Safety
of a Guidewire
Technique
for Replacement
of Pulmonary
Alfons Bach, MD, Hubert B6hrer, MD, DEAA, Heinrich The purpose of this study was to determine if a guidewire change from a pulmonary artery catheter (PAC) to a central venous catheter (CVC) poses a significant infection risk. A total of 128 consecutive cardiac surgical patients with PACs inserted in the operating room were entered into this study. Postoperatively, patients were randomly allocated to receive a double-lumen CVC, either at the initial introducer insertion site over a guidewire, or at a new site with de novo catheterization. The tips of all introducers, PACs, and CVCs were cut off, cultured, and semi-quantatively analyzed. The results show that insertion of CVCs over a guidewire within 48 hours after initial venapuncture is no more likely to be associated with catheter colonization than is de novo percutaneous insertion at a different site. From 48 hours up to 72
C
ARDIAC surgical patients often require invasive monitoring with a pulmonary artery catheter (PAC).l,z With postoperative cardiovascular stabilization, the PAC may be removed during the following days. However, many patients still require central venous access for diagnostic and therapeutic purposes. De novo insertion of a central venous catheter (CVC) exposes the patient to the wellknown risks of percutaneous catheter placement, such as arterial puncture, hematoma formation, or pneumothorax.3-6 The goal of this study was to determine if changing PACs to CVCs over a guidewire through the same site of insertion poses a significant microbiologic risk. In a previous study, significant differences were found between the guidewire and the de novo technique,’ but due to the study design the rate of catheter-related bacteremia could not be assessed. In the present study, routine blood cultures were included at the time of catheter withdrawal in order to detect catheter-related bacteremia, as suggested by Plit et al8
METHODS A total of 128 cardiac surgical patients with PACs (Edwards Lab, Santa Ana, CA) were entered into this study after the University’s Institutional Review Board for Human Investigations approved the protocol. Informed consent was obtained from each patient or from the person legally responsible for the patient. All consecutive patients with a PAC for cardiac surgical procedures in whom this monitoring could be replaced by a CVC within the first 3 postoperative days according to the judgment of the intensivists were studied. Patients who required prolonged PAC monitoring beyond 72 hours were excluded. Patients showing clinical signs of infection before initial PAC placement were excluded. Thus, 4 patients with endocarditis could not be entered in this study. All participants, ie, anesthesiologists, intensivists, microbiologists, and nursing staff were informed about the study protocol. All PACs had been inserted percutaneously via the right or left internal jugular vein by experienced anesthesiologists in the operating room. Special attention was given to aseptic insertion conditions, ie, washing the puncture site three times with povidoneiodine and having the operator wear gloves, gown, hat, and mask. Catheters were introduced through 8.5-F sideport sheath introducers (Arrow Inc, Reading, PA), which were left in place. The protective anticontamination sleeve of the introducer set was attached to the introducer hub and extended to cover as much of the catheters as possible. Postoperatively, patients were treated in
Artery Catheters
K. Geiss, MD
hours following initial insertion of the PAC. an incidence of catheter-related infection of 35.3% was observed in the guidewire group, as opposed to 12.5% in the de novo group. It is recommended that the use of a guidewire technique for catheter replacement (PAC to CVC) is a safe alternative to de novo insertion of a CVC within 48 hours after initial insertion of the PAC. In order to minimize the potential risk of catheter-related infection and bacteremia in cardiac surgical patients, de novo catheterization beyond 48 hours after initial venapuncture is suggested. Copyright 0 1992 by W.B. Saunders Company KEY WORDS: catheter-related infection, ment, pulmonary artery catheters
guidewire
replace-
an open intensive care unit (ICU), with adherence to a strict catheter care protocol. Dressing with nonocclusive sterile gauze was changed daily and the skin puncture site was thoroughly inspected for signs of local infection. Soiled dressings were changed immediately. No topical antiseptic or antibiotic was applied. Manipulation of catheters was allowed only after desinfecting hands with an alcoholic solution or wearing sterile gloves. Intravenous tubings were changed every 48 hours by ICU nurses wearing sterile gloves. Cefuroxime was given to patients for routine prophylaxis during the 24-hour perioperative period. When PAC monitoring was no longer required, the catheter was removed within 72 hours after primary placement. Patients were randomly allocated to receive a double-lumen CVC (Arrow Inc, Reading, PA) either at the initial introducer insertion site over a guidewire, or at a new site with de novo placement using a sterile technique. The skin was swabbed three times with povidone-iodine solution and allowed to dry. Sterile covers were placed around the site of insertion. The intensivist wore sterile gloves, hat, mask, and gown after desinfecting his hands. All parts of the opened catheter set were handled sterile by the intensivist and the assisting nurse. After puncture of the opposite jugular vein the catheter was introduced and secured by skin sutures. A nonocclusive sterile gauze dressing covered the puncture site. The procedure for changing catheters over a guidewire was as follows. The PAC and the protective sleeve with all dressings were removed completely while the intensivist was wearing sterile clothes. A guidewire (Arrow Inc, Reading, PA) was then inserted into the introducer sheath. The skin suture was cut off with sterile scissors and the introducer was withdrawn, leaving the guidewire in place. After swabbing the wire with povidone-iodine solution, the double-lumen catheter was advanced over it. The correct position of the catheter was confirmed by a chest film, and the catheter was secured with sutures. A nonocclusive sterile dressing was used to cover the puncture site. All CVCs were removed 7 days after secondary placement according to the standards of care in the ICU. The puncture site was cleaned with an :alcohol disinfectant, and the catheters were withdrawn with sterile forceps. The distal 3 cm of all introducers, PACs and CVCs were cut off with sterile scissors immediately after removal. The tips were collected in sterile tubes, labelled, and immediately transported to the microbiology laboratory. The
From the Departments OfAnesthesiology and Microbiology University of Heidelberg, Heidelberg, Germany. Reprints are not available. Copyright Q 1992 by WB. Saunders Company 1053-0770192/0606-0014$03.0010
Journalof Cardiothoracic and VascularAnesthesia, Vol6, No 6 (December),1992: pp 711-714
711
BACH, BGHRER, AND GEISS
712
Table 1. Semiquantitative
Results of Introducer and Pulmonary Artery Catheter Tips PACS
Introducers PAC in
15cfu
il5CfU
0% (O/87)
1.1% (l/87)
24 hours
0% (O/28)
2.3% (2187)
0% (O/87)
3.6% (l/28)
0% (O/28) 9.1% (3133)
24.2% (8/33)
6.1% (2/33)
NOTE. Figures in brackets are the ratio of infected catheters with greater than 15 cfu per plate to all catheters
Introducers and PACs removed at various intervals after initial insertion showed comparable microbiological results (Table l), the rate of catheter tips growing more than 15 cfu depending on the period in situ. CVCs placed via guidewire change, within 48 hours after initial placement, exhibited similar rates of infected catheter tips as compared to de novo placed catheters (Table 2). However, after 48 hours, rates were 35.3% of catheter tips with more than 15 cfu in the guidewire group, as opposed to 12.5% in the de novo group (difference significant, x2 analysis, P < 0.05). Similar results were obtained in catheter segments growing less than 15 cfu (Table 3). In the de novo group, in which catheter change took place more than 48 hours after PAC insertion, no positive blood cultures with the same microorganism as the catheter tip were obtained at catheter withdrawal after 7 days. In the corresponding guidewire subgroup, blood cultures were positive at the end of the catheter life in two patients showing intermittent fever greater than 38.5”C with leucocy tosis greater than 15,0OO/l.~L. In two other patients, blood cultures were positive with the same microorganism at days 5 and 6, these patients showing signs of infection with rectal body temperature greater than 38.5”C and leukocytes greater than 15,OOO/uL. In all 4 of these patients, temperature dropped below 37.5”C within 12 hours after catheter
removal and leucocytosis resolved, supporting the assumption of catheter-related infection in these cases. In these patients, PAC to CVC change took place 49,50,69, and 71 hours after initial insertion of the PAC. In this study, there were 38 microbiological isolates from PAC and CVC tips, with no catheter yielding more than one organism. The most frequent organisms cultured from catheters were Staphylococcus epidemidis (26 cultures), followed by S aureus (7 cultures). In the guidewire group with secondary catheter placement after 48 hours. a total of 6 patients showed catheter tips with greater than 15 cfu. Organisms isolated were S aureus in 2, S epidermidis in 2, Enterococcus in 1. and Klebsiella pneumoniae in 1 patient. In this subgroup of patients, S aureus could be isolated from peripheral blood in one case, S epidemidis in two cases, and Enterococcus in one case. The identical microorganisms could be cultured in the corresponding catheter tips with more than 15 cfu on blood agar plate. All these patients presented signs of catheter-related infections, ie, fever and leucocytosis. All other patients with positive tips, in either group, showed no clinical or microbiological signs of catheterrelated infection. Daily inspection of the insertion sites revealed only minor inflammation with no signs of local infection, ie, purulent discharge or cellulitis. Routine cultures of sputum, wound swabs, and urine samples were made twice a week. In the 4 patients with clinical signs of catheter-related infection and catheterrelated bacteremia, these cultures remained sterile or produced microorganisms differing from those found in catheter tip or blood cultures. However, none of these patients exhibited signs of septic shock. Furthermore, no patient developed signs of endocarditis, ie, remittent fever in association with a significant new cardiac murmur or clinical and echocardiographic signs of valve insufficiency. During initial and secondary placement of catheters, no serious complications such as pneumothorax, hemothorax, life-threatening arrhythmias, or other problems occurred. Minor complications such as local hematoma, transient
Table 2. Central Venous Catheter Tips Having Greater Than 15 cfu
Table 3. Central Venous Catheter Tips Having Less Than 15 cfu on
segments were rolled on blood agar plates and analyzed semiquantitively according to the method of Maki.’ A threshold of 15 colony-forming units (cfu) per agar plate was assumed to distinguish colonized from infected catheters. Organisms were identified according to standard microbiological techniques. Blood samples for culture were taken in duplicate from a peripheral vein within 15 minutes after catheter removal and cultured. Further blood samples for blood cultures were obtained by venapuncture if the patient exhibited signs of systemic infection, eg, rectal temperature greater than 38.5”C, or leukocyte count > 15,OOO/uL with no apparent cause within 24 hours after their occurrence. Statistical interpretation of the data was achieved using x’ analysis, with Yates correction factor, where appropriate. Statistical significance was accepted at P < 0.05. RESULTS
on the Plate PAC in
place
Guidewlre
the Agar Plate and/or PAC in
De Now
place
Positive in Broth
Guidewire
De Now
4.7% (2144)
24 hours
2.3% (l/43)
2.3% (1144)
of a Guidewire
Technique
for Replacement
of Pulmonary
Alfons Bach, MD, Hubert B6hrer, MD, DEAA, Heinrich The purpose of this study was to determine if a guidewire change from a pulmonary artery catheter (PAC) to a central venous catheter (CVC) poses a significant infection risk. A total of 128 consecutive cardiac surgical patients with PACs inserted in the operating room were entered into this study. Postoperatively, patients were randomly allocated to receive a double-lumen CVC, either at the initial introducer insertion site over a guidewire, or at a new site with de novo catheterization. The tips of all introducers, PACs, and CVCs were cut off, cultured, and semi-quantatively analyzed. The results show that insertion of CVCs over a guidewire within 48 hours after initial venapuncture is no more likely to be associated with catheter colonization than is de novo percutaneous insertion at a different site. From 48 hours up to 72
C
ARDIAC surgical patients often require invasive monitoring with a pulmonary artery catheter (PAC).l,z With postoperative cardiovascular stabilization, the PAC may be removed during the following days. However, many patients still require central venous access for diagnostic and therapeutic purposes. De novo insertion of a central venous catheter (CVC) exposes the patient to the wellknown risks of percutaneous catheter placement, such as arterial puncture, hematoma formation, or pneumothorax.3-6 The goal of this study was to determine if changing PACs to CVCs over a guidewire through the same site of insertion poses a significant microbiologic risk. In a previous study, significant differences were found between the guidewire and the de novo technique,’ but due to the study design the rate of catheter-related bacteremia could not be assessed. In the present study, routine blood cultures were included at the time of catheter withdrawal in order to detect catheter-related bacteremia, as suggested by Plit et al8
METHODS A total of 128 cardiac surgical patients with PACs (Edwards Lab, Santa Ana, CA) were entered into this study after the University’s Institutional Review Board for Human Investigations approved the protocol. Informed consent was obtained from each patient or from the person legally responsible for the patient. All consecutive patients with a PAC for cardiac surgical procedures in whom this monitoring could be replaced by a CVC within the first 3 postoperative days according to the judgment of the intensivists were studied. Patients who required prolonged PAC monitoring beyond 72 hours were excluded. Patients showing clinical signs of infection before initial PAC placement were excluded. Thus, 4 patients with endocarditis could not be entered in this study. All participants, ie, anesthesiologists, intensivists, microbiologists, and nursing staff were informed about the study protocol. All PACs had been inserted percutaneously via the right or left internal jugular vein by experienced anesthesiologists in the operating room. Special attention was given to aseptic insertion conditions, ie, washing the puncture site three times with povidoneiodine and having the operator wear gloves, gown, hat, and mask. Catheters were introduced through 8.5-F sideport sheath introducers (Arrow Inc, Reading, PA), which were left in place. The protective anticontamination sleeve of the introducer set was attached to the introducer hub and extended to cover as much of the catheters as possible. Postoperatively, patients were treated in
Artery Catheters
K. Geiss, MD
hours following initial insertion of the PAC. an incidence of catheter-related infection of 35.3% was observed in the guidewire group, as opposed to 12.5% in the de novo group. It is recommended that the use of a guidewire technique for catheter replacement (PAC to CVC) is a safe alternative to de novo insertion of a CVC within 48 hours after initial insertion of the PAC. In order to minimize the potential risk of catheter-related infection and bacteremia in cardiac surgical patients, de novo catheterization beyond 48 hours after initial venapuncture is suggested. Copyright 0 1992 by W.B. Saunders Company KEY WORDS: catheter-related infection, ment, pulmonary artery catheters
guidewire
replace-
an open intensive care unit (ICU), with adherence to a strict catheter care protocol. Dressing with nonocclusive sterile gauze was changed daily and the skin puncture site was thoroughly inspected for signs of local infection. Soiled dressings were changed immediately. No topical antiseptic or antibiotic was applied. Manipulation of catheters was allowed only after desinfecting hands with an alcoholic solution or wearing sterile gloves. Intravenous tubings were changed every 48 hours by ICU nurses wearing sterile gloves. Cefuroxime was given to patients for routine prophylaxis during the 24-hour perioperative period. When PAC monitoring was no longer required, the catheter was removed within 72 hours after primary placement. Patients were randomly allocated to receive a double-lumen CVC (Arrow Inc, Reading, PA) either at the initial introducer insertion site over a guidewire, or at a new site with de novo placement using a sterile technique. The skin was swabbed three times with povidone-iodine solution and allowed to dry. Sterile covers were placed around the site of insertion. The intensivist wore sterile gloves, hat, mask, and gown after desinfecting his hands. All parts of the opened catheter set were handled sterile by the intensivist and the assisting nurse. After puncture of the opposite jugular vein the catheter was introduced and secured by skin sutures. A nonocclusive sterile gauze dressing covered the puncture site. The procedure for changing catheters over a guidewire was as follows. The PAC and the protective sleeve with all dressings were removed completely while the intensivist was wearing sterile clothes. A guidewire (Arrow Inc, Reading, PA) was then inserted into the introducer sheath. The skin suture was cut off with sterile scissors and the introducer was withdrawn, leaving the guidewire in place. After swabbing the wire with povidone-iodine solution, the double-lumen catheter was advanced over it. The correct position of the catheter was confirmed by a chest film, and the catheter was secured with sutures. A nonocclusive sterile dressing was used to cover the puncture site. All CVCs were removed 7 days after secondary placement according to the standards of care in the ICU. The puncture site was cleaned with an :alcohol disinfectant, and the catheters were withdrawn with sterile forceps. The distal 3 cm of all introducers, PACs and CVCs were cut off with sterile scissors immediately after removal. The tips were collected in sterile tubes, labelled, and immediately transported to the microbiology laboratory. The
From the Departments OfAnesthesiology and Microbiology University of Heidelberg, Heidelberg, Germany. Reprints are not available. Copyright Q 1992 by WB. Saunders Company 1053-0770192/0606-0014$03.0010
Journalof Cardiothoracic and VascularAnesthesia, Vol6, No 6 (December),1992: pp 711-714
711
BACH, BGHRER, AND GEISS
712
Table 1. Semiquantitative
Results of Introducer and Pulmonary Artery Catheter Tips PACS
Introducers PAC in
15cfu
il5CfU
0% (O/87)
1.1% (l/87)
24 hours
0% (O/28)
2.3% (2187)
0% (O/87)
3.6% (l/28)
0% (O/28) 9.1% (3133)
24.2% (8/33)
6.1% (2/33)
NOTE. Figures in brackets are the ratio of infected catheters with greater than 15 cfu per plate to all catheters
Introducers and PACs removed at various intervals after initial insertion showed comparable microbiological results (Table l), the rate of catheter tips growing more than 15 cfu depending on the period in situ. CVCs placed via guidewire change, within 48 hours after initial placement, exhibited similar rates of infected catheter tips as compared to de novo placed catheters (Table 2). However, after 48 hours, rates were 35.3% of catheter tips with more than 15 cfu in the guidewire group, as opposed to 12.5% in the de novo group (difference significant, x2 analysis, P < 0.05). Similar results were obtained in catheter segments growing less than 15 cfu (Table 3). In the de novo group, in which catheter change took place more than 48 hours after PAC insertion, no positive blood cultures with the same microorganism as the catheter tip were obtained at catheter withdrawal after 7 days. In the corresponding guidewire subgroup, blood cultures were positive at the end of the catheter life in two patients showing intermittent fever greater than 38.5”C with leucocy tosis greater than 15,0OO/l.~L. In two other patients, blood cultures were positive with the same microorganism at days 5 and 6, these patients showing signs of infection with rectal body temperature greater than 38.5”C and leukocytes greater than 15,OOO/uL. In all 4 of these patients, temperature dropped below 37.5”C within 12 hours after catheter
removal and leucocytosis resolved, supporting the assumption of catheter-related infection in these cases. In these patients, PAC to CVC change took place 49,50,69, and 71 hours after initial insertion of the PAC. In this study, there were 38 microbiological isolates from PAC and CVC tips, with no catheter yielding more than one organism. The most frequent organisms cultured from catheters were Staphylococcus epidemidis (26 cultures), followed by S aureus (7 cultures). In the guidewire group with secondary catheter placement after 48 hours. a total of 6 patients showed catheter tips with greater than 15 cfu. Organisms isolated were S aureus in 2, S epidermidis in 2, Enterococcus in 1. and Klebsiella pneumoniae in 1 patient. In this subgroup of patients, S aureus could be isolated from peripheral blood in one case, S epidemidis in two cases, and Enterococcus in one case. The identical microorganisms could be cultured in the corresponding catheter tips with more than 15 cfu on blood agar plate. All these patients presented signs of catheter-related infections, ie, fever and leucocytosis. All other patients with positive tips, in either group, showed no clinical or microbiological signs of catheterrelated infection. Daily inspection of the insertion sites revealed only minor inflammation with no signs of local infection, ie, purulent discharge or cellulitis. Routine cultures of sputum, wound swabs, and urine samples were made twice a week. In the 4 patients with clinical signs of catheter-related infection and catheterrelated bacteremia, these cultures remained sterile or produced microorganisms differing from those found in catheter tip or blood cultures. However, none of these patients exhibited signs of septic shock. Furthermore, no patient developed signs of endocarditis, ie, remittent fever in association with a significant new cardiac murmur or clinical and echocardiographic signs of valve insufficiency. During initial and secondary placement of catheters, no serious complications such as pneumothorax, hemothorax, life-threatening arrhythmias, or other problems occurred. Minor complications such as local hematoma, transient
Table 2. Central Venous Catheter Tips Having Greater Than 15 cfu
Table 3. Central Venous Catheter Tips Having Less Than 15 cfu on
segments were rolled on blood agar plates and analyzed semiquantitively according to the method of Maki.’ A threshold of 15 colony-forming units (cfu) per agar plate was assumed to distinguish colonized from infected catheters. Organisms were identified according to standard microbiological techniques. Blood samples for culture were taken in duplicate from a peripheral vein within 15 minutes after catheter removal and cultured. Further blood samples for blood cultures were obtained by venapuncture if the patient exhibited signs of systemic infection, eg, rectal temperature greater than 38.5”C, or leukocyte count > 15,OOO/uL with no apparent cause within 24 hours after their occurrence. Statistical interpretation of the data was achieved using x’ analysis, with Yates correction factor, where appropriate. Statistical significance was accepted at P < 0.05. RESULTS
on the Plate PAC in
place
Guidewlre
the Agar Plate and/or PAC in
De Now
place
Positive in Broth
Guidewire
De Now
4.7% (2144)
24 hours
2.3% (l/43)
2.3% (1144)
