- Email: [email protected]
The effect of the arterial catheter insertion technique on the success of radial artery cannulation: A prospective and randomized study
Journal of Critical Care 29 (2014) 475.e7–475.e10
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
The effect of the arterial catheter insertion technique on the success of radial artery cannulation: A prospective and randomized study☆ Gildasio S. De Oliveira Jr., MD, MSC a,⁎, Katharina Beckmann, MD a, Alain Salvacion, MD a, John Kim, MD b, Saadia Sherwani, MD a, Robert J. McCarthy, PharmD a a b
Department of Anesthesiology, Northwestern University, Chicago, IL, USA Department of Surgery, Northwestern University, Chicago, IL, USA
a r t i c l e
i n f o
Keywords: Arterial line cannulation
a b s t r a c t Purpose: The main objective of the current investigation was to compare a single wall puncture to vessel transfixing on the success of radial artery cannulation by resident physicians. Material and Methods: The study was a prospective and randomized investigation. Twelve anesthesiology residents performed radial arterial insertions in 126 patients using both the single wall and vessel transfixing technique in random order. The primary outcome was successful cannulation of the radial artery in 4 or less attempts. Other data collected included the total number of attempts and total time to catheter cannulation. Results: Successful radial artery cannulation was achieved in 88% and 86% of patients using the transfixing technique and single wall group, respectively (difference 2%; 95% CI, 14-9, P = 0.8, Fisher exact test). Cannulation was successfully on the first attempt in 38% of the transfixing compared to 54% using the single wall technique (difference − 16%; 95% CI, 32-2, P = 0.1, Fisher Exact test). The median (interquartile range) time to successful cannulation was longer in the transfixing group, 105 (69-176) seconds compared to 65 (25-114) seconds in the single puncture group (P = .009, log-rank test). Conclusions: Our findings suggest that there does not appear to be an advantage of the transfixing technique over the single wall puncture method for cannulating the radial artery by resident physicians. Cannulation was achieved in shorter time using the single wall puncture technique even after accounting for differences between residents and prior levels of experience. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Invasive hemodynamic monitoring is an essential component in the care of critically ill patient [1–3]. Indications to continuous blood pressure monitoring using an arterial catheter include the need to administer vasoactive or inotropic drugs and patients requiring frequent arterial blood gas analysis [4–6]. Among the different body sites, the radial artery is the most commonly used one due to low complication rates compared to other locations [7–9]. Recently, the radial arterial site has been also used to perform cardiac catheterization as a safer alternative to the traditional femoral artery approach [10]. Cannulation of the radial artery can sometimes be difficult and require multiple attempts. Two different insertion techniques are commonly used in order to cannulate the radial artery: the single wall puncture and the transfixing technique [11]. The single wall puncture follows the principles described by Seldinger in which the catheter is threaded over a guide wire after obtaining blood flow [12]. In the ☆ Funding: Department of Anesthesiology, Northwestern University. ⁎ Corresponding author. Department of Anesthesiology, Northwestern University, F5-704, Chicago, Illinois. Tel.: +1 312 472 3573. E-mail address: [email protected] (G.S. De Oliveira). 0883-9441/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcrc.2014.01.001
transfixing technique, the posterior wall of the artery is also deliberately punctured, and the needle/catheter is then withdrawn slowly until a steady or pulsatile flow of blood is observed. The guide wire is then advanced through the needle/catheter, and the catheter is advanced over the needle and guide wire. The choice of arterial line insertion technique is frequently based on provider preference due to the lack of studies comparing the success of these methods. It is also currently unknown if a specific technique can result in a greater success rate of arterial line cannulation by trainees. This question is important because multiple attempts can not only lead to patient discomfort but should be avoided to minimize complications [13]. The major objective of the current investigation was to compare the single wall puncture method to the transfixing method on the success of radial artery cannulation. We hypothesized that the transfixing technique would lead to a greater success rate of radial artery cannulation when compared to the single wall technique. 2. Methods This study was a prospective and randomized investigation. Study approval was obtained from the Northwestern University Institutional
475.e8
G.S. De Oliveira Jr. et al. / Journal of Critical Care 29 (2014) 475.e7–475.e10
Review Board (IRB identifier: STU00032387), and written informed consent was obtained from all the study participants (anesthesiology residents and patients). Anesthesiolgy residents during a cardiothoracic anesthesia rotation that had performed a minimum of thirty radial artery cannulations using both study methods were eligible for participation. Residents' log of procedures were utilized to determine the number of catheter cannulations performed. Each resident was scheduled to sequentially perform radial arterial cannulation in twelve patients. The method of insertion was randomized using a computer generated table of random numbers for each resident such that an equal number of attempts with each method would be performend. Prior to participation residents were required to watch a short video developed by one of the investigators (AS) demonstrating the single puncture and transfixing methods. Eligible patients were males and females undergoing cardiac or thoracic surgery. Patients younger than 18 years, with a history of renal failure requiring dialysis, pregnant patients and history of radial arterial line insertion in the last month were not recruited. Method or cannulation assignments were sealed in sequentially numbered opaque envelopes that were opened by one of the investigators after written informed consent was obtained from the patient just prior to catheter insertion. Reason for exclusion from the study following randomization were resident and/or patient request. Patients were brought to the operating room and standard American Society of Anesthesiology monitors were applied. All patients received 1 to 2 mg of intravenous midazolam as a sedative. The patients’ wrist were placed in extension (using one small towel rolled under the wrist, the same for all patients) and the fingers were taped to an arm board for radial exposure which provided a 30 to 45 degrees wrist extension and immobilization. Following sterile field preparation, palpation of radial artery was performed and 2 ml of lidocaine was used to anesthetize the skin. A 20-gauge polyurethane catheter with a built-in guide wire mechanism was used for radial cannulation in all subjects (Radial Artery Catheterization Set; Arrow International, Reading, PA). After palpation of the radial arterial pulse, the arterial catheter was inserted for both groups at an angle of 30 to 45 degrees. In the single wall group, after arterial blood flow was observed in the wire chamber, the arterial catheterization set was dropped to an angle of 5° to 20° and the guide wire was advanced. The catheter was then threaded over the guide wire. In the transfixing group, after blood flow was seen in the wire chamber, the arterial catheter was advanced through the posterior wall of the radial artery. The catheter was then withdrawn slowly until arterial blood flow was again visualized. The guide wire was then inserted and the catheter threaded over the guide wire. Confirmation of correct catheter placement was achieved by observing the presence of an arterial waveform upon connecting the catheter to a transducing system (Pressure Monitoring Kit; Edwards Lifesciences, Irvine, CA). An attempt at radial artery cannulation was defined by the attempt to advance the catheter only after the presence of blood within the wire chamber of the catheterization set was identified. Each subsequent attempt was performed with a new catheterization set. No feedback was provided by the attending anesthesiolgy during the performance. A maximum of 2 cannulation attempts per radial site per resident (4 total) was allowed for patient safety and minimize potential morbidity. Total time to cannulation was defined as the total time (seconds) between first skin puncture to the time to obtain a stable arterial waveform. If more than one cannulation attempt was required, the time of individual skin puncture until the operator withdraw the catheter from the skin or a stable waveform was obtained were recorded and added. The time between attempts needed to obtain an prepare a new catheter were not included in the time determination. Other data collected include patient characteristics, American Society of Anesthesiologists physical status, and history of coronary artery dissease. Patients were also followed up by an investigator blinded to the group allocation durging hospital stay
to assess the development of ischemic complications or nerve damage related to arterial catheter cannulation. All data were collected by an independent observer (KB or AS) not involved with patient care. The primary outcome was defined as a successful cannulation of the radial artery confirmed by the presence of a stable artery wave form achieved in 4 or less attempts. Residents entrance on the study was based on a convenience sample of the number of eligible anesthesiology residents assigned to the cardiothoracic surgery rotation a six month study period. The number of repetitions per resident was based on the number of arterial lines placed by a resident during the cadfriothoracic rotation assuming that patient eligibility, patient consent, and availability of study personnel would be achieved in 50% of available patients. One hundred twenty-six of a possible 144 attempts were studied with 65 attempts allocated to transfixation and 61 attempots allocated to the direct puncture technique. Group sample sizes of 65 subjects per transfixation and 61 subjects in the direct puncture group would achieve 89% power at α = .05 to detect a difference between groups of 20% using the Fisher’s Exact Test assuming that 95% of pateints in the transfixation group and 75% in the direct puncture group would be successfully cannulated. The difference of 20% in successful cannulation was selected as it represents a clinically important difference. The sample size calculation was made using PASS version 11.0.10 release date August, 9, 2012 (NCSS, LLC, Kaysville, UT). The Shapiro-Wilk test was used to evaluate the hypothesis of normal distribution. Normally distributed interval data are reported as mean (SD) and non-normally distributed interval and ordinal data are reported as median (range or interquartile range [IQR]). The primary outcome successful radial artery cannulation was compared between direct puncture and transfixing method using Fisher exact test. Successful cannulation on the first attempt was compared using Fisher exact test. A Kaplan-Meyer plot was constructed and the time to successful arterial cannulation was compared using the log rank test. The time to catheter placement and number of catheterization attempts was analyzed using a generalized linear mixed model with technique, resident and resident and cannulation technique as fixed factors and the total number of patients including the current patient that the resident had attempted to place an arterial catheter as a covariate. We choose the total number of attempt for each resident rather than the sequence number in the current study to assess for a training effect since there was variability in prior experience among resident in the study. All reported P values are 2 tailed. The criterion for rejection of the null hypothesis was a 2-tailed P ≤ .05. Statistical analysis was performed using R version 3.0.1, release date 5/16/2013 (The R Foundation for Statistical Computing, Vienna, Austria). 3. Results Twelve anesthesiology residents and 126 patients participated in the study. Baseline characteristics of the patients in the two study groups were not different (Table 1). The method of cannulation, success by attempt and time for cannulation for each resident participant is shown in Fig. 1. The transfixing technique was performed in 65 patients and the single wall puncture method in 61 patients. One resident performed arterial cannulation in only 3 patients. Resident participants had performed a median (IQR) of 50 (50-80) arterial line insertions before participating. The radial artery was successfully cannulated in 56 of 65 (86%) patients in the transfixing and 54 of 61 (88%) patients in the single wall technique group (difference 2%; 95% CI, 14-9; P = 0.8). Cannulation was successfully on the first attempt in 25 out of 65 (38%) patients in the transfixing group compared to 33 out of 61 (54%) patients in the single wall group (difference 16%; 95% CI, 32-2; P = 0.1). Three resident physicians successfully cannulated the radial artery on all attempts. In one or more attempts, 3 resident physicians failed using the direct puncture but not the transfixing technique, 3 failed
G.S. De Oliveira Jr. et al. / Journal of Critical Care 29 (2014) 475.e7–475.e10 Table 1 Baseline characteristics of the study groups
Age (years) Gender Male Female Body Mass Index (Kg/m2) ASA Class II III IV MAP before insertion(mmHg) Coronary artery disease Yes No
Transfixing (n = 65)
Single wall (n = 61)
P value
65.5 ± 12.9
62.2 ± 13.0
0.15 0.15
43 22 28.8 ± 6.5 4 55 6 84 (75 to 96) 30 35
32 29 28.0 ± 5.6 6 44 11 84 (72 to 96)
0.52 0.24
0.72 0.48
24 37
Data presented as mean ± SD, median (interquartile range) or counts (n).
Fig. 1. Figure depicts the resident allocation by sequence to transfixing or single wall puncture technique. Red bars represent attempts using the transfixing technique while blue bars represent attempts using the single wall technique. The height of the bar is the total time required for successful of failed insertion. Tick marks on the y axis represent 120 seconds. Attempts marked with an asterisk (*) were greater than 240 seconds.
475.e9
using the transfixing but not the direct puncture technique and 3 failed using both the direct puncture and transfixing techniques. The percentage of successful cannulations by the total time to cannulate and the number of subjects exposed at time intervals is shown in Fig. 2. The median time for cannulation was longer in the transfixing group, median (IQR) of 105 (69-176) seconds compared to median (IQR) of 65 (25-114) seconds in the direct puncture group, P = .009. Generalized linear modeling of the total time attempting cannulation identified technique (P = .003) and resident (P = .004) but not resident by technique (P = 0.17) or total number of arterial catheterizations attempts performed by the resident (P = 0.78) as significant variables in the model. The fitted model was significantly better than the intercept only model (P = .002). The difference in estimated marginal means for the total attempt time for the transfixing compared to direct puncture technique was 41 (95% CI, 14-68) seconds. Examination of the total attempt times for the 12 residents revealed that resident #5 (mean time per attempt 185 [95% CI, 144-226]) seconds required more time per attempt than the other resident in the study. The difference in estimated marginal means for the total attempt time for the transfixing compared to direct puncture technique was 33 (95% CI, 8-59) seconds (P = .01) even when resident #5 was removed from the analysis. There was no difference among residents (P = .13), the interaction of resident by technique (P = .22) or total number of prior cannulation attempts performed by the resident (P = .74) when resident 5 was removed from the model. Cannulation technique (P = .04) and the interaction of cannulation technique and resident (P = .01) were identified as significant factors in the model of the number of cannulation attempts. Resident (P = .11) and total number of prior cannulation attempts performed by the resident (0.70) were not significant variables in the model. The fitted model was significantly better than the intercept only model (P = .01). Examination of the data revealed that resident #4 required an average of 3.8 attempt using the transfixing but only 1.5 using the direct puncture technique. With resident #4 removed from the analysis the difference in estimated marginal differences were not different for cannulation technique (P = .33) or the interaction of cannulation technique and resident (P = .46). The number of cannulation attempts were not correlated to patients’ age (P = 0.76), were not different by patients’ gender
Fig. 2. Kaplan-Meier plot of proportion of subjects with successful inserted the arterial line by study groups. The median total time to cannulation was longer in the transfixing group, median (IQR) of 105 (69-176) seconds compared to median (IQR) of 65 (25-114) seconds in the direct puncture group, P = .009. Data were analyzed using the log-rank test.
475.e10
G.S. De Oliveira Jr. et al. / Journal of Critical Care 29 (2014) 475.e7–475.e10
(P = 0.78) and were not correlated with values of baseline mean arterial pressure (P = 0.97). In addition, the number of attempts was not correlated with patients’ body mass index (P = .77). No ischemic complications or cases of nerve damage were detected in any of the subjects participating in the study. 4. Discussion The most important finding of the current investigation was that, in contrast to what we originally hypothesized, successful arterial cannulation performed either with a transfixing or the single wall puncture technique resulted in similar proportion of successful cannulations even when performed by resident physicians with limited experience. The total time required for cannulation was longer using the transfixing technique compared to the single wall technique even when corrected for outliers. Although there appeared to be a decrease in the number of cannulation attempts using the single wall technique, this finding was influenced by the difference in a single participant. In our clinical practice, it is believed that the transfixing technique is more successful to cannulate the radial artery than the single wall technique. In several circumstances, a backflow can be seen but the catheter cannot be threaded successfully into the vessel. The cause of failure can be a tangential orientation of the catheter in relation to the artery, tortuosity of the vessel, or most commonly, arterial vasospasm. In these circumstances, the transfixing technique might improve success rate by allowing optimal blood flow before the guide wire is advanced. We did not use ultrasound to identify the artery before the vessel was punctured. The main objective of our investigation was to examine if the insertion technique would result in a more successful cannulation after the artery was identified. This is the reason why we defined an attempt by the presence of blood within the wire chamber of the catheterization set and not by the number of skin punctures. Since we performed the arterial cannulation in awake patients with normal mean arterial pressures, the identification of the artery by palpation was not a barrier to successful cannulation in the current study. We did not detect any clinically significant complication that could have been attributed to the arterial line insertion. However, it is important to note that we limited each radial artery to two attempts by resident physicians to minimize complications. Russel et al have previously determined that difficulty in percutaneous insertion in radial artery site was associated with much greater rates of complications [14]. In contrast, Jones et al compared patients who underwent arterial cannulation using a single wall versus a transfixing technique using ultrasound at 1 hour and 5 days after catheter removal and did not find differences in the quality of post-cannulation blood flow or the incidence of thrombus formation [15]. Our study can only be interpreted with the context of its limitations. We evaluated only anesthesia trainees and it is possible that the examination of more experienced anesthesiologists would lead to different results. We attempted to minimize an initial learning effect by only recruiting residents with some experience in arterial line insertions (N30 arterial line insertions). In addition, we also attempted to minimize a learning effect of randomly exposing each resident to both techniques. Another limitation is that we only examined awake and normotensive patients. It remains to be further examined if the single wall technique would show similar results than the transfixing technique in hypotensive patients. We did not add the
time between attempts when we calculated the total time to cannulation in cases requiring more than one attempt to successful cannulation. The inclusion of the time between attempts would likely represent a more clinically meaningful outcome; however, it could also introduce significant confounding bias and potentially result in an invalid comparison. Our study had a relatively small sample size, and it is possible that variations in the sampling process could have resulted in the lack of significant differences between the direct puncture and the transfixing technique. Future larger studies are needed in order to confirm or refute the comparison of the direct puncture and transfixing techniques. In contrast, our findings strongly support the lack of superiority of the transfixing technique compared to the direct puncture technique even in resident physicians with limited prior experience. In summary, contrary to what we hypothesized, the single wall technique resulted in a similar success rate but required less time to successful cannulate the radial artery compared with the transfixing technique. The choice of the single wall technique seems to be an acceptable option for cannulation of the radial artery in hemodynamically stable patients. References [1] Joshi B, Ono M, Brown C, Brady K, Easley RB, Yenokyan G, et al. Predicting the limits of cerebral autoregulation during cardiopulmonary bypass. Anesth Analg 2012;114:503–10. [2] Garnier RP, van der Spoel AG, Sibarani-Ponsen R, Markhorst DG, Boer C. Level of agreement between Nexfin non-invasive arterial pressure with invasive arterial pressure measurements in children. Br J Anaesth 2012;109:609–15. [3] Yamada T, Tsutsui M, Sugo Y, Sato T, Akazawa T, Sato N, et al. Multicenter study verifying a method of noninvasive continuous cardiac output measurement using pulse wave transit time: a comparison with intermittent bolus thermodilution cardiac output. Anesth Analg 2012;115:82–7. [4] Unzueta C, Tusman G, Suarez-Sipmann F, Böhm S, Moral V. Alveolar recruitment improves ventilation during thoracic surgery: a randomized controlled trial. Br J Anaesth 2012;108:517–24. [5] Kim SY, Lee JS, Kim WO, Sun JM, Kwon MK, Kil HK. Evaluation of radial and ulnar blood flow after radial artery cannulation with 20- and 22-gauge cannulae using duplex Doppler ultrasound. Anaesthesia 2012;67:1138–45. [6] Weiskopf RB, Feiner J, Toy P, Twiford J, Shimabukuro D, Lieberman J, et al. Fresh and stored red blood cell transfusion equivalently induce subclinical pulmonary gas exchange deficit in normal humans. Anesth Analg 2012;114:511–9. [7] Fischer MO, Avram R, Cârjaliu I, Massetti M, Gérard JL, Hanouz JL, et al. Noninvasive continuous arterial pressure and cardiac index monitoring with Nexfin after cardiac surgery. Br J Anaesth 2012;109:514–21. [8] Meidert AS, Huber W, Hapfelmeier A, Schöfthaler M, Müller JN, Langwieser N, et al. Evaluation of the radial artery applanation tonometry technology for continuous noninvasive blood pressure monitoring compared with central aortic blood pressure measurements in patients with multiple organ dysfunction syndrome. J Crit Care 2013;28:908–12. [9] Ilies C, Bauer M, Berg P, Rosenberg J, Hedderich J, Bein B, et al. Investigation of the agreement of a continuous non-invasive arterial pressure device in comparison with invasive radial artery measurement. Br J Anaesth 2012;108:202–10. [10] Byrne RA, Cassese S, Linhardt M, Kastrati A. Vascular access and closure in coronary angiography and percutaneous intervention. Nat Rev Cardiol 2013;10:27–40. [11] Beards SC, Doedens L, Jackson A, Lipman J. A comparison of arterial lines and insertion techniques in critically ill patients. Anaesthesia 1994;49:968–73. [12] Seldinger SI. Catheter replacement of the needle in percutaneous arteriography; a new technique. Acta radiol 1953;39:368–76. [13] Kanei Y, Kwan T, Nakra NC, Liou M, Huang Y, Vales LL. Saito S Transradial cardiac catheterization: a review of access site complications. Catheter Cardiovasc Interv 2011;78:840–6. [14] Russell JA, Joel M, Hudson RJ, Mangano DT, Schlobohm RM. Prospective evaluation of radial and femoral artery catheterization sites in critically ill adults. Crit Care Med 1983;11:936–9. [15] Jones RM, Hill AB, Nahrwold ML, Bolles RE. The effect of method of radial artery cannulation on postcannulation blood flow and thrombus formation. Anesthesiology 1981;55:76–8.
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
The effect of the arterial catheter insertion technique on the success of radial artery cannulation: A prospective and randomized study☆ Gildasio S. De Oliveira Jr., MD, MSC a,⁎, Katharina Beckmann, MD a, Alain Salvacion, MD a, John Kim, MD b, Saadia Sherwani, MD a, Robert J. McCarthy, PharmD a a b
Department of Anesthesiology, Northwestern University, Chicago, IL, USA Department of Surgery, Northwestern University, Chicago, IL, USA
a r t i c l e
i n f o
Keywords: Arterial line cannulation
a b s t r a c t Purpose: The main objective of the current investigation was to compare a single wall puncture to vessel transfixing on the success of radial artery cannulation by resident physicians. Material and Methods: The study was a prospective and randomized investigation. Twelve anesthesiology residents performed radial arterial insertions in 126 patients using both the single wall and vessel transfixing technique in random order. The primary outcome was successful cannulation of the radial artery in 4 or less attempts. Other data collected included the total number of attempts and total time to catheter cannulation. Results: Successful radial artery cannulation was achieved in 88% and 86% of patients using the transfixing technique and single wall group, respectively (difference 2%; 95% CI, 14-9, P = 0.8, Fisher exact test). Cannulation was successfully on the first attempt in 38% of the transfixing compared to 54% using the single wall technique (difference − 16%; 95% CI, 32-2, P = 0.1, Fisher Exact test). The median (interquartile range) time to successful cannulation was longer in the transfixing group, 105 (69-176) seconds compared to 65 (25-114) seconds in the single puncture group (P = .009, log-rank test). Conclusions: Our findings suggest that there does not appear to be an advantage of the transfixing technique over the single wall puncture method for cannulating the radial artery by resident physicians. Cannulation was achieved in shorter time using the single wall puncture technique even after accounting for differences between residents and prior levels of experience. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Invasive hemodynamic monitoring is an essential component in the care of critically ill patient [1–3]. Indications to continuous blood pressure monitoring using an arterial catheter include the need to administer vasoactive or inotropic drugs and patients requiring frequent arterial blood gas analysis [4–6]. Among the different body sites, the radial artery is the most commonly used one due to low complication rates compared to other locations [7–9]. Recently, the radial arterial site has been also used to perform cardiac catheterization as a safer alternative to the traditional femoral artery approach [10]. Cannulation of the radial artery can sometimes be difficult and require multiple attempts. Two different insertion techniques are commonly used in order to cannulate the radial artery: the single wall puncture and the transfixing technique [11]. The single wall puncture follows the principles described by Seldinger in which the catheter is threaded over a guide wire after obtaining blood flow [12]. In the ☆ Funding: Department of Anesthesiology, Northwestern University. ⁎ Corresponding author. Department of Anesthesiology, Northwestern University, F5-704, Chicago, Illinois. Tel.: +1 312 472 3573. E-mail address: [email protected] (G.S. De Oliveira). 0883-9441/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcrc.2014.01.001
transfixing technique, the posterior wall of the artery is also deliberately punctured, and the needle/catheter is then withdrawn slowly until a steady or pulsatile flow of blood is observed. The guide wire is then advanced through the needle/catheter, and the catheter is advanced over the needle and guide wire. The choice of arterial line insertion technique is frequently based on provider preference due to the lack of studies comparing the success of these methods. It is also currently unknown if a specific technique can result in a greater success rate of arterial line cannulation by trainees. This question is important because multiple attempts can not only lead to patient discomfort but should be avoided to minimize complications [13]. The major objective of the current investigation was to compare the single wall puncture method to the transfixing method on the success of radial artery cannulation. We hypothesized that the transfixing technique would lead to a greater success rate of radial artery cannulation when compared to the single wall technique. 2. Methods This study was a prospective and randomized investigation. Study approval was obtained from the Northwestern University Institutional
475.e8
G.S. De Oliveira Jr. et al. / Journal of Critical Care 29 (2014) 475.e7–475.e10
Review Board (IRB identifier: STU00032387), and written informed consent was obtained from all the study participants (anesthesiology residents and patients). Anesthesiolgy residents during a cardiothoracic anesthesia rotation that had performed a minimum of thirty radial artery cannulations using both study methods were eligible for participation. Residents' log of procedures were utilized to determine the number of catheter cannulations performed. Each resident was scheduled to sequentially perform radial arterial cannulation in twelve patients. The method of insertion was randomized using a computer generated table of random numbers for each resident such that an equal number of attempts with each method would be performend. Prior to participation residents were required to watch a short video developed by one of the investigators (AS) demonstrating the single puncture and transfixing methods. Eligible patients were males and females undergoing cardiac or thoracic surgery. Patients younger than 18 years, with a history of renal failure requiring dialysis, pregnant patients and history of radial arterial line insertion in the last month were not recruited. Method or cannulation assignments were sealed in sequentially numbered opaque envelopes that were opened by one of the investigators after written informed consent was obtained from the patient just prior to catheter insertion. Reason for exclusion from the study following randomization were resident and/or patient request. Patients were brought to the operating room and standard American Society of Anesthesiology monitors were applied. All patients received 1 to 2 mg of intravenous midazolam as a sedative. The patients’ wrist were placed in extension (using one small towel rolled under the wrist, the same for all patients) and the fingers were taped to an arm board for radial exposure which provided a 30 to 45 degrees wrist extension and immobilization. Following sterile field preparation, palpation of radial artery was performed and 2 ml of lidocaine was used to anesthetize the skin. A 20-gauge polyurethane catheter with a built-in guide wire mechanism was used for radial cannulation in all subjects (Radial Artery Catheterization Set; Arrow International, Reading, PA). After palpation of the radial arterial pulse, the arterial catheter was inserted for both groups at an angle of 30 to 45 degrees. In the single wall group, after arterial blood flow was observed in the wire chamber, the arterial catheterization set was dropped to an angle of 5° to 20° and the guide wire was advanced. The catheter was then threaded over the guide wire. In the transfixing group, after blood flow was seen in the wire chamber, the arterial catheter was advanced through the posterior wall of the radial artery. The catheter was then withdrawn slowly until arterial blood flow was again visualized. The guide wire was then inserted and the catheter threaded over the guide wire. Confirmation of correct catheter placement was achieved by observing the presence of an arterial waveform upon connecting the catheter to a transducing system (Pressure Monitoring Kit; Edwards Lifesciences, Irvine, CA). An attempt at radial artery cannulation was defined by the attempt to advance the catheter only after the presence of blood within the wire chamber of the catheterization set was identified. Each subsequent attempt was performed with a new catheterization set. No feedback was provided by the attending anesthesiolgy during the performance. A maximum of 2 cannulation attempts per radial site per resident (4 total) was allowed for patient safety and minimize potential morbidity. Total time to cannulation was defined as the total time (seconds) between first skin puncture to the time to obtain a stable arterial waveform. If more than one cannulation attempt was required, the time of individual skin puncture until the operator withdraw the catheter from the skin or a stable waveform was obtained were recorded and added. The time between attempts needed to obtain an prepare a new catheter were not included in the time determination. Other data collected include patient characteristics, American Society of Anesthesiologists physical status, and history of coronary artery dissease. Patients were also followed up by an investigator blinded to the group allocation durging hospital stay
to assess the development of ischemic complications or nerve damage related to arterial catheter cannulation. All data were collected by an independent observer (KB or AS) not involved with patient care. The primary outcome was defined as a successful cannulation of the radial artery confirmed by the presence of a stable artery wave form achieved in 4 or less attempts. Residents entrance on the study was based on a convenience sample of the number of eligible anesthesiology residents assigned to the cardiothoracic surgery rotation a six month study period. The number of repetitions per resident was based on the number of arterial lines placed by a resident during the cadfriothoracic rotation assuming that patient eligibility, patient consent, and availability of study personnel would be achieved in 50% of available patients. One hundred twenty-six of a possible 144 attempts were studied with 65 attempts allocated to transfixation and 61 attempots allocated to the direct puncture technique. Group sample sizes of 65 subjects per transfixation and 61 subjects in the direct puncture group would achieve 89% power at α = .05 to detect a difference between groups of 20% using the Fisher’s Exact Test assuming that 95% of pateints in the transfixation group and 75% in the direct puncture group would be successfully cannulated. The difference of 20% in successful cannulation was selected as it represents a clinically important difference. The sample size calculation was made using PASS version 11.0.10 release date August, 9, 2012 (NCSS, LLC, Kaysville, UT). The Shapiro-Wilk test was used to evaluate the hypothesis of normal distribution. Normally distributed interval data are reported as mean (SD) and non-normally distributed interval and ordinal data are reported as median (range or interquartile range [IQR]). The primary outcome successful radial artery cannulation was compared between direct puncture and transfixing method using Fisher exact test. Successful cannulation on the first attempt was compared using Fisher exact test. A Kaplan-Meyer plot was constructed and the time to successful arterial cannulation was compared using the log rank test. The time to catheter placement and number of catheterization attempts was analyzed using a generalized linear mixed model with technique, resident and resident and cannulation technique as fixed factors and the total number of patients including the current patient that the resident had attempted to place an arterial catheter as a covariate. We choose the total number of attempt for each resident rather than the sequence number in the current study to assess for a training effect since there was variability in prior experience among resident in the study. All reported P values are 2 tailed. The criterion for rejection of the null hypothesis was a 2-tailed P ≤ .05. Statistical analysis was performed using R version 3.0.1, release date 5/16/2013 (The R Foundation for Statistical Computing, Vienna, Austria). 3. Results Twelve anesthesiology residents and 126 patients participated in the study. Baseline characteristics of the patients in the two study groups were not different (Table 1). The method of cannulation, success by attempt and time for cannulation for each resident participant is shown in Fig. 1. The transfixing technique was performed in 65 patients and the single wall puncture method in 61 patients. One resident performed arterial cannulation in only 3 patients. Resident participants had performed a median (IQR) of 50 (50-80) arterial line insertions before participating. The radial artery was successfully cannulated in 56 of 65 (86%) patients in the transfixing and 54 of 61 (88%) patients in the single wall technique group (difference 2%; 95% CI, 14-9; P = 0.8). Cannulation was successfully on the first attempt in 25 out of 65 (38%) patients in the transfixing group compared to 33 out of 61 (54%) patients in the single wall group (difference 16%; 95% CI, 32-2; P = 0.1). Three resident physicians successfully cannulated the radial artery on all attempts. In one or more attempts, 3 resident physicians failed using the direct puncture but not the transfixing technique, 3 failed
G.S. De Oliveira Jr. et al. / Journal of Critical Care 29 (2014) 475.e7–475.e10 Table 1 Baseline characteristics of the study groups
Age (years) Gender Male Female Body Mass Index (Kg/m2) ASA Class II III IV MAP before insertion(mmHg) Coronary artery disease Yes No
Transfixing (n = 65)
Single wall (n = 61)
P value
65.5 ± 12.9
62.2 ± 13.0
0.15 0.15
43 22 28.8 ± 6.5 4 55 6 84 (75 to 96) 30 35
32 29 28.0 ± 5.6 6 44 11 84 (72 to 96)
0.52 0.24
0.72 0.48
24 37
Data presented as mean ± SD, median (interquartile range) or counts (n).
Fig. 1. Figure depicts the resident allocation by sequence to transfixing or single wall puncture technique. Red bars represent attempts using the transfixing technique while blue bars represent attempts using the single wall technique. The height of the bar is the total time required for successful of failed insertion. Tick marks on the y axis represent 120 seconds. Attempts marked with an asterisk (*) were greater than 240 seconds.
475.e9
using the transfixing but not the direct puncture technique and 3 failed using both the direct puncture and transfixing techniques. The percentage of successful cannulations by the total time to cannulate and the number of subjects exposed at time intervals is shown in Fig. 2. The median time for cannulation was longer in the transfixing group, median (IQR) of 105 (69-176) seconds compared to median (IQR) of 65 (25-114) seconds in the direct puncture group, P = .009. Generalized linear modeling of the total time attempting cannulation identified technique (P = .003) and resident (P = .004) but not resident by technique (P = 0.17) or total number of arterial catheterizations attempts performed by the resident (P = 0.78) as significant variables in the model. The fitted model was significantly better than the intercept only model (P = .002). The difference in estimated marginal means for the total attempt time for the transfixing compared to direct puncture technique was 41 (95% CI, 14-68) seconds. Examination of the total attempt times for the 12 residents revealed that resident #5 (mean time per attempt 185 [95% CI, 144-226]) seconds required more time per attempt than the other resident in the study. The difference in estimated marginal means for the total attempt time for the transfixing compared to direct puncture technique was 33 (95% CI, 8-59) seconds (P = .01) even when resident #5 was removed from the analysis. There was no difference among residents (P = .13), the interaction of resident by technique (P = .22) or total number of prior cannulation attempts performed by the resident (P = .74) when resident 5 was removed from the model. Cannulation technique (P = .04) and the interaction of cannulation technique and resident (P = .01) were identified as significant factors in the model of the number of cannulation attempts. Resident (P = .11) and total number of prior cannulation attempts performed by the resident (0.70) were not significant variables in the model. The fitted model was significantly better than the intercept only model (P = .01). Examination of the data revealed that resident #4 required an average of 3.8 attempt using the transfixing but only 1.5 using the direct puncture technique. With resident #4 removed from the analysis the difference in estimated marginal differences were not different for cannulation technique (P = .33) or the interaction of cannulation technique and resident (P = .46). The number of cannulation attempts were not correlated to patients’ age (P = 0.76), were not different by patients’ gender
Fig. 2. Kaplan-Meier plot of proportion of subjects with successful inserted the arterial line by study groups. The median total time to cannulation was longer in the transfixing group, median (IQR) of 105 (69-176) seconds compared to median (IQR) of 65 (25-114) seconds in the direct puncture group, P = .009. Data were analyzed using the log-rank test.
475.e10
G.S. De Oliveira Jr. et al. / Journal of Critical Care 29 (2014) 475.e7–475.e10
(P = 0.78) and were not correlated with values of baseline mean arterial pressure (P = 0.97). In addition, the number of attempts was not correlated with patients’ body mass index (P = .77). No ischemic complications or cases of nerve damage were detected in any of the subjects participating in the study. 4. Discussion The most important finding of the current investigation was that, in contrast to what we originally hypothesized, successful arterial cannulation performed either with a transfixing or the single wall puncture technique resulted in similar proportion of successful cannulations even when performed by resident physicians with limited experience. The total time required for cannulation was longer using the transfixing technique compared to the single wall technique even when corrected for outliers. Although there appeared to be a decrease in the number of cannulation attempts using the single wall technique, this finding was influenced by the difference in a single participant. In our clinical practice, it is believed that the transfixing technique is more successful to cannulate the radial artery than the single wall technique. In several circumstances, a backflow can be seen but the catheter cannot be threaded successfully into the vessel. The cause of failure can be a tangential orientation of the catheter in relation to the artery, tortuosity of the vessel, or most commonly, arterial vasospasm. In these circumstances, the transfixing technique might improve success rate by allowing optimal blood flow before the guide wire is advanced. We did not use ultrasound to identify the artery before the vessel was punctured. The main objective of our investigation was to examine if the insertion technique would result in a more successful cannulation after the artery was identified. This is the reason why we defined an attempt by the presence of blood within the wire chamber of the catheterization set and not by the number of skin punctures. Since we performed the arterial cannulation in awake patients with normal mean arterial pressures, the identification of the artery by palpation was not a barrier to successful cannulation in the current study. We did not detect any clinically significant complication that could have been attributed to the arterial line insertion. However, it is important to note that we limited each radial artery to two attempts by resident physicians to minimize complications. Russel et al have previously determined that difficulty in percutaneous insertion in radial artery site was associated with much greater rates of complications [14]. In contrast, Jones et al compared patients who underwent arterial cannulation using a single wall versus a transfixing technique using ultrasound at 1 hour and 5 days after catheter removal and did not find differences in the quality of post-cannulation blood flow or the incidence of thrombus formation [15]. Our study can only be interpreted with the context of its limitations. We evaluated only anesthesia trainees and it is possible that the examination of more experienced anesthesiologists would lead to different results. We attempted to minimize an initial learning effect by only recruiting residents with some experience in arterial line insertions (N30 arterial line insertions). In addition, we also attempted to minimize a learning effect of randomly exposing each resident to both techniques. Another limitation is that we only examined awake and normotensive patients. It remains to be further examined if the single wall technique would show similar results than the transfixing technique in hypotensive patients. We did not add the
time between attempts when we calculated the total time to cannulation in cases requiring more than one attempt to successful cannulation. The inclusion of the time between attempts would likely represent a more clinically meaningful outcome; however, it could also introduce significant confounding bias and potentially result in an invalid comparison. Our study had a relatively small sample size, and it is possible that variations in the sampling process could have resulted in the lack of significant differences between the direct puncture and the transfixing technique. Future larger studies are needed in order to confirm or refute the comparison of the direct puncture and transfixing techniques. In contrast, our findings strongly support the lack of superiority of the transfixing technique compared to the direct puncture technique even in resident physicians with limited prior experience. In summary, contrary to what we hypothesized, the single wall technique resulted in a similar success rate but required less time to successful cannulate the radial artery compared with the transfixing technique. The choice of the single wall technique seems to be an acceptable option for cannulation of the radial artery in hemodynamically stable patients. References [1] Joshi B, Ono M, Brown C, Brady K, Easley RB, Yenokyan G, et al. Predicting the limits of cerebral autoregulation during cardiopulmonary bypass. Anesth Analg 2012;114:503–10. [2] Garnier RP, van der Spoel AG, Sibarani-Ponsen R, Markhorst DG, Boer C. Level of agreement between Nexfin non-invasive arterial pressure with invasive arterial pressure measurements in children. Br J Anaesth 2012;109:609–15. [3] Yamada T, Tsutsui M, Sugo Y, Sato T, Akazawa T, Sato N, et al. Multicenter study verifying a method of noninvasive continuous cardiac output measurement using pulse wave transit time: a comparison with intermittent bolus thermodilution cardiac output. Anesth Analg 2012;115:82–7. [4] Unzueta C, Tusman G, Suarez-Sipmann F, Böhm S, Moral V. Alveolar recruitment improves ventilation during thoracic surgery: a randomized controlled trial. Br J Anaesth 2012;108:517–24. [5] Kim SY, Lee JS, Kim WO, Sun JM, Kwon MK, Kil HK. Evaluation of radial and ulnar blood flow after radial artery cannulation with 20- and 22-gauge cannulae using duplex Doppler ultrasound. Anaesthesia 2012;67:1138–45. [6] Weiskopf RB, Feiner J, Toy P, Twiford J, Shimabukuro D, Lieberman J, et al. Fresh and stored red blood cell transfusion equivalently induce subclinical pulmonary gas exchange deficit in normal humans. Anesth Analg 2012;114:511–9. [7] Fischer MO, Avram R, Cârjaliu I, Massetti M, Gérard JL, Hanouz JL, et al. Noninvasive continuous arterial pressure and cardiac index monitoring with Nexfin after cardiac surgery. Br J Anaesth 2012;109:514–21. [8] Meidert AS, Huber W, Hapfelmeier A, Schöfthaler M, Müller JN, Langwieser N, et al. Evaluation of the radial artery applanation tonometry technology for continuous noninvasive blood pressure monitoring compared with central aortic blood pressure measurements in patients with multiple organ dysfunction syndrome. J Crit Care 2013;28:908–12. [9] Ilies C, Bauer M, Berg P, Rosenberg J, Hedderich J, Bein B, et al. Investigation of the agreement of a continuous non-invasive arterial pressure device in comparison with invasive radial artery measurement. Br J Anaesth 2012;108:202–10. [10] Byrne RA, Cassese S, Linhardt M, Kastrati A. Vascular access and closure in coronary angiography and percutaneous intervention. Nat Rev Cardiol 2013;10:27–40. [11] Beards SC, Doedens L, Jackson A, Lipman J. A comparison of arterial lines and insertion techniques in critically ill patients. Anaesthesia 1994;49:968–73. [12] Seldinger SI. Catheter replacement of the needle in percutaneous arteriography; a new technique. Acta radiol 1953;39:368–76. [13] Kanei Y, Kwan T, Nakra NC, Liou M, Huang Y, Vales LL. Saito S Transradial cardiac catheterization: a review of access site complications. Catheter Cardiovasc Interv 2011;78:840–6. [14] Russell JA, Joel M, Hudson RJ, Mangano DT, Schlobohm RM. Prospective evaluation of radial and femoral artery catheterization sites in critically ill adults. Crit Care Med 1983;11:936–9. [15] Jones RM, Hill AB, Nahrwold ML, Bolles RE. The effect of method of radial artery cannulation on postcannulation blood flow and thrombus formation. Anesthesiology 1981;55:76–8.