at risk. Kaplan-Meier curves were used to estimate the occurrence of CVC-related infection in patients with and without PN. A multivariate Cox proportional hazards model was used to analyze possible risk factors for CVC-related infection (dependent variable). Multiple logistic regression was used to evaluate factors associated with PN use at baseline. The independent variables in these models were selected according to their significance in univariate analyses (p < .10) or according to their biologic relevance, p Values <0.05 were considered to be statistically significant.

Results were expressed as means SD, as median (range) for variables without normal distribution, or as percentage of patients with the analyzed characteristic. The SPSS 10.0 Professional software was used for the analyses.

RESULTS

From May 2000 to December 2001, 212 adult patients hospitalized in the ICU and medical-surgical wards were evaluated. From the 212 patients followed, 59 were excluded (13 patients receiving PN and 46 without PN) due to the absence of microbiological analysis of catheter tip in any of the CVCs used. None of the patients excluded presented any clinical signs (local or systemic) of CVC-related infection. The main baseline characteristics of excluded and nonexcluded patients were similar (male:female ratio, BMI, days of hospitalization before CVC insertion, serum albumin and ASIS score), but excluded patients were older (57 vs 62 years; ρ = .034) and less exposed to PN (22.0% vs 39.2%; p = .018) than nonexcluded patients. Then, data from 153 patients (60 exposed and 93 nonexposed to PN) followed for 1-99 days (days at risk per patient) were analyzed.

Twenty-eight patients presented CVC-related infection, which represents a cumulative incidence of 18.3% and an incidence density of 11.7 infections per 1000 catheter-days. Signs of infection at exit site with positive catheter-tip culture were observed in 22 patients. Purulence at the catheter exit site was present in 14 patients. In 8 patients, signs of infection at exit site and purulence were present. Mortality was 26.8% at 28 days (41 patients) and 39.9% during the follow-up period (61 patients).

Regarding the 286 inserted catheters (1.9 1.4 catheters per patient), 60.6% were single-lumen, 78.0% were inserted into the subclavian vein (right or left), and 73.5% were inserted at the right side (subclavian or jugular vein). Four mechanical complications occurred on CVC insertion: 3 pneumothorax events and 1 case of CVC occlusion. Double-lumen catheters (59.4% vs 36.6%; p = .013), jugular access (80.8% vs 63.0%; p = .036), and catheters used for PN (all types of CVC) (59.4% vs 37.7%; p = .018) were more frequent in patients with than in patients without CVC-related infection. No difference was observed regarding rightside access (70.4% vs 74.4%; p = .656) and days of CVC use (10.4 5.1 vs 9.4 6.1; p = .362) between the 2 groups.

CVC dressings were changed every 3.4 1.7 days, with no difference between patients with and without CVC-related infection (3.5 1.2 vs 3.4 1.9 days; p = .737). The dressing change interval was >7 days in 2 patients in the group without CVC-related infection and in none in the group with CVC-related infection (2.5% us 0%; p = 1.00).

TABLE I

Baseline clinical and laboratory characteristics of patients according to the presence of central venous catheter-related infection

In the 37 positive catheter-tip cultures (from 28 patients), 45 pathogens were isolated: 62.2% were Gram-positive microorganisms (coagulase-negative staphylococci, n = 13; Staphylococcus aureus, n = 12; others, n = 3); 33.3% were Gram-negative microorganisms (Klebsiellapneumoniae, n = 6; Acinetobacter, n = 3; and other microorganisms, n = 6). Candida albicans was cultured in 2 catheters (4.4%).

The baseline clinical and laboratory characteristics of patients with and without CVC-related infection are shown in Table I. The only difference between the groups was a higher frequency of PN exposure in patients with CVC-related infection. The variables analyzed during the study were similar in patients with and without CVC-related infection (Table II).

The incidence density of CVC-related infection was higher in patients with PN (16.02 infections/1000 days of CVC) than in patients without PN (8.31 infections/1000 days of CVC). The total number of days at risk per patient (interval between the day of catheter insertion [patient inclusion] and CVC-related infection [outcome] or catheter removal) for patients with CVC-related infection was 7 (range: 1-41) and for patients without CVC-related infection, 11 (range: 1-99). The cumulative incidence of CVC- related infection was 28.3% in exposed patients and 11.8% nonexposed patients (p = .010).

PN-exposed patients presented 17 episodes of CVC-related infection us 11 episodes in nonexposed patients (Figure 1; log rank = 3.01; p = .082).

As shown by multivariate analysis, PN (RR 3.30; 95% confidence interval [CI], 1.30-8.34; p = .012) was the only risk factor for CVC- related infection (dependent variable). Baseline malnutrition (RR = 0.45; 95% CI, 0.15-1.34; p = .152), time (days) before CVC insertion (relative risk (RR) = 1.00; 95% CI, 0.98-1.02; p = .801), and sustained hyperglycemia during the study (RR = 0.49; 95% CI, 0.98- 1.21; p = .091) were not significant in the model.

From 153 patients, 10 (6.5%) presented CVC-related BSI (bacteremia or sepsis). From those, 7 were receiving PN therapy. The incidence density of sepsis in patients receiving PN was 6.6 episodes of BSI/1000 catheter-days vs 2.3 episodes of BSI/1000 catheterdays in patients not exposed to PN.

Indications for PN (n = 60) in the present study were >7 days receiving NPO during the postoperative period (n = 23), postoperative fistulas (n = 19), pancreatitis (n = 7), short-bowel syndrome (n = 4), neoplasm (n = 4), chronic malnutrition (n = 1), or other conditions (n = 2). The mean number of days receiving PN was 15 (range: 1-101). Standard PN formulas were used for 48 out of 60 patients receiving PN (formula 1 = 36.5%; formula 2 = 32.7%; formula 3 = 23.1%), and in 12 patients, physicians prescribed PN acco\rding to each patient's estimated needs.

Table III describes the clinical and laboratory characteristics of patients at baseline and during the study according to PN use. The following were more frequent in PN patients: malnutrition, serum albumin levels <2.0 g/dL, sustained hyperglycemia, insulin use, CVC indication for surgical-related pathologies, and less severe clinical status (ASIS score). The number of inpatient days before CVC insertion was also higher in PN patients. On multiple logistic regression, after adjustment for ASIS score (OR = 0.33; 95% CI, 0.14- 0.80; p = .014), the following factors were associated with PN use: malnutrition at baseline (OR = 8.05; 95% CI, 1.85-35.03; p = .005), CVC indication for surgical-related pathology at baseline (OR = 7.26; 95% CI, 2.51-21.04; p < .001), sustained hyperglycemia during the study (OR = 4.34; 95% CI, 1.79-10.52; p = .001), and inpatient days before CVC insertion (OR = 1.04; 95% CI, 1.01-1.07; p = .004).

TABLE II

Clinical characteristics of patients according to the presence of central venous catheter-related infection

FIGURE 1. Central venous catheter-related infection in patients exposed and not exposed to parenteral nutrition therapy.

DISCUSSION

In the present study, PN was an independent risk factor for the development of CVC-related infection in adult hospitalized patients. The relative risk for development of infection was 3.30 in nonselected patients.

The role of PN as a risk factor for CVC-related infection has not been clearly established because data from noncontrolled5,6,11,25 or retrospective11,13,15,17,18,25 studies, or from studies focusing on specific groups of patients,9,11,13,14,10,17,19,26 cannot be generalized. Furthermore, as far as we know, the studies analyzing PN as a risk factor for infection using multivariate analysis13,17 were not designed to specifically evaluate CVC-related infection and included only ICU patients. One study17 compared 28 patients with bacteremia or sepsis resulting from various causes with 99 patients without infection. PN, CVC use, and ICU admission when nursing staff was less available were identified as risk factors for bacteremia and sepsis. However, in that retrospective study, the differences between patients with and without infection (such as antibiotic use and invasive procedures) were not taken into account in the multivariate analysis. Another retrospective study13 performed in a trauma ICU demonstrated that PN was a risk factor for nonspecific fungemia. In that study, the statistical analysis was not adjusted for the more severe clinical status and longer period receiving PN of patients with Candida infection (n = 20) when compared with control patients (n = 45).

In the present study, the cumulative incidence of 28.3% of CVC- related infection in patients receiving PN was similar to that observed in studies including PN patients only.8,27 There is a recommendation to express incidence as incidence density because this approach takes into account the number of days at risk for each studied patient and simplifies the comparison between studies.20 However, studies describing incidence density are scarce. An incidence density of 5.2 episodes of sepsis/1000 catheter-days was observed in one study.5 The difference in relation to the present results (4.2 episodes of sepsis/1000 catheter-days) could be explained by the fact that in that study, incidence density was determined according to number of catheters rather than number of patients (as in the present study). Furthermore, the almost exclusive use of multilumen catheters, which has been considered as a risk factor for CVC-related infection,16,19,25,28,29 may also have influenced those results.

TABLE III

Clinical and laboratory characteristics of patients with or without PN therapy use

In the present study, nutrition status was evaluated by the subjective global assessment method,22 according to which >80% of our patients were classified as malnourished. This evaluation seems to be more sensitive than a simple measure of BMI,30 which is influenced by other variables, such as edema or overhydration, and does not incorporate specific clinical aspects. In fact, the BMI of our patients was within the normal range (24.7 5.9 kg/m^sup 2^). Malnutrition was not a risk factor for CVC-related infection in this study, even on multivariate analysis. This observation is explained by the high frequency of malnutrition in patients both with and without CVC-related infection. Furthermore, although the presence of malnutrition can be associated with infectious complications in hospitalized patients,31 the relationship between infectious complications and nutrition status is not straightforward, because both might be a consequence of the primary disease.32

Increase in blood glucose levels is associated with higher frequency of infection,33 especially in critically ill patients.34 The intense contraregulatory hormone and cytokine responses to severe disease and the excessive administration of glucose, usually as PN, could contribute to hyperglycemia and associated infectious complications.34,35 In fact, we observed a higher frequency of sustained hyperglycemia and consequent insulin use in patients with PN compared with patients without PN. However, the frequency of sustained hyperglycemia was not different in patients with or without CVC-related infection, and sustained hyperglycemia was not an independent risk factor for CVC-related infection in multivariate Cox analysis.

One possible drawback of the present study could be that the CVC dressing was not changed at least weekly in all patients. However, this occurred in only 2 patients in the group without CVC-related infection (and in none of the patients in the infection group), and thus it is unlikely that this aspect had any influence on our results.

Unsteady clinical status (ASIS score 4 or 5) was not a risk factor for CVC-related infection in our study. The adopted study design, including patients from medical and surgical wards in addition to ICU patients, explains and underscores the relevance of our negative findings. Other investigators have reported similar results.5,6,13,18

The clinical and laboratory profile of our PN patients and the indications for this therapy seem to be representative of patients receiving PN.1,2,36 In the present study, CVC indication for surgical-related pathologies was associated with PN because in 85% (51/60) of patients, this therapy was initiated in the perioperative period. As expected, malnutrition was more frequent in PN patients. Also, the clinical status of our PN patients was less severe than that of the patients without PN. This is related to the fact that patients with unstable clinical status are not suitable for PN.37

The present study shows that PN was an independent risk factor for CVC-related infection. Some of the factors that could be hypothesized to explain the risk conferred by PN for CVC-related infection were probably ruled out: malnutrition, hyperglycemia, and severity of clinical status in PN patients. It is possible that the PN formula composition itself constitutes a risk factor for CVC- related infection. One hypothesis is that the contamination of particulates such as trace elements, not detected by routine microbiological methods, could be associated to CVC-related infection in patients receiving PN therapy.38,39

Finally, the high incidence density of CVC-related infection associated with PN could indicate the use of impregnated chlorhexidine/silver sulfadiazine or minocycline/rifampin catheters.20,40-42 In general, however, these impregnated catheters lose their antimicrobial activity within 14 days, whereas PN is used for longer periods of time.20

CONCLUSION

PN is an independent risk factor for CVC-related infection in nonselected adult hospitalized patients. Novel therapeutic strategies must be developed in order to prevent and reduce the incidence of CVC-related infection in patients receiving PN.

ACKNOWLEDGMENTS

The study was supported by grants from Programa de Apoio a Ncleos de Excelncia (PRONEX) and Hospital de Clnicas de Porto Alegre. LT was the recipient of an academic scholarship from CNPq.

REFERENCES

1. American Gastroenterological Association. American Gastroenterological Association medical position statement: parenteral nutrition. Gastroenterology. 2001;121:966-969.

2. Jeejeebhoy KN. Total parenteral nutrition: potion or poison? Am J Clin Nutr. 2001;74:160-163.

3. Braunschweig CL, Levy P, Sheean PM, Wang X. Enteral compared with parenteral nutrition: a meta-analysis. Am J Clin Nutr. 2001;74:534-542.

4. Fuhrman MP. Management of complications of parenteral nutrition. In: Matarese LE, Gottschlich MM, eds. Contemporary Nutrition Support Practice: A Clinical Guide. Philadelphia, PA: W.B. Saunders Company; 1998:243-263.

5. Savage AP, Picard M, Hopkins CC, Malt RA. Complications and survival of multilumen central venous catheters used for total parenteral nutrition. Br J Surg. 1993;80:1287-1290.

6. Freund HR, Rimon B. Sepsis during total parenteral nutrition. JPEN J Parenter Enteral Nutr. 1990;14:39-41.

7. Mitchell A, Atkins S, Royle GT, Kettlewel MGW. Reduced catheter sepsis and prolonged catheter life using a tunneled silicone rubber catheter for total parenteral nutrition. Br J Surg. 1982; 69:420-422.

8. Nehme AE. Nutritional support of the hospitalized patient: the team concept. JAMA. 1980;243:1906-1908.

9. Yeung CY, Lee HC, Huang FY, Wang CS. Sepsis during total parenteral nutrition: exploration of risk factors and determination of the effectiveness of peripherally inserted central venous catheters. Pediatr Infect Dis J. 1998; 17:135-142.

10. Ryan JA, Abel RM, Abbott WM, et al. Catheter complications in total parenteral nutrition: a prospective study of 200 consecutive patients. N Engl J Med. 1974;290:757-761.

11. Curry CR, Quie PG. Fungal septicemia in patients receiving parenteral hyperalimentat\ion. N Engl J Med. 1971;285:1221-1225.

12. Goldmann DA, Maki DG. Infection control in total parenteral nutrition. JAMA. 1973;223:1360-1364.

13. Borzotta AP, Beardsley K. Candida infection in critically ill trauma patients: a retrospective case-control study. Arch Surg. 1999;134:657-665.

14. Fridkin SK, Pear SM, Williamson TH, Galgiani JN, Jarwis WR. The role of understaffing in central venous catheter-associated bloodstream infections. Infect Control Hosp Epidemiol. 1996; 17: 150- 158.

15. Gumbo T, Isada CM, Hall G, Karafa MT, Gordon SM. Candida glabata fungemia: clinical features of 139 patients. Medicine. 1999;78:220-227.

16. Hilton E, Haslett TM, Borenstein MT, Tucci V, Isemberg HD, Singer C. Central catheter infections: single- versus triple-lumen catheters: influence of guide wires on infection rates when used for replacement of catheters. Am J Med. 1988;84:667-672.

17. Robert J, Fridkin SK, Blumberg H, et al. The influence of the composition of the nursing staff on primary bloodstream infection rates in a surgical intensive care unit. Infect Control Hosp Epidemiol. 2000;21:12-27.

18. Rose SG, Pitsch RJ, Karrer FW, Moor BJ. Subclavian catheter infections. JPEN J Parenter Enteral Nutr. 1986;12:511-512.

19. Yeung C, May J, Hughes R. Infections rate for single lumen v triple lumen subclavian catheters. Infect Control Hasp Epidemiol. 1988;9:154-158.

20. Centers for Disease Control and Prevention. Guidelines for the prevention of intravascular catheter-related infections. MMWR Morb Mortal Wkly Rep. 2002;51:1-29.

21. Maki DG. Skin as a source of nosocomial infection: directions for future research. Infect Control. 1986;7(suppl 2):S113-S117.

22. Detski AS, McLaughlin JR, Baker JP, et al. What is subjective global assessment of nutritional status? JPEN J Parenter Enteral Nutr. 1987;11:8-13.

23. American Diabetes Association. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. 2003;26(suppl 1):S5-S20.

24. Emori TG, Culver DH, Horan TC, et al. National nosocomial infections surveillance system (NNIS): description of surveillance methods. Am J Infect Control. 1991;19:19-35.

25. Kemp L, Burge J, Choban P, Harden J, Mirtallo J, Flancbaum L. The effect of catheter type and site on infection rates in total parenteral nutrition patients. JPEN J Parenter Enteral Nutr. 1994;18:71-74.

26. McKinnon PS, Goff DA, Kern JW, et al. Temporal assessment of Candida risk factors in the surgical intensive care unit. Arch Surg. 2001;136:1401-1409.

27. Faubion WC, Wesley JR, Khalidi N, Suva J. Total parenteral nutrition catheter sepsis: impact of the team approach. JPEN J Parenter Enteral Nutr. 1986;10:642-645.

28. Gil RT, Kruse JA, Thill-Baharozian MC, Carlson RW. Triple- us single-lumen central venous catheters. Arch Intern Med. 1989; 149:1139-1143.

29. Gowardman JR, Montgomery C, Thirlwell S, et al. Central venous catheter-related bloodstream infections: an analysis of incidence and risk factors in a cohort of 400 patients. Intensive Care Med. 1998;24:1034-1039.

30. Detsky AS, Baker JP, Mendelson RA, Wolman SL, Wesson DE, Jeejeebhoy KN. Evaluating the accuracy of nutritional assessment techniques applied to hospitalized patients: methodology and comparisons. JPEN J Parenter Enteral Nutr 1984;8:153-159.

31. Correia MITD, Waitzberg DL. The impact of malnutrition on morbidity, mortality, length of hospital stay and costs evaluated through a multivariate model analysis. Clin Nutr. 2003;22:235-239.

32. Naber THJ, Schermer T, Bree A, et al. Prevalence of malnutrition in nonsurgical hospitalized patients and its association with disease complications. Am J Clin Nutr. 1997;66:1232- 1239.

33. Joshi N, Mahajan N. Infection and diabetes. In: Pickup JC, Williams G, eds. Textbook of Diabetes. Oxford: Blackwell Science Ltd.; 2003:1-16.

34. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med. 2001;345:1359-1367.

35. van den Berghe G, Wouters PJ, Boullion R, et al. Outcome benefit of intensive insulin therapy in the critically ill: insulin dose versus glycemic control. Crit Care Med. 2003;31:359-366.

36. Heyland DH, MacDonald S, Keefe L, Drover JW. Total parenteral nutrition in the critically ill patient: a meta-analysis. JAMA. 1998;280:2013-2019.

37. Chan S, McCowen KC, Blackburn GL. Nutritional management in the ICU. Chest. 1999;115(suppl):145S-148S.

38. Ball PA, Bethune K, Fox J, Ledger R, Barnett M. Particulate contamination in parenteral nutrition solutions: still a cause for concern? Nutrition. 2001;17:926-929.

39. Failla ML, Benedict CD, Weinberg ED. Bacterial and fungal growth in total parenteral solutions. Antonie van Leeuwenhoek. 1975;41:319-328.

40. Pluhator-Murton MM, Fedorak RN, Audette RJ, Marriage BJ, Yatscoff RW, Gramlich LM. Trace element contamination of total parenteral nutrition. 1. Contribution of component solutions. JPEN J Parenter Enterai Nutr. 1999;23:222-227.

41. O'Grandy NP, Gerberding JL, Weintein RA, Masur H. Patients' safety and the science of prevention: the time for implementing the Guidelines for Prevention of Intravascular Catheter-Related Infections is now. Crit Care Med. 2003;31:291-292.

42. Veenstra DL, Saint S, Saha S, Lumley T, Sullivan SD. Efficacy of antiseptic-impregnated central venous catheters in preventing catheter-related bloodstream infection: a meta-analysis. JAMA. 1999;281:261-267.

Mariur G. Beghetto, RN*; Josu Victorino, MD[dagger]; Luciana Teixeira, RN*; and Mirela J. de Azevedo, MD[double dagger]

From the * Nursing, [dagger] Intensive Care, and [double dagger] Endocrine Divisions, Hospital de Clnicas de Porto Alegre, Federal University of Rio Grande do Sul, Porto Alegre, Brazil

Received for publication March 10, 2004.

Accepted for publication April 22, 2005.

Correspondence: Mirela J. de Azevedo, MD, Service de Endocrinologia, Hospital de Clnicas de Porto Alegre, Rua Ramiro Barcelos 2350/Prdio 12/4 andar, 90035-003 Porto Alegre, RS, Brazil. Electronic mail may be sent to mirelaazevedo@terra.com.br.

Copyright American Society for Parenteral and Enteral Nutrition Sep/ Oct 2005

Source: JPEN, Journal of Parenteral and Enteral Nutrition

More News in this Category