Prospective Study of the Value of Quantitative Culture of Organisms from Blood Collected through Central Venous Catheters in Differentiating
http://www.100md.com
微生物临床杂志 2006年第5期
University of Texas M. D. Anderson Cancer Center
Baylor College of Medicine and Veterans Affairs Medical Center, Houston, Texas
Division of Internal Medicine, Department of Infectious Diseases, School of Medicine, University of Crete, Crete, Greece
ABSTRACT
Collection of blood through a central venous catheter for the diagnosis of bacteremia is a debated topic. Quantitative cultures of organisms from blood collected through central venous catheters were found to be highly sensitive, specific, and predictive of bacteremia, especially when a cutoff point of 15 colonies of skin organisms was used.
TEXT
Drawing of blood through a central venous catheter (CVC) for the diagnosis of bacteremia is highly debated (2, 4, 11-13, 15) due to the possibility of culturing blood contaminated by organisms adhering to CVC lumen. Although quantitative blood cultures (QBC) collected simultaneously through a CVC and peripheral venipuncture (PV) have been used for the diagnosis of catheter-related bloodstream infections (3, 6, 14), the usefulness of QBC collected through CVC for the diagnosis of bacteremia from any source has not been thoroughly investigated.
Between September 1999 and May 2002, we followed up adult cancer patients who required new insertions of a peripherally inserted or subclavian silicone CVC at the M. D. Anderson Cancer Center in order to evaluate the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of QBC collected through CVC for the diagnosis of bacteremia. Maximal sterile barrier precautions were followed during catheter insertions (9). Patients were followed up until catheter removal or up to 100 days, whichever occurred first. At the onset of fever or the suspicion of catheter-related bloodstream infections, simultaneous QBC collected through CVC and PV were obtained. When the CVC remained in place beyond 100 days, blood cultures were performed, even in the absence of signs and symptoms of infection (SSI). Microorganisms were identified according to standard methods (8). An initial 10 ml of CVC-collected blood was discarded to avoid contact with previously infused drugs with antimicrobial activity. Subsequently, two 20-ml blood samples were drawn for QBC, one through the CVC and the other through PV, and a 10-ml portion of each was cultured aerobically (Bactec 9240, Bactec Plus Aerobic/F; BD Diagnostic Systems, Sparks, MD). At our institution, anaerobes constitute less than 0.1% of positive cultures; most of these are found to be possible contaminants. The remaining 10 ml was placed into isolator tubes (Isolator 10; Wampole, Cranbury, NJ) to be cultured quantitatively (lysis centrifugation method) (5). We defined bacteremia according to guidelines from the Centers for Disease Control and Prevention (CDC) (7). Possible contaminating microorganisms included skin microorganisms such as coagulase-negative staphylococci (CNS), diphtheroids, and Bacillus spp., whereas pathogenic microorganisms included Staphylococcus aureus, alpha-hemolytic Streptococcus spp., gram-negative bacilli, and Candida spp. A QBC collected through CVC was considered a true positive if (i) a pathogen was isolated in the presence of SSI such as fever, hypotension, and rigors or when the same pathogen was isolated from a peripheral blood culture within 48 h or (ii) a skin contaminant was isolated in the presence of SSI and the successful use of appropriate antibiotics or in the presence of SSI and the isolation of the same microorganism from a peripheral blood culture. A QBC collected through CVC was considered a false positive if (i) a pathogen was isolated in the absence of SSI and in the absence of a positive peripheral blood culture or (ii) a skin contaminant was isolated in the absence of SSI. A negative QBC collected through CVC was considered a true negative in the presence of a negative QBC of PV. A QBC collected through CVC was considered a false negative if (i) a QBC of PV was positive, within 48 h, for a pathogen in the presence of SSI or if at least two QBC of PV were positive for the same pathogen or (ii) a single QBC of PV was positive, within 48 h, for a contaminant in the presence of SSI and response to appropriate antibiotics or if two QBC of PV were positive for the same contaminant in the presence of SSI.
We evaluated 165 QBC collected through CVC. Most patients were males (65%) with a mean age of 51 years (±14.7 years), 57% had solid tumors, 23% were neutropenic (absolute neutrophil count, 500 cells/μl), and 76% of QBC collected through CVC were collected through subclavian CVC. Antibiotics were administered to 38% of the patients according to their blood culture results. True-positive QBC collected through CVC were identified in 96% (43/45) of patients with bacteremia (positive QBC of PV), compared with 6% (7/120) of those without a positive QBC of PV (false positive; P < 0.001). The majority (27; 60%) of single pathogens responsible for bacteremia were gram positive, 18 (67%) of which were CNS and 4 (15%) of which were S. aureus isolates. Single gram-negative microorganisms were responsible for eight (18%) cases of bacteremia, while nine (20%) cases were caused by polymicrobes. There was only one case of fungemia (Candida albicans). All seven false-positive QBC collected through CVC were due to skin microorganisms, six of which were CNS and one of which was Corynebacterium species, and five out of seven cultures had <10 CFU. Of the 115 (70%) negative QBC collected through CVC, 113 (98%) were true negatives. In the two cases with false-negative QBC collected through CVC, QBC of PV revealed CNS in the presence of SSI, with response to antibiotic therapy. Therefore, QBC collected through CVC were shown to have 96% sensitivity, 94% specificity, 86% PPV, and 98% NPV for diagnosing true bacteremia (Table 1). A QBC positive for CNS with 15 CFU was associated with 96% sensitivity and 99% specificity for diagnosing true bacteremia, with a PPV and NPV of 98% each.
Blood culture is the most widely used test to detect bacteremia in a clinical setting, where the microbial inoculum may affect its positivity. Blood cultures collected through CVC are associated with higher rates of contamination (1, 11) and lower specificity and PPV (4) compared with those collected via PV. A high probability of contamination is usually associated with the isolation of organisms such as Corynebacterium spp., Propionibacterium spp., and CNS, whereas low probability (<10%) is associated with such organism as S. aureus, Pseudomonas aeruginosa, and C. albicans (14). In particular, CNS have been proven to be the most frequent contaminant (10). Since the contamination of a blood sample is usually associated with a low inoculum, QBC could be useful in differentiating contamination from bacteremia. Therefore, the utilization of clinical findings with stricter laboratory criteria may contribute to a more accurate interpretation of blood culture results, especially those that are positive for CNS. Our study indicates that QBC collected through CVC, with a cutoff point of >15 CFU/ml, could be a useful laboratory criterion, together with positive clinical findings, for differentiating true bacteremia from false-positive contaminated blood cultures. The sensitivity (96%), specificity (94%), PPV (86%), and NPV (98%) of QBC collected through CVC in this study are of similar value to those obtained through peripherally collected samples. However, the common practice of performing simultaneous quantitative and qualitative cultures of blood drawn through the CVC and peripheral vein should continue until these findings are confirmed through larger studies.
(Part of the scientific work presented in this paper was presented as a poster at the 43rd Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, Ill., 14 to 17 September 2003.)
ACKNOWLEDGMENTS
There is no financial support associated with the manuscript, and none of the authors claim any conflict of interest.
REFERENCES
Bates, D. W., L. Goldman, and T. H. Lee. 1991. Contaminant blood cultures and resource utilization. The true consequences of false-positive results. JAMA 265:365-369.
Bryant, J. K., and C. L. Strand. 1987. Reliability of blood cultures collected from intravascular catheter versus venipuncture. Am. J. Clin. Pathol. 88:113-116.
Capdevila, J. A., A. M. Planes, M. Palomar, I. Gasser, B. Almirante, A. Pahissa, E. Crespo, and J. M. Martinez-Vazquez. 1992. Value of differential quantitative blood cultures in the diagnosis of catheter-related sepsis. Eur. J. Clin. Microbiol. Infect. Dis. 11:403-407.
DesJardin, J. A., M. E. Falagas, R. Ruthazer, J. Griffith, D. Wawrose, D. Schenkein, K. Miller, and D. R. Snydman. 1999. Clinical utility of blood cultures drawn from indwelling central venous catheters in hospitalized patients with cancer. Ann. Intern. Med. 131:641-647.
Dorn, G. L., G. A. Land, and G. E. Wilson. 1979. Improved blood culture technique based on centrifugation: clinical evaluation. J. Clin. Microbiol. 9:391-396.
Douard, M. C., G. Arlet, G. Leverger, R. Paulien, C. Waintrop, E. Clementi, B. Eurin, and G. Schaison. 1991. Quantitative blood cultures for diagnosis and management of catheter-related sepsis in pediatric hematology and oncology patients. Intensive Care Med. 17:30-35.
Garner, J. S., W. R. Jarvis, T. G. Emori, T. C. Horan, and J. M. Hughes. 1988. CDC definitions for nosocomial infections, 1988. Am. J. Infect. Control 16:128-140.
Isenberg, H. D. (ed.). 1992. Clinical microbiology procedures handbook, vol. 1, p. 1.1.1-1.20.44. ASM Press, Washington, D.C.
Raad, I. I., D. C. Hohn, B. J. Gilbreath, N. Suleiman, L. A. Hill, P. A. Bruso, K. Marts, P. F. Mansfield, and G. P. Bodey. 1994. Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during insertion. Infect. Control Hosp. Epidemiol. 15:231-238.
Rogers, M. S., and B. A. Oppenheim. 1998. The use of continuous monitoring blood culture systems in the diagnosis of catheter related sepsis. J. Clin. Pathol. 51:635-637.
Tafuro, P., D. Colbourn, I. Gurevich, P. Schoch, H. Wachs, S. Krystofiak, and B. A. Cunha. 1986. Comparison of blood cultures obtained simultaneously by venepuncture and from vascular lines. J. Hosp. Infect. 7:283-288.
Tonnesen, A., M. Peuler, and W. R. Lockwood. 1976. Cultures of blood drawn by catheters vs venipuncture. JAMA 235:1877.
Weinstein, M. P. 1996. Current blood culture methods and systems: clinical concepts, technology, and interpretation of results. Clin. Infect. Dis. 23:40-46.
Wing, E. J., C. W. Norden, R. K. Shadduck, and A. Winkelstein. 1979. Use of quantitative bacteriologic techniques to diagnose catheter-related sepsis. Arch. Intern. Med. 139:482-483.
Wormser, G. P., I. M. Onorato, T. J. Preminger, D. Culver, and W. J. Martone. 1990. Sensitivity and specificity of blood cultures obtained through intravascular catheters. Crit. Care Med. 18:152-156.(Ioannis Chatzinikolaou, H)
Baylor College of Medicine and Veterans Affairs Medical Center, Houston, Texas
Division of Internal Medicine, Department of Infectious Diseases, School of Medicine, University of Crete, Crete, Greece
ABSTRACT
Collection of blood through a central venous catheter for the diagnosis of bacteremia is a debated topic. Quantitative cultures of organisms from blood collected through central venous catheters were found to be highly sensitive, specific, and predictive of bacteremia, especially when a cutoff point of 15 colonies of skin organisms was used.
TEXT
Drawing of blood through a central venous catheter (CVC) for the diagnosis of bacteremia is highly debated (2, 4, 11-13, 15) due to the possibility of culturing blood contaminated by organisms adhering to CVC lumen. Although quantitative blood cultures (QBC) collected simultaneously through a CVC and peripheral venipuncture (PV) have been used for the diagnosis of catheter-related bloodstream infections (3, 6, 14), the usefulness of QBC collected through CVC for the diagnosis of bacteremia from any source has not been thoroughly investigated.
Between September 1999 and May 2002, we followed up adult cancer patients who required new insertions of a peripherally inserted or subclavian silicone CVC at the M. D. Anderson Cancer Center in order to evaluate the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of QBC collected through CVC for the diagnosis of bacteremia. Maximal sterile barrier precautions were followed during catheter insertions (9). Patients were followed up until catheter removal or up to 100 days, whichever occurred first. At the onset of fever or the suspicion of catheter-related bloodstream infections, simultaneous QBC collected through CVC and PV were obtained. When the CVC remained in place beyond 100 days, blood cultures were performed, even in the absence of signs and symptoms of infection (SSI). Microorganisms were identified according to standard methods (8). An initial 10 ml of CVC-collected blood was discarded to avoid contact with previously infused drugs with antimicrobial activity. Subsequently, two 20-ml blood samples were drawn for QBC, one through the CVC and the other through PV, and a 10-ml portion of each was cultured aerobically (Bactec 9240, Bactec Plus Aerobic/F; BD Diagnostic Systems, Sparks, MD). At our institution, anaerobes constitute less than 0.1% of positive cultures; most of these are found to be possible contaminants. The remaining 10 ml was placed into isolator tubes (Isolator 10; Wampole, Cranbury, NJ) to be cultured quantitatively (lysis centrifugation method) (5). We defined bacteremia according to guidelines from the Centers for Disease Control and Prevention (CDC) (7). Possible contaminating microorganisms included skin microorganisms such as coagulase-negative staphylococci (CNS), diphtheroids, and Bacillus spp., whereas pathogenic microorganisms included Staphylococcus aureus, alpha-hemolytic Streptococcus spp., gram-negative bacilli, and Candida spp. A QBC collected through CVC was considered a true positive if (i) a pathogen was isolated in the presence of SSI such as fever, hypotension, and rigors or when the same pathogen was isolated from a peripheral blood culture within 48 h or (ii) a skin contaminant was isolated in the presence of SSI and the successful use of appropriate antibiotics or in the presence of SSI and the isolation of the same microorganism from a peripheral blood culture. A QBC collected through CVC was considered a false positive if (i) a pathogen was isolated in the absence of SSI and in the absence of a positive peripheral blood culture or (ii) a skin contaminant was isolated in the absence of SSI. A negative QBC collected through CVC was considered a true negative in the presence of a negative QBC of PV. A QBC collected through CVC was considered a false negative if (i) a QBC of PV was positive, within 48 h, for a pathogen in the presence of SSI or if at least two QBC of PV were positive for the same pathogen or (ii) a single QBC of PV was positive, within 48 h, for a contaminant in the presence of SSI and response to appropriate antibiotics or if two QBC of PV were positive for the same contaminant in the presence of SSI.
We evaluated 165 QBC collected through CVC. Most patients were males (65%) with a mean age of 51 years (±14.7 years), 57% had solid tumors, 23% were neutropenic (absolute neutrophil count, 500 cells/μl), and 76% of QBC collected through CVC were collected through subclavian CVC. Antibiotics were administered to 38% of the patients according to their blood culture results. True-positive QBC collected through CVC were identified in 96% (43/45) of patients with bacteremia (positive QBC of PV), compared with 6% (7/120) of those without a positive QBC of PV (false positive; P < 0.001). The majority (27; 60%) of single pathogens responsible for bacteremia were gram positive, 18 (67%) of which were CNS and 4 (15%) of which were S. aureus isolates. Single gram-negative microorganisms were responsible for eight (18%) cases of bacteremia, while nine (20%) cases were caused by polymicrobes. There was only one case of fungemia (Candida albicans). All seven false-positive QBC collected through CVC were due to skin microorganisms, six of which were CNS and one of which was Corynebacterium species, and five out of seven cultures had <10 CFU. Of the 115 (70%) negative QBC collected through CVC, 113 (98%) were true negatives. In the two cases with false-negative QBC collected through CVC, QBC of PV revealed CNS in the presence of SSI, with response to antibiotic therapy. Therefore, QBC collected through CVC were shown to have 96% sensitivity, 94% specificity, 86% PPV, and 98% NPV for diagnosing true bacteremia (Table 1). A QBC positive for CNS with 15 CFU was associated with 96% sensitivity and 99% specificity for diagnosing true bacteremia, with a PPV and NPV of 98% each.
Blood culture is the most widely used test to detect bacteremia in a clinical setting, where the microbial inoculum may affect its positivity. Blood cultures collected through CVC are associated with higher rates of contamination (1, 11) and lower specificity and PPV (4) compared with those collected via PV. A high probability of contamination is usually associated with the isolation of organisms such as Corynebacterium spp., Propionibacterium spp., and CNS, whereas low probability (<10%) is associated with such organism as S. aureus, Pseudomonas aeruginosa, and C. albicans (14). In particular, CNS have been proven to be the most frequent contaminant (10). Since the contamination of a blood sample is usually associated with a low inoculum, QBC could be useful in differentiating contamination from bacteremia. Therefore, the utilization of clinical findings with stricter laboratory criteria may contribute to a more accurate interpretation of blood culture results, especially those that are positive for CNS. Our study indicates that QBC collected through CVC, with a cutoff point of >15 CFU/ml, could be a useful laboratory criterion, together with positive clinical findings, for differentiating true bacteremia from false-positive contaminated blood cultures. The sensitivity (96%), specificity (94%), PPV (86%), and NPV (98%) of QBC collected through CVC in this study are of similar value to those obtained through peripherally collected samples. However, the common practice of performing simultaneous quantitative and qualitative cultures of blood drawn through the CVC and peripheral vein should continue until these findings are confirmed through larger studies.
(Part of the scientific work presented in this paper was presented as a poster at the 43rd Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, Ill., 14 to 17 September 2003.)
ACKNOWLEDGMENTS
There is no financial support associated with the manuscript, and none of the authors claim any conflict of interest.
REFERENCES
Bates, D. W., L. Goldman, and T. H. Lee. 1991. Contaminant blood cultures and resource utilization. The true consequences of false-positive results. JAMA 265:365-369.
Bryant, J. K., and C. L. Strand. 1987. Reliability of blood cultures collected from intravascular catheter versus venipuncture. Am. J. Clin. Pathol. 88:113-116.
Capdevila, J. A., A. M. Planes, M. Palomar, I. Gasser, B. Almirante, A. Pahissa, E. Crespo, and J. M. Martinez-Vazquez. 1992. Value of differential quantitative blood cultures in the diagnosis of catheter-related sepsis. Eur. J. Clin. Microbiol. Infect. Dis. 11:403-407.
DesJardin, J. A., M. E. Falagas, R. Ruthazer, J. Griffith, D. Wawrose, D. Schenkein, K. Miller, and D. R. Snydman. 1999. Clinical utility of blood cultures drawn from indwelling central venous catheters in hospitalized patients with cancer. Ann. Intern. Med. 131:641-647.
Dorn, G. L., G. A. Land, and G. E. Wilson. 1979. Improved blood culture technique based on centrifugation: clinical evaluation. J. Clin. Microbiol. 9:391-396.
Douard, M. C., G. Arlet, G. Leverger, R. Paulien, C. Waintrop, E. Clementi, B. Eurin, and G. Schaison. 1991. Quantitative blood cultures for diagnosis and management of catheter-related sepsis in pediatric hematology and oncology patients. Intensive Care Med. 17:30-35.
Garner, J. S., W. R. Jarvis, T. G. Emori, T. C. Horan, and J. M. Hughes. 1988. CDC definitions for nosocomial infections, 1988. Am. J. Infect. Control 16:128-140.
Isenberg, H. D. (ed.). 1992. Clinical microbiology procedures handbook, vol. 1, p. 1.1.1-1.20.44. ASM Press, Washington, D.C.
Raad, I. I., D. C. Hohn, B. J. Gilbreath, N. Suleiman, L. A. Hill, P. A. Bruso, K. Marts, P. F. Mansfield, and G. P. Bodey. 1994. Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during insertion. Infect. Control Hosp. Epidemiol. 15:231-238.
Rogers, M. S., and B. A. Oppenheim. 1998. The use of continuous monitoring blood culture systems in the diagnosis of catheter related sepsis. J. Clin. Pathol. 51:635-637.
Tafuro, P., D. Colbourn, I. Gurevich, P. Schoch, H. Wachs, S. Krystofiak, and B. A. Cunha. 1986. Comparison of blood cultures obtained simultaneously by venepuncture and from vascular lines. J. Hosp. Infect. 7:283-288.
Tonnesen, A., M. Peuler, and W. R. Lockwood. 1976. Cultures of blood drawn by catheters vs venipuncture. JAMA 235:1877.
Weinstein, M. P. 1996. Current blood culture methods and systems: clinical concepts, technology, and interpretation of results. Clin. Infect. Dis. 23:40-46.
Wing, E. J., C. W. Norden, R. K. Shadduck, and A. Winkelstein. 1979. Use of quantitative bacteriologic techniques to diagnose catheter-related sepsis. Arch. Intern. Med. 139:482-483.
Wormser, G. P., I. M. Onorato, T. J. Preminger, D. Culver, and W. J. Martone. 1990. Sensitivity and specificity of blood cultures obtained through intravascular catheters. Crit. Care Med. 18:152-156.(Ioannis Chatzinikolaou, H)