- Email: [email protected]
Central venous catheter placement in patients with disorders of hemostasis
Central Venous Catheter Placement in Patients with Disorders of Hemostasis Hamid Mumtaz, MD, Victor Williams, MD, Martin Hauer-Jensen, MD, Mark Rowe, RNP, Rhonda S. Henry-Tillman, MD, Keith Heaton, MD, Anne T. Mancino, MD, Roberta L. Muldoon, MD, V. Suzanne Klimberg, MD, J. Ralph Broadwater, MD, Kent C. Westbrook, MD, Nicholas P. Lang, MD, Little Rock, Arkansas
BACKGROUND: Patients requiring central venous access frequently have disorders of hemostasis. The aim of this study was to identify factors predictive of bleeding complications after central venous catheterization in this group of patients. METHODS: A retrospective analysis of all central venous catheters placed over a 2-year period (1997 to 1999) at our institution were performed. The age, sex, clinical diagnosis, most recent platelet count, prothrombin international normalized ratio (INR), activated partial thromboplastin time (aPTT), catheter type, the number of passes to complete the procedure, and bleeding complications were retrieved from the medical records. RESULTS: In a 2-year period, 2,010 central venous catheters were placed in 1,825 patients. Three hundred and thirty placements were in patients with disorders of hemostasis. In 88 of the 330 patients, the underlying coagulopathy was not corrected before catheter placement. In these patients, there were 3 bleeding complications requiring placement of a purse string suture at the catheter entry site. In the remaining 242 patients, there was 1 bleeding complication. Of the variables analyzed, only a low platelet count (<50 ⴛ 109/L) was significantly associated with bleeding complications. CONCLUSION: Central venous access procedures can be safely performed in patients with underlying disorders of hemostasis. Even patients with low platelet counts have infrequent (3 of 88) bleeding complications, and these problems are easily managed. Am J Surg. 2001;180:503–506. © 2001 by Excerpta Medica, Inc.
From the Department of Surgery (HM, VW, MHJ, RSHT, KH, ATM, RLM, VSK, JRB, KCW, NPL) and the Division of Surgical Oncology (MHJ, MR, RSHT, KH, ATM, RLM, VSK, JRB, KCW, NPL), University of Arkansas for Medical Science, Arkansas Cancer Research Center (MHJ, RSHT, KH, ATM, RLM, VSK, JRB, KCW, NPL), and the Central Arkansas Veterans Health Care System (MHJ, RSHT, KH, ATM, RLM, VSK, JRB, KCW, NPL), Little Rock, Arkansas Requests for reprints should be addressed to Nicholas P. Lang, MD, Division of Surgical Oncology, Department of Surgery, 4301 West Markham, Slot 725, Little Rock, Arkansas 72205-7199. Presented at the 52nd Annual Meeting of the Southwestern Surgical Congress, Colorado Springs, Colorado, April 9 –12, 2000.
© 2001 by Excerpta Medica, Inc. All rights reserved.
T
he need for central venous access in the management of patients has increased in current medical practice. In addition to its role in hemodynamic monitoring, central venous access is frequently required for the administration of chemotherapeutic agents, parenteral nutrition, hemodialysis, stem cell transfusions, and in situations where peripheral venous access is not available. Disorders of hemostasis are present in a large number of patients requiring central venous access. Consequently, physicians are hesitant to perform these procedures owing to the perceived increased risk of bleeding. Bleeding complications after central venous catheterization may present as insertional site bleeding, subcutaneous hematoma, mediastinal hematoma, or hemothorax. There are at present no firm guidelines on central venous catheterization in the presence of coagulopathy, although some series have suggested a low incidence of vascular complications.1– 4 Furthermore, it is unclear whether fresh frozen plasma (FFP) or platelet concentrate should be given prior to attempted catheterization. Finally, although correction of coagulopathy may be possible, it may not be beneficial, it may be impossible to administer the corrective transfusion factor owing to lack of venous access, or the condition may not be correctable by transfusion alone. The purpose of this study was to identify factors predictive of bleeding complications from central venous catheterization in patients with underlying disorders of hemostasis, and to determine the frequency and severity of these complications.
MATERIALS AND METHODS In our institution, central venous catheterizations performed under the care of a surgical attending are monitored by a dedicated central venous line clinic. This clinic is staffed with three nurses and has two dedicated rooms. A surgical intern with the supervision of surgical attending physician places the catheters in the majority of the cases. In this study, a retrospective analysis of the catheters placed over a 2-year period (September 1997 to August 1999) was undertaken. This analysis was undertaken as part of a continuous quality control evaluation of central venous catheters placed by the surgical service in our institution. During this period, 2,010 catheters were placed in 1,825 patients. All catheters were placed percutaneously using Seldinger’s technique. The choice of right versus left side placement was based upon the patient’s preference and was modified by any history of previous cannulation difficulty, vascular stenosis, or thrombosis. In situations requiring 0002-9610/01/$–see front matter PII S0002-9610(00)00552-3
503
CENTRAL VENOUS CATHETER PLACEMENT WITH HEMOSTASIS DISORDERS/MUMTAZ ET AL
long-term venous access, as in the majority of our patient population, we have preferred subclavian cannulation. The latter appears to be more comfortable for the patient and easier to take care of. The choice of catheter type was based upon the clinical indication. Patients who had subcutaneous infusaports or cuffed catheters and patients whose catheter placement was performed by another service were not included in the study. The data extracted retrospectively from the medical records included age, sex, clinical diagnosis, most recent platelet count, activated partial thromboplastin time (aPTT), prothrombin time (PT), international normalized ratio (INR), catheter type, the number of passes required to complete the procedure, and the occurrence of bleeding complications. The aPTT/PT were measured using the Electra 1600 MLA analyzer (Coulter-Beckmann, Brea, California). The range for the normal values for PT and PTT in our laboratory is 10 to 13 sec, and 23 to 37 sec, respectively. The platelet count was measured using the automated CellDynn 400 (Abbott, South Pasadena, California) system. Phase microscopy with a Neubauer hemocytometer was performed if the platelet count was less than 35 ⫻ 109/L by the automated method. Normal platelet values in our laboratory range from 150 to 500 ⫻ 109/L. Patients with an INR ⬎1.3, PTT ⬎37 sec, or platelets ⬍150 ⫻ 109/L were considered to have disorders of hemostasis. Bleeding complications were recorded as significant when an intervention necessary to stop the hemorrhage was undertaken. Bleeding that was arrested with digital manual pressure was not considered significant for the analysis in this study. Subcutaneous hematomas that increased in size on serial examination were considered significant. A postprocedure chest roentgenogram was obtained in each case, and any new opacity in the hemithorax or the mediastinum was also considered a significant complication. The variables PTT, INR, platelet count, catheter type, number of passes to complete the procedure, and precatheter correction of coagulopathy were analyzed as potential factors predictive of bleeding complications. Univariate statistical analysis was done using the Fisher exact test, analysis of variance (ANOVA), and the Mann-Whitney U test as appropriate. The level of test significance was taken as P value of less than or equal to 0.05.
RESULTS Of the 2,010 catheters (1,825 patients), 96% were placed in the subclavian vein (SV) and 4% in the internal jugular vein (IJV). These included 8 Fr Cook catheters (64%), 11.5 Fr Quinton catheters (22%), and 7 Fr Arrow catheters (14%). Table I summarizes the clinical indications of central venous catheterizations in the 2,010 procedures reviewed. An underlying disorder of hemostasis was present in 330 procedures (330 patients). In 88 of the 330 patients, the hemostatic disorder had not been corrected prior to catheter placement. The clinical indications for central venous catheterization in this group of 88 patients (88 procedures) are also shown in Table I. In the 88 patients with abnormal hemostasis, the median value and range for the platelet count, PTT, and INR were 95 ⫻ 109/L (12 to 330 ⫻ 109/L), 54 sec (22 to 100 sec), and 1.8 (1.2 to 3.5), respectively. The number of passes required to successfully gain central venous access were 1 504
TABLE I Clinical Indications for Central Venous Catheterizations in All Patients and in Patients with Abnormalities of Hemostasis
Hematologic malignancies Other malignancies Intensive care patients Total parenteral nutrition Renal failure Lack of peripheral access
All Patients (n ⴝ 2,010 Catheters)
Abnormal Hemostasis (n ⴝ 88 Catheters)
1608 108 85 65 42 102
42 0 21 0 11 14
(64 patients), 2 (16 patients), 3 (5 patients), and ⬎4 (3 patients). Bleeding complications were seen in 4 patients only (Table II). Of these 4 patients, 3 had abnormal hemostasis at the time of central venous catheter placement. In the remaining patient, the abnormal hemostasis had been corrected with FFP and platelet concentrate. In all 4 patients, bleeding was seen at the catheter insertion site that could not be controlled with digital pressure applied for approximately 20 minutes. In all cases, hemostasis was obtained by placing a purse string suture around the catheter entry site. No other evidence of bleeding was seen in these patients. Univariate analysis revealed no significant difference in the outcome of bleeding complications in patients with abnormal hemostasis (versus those that had precatheter placement correction of hemostasis). In the 88 patients with abnormal hemostasis, only a low platelet count ⬍50 ⫻ 109/L (versus platelet count ⬎51 ⫻ 109/L, P ⬍0.02) was a significant predictor of bleeding complications. The variables PTT, INR, catheter types, and the number of passes required for successful catheterizations were not significantly associated with bleeding complication.
COMMENTS The incidence of bleeding complication in the present study of central venous catheterization in patients with abnormal hemostasis was 3%. In this study, we have defined bleeding complication as being significant when an intervention other than digital pressure was necessary to secure hemostasis. This is in contrast to other recent studies that have also included minor vascular complications such as superficial oozing, nonexpanding hematomas, and bleeding at the suture site in their analysis. However, even using these criteria, a low incidence of bleeding complications has been reported.1– 4 In this study we did not identify any reports of inadvertent arterial puncture. This may be related to inaccurate reporting and limitations of a retrospective data analysis. However, a significant finding in the present study was the absence of any major arterial vascular complications requiring surgical intervention. Most significant bleeding complications occur mainly as a result of inadvertent arterial puncture especially with a large-caliber catheter. In the absence of such arterial puncture, bleeding that is not
THE AMERICAN JOURNAL OF SURGERY® VOLUME 180 DECEMBER 2000
CENTRAL VENOUS CATHETER PLACEMENT WITH HEMOSTASIS DISORDERS/MUMTAZ ET AL
TABLE II Details of Four Patients with Bleeding Complications
Patient Patient Patient Patient
1 2 3 4*
Diagnosis
PLT Count (ⴛ109/L)
PTT (sec)
INR
Number of Passes
Catheter Type (Fr)
Multiple myeloma Septic shock Renal failure Multiple myeloma
12 31 46 154
24 34 42 35
1.2 1.5 1.1 1.1
1 2 1 2
8 8 11.5 8
* Patient with abnormal hemostasis corrected before central venous cannulation. PLT ⫽ platelet; PTT ⫽ partial thromboplastin time; INR ⫽ international normalized ratio.
controlled with digital pressure can be easily managed by a local method using cutaneous purse string suture at the catheter entry site. Although radiological guidance can aid in accurate central venous cannulation, it is not a foolproof technique.5,6 In the former study, using subtraction angiographic technique for venous mapping and ultrasound localization, the incidence of bleeding complications seen as hematoma or prolonged oozing at the puncture site was 6.1%. In this study, a low platelet count (⬍50 ⫻ 109/L) was the only significant predictor of bleeding complication in patients with abnormal hemostasis. In particular, an abnormal PTT and INR were not associated with bleeding complications. This observation is similar to that reported in other recent studies.3,4 Thus, transfusion of fresh frozen plasma for correction of PT or INR prior to central venous cannulation is of doubtful benefit. Even the use of platelet transfusion is of minimal benefit to correct a problem so easily addressed with a suture. It would seem logical to assume that the more passes done to achieve successful cannulation, the higher would be the incidence of bleeding complications. However, with careful attention to detail, most central venous cannulation can be done at the first attempt. In the present study, 64 of 88 (72%) patients were successfully catheterized at the first attempt. The site of venous cannulation and operator experience has been reported to influence the occurrence of bleeding complications.4 However, in our study we are unable to draw conclusions on these variables as, in the majority of the cases, the subclavian vein was used for central venous access and the majority of the lines were placed by a surgical intern. We would stress that dedicated staff and meticulous adherence to technique can achieve the outcomes shown in this study. There are no previous studies available to support the benefit of treatment of coagulopathy prior to cannulation or to suggest the extent and duration of this treatment to avoid bleeding complications. In the present study, we did not observe a significant difference in the outcome of bleeding complications in patients with abnormal hemostasis versus those who had precatheter placement correction of hemostasis. Central venous cannulation has been reported to be safe in the fibrinolytic state associated with
thrombolytic therapy.7 Precannulation correction of coagulopathy can, in certain situations, be detrimental to the patient. Critically ill patients may not tolerate infusion of large volumes of blood products to correct coagulopathy. Also, blood-component therapy is associated with a small but definite risk of disease transmission. Although a potentially suitable alternative to central venous cannulation in patients with abnormal hemostasis is the use of peripherally inserted central catheters (PICC lines), it has limitations. The latter requires a suitable peripheral venous access that is frequently not available in patients on long-term chemotherapy. In addition, PICC lines are often associated with difficulty in drawing blood. In conclusion, the low incidence of bleeding complications demonstrated in this and other recent studies suggest the experienced clinician using simple precautions, good technique, and correct patient positioning can safely undertake percutaneous central venous catheterization in patients with disorders of hemostasis. Furthermore, these attempts need not be preceded by infusion of blood components to correct the coagulopathy. In the absence of an arterial puncture, most bleeding complications can be managed by simple local methods.
REFERENCES 1. Goldfarb G, Lebrec D. Percutaneous cannulation of the internal jugular vein in patients with coagulopathies: an experience based on 1000 attempts. Anesthesiology. 1982;56:321–323. 2. Foster PF, Moore LR, Sankary HN. Central venous catheterization in patients with coagulopathy. Arch Surg. 1992;127:273–275. 3. Doerfler MD, Kaufmann B, Goldenberg AS. Central venous catheter placement in patients with disorders of hemostasis. Chest. 1996;110:185–188. 4. Fisher NC, Mutimer DJ. Central venous cannulation in patients with liver disease and coagulpathy—a prospective audit. Intens Care Med. 1999;25:481– 485. 5. Chan LL, Tan BS, Kaw GJ, et al. Radiological placement of 211 central venous catheters: outcome and complications. Ann Acad Med (Singapore). 1999;28:481– 487. 6. Fry WS, Clagett GC, O’Rourke PT. Ultrasound guided central venous access. Arch Surg. 1999;134:738 –741. 7. Lee HS, Quinn T, Boyle RM. Safety of thrombolytic treatment in patients with central venous cannulation. Br Heart J. 1995;73:359 –362.
THE AMERICAN JOURNAL OF SURGERY® VOLUME 180 DECEMBER 2000
505
CENTRAL VENOUS CATHETER PLACEMENT WITH HEMOSTASIS DISORDERS/MUMTAZ ET AL
DISCUSSION Edward W. Nelson, MD (Salt Lake City, Utah): I was very pleased and a little bit surprised to see that there are three papers on the topic of central venous catheter placement at our current meeting. All these papers deal with specific aspects of line placement and secondary complications, which have occurred in very large numbers of patients. This nice study from Arkansas strikes me as unique. Not for its conclusion regarding the low incidence of bleeding complications, regardless of abnormal coagulation studies, but, rather, it is unique because the authors appear to have developed a dedicated central line clinic, where interns are supervised by attendings as they insert lines for a wide variety of indications. Assuming such an ideal situation, free from intrusion by such aberrations as poorly supervised medical house staff and financially motivated interventional radiologists, I am surprised that this review was limited to such a single and simple question with such a predictable answer as the low rate of bleeding complications. Perhaps more data will be presented next year. Regardless, I have several questions regarding this dedicated central line clinic. First, what percentage of all line placements at your institution does this represent? How many patients escape your control? How is attending coverage arranged? Specifically, do your coauthors honestly take a turn watching interns insert central lines? Second,
506
are tunnel lines placed in sick oncology patients done in this clinic, or is this group still more appropriately managed in the operating room? And third, what about your overall complication rate? Specifically, if no radiologic guidance is used, as you mentioned in the manuscript, how often are repeat procedures needed to redirect misplaced central catheters? My strong bias toward surgeons keeping control of central line placement has been well validated by this year’s meeting.
CLOSING Hamid Mumtaz, MD: At the University of Arkansas of Medical Sciences and the Arkansas Cancer Research Center, we have a dedicated central venous line clinic. Each day the lines are staffed by one surgery attending physician. No patients who had lines placed outside this area were included in this study. All the catheters are placed by the surgical intern under the direct supervision of an attending. At our center, we do not place tunnel catheters. We routinely place Cook catheters for long-term chemotherapy purposes. Briefly, we have a 1% to 2% pneumothorax rate, and a malposition rate of about 3% to 4%. As soon as we identify such malposition, these patients are referred to radiology for appropriate placement and correct positioning.
THE AMERICAN JOURNAL OF SURGERY® VOLUME 180 DECEMBER 2000
BACKGROUND: Patients requiring central venous access frequently have disorders of hemostasis. The aim of this study was to identify factors predictive of bleeding complications after central venous catheterization in this group of patients. METHODS: A retrospective analysis of all central venous catheters placed over a 2-year period (1997 to 1999) at our institution were performed. The age, sex, clinical diagnosis, most recent platelet count, prothrombin international normalized ratio (INR), activated partial thromboplastin time (aPTT), catheter type, the number of passes to complete the procedure, and bleeding complications were retrieved from the medical records. RESULTS: In a 2-year period, 2,010 central venous catheters were placed in 1,825 patients. Three hundred and thirty placements were in patients with disorders of hemostasis. In 88 of the 330 patients, the underlying coagulopathy was not corrected before catheter placement. In these patients, there were 3 bleeding complications requiring placement of a purse string suture at the catheter entry site. In the remaining 242 patients, there was 1 bleeding complication. Of the variables analyzed, only a low platelet count (<50 ⴛ 109/L) was significantly associated with bleeding complications. CONCLUSION: Central venous access procedures can be safely performed in patients with underlying disorders of hemostasis. Even patients with low platelet counts have infrequent (3 of 88) bleeding complications, and these problems are easily managed. Am J Surg. 2001;180:503–506. © 2001 by Excerpta Medica, Inc.
From the Department of Surgery (HM, VW, MHJ, RSHT, KH, ATM, RLM, VSK, JRB, KCW, NPL) and the Division of Surgical Oncology (MHJ, MR, RSHT, KH, ATM, RLM, VSK, JRB, KCW, NPL), University of Arkansas for Medical Science, Arkansas Cancer Research Center (MHJ, RSHT, KH, ATM, RLM, VSK, JRB, KCW, NPL), and the Central Arkansas Veterans Health Care System (MHJ, RSHT, KH, ATM, RLM, VSK, JRB, KCW, NPL), Little Rock, Arkansas Requests for reprints should be addressed to Nicholas P. Lang, MD, Division of Surgical Oncology, Department of Surgery, 4301 West Markham, Slot 725, Little Rock, Arkansas 72205-7199. Presented at the 52nd Annual Meeting of the Southwestern Surgical Congress, Colorado Springs, Colorado, April 9 –12, 2000.
© 2001 by Excerpta Medica, Inc. All rights reserved.
T
he need for central venous access in the management of patients has increased in current medical practice. In addition to its role in hemodynamic monitoring, central venous access is frequently required for the administration of chemotherapeutic agents, parenteral nutrition, hemodialysis, stem cell transfusions, and in situations where peripheral venous access is not available. Disorders of hemostasis are present in a large number of patients requiring central venous access. Consequently, physicians are hesitant to perform these procedures owing to the perceived increased risk of bleeding. Bleeding complications after central venous catheterization may present as insertional site bleeding, subcutaneous hematoma, mediastinal hematoma, or hemothorax. There are at present no firm guidelines on central venous catheterization in the presence of coagulopathy, although some series have suggested a low incidence of vascular complications.1– 4 Furthermore, it is unclear whether fresh frozen plasma (FFP) or platelet concentrate should be given prior to attempted catheterization. Finally, although correction of coagulopathy may be possible, it may not be beneficial, it may be impossible to administer the corrective transfusion factor owing to lack of venous access, or the condition may not be correctable by transfusion alone. The purpose of this study was to identify factors predictive of bleeding complications from central venous catheterization in patients with underlying disorders of hemostasis, and to determine the frequency and severity of these complications.
MATERIALS AND METHODS In our institution, central venous catheterizations performed under the care of a surgical attending are monitored by a dedicated central venous line clinic. This clinic is staffed with three nurses and has two dedicated rooms. A surgical intern with the supervision of surgical attending physician places the catheters in the majority of the cases. In this study, a retrospective analysis of the catheters placed over a 2-year period (September 1997 to August 1999) was undertaken. This analysis was undertaken as part of a continuous quality control evaluation of central venous catheters placed by the surgical service in our institution. During this period, 2,010 catheters were placed in 1,825 patients. All catheters were placed percutaneously using Seldinger’s technique. The choice of right versus left side placement was based upon the patient’s preference and was modified by any history of previous cannulation difficulty, vascular stenosis, or thrombosis. In situations requiring 0002-9610/01/$–see front matter PII S0002-9610(00)00552-3
503
CENTRAL VENOUS CATHETER PLACEMENT WITH HEMOSTASIS DISORDERS/MUMTAZ ET AL
long-term venous access, as in the majority of our patient population, we have preferred subclavian cannulation. The latter appears to be more comfortable for the patient and easier to take care of. The choice of catheter type was based upon the clinical indication. Patients who had subcutaneous infusaports or cuffed catheters and patients whose catheter placement was performed by another service were not included in the study. The data extracted retrospectively from the medical records included age, sex, clinical diagnosis, most recent platelet count, activated partial thromboplastin time (aPTT), prothrombin time (PT), international normalized ratio (INR), catheter type, the number of passes required to complete the procedure, and the occurrence of bleeding complications. The aPTT/PT were measured using the Electra 1600 MLA analyzer (Coulter-Beckmann, Brea, California). The range for the normal values for PT and PTT in our laboratory is 10 to 13 sec, and 23 to 37 sec, respectively. The platelet count was measured using the automated CellDynn 400 (Abbott, South Pasadena, California) system. Phase microscopy with a Neubauer hemocytometer was performed if the platelet count was less than 35 ⫻ 109/L by the automated method. Normal platelet values in our laboratory range from 150 to 500 ⫻ 109/L. Patients with an INR ⬎1.3, PTT ⬎37 sec, or platelets ⬍150 ⫻ 109/L were considered to have disorders of hemostasis. Bleeding complications were recorded as significant when an intervention necessary to stop the hemorrhage was undertaken. Bleeding that was arrested with digital manual pressure was not considered significant for the analysis in this study. Subcutaneous hematomas that increased in size on serial examination were considered significant. A postprocedure chest roentgenogram was obtained in each case, and any new opacity in the hemithorax or the mediastinum was also considered a significant complication. The variables PTT, INR, platelet count, catheter type, number of passes to complete the procedure, and precatheter correction of coagulopathy were analyzed as potential factors predictive of bleeding complications. Univariate statistical analysis was done using the Fisher exact test, analysis of variance (ANOVA), and the Mann-Whitney U test as appropriate. The level of test significance was taken as P value of less than or equal to 0.05.
RESULTS Of the 2,010 catheters (1,825 patients), 96% were placed in the subclavian vein (SV) and 4% in the internal jugular vein (IJV). These included 8 Fr Cook catheters (64%), 11.5 Fr Quinton catheters (22%), and 7 Fr Arrow catheters (14%). Table I summarizes the clinical indications of central venous catheterizations in the 2,010 procedures reviewed. An underlying disorder of hemostasis was present in 330 procedures (330 patients). In 88 of the 330 patients, the hemostatic disorder had not been corrected prior to catheter placement. The clinical indications for central venous catheterization in this group of 88 patients (88 procedures) are also shown in Table I. In the 88 patients with abnormal hemostasis, the median value and range for the platelet count, PTT, and INR were 95 ⫻ 109/L (12 to 330 ⫻ 109/L), 54 sec (22 to 100 sec), and 1.8 (1.2 to 3.5), respectively. The number of passes required to successfully gain central venous access were 1 504
TABLE I Clinical Indications for Central Venous Catheterizations in All Patients and in Patients with Abnormalities of Hemostasis
Hematologic malignancies Other malignancies Intensive care patients Total parenteral nutrition Renal failure Lack of peripheral access
All Patients (n ⴝ 2,010 Catheters)
Abnormal Hemostasis (n ⴝ 88 Catheters)
1608 108 85 65 42 102
42 0 21 0 11 14
(64 patients), 2 (16 patients), 3 (5 patients), and ⬎4 (3 patients). Bleeding complications were seen in 4 patients only (Table II). Of these 4 patients, 3 had abnormal hemostasis at the time of central venous catheter placement. In the remaining patient, the abnormal hemostasis had been corrected with FFP and platelet concentrate. In all 4 patients, bleeding was seen at the catheter insertion site that could not be controlled with digital pressure applied for approximately 20 minutes. In all cases, hemostasis was obtained by placing a purse string suture around the catheter entry site. No other evidence of bleeding was seen in these patients. Univariate analysis revealed no significant difference in the outcome of bleeding complications in patients with abnormal hemostasis (versus those that had precatheter placement correction of hemostasis). In the 88 patients with abnormal hemostasis, only a low platelet count ⬍50 ⫻ 109/L (versus platelet count ⬎51 ⫻ 109/L, P ⬍0.02) was a significant predictor of bleeding complications. The variables PTT, INR, catheter types, and the number of passes required for successful catheterizations were not significantly associated with bleeding complication.
COMMENTS The incidence of bleeding complication in the present study of central venous catheterization in patients with abnormal hemostasis was 3%. In this study, we have defined bleeding complication as being significant when an intervention other than digital pressure was necessary to secure hemostasis. This is in contrast to other recent studies that have also included minor vascular complications such as superficial oozing, nonexpanding hematomas, and bleeding at the suture site in their analysis. However, even using these criteria, a low incidence of bleeding complications has been reported.1– 4 In this study we did not identify any reports of inadvertent arterial puncture. This may be related to inaccurate reporting and limitations of a retrospective data analysis. However, a significant finding in the present study was the absence of any major arterial vascular complications requiring surgical intervention. Most significant bleeding complications occur mainly as a result of inadvertent arterial puncture especially with a large-caliber catheter. In the absence of such arterial puncture, bleeding that is not
THE AMERICAN JOURNAL OF SURGERY® VOLUME 180 DECEMBER 2000
CENTRAL VENOUS CATHETER PLACEMENT WITH HEMOSTASIS DISORDERS/MUMTAZ ET AL
TABLE II Details of Four Patients with Bleeding Complications
Patient Patient Patient Patient
1 2 3 4*
Diagnosis
PLT Count (ⴛ109/L)
PTT (sec)
INR
Number of Passes
Catheter Type (Fr)
Multiple myeloma Septic shock Renal failure Multiple myeloma
12 31 46 154
24 34 42 35
1.2 1.5 1.1 1.1
1 2 1 2
8 8 11.5 8
* Patient with abnormal hemostasis corrected before central venous cannulation. PLT ⫽ platelet; PTT ⫽ partial thromboplastin time; INR ⫽ international normalized ratio.
controlled with digital pressure can be easily managed by a local method using cutaneous purse string suture at the catheter entry site. Although radiological guidance can aid in accurate central venous cannulation, it is not a foolproof technique.5,6 In the former study, using subtraction angiographic technique for venous mapping and ultrasound localization, the incidence of bleeding complications seen as hematoma or prolonged oozing at the puncture site was 6.1%. In this study, a low platelet count (⬍50 ⫻ 109/L) was the only significant predictor of bleeding complication in patients with abnormal hemostasis. In particular, an abnormal PTT and INR were not associated with bleeding complications. This observation is similar to that reported in other recent studies.3,4 Thus, transfusion of fresh frozen plasma for correction of PT or INR prior to central venous cannulation is of doubtful benefit. Even the use of platelet transfusion is of minimal benefit to correct a problem so easily addressed with a suture. It would seem logical to assume that the more passes done to achieve successful cannulation, the higher would be the incidence of bleeding complications. However, with careful attention to detail, most central venous cannulation can be done at the first attempt. In the present study, 64 of 88 (72%) patients were successfully catheterized at the first attempt. The site of venous cannulation and operator experience has been reported to influence the occurrence of bleeding complications.4 However, in our study we are unable to draw conclusions on these variables as, in the majority of the cases, the subclavian vein was used for central venous access and the majority of the lines were placed by a surgical intern. We would stress that dedicated staff and meticulous adherence to technique can achieve the outcomes shown in this study. There are no previous studies available to support the benefit of treatment of coagulopathy prior to cannulation or to suggest the extent and duration of this treatment to avoid bleeding complications. In the present study, we did not observe a significant difference in the outcome of bleeding complications in patients with abnormal hemostasis versus those who had precatheter placement correction of hemostasis. Central venous cannulation has been reported to be safe in the fibrinolytic state associated with
thrombolytic therapy.7 Precannulation correction of coagulopathy can, in certain situations, be detrimental to the patient. Critically ill patients may not tolerate infusion of large volumes of blood products to correct coagulopathy. Also, blood-component therapy is associated with a small but definite risk of disease transmission. Although a potentially suitable alternative to central venous cannulation in patients with abnormal hemostasis is the use of peripherally inserted central catheters (PICC lines), it has limitations. The latter requires a suitable peripheral venous access that is frequently not available in patients on long-term chemotherapy. In addition, PICC lines are often associated with difficulty in drawing blood. In conclusion, the low incidence of bleeding complications demonstrated in this and other recent studies suggest the experienced clinician using simple precautions, good technique, and correct patient positioning can safely undertake percutaneous central venous catheterization in patients with disorders of hemostasis. Furthermore, these attempts need not be preceded by infusion of blood components to correct the coagulopathy. In the absence of an arterial puncture, most bleeding complications can be managed by simple local methods.
REFERENCES 1. Goldfarb G, Lebrec D. Percutaneous cannulation of the internal jugular vein in patients with coagulopathies: an experience based on 1000 attempts. Anesthesiology. 1982;56:321–323. 2. Foster PF, Moore LR, Sankary HN. Central venous catheterization in patients with coagulopathy. Arch Surg. 1992;127:273–275. 3. Doerfler MD, Kaufmann B, Goldenberg AS. Central venous catheter placement in patients with disorders of hemostasis. Chest. 1996;110:185–188. 4. Fisher NC, Mutimer DJ. Central venous cannulation in patients with liver disease and coagulpathy—a prospective audit. Intens Care Med. 1999;25:481– 485. 5. Chan LL, Tan BS, Kaw GJ, et al. Radiological placement of 211 central venous catheters: outcome and complications. Ann Acad Med (Singapore). 1999;28:481– 487. 6. Fry WS, Clagett GC, O’Rourke PT. Ultrasound guided central venous access. Arch Surg. 1999;134:738 –741. 7. Lee HS, Quinn T, Boyle RM. Safety of thrombolytic treatment in patients with central venous cannulation. Br Heart J. 1995;73:359 –362.
THE AMERICAN JOURNAL OF SURGERY® VOLUME 180 DECEMBER 2000
505
CENTRAL VENOUS CATHETER PLACEMENT WITH HEMOSTASIS DISORDERS/MUMTAZ ET AL
DISCUSSION Edward W. Nelson, MD (Salt Lake City, Utah): I was very pleased and a little bit surprised to see that there are three papers on the topic of central venous catheter placement at our current meeting. All these papers deal with specific aspects of line placement and secondary complications, which have occurred in very large numbers of patients. This nice study from Arkansas strikes me as unique. Not for its conclusion regarding the low incidence of bleeding complications, regardless of abnormal coagulation studies, but, rather, it is unique because the authors appear to have developed a dedicated central line clinic, where interns are supervised by attendings as they insert lines for a wide variety of indications. Assuming such an ideal situation, free from intrusion by such aberrations as poorly supervised medical house staff and financially motivated interventional radiologists, I am surprised that this review was limited to such a single and simple question with such a predictable answer as the low rate of bleeding complications. Perhaps more data will be presented next year. Regardless, I have several questions regarding this dedicated central line clinic. First, what percentage of all line placements at your institution does this represent? How many patients escape your control? How is attending coverage arranged? Specifically, do your coauthors honestly take a turn watching interns insert central lines? Second,
506
are tunnel lines placed in sick oncology patients done in this clinic, or is this group still more appropriately managed in the operating room? And third, what about your overall complication rate? Specifically, if no radiologic guidance is used, as you mentioned in the manuscript, how often are repeat procedures needed to redirect misplaced central catheters? My strong bias toward surgeons keeping control of central line placement has been well validated by this year’s meeting.
CLOSING Hamid Mumtaz, MD: At the University of Arkansas of Medical Sciences and the Arkansas Cancer Research Center, we have a dedicated central venous line clinic. Each day the lines are staffed by one surgery attending physician. No patients who had lines placed outside this area were included in this study. All the catheters are placed by the surgical intern under the direct supervision of an attending. At our center, we do not place tunnel catheters. We routinely place Cook catheters for long-term chemotherapy purposes. Briefly, we have a 1% to 2% pneumothorax rate, and a malposition rate of about 3% to 4%. As soon as we identify such malposition, these patients are referred to radiology for appropriate placement and correct positioning.
THE AMERICAN JOURNAL OF SURGERY® VOLUME 180 DECEMBER 2000