| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Department of Anesthesiology and Reanimation, Dokuz Eylul University, Izmir, Turkey
Address correspondence and reprint requests to Dr. Binnur Erdalkiran Tuncali, Huzur mah. Sumbul sok. No:42/11, Narlidere Izmir, Turkey. Address e-mail to binnur.tuncali{at}deu.edu.tr.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The anticoagulant properties of heparin has led clinicians to use heparin flushes or heparinized infusion in an attempt to prevent thrombus formation and to prolong the duration of catheter patency. Use of heparin as an antithrombotic drug in catheters has been widespread for over 25 yr. Despite almost universal use, the benefit of heparin has not been firmly established, and there are some risks (3). The most important risk is heparin-induced thrombocytopenia after multiple heparin flushes (4,5). Decreasing unnecessary exposure to heparin is important to minimize the complications from sensitization.
The primary purpose of this study was to compare the patency of radial arterial catheters maintained with heparinized and nonheparinized infusions for perioperative arterial blood pressure monitoring. Second, we wanted to search for factors predisposing to subsequent arterial occlusion.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Patients were randomized to receive 0.9% NaCl (group S, n = 100) or 0.9% NaCl containing heparin 1 U/mL (group H, n = 100) flush solutions. Random numbers were put into sealed envelopes and opened before each anesthesia induction. Radial and ulnar blood flows were assessed by the blinded investigator using a Doppler probe (Japan Standart Corp, Shiki-City, Saitama, Japan) and pulse oximetry with and without manual occlusion of the radial and ulnar arteries.
On arrival at the operating room, an IV catheter was inserted. In addition to standard monitoring, invasive arterial blood pressure was monitored. Radial artery cannulation was performed after the induction of general anesthesia or after local anesthetic application for patients receiving regional anesthesia. In all cases, a 20-gauge polytetrafluoroethylene catheter (Metdem Medical Instruments, Ankara, Turkey) was inserted into the radial artery on the nondominant upper extremity using either direct puncture or the transfixation technique. Cannulations were performed after skin preparation with povidone iodine by anesthesia trainees and staff anesthesiologists. Age, sex, and the number of puncture attempts were recorded for each patient.
During the perioperative period, patency of the arterial cannulas was maintained by the continuous infusion of heparinized or nonheparinized flush solutions. In both groups, a continuous pressure of 300 mm Hg was applied with a pressure bag, and a flow rate of approximately 3 mL/h was maintained through a pressure monitoring kit with a Uniflow disposable transducer (Baxter International Inc, Biçakcilar, Turkey).
Because dampening of the pressure wave often precedes catheter occlusion, its occurrence, and interventions required to correct it such as positional changes and manual flushes, were closely monitored and recorded. The number of times blood was withdrawn from the system was also recorded. The length of time (in minutes) that the arterial catheter was in place was documented. The use of vasodilating and vasoactive drugs was recorded.
After arterial catheter removal, the site was manually compressed for 3–5 min until no bleeding occurred. All patients were examined by a blinded investigator just after and 24 h after decannulation. The investigator checked for infection and hematoma at the puncture site, sensory abnormality of the hand, and arterial flow estimated by Doppler and pulse oximetry. Patients who were still in the hospital 7 days after decannulation were reassessed for infection and hematoma at the puncture site and sensory abnormalities of the hand. Patients who were discharged were interviewed by telephone for any complaint about the puncture site or hand.
Arterial blood flow was evaluated at the site of cannulation, proximal and distal to it, and with and without manual occlusion of the ulnar artery. If the sound of flow over the puncture site was normal relative to the normal proximal vessel by Doppler and flow was present by pulse oximetry, no occlusion was considered. If the sound of flow over the puncture site was absent or obviously reduced relative to the normal proximal vessel by Doppler and flow was present by pulse oximetry, partial occlusion was considered. If the sound of flow over the puncture site was absent or obviously reduced relative to the normal proximal vessel by Doppler and flow was absent by pulse oximetry, complete occlusion was considered.
At the onset of this study, a sample size of 200 subjects was chosen, which was estimated to provide 80% power for detecting a 15% reduction in an estimated failure rate in controls of 22%. In the statistical analysis, the t-test was used for parametric data and the 
2 test was used for nonparametric data. Predisposing risk factors for the development of arterial occlusion such as age (older or younger than 65 yr old), sex, puncture attempts (one or more), duration of cannulation (less or more than 240 min), presence of perioperative hypotension requiring IV inotropic drug infusion, and presence of hematoma at the puncture site after decannulation were evaluated using regression analysis. P < 0.05 was taken as a significant value.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
The mean number of puncture attempts, the mean duration of cannulation, and the mean number of times blood was withdrawn from the system were similar in both groups. The mean number of corrective interventions because of dampening of the pressure wave, such as positional changes and manual flushes, was not significantly different in both groups (Table 2).
|
After decannulation, partial occlusion developed in 14 patients in each group. Total occlusion developed in six patients in group H and two patients in group S. At 24 h, partial occlusion had developed in one patient in each group. These differences were not statistically significant (Fig. 1).
|
Seven patients in group S and 15 patients in group H received IV 5–10 mg of ephedrine for treatment of hypotension associated with spinal anesthesia. One patient in group H received an IV dobutamine infusion, and 13 patients in each group received a prophylactic nitroglycerine infusion to reduce the incidence of ischemic episodes throughout the surgical procedure. These differences were not significant between groups.
After decannulation, hematoma developed in 11 patients in each group. At 24 h, hematoma was present in one patient in group S and two patients in group H. The incidence of hematoma after decannulation and at 24 h was similar in both groups.
Twenty-four hours and 7 days after decannulation all patients were reevaluated. None of the patients experienced distal ischemia, infection at the site, or sensory abnormalities of the hand.
Predisposing risk factors for the development of arterial occlusion were evaluated using logistic regression analysis. Longer duration of cannulation (P = 0.04), age younger than 65 yr (P = 0.009), and the presence of hematoma at the site (P = 0.013) were significant factors for arterial occlusion. The relative risk of arterial occlusion was 4.9 (95% confidence interval [CI], 0.07–1.02) for longer duration of cannulation, 2.8 (95% CI, 1.0–4.6) for age younger than 65 yr, and 3.7 (95% CI, 1.1–11.9) for the presence of hematoma at the site.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In our study, mean age and the sex of patients were similar in both groups, and we used 20-gauge polytetrafluoroethylene catheters in all patients. Furthermore, the number of puncture attempts, duration of cannulations, and number of blood samplings from the system were similar in both groups. Therefore, we propose that our results were related to the type of flush solutions only.
Heparinized arterial catheter infusions have long been used to maintain catheter patency. There is controversy over the type of flush solutions required to maintain patency of arterial catheters (13). Clifton et al. (14) suggested that nonheparinized flush solutions are associated with a significant increase in frequency of catheter malfunction compared with heparinized flush solutions. But this difference seems to develop after 10 hours of cannulation. In 50 cardiovascular ICU patients, Hook et al. (15) compared the patency of arterial catheters maintained with heparinized (2.5 U/mL) and nonheparinized flush solutions and found no significant differences between groups. The authors recommended that heparinized flush solutions not be used to prevent the risks associated with heparin and to reduce costs. Similarly, Kulkarni et al. (16) found no significant differences between heparinized (2 U/mL) and nonheparinized flush solutions in the maintenance of radial arterial catheters. Leighton (17) reported that after a small feasibility study on arterial catheters using 0.9% sodium chloride, they changed their unit policy to include using nonheparinized flush solutions for maintaining arterial catheter patency. A group of 13 experts appointed by the French Society of Anesthesia and Intensive Care has produced a guideline for arterial catheterization in adults, and they recommended the addition of heparin for catheterizations of more than 24 hours in duration (18). These studies were generally performed in ICU patients and focused on long-term arterial monitoring, but no empirical evidence has been found that verifies the need for heparinized arterial catheter infusion solutions when the catheters are placed in adults for short-term monitoring, such as during the perioperative period.
This double-blind, randomized study demonstrated that there was no significant difference in arterial catheter patency and the subsequent arterial occlusion between lines maintained with heparinized infusions and those maintained with nonheparinized infusions in adults for perioperative monitoring. The number of interventions to correct the dampening of the arterial waveform, such as positional changes and manual flushes, was not significantly different between groups. Arterial catheter waveforms were not shown to be more frequently dampened in the nonheparinized catheters. In other words, arterial monitoring could be performed successfully with both heparinized and nonheparinized flush solutions. After this study, we changed our policy and began to use nonheparinized flush solutions in the maintenance of perioperative radial arterial monitoring. Seven months have passed since this study and we have not seen any catheter malfunction or ischemic complications after radial arterial cannulations in more than 280 patients.
Slogoff et al. (6) evaluated radial artery occlusion after decannulation in 1699 cardiovascular surgical patients and found that occlusion occurred in more than 25% of the patients without ischemic damage. Cederholm et al. (11) reported that the incidence of radial arterial occlusion after decannulation was 33%. Davis (9) showed that the incidence of occlusion after decannulation was 42%. However, resolution of the thrombus always seems to occur after decannulation. In our study, the incidence of complete and partial occlusion was similar between groups. Although approximately 18% of these patients suffered partial or complete occlusion of the radial artery after decannulation, no permanent ischemic damage to the hand or forearm occurred, and there were no clinical consequences of arterial thrombosis when it occurred. The incidence of occlusion was less than that seen in the literature. In our study, mean duration of cannulations was as short as six to seven hours. This may explain why the incidence of occlusion was less than seen in the literature.
The widespread use of heparin flushes suggests that practitioners have generally assumed that the small doses used in flushing procedures are unlikely to cause any problems. However, side effects of heparin may occur, even with small amounts of heparin. Clinical experience and research have suggested that even small amounts of heparin may result in a heparin-related hematologic crisis, which is most frequently associated with thrombocytopenia (4,5). Therefore, evaluation of the need for further heparin exposure when heparin is used to maintain patency of arterial catheters is indicated. The most valuable benefit of changing to saline as the flush solution will be elimination of risks such as heparin-induced thrombocytopenia. In addition, cost savings in supplies, drugs, and staff time can be realized by switching to saline flush.
In this study, the incidence of arterial occlusion was correlated with longer duration of cannulation, so we agree with the authors who suggest that the longer the duration of cannulation the more frequent the incidence of subsequent occlusion. We also found that the incidence of arterial occlusion was correlated with the development of hematoma at the site. This finding is consistent with Cederholm et al. (11). Finally, our study showed that the incidence of arterial occlusion was correlated with ages younger than 65 years.
In conclusion, there are no significant differences between heparinized and nonheparinized flush solutions in the maintenance of perioperative arterial catheter patency and the incidence of subsequent radial arterial occlusion. These findings support the elimination of heparin as a flush solution. Not only will this change in practice provide a safer therapy for the patient, but it will also result in significant cost savings.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. A. Halm Flushing Hemodynamic Catheters: What Does the Science Tell Us? Am. J. Crit. Care., January 1, 2008; 17(1): 73 - 76. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
