ࡱ >
_ bjbjzz ., B\B\
4 4 4 4 4 s! d ! 4 " $ $ $ $ $ $ $ ; j H " " " H ] g# g# g# " " g# " " g# g# J| uK\ )# . 2 P s 0 ` W# , " " g# " " " " " H H g# " " " " " " " " " " " " " " " "
> : Title
Radial Artery Blood Gas Sampling: A Randomised Controlled Trial of Lidocaine Local Anaesthesia
Running Head
A RCT of lidocaine for arterial puncture
Keywords
Pain; success; outcome; local; injected; subcutaneous; anaesthesia; anaesthetic; lidocaine; radial; artery; arterial; blood; gas; randomised; clinical; controlled; control; trial; needle.
Authors
Mr Ryckie G Wade MBBS MClinEd MRCS FHEA1, Dr Jim Crawfurd MBBS MRCP FCEM2, Dr Donna Wade BM, BCh, MRCP, FCEM2, Professor Richard Holland BA BM BCh DA DPH FFPH PhD3,4
Institutions
James Paget University Hospitals NHS Foundation Trust, Gorleston, UK, NR31 6LA
Consultant in Accident and Emergency, James Paget University Hospitals NHS Foundation Trust, Gorleston, UK, NR31 6LA
Professor of Public Health Medicine and Course Director, Norwich Medical School, University of East Anglia, Norfolk, UK, NR4 7TJ
Co-Director of the Norwich Research and Clnical Trials Unit, Norfolk and Norwich University Hospital NHS Foundation Trust, Norwich, NR4 7TJ
Correspondence
Ryckie G. Wade
4 Ashmead, Clifford, LS23 6PW, UK
E-mail: ryckiewade@gmail.com
Abstract
Introduction: Radial artery puncture is a common procedure and yet the role of local anaesthesia for reducing the pain of this procedure continues to be debated. Clinical practice is variable and there is potential for substantial financial savings. This is the first randomised trial to investigate the effectiveness of subcutaneously injected lidocaine anaesthesia on the perceived pain of radial artery puncture and the financial impact.
Methods: Between December 2012 and April 2013, 43 patients in the Emergency Department were randomised into the intervention group to receive lidocaine 1% 1ml subcutaneously or the control group (to receive no local anaesthesia) prior to radial artery puncture for blood gas sampling. Pain was rated on a 10cm visual analogue scale (VAS) and procedural variables collected for between group analyses.
Results: Overall, 41 participants were included. Subcutaneously injected lidocaine anaesthesia did not reduce the median pain of radial artery puncture (control 1.8 vs. intervention 1.6 cm, p=0.938). Those patients who had other systemically acting analgesia appeared to report reduced pain for radial artery puncture (0.60 vs. 2.30 cm, p=0.105) as did those where a smaller 25-gauge needle was used compared to the standard 22-gauge reported (1.40 vs. 4.35 cm, p=0.150), although these were not statistically significant. Anxious patients and those requesting local anaesthesia experienced relatively higher levels of pain.
Conclusion: Local anaesthesia did not reduce the perceived pain of radial artery puncture.
Abstract word count = 228
Article word count = 2243
Introduction
Radial artery puncture is a common method of obtaining blood for monitoring respiratory and metabolic parameters. The procedure, which many patients dread ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_1" \o "Lightower, 1977 #86" 1-3), is associated with considerable pain and anxiety. Therefore, it is reported that up to 60% of clinicians use subcutaneous local anaesthetic agents ADDIN EN.CITE Sado200589(4, 5)898917Sado, D.M.,Deakin, C.D.Local anaesthesia for venous cannulation and arterial blood gas sampling: are doctors using it?J R Soc MedJ R Soc Med158-160982005Hudon200610610610617Hudon, T.L.,Dukes, F.S.,Reilly, K.Use of Local Anesthesia for Arterial PuncturesAm J Crit CareAm J Crit Care595-599152006( HYPERLINK \l "_ENREF_4" \o "Sado, 2005 #89" 4, HYPERLINK \l "_ENREF_5" \o "Hudon, 2006 #106" 5) in an effort to reduce the pain of the procedure, especially in patients requiring multiple punctures. Conversely, clinicians who avoid using subcutaneous local anaesthesia purport that it leads to increased pain from two stabs (one for the anaesthetic, one to obtain arterial blood), tissue distension, the acidic pH and lower temperature of the solution. Also, some clinicians find that after injecting local anaesthesia that the pulse is more difficult to palpate so the procedure more difficult to complete. Clearly, inadvertent intravascular injection is another reason for caution. This is an important clinical dilemma which doctors face on a day-to-day basis and also carries potentially significant economic implications for health services.
In 1977, Lightower and Elliot ADDIN EN.CITE Lightower197786(1)868617Lightower, J.V.J.,Elliott, M.W.Local anaesthetic infiltration prior to arterial puncture for blood gas analysis: a survey of current practice and a randomized double blind placebo controlled trialJ R Coll Physicians LondJ R Coll Physicians Lond645-646311977( HYPERLINK \l "_ENREF_1" \o "Lightower, 1977 #86" 1) reported their quasi-randomised trial of subcutaneous local anaesthesia for arterial puncture although their article: lacked details of numerous important methodological steps, inadequately described the statistical analysis, was arguably underpowered for a three-group study and therefore we suggest that their conclusions are unreliable. More recent literature is still not in agreement as one group suggested injected mepivacine was superior to topical EMLA or control ADDIN EN.CITE Giner20001347(6)1347134717Giner, J.,Casan, P.,Belda, J.,Litvan, H.,Sanchis, J.[Use of the anesthetic cream EMLA in arterial punction]Rev Esp Anestesiol ReanimRev Esp Anestesiol Reanim63-664722000( HYPERLINK \l "_ENREF_6" \o "Giner, 2000 #1347" 6), whilst other groups have shown the reverse ADDIN EN.CITE Joly19981345(7, 8)1345134517Joly, L.M.,Spaulding, C.,Monchi, M.,Ali, O.S.,Weber, S.,Benhamou, D.Topical lidocaine-prilocaine cream (EMLA) versus local infiltration anesthesia for radial artery cannulationAnesth AnalgAnesth Analg403-406821998Kim200713461346134617Kim, J.Y.,Yoon, J.,Yoo, B.S.,Lee, S.H.,Choe, K.H.The effect of a eutectic mixture of local anesthetic cream on wrist pain during transradial coronary proceduresJ Invasive CardiolJ Invasive Cardiol6-91912007( HYPERLINK \l "_ENREF_7" \o "Joly, 1998 #1345" 7, HYPERLINK \l "_ENREF_8" \o "Kim, 2007 #1346" 8). Also, trials comparing different brands of topical anaesthetic agents for radial artery puncture have not demonstrated a significant reduction in pain ADDIN EN.CITE Olday200291(9, 10)919117Olday, S.J.,Walpole, R.,Wang, J.Y.Y.Radial artery cannulation: topical amethocaine versus lidocaine infiltrationBritish Journal of AnaesthesiaBritish Journal of Anaesthesia580-5828842002Aaron200392929217Aaron, S.D.,Vandemheen, K.L.,Naftel, S.A.,Lewis, M.J.,Rodger, M.A.Topical tetracaine prior to arterial puncture: a randomized, placebo-controlled clinical trialRespiratory MedicineRespiratory Medicine1195-1199972003( HYPERLINK \l "_ENREF_9" \o "Olday, 2002 #91" 9, HYPERLINK \l "_ENREF_10" \o "Aaron, 2003 #92" 10). In summary, we suggest that the literature is unclear on the role of local anaesthetic for radial artery puncture and this is reflected in ongoing clinical debate. To-date there is no reliable data comparing injected local anaesthetic to nothing. Therefore, our primary objective was to evaluate the effectiveness of subcutaneous local anaesthesia on the perceived pain of radial artery puncture. Secondary objectives included the assessment of the impact of patient characteristics, biochemical status, as well as technical aspects of the arterial puncture on patients reported level of pain. We also discuss direct and indirect economic implications.
Materials and Methods
This was a pragmatic two-group parallel randomised controlled trial which took place between December 5th 2012 and April 3rd 2013 at the James Paget University Hospital NHS Foundation Trust, UK.
Participants
We included all adult patients over 18 years old undergoing their first radial artery puncture by the 1st author (RGW), as part of their normal care and assessment in the Emergency Department (ED). Patients were excluded from the trial if: the Allens test was negative (no clinical collateral circulation to the hand), an urgent arterial blood gas result was needed (ie. where a delay in treatment to gain consent to enter the study would be clinically unacceptable) and the patient lacked capacity to give informed consent. In addition, those previously enrolled in the study were also excluded (ie. patients who attended the ED multiple times during the study period were only randomised once). Finally, in patients where the first attempt at obtaining an arterial blood sample failed, we only included data from the first (failed) attempt.
Recruitment
Eligible patients were given an information sheet and allowed sufficient time to consider participation prior to providing informed written consent. Withdrawal was permitted at any point.
Randomisation
Participants were randomised to two groups: intervention or control, according to a computer generated random number table with random block sizes of 4 and 6. There was no stratification variable as there are no factors known to influence the outcome of radial artery puncture HYPERLINK \l "_ENREF_8" \o "Zwarenstein. M., 2008 #95" . Once recruited, allocation was conferred by a 3rd party (the ED receptionist) who had no direct contact with the patient, using sequentially numbered, sealed opaque envelopes. This ensured allocation was concealed from the researcher recruiting subjects.
Intervention
Intervention group: Participants received a subcutaneous injection of lidocaine 1% 1ml directly volar to the radial artery (proximal to the wrist creases and lateral to the flexor carpi radialis tendon, where the artery was easily palpable), in their non-dominant limb, using a standard insulin (30 gauge) needle and syringe. Local anaesthesia was stored at room temperature and not mixed with any other solution. Injected lidocaine was massaged for approximately 10 seconds with a swab. Thereafter, participants underwent arterial puncture with preferably a 25-gauge needle (or the pre-packaged 22-gauge depending on availability), attached to a standard 5ml heparinised syringe. The skin and artery were punctured in a retrograde fashion at 30 degrees, through the previously anaesthetised skin. The syringe was permitted to fill under the systolic pressure. Upon withdrawing the needle, immediate compression with a sterile gauze was provided by the patient and continued for approximately 1 minute.
Control group: Participants did not receive any local anaesthesia but were otherwise treated identically to intervention subjects, with the ABG sample being taken from their non-dominant radial artery exactly as described above.
Blinding
As this was a pragmatic trial of technique we were unable to blind either clinicians or patients. However, in order to reduce researcher bias all data were collected by the same researcher/clinician (RGW) who aimed not to influence respondents ratings by maintaining the same technique for all patients irrespective of group.
Outcome measures
Our primary outcome was pain experienced by patients during radial artery puncture, reported on a 10 cm Visual Analogue Scale (VAS). A VAS allows accurate pain assessment (being able to differentiate up to 1 mm or 1%). Patients recorded their pain on this scale immediately after they had completed holding sterile gauze over their puncture wound (i.e. approximately 1 minute after their ABG sample was taken). The character of the radial pulse was subjectively assessed by RGW and recorded before the procedure. An estimation of the movement of the patients wrist was made during the attempted puncture and classified as none, minor (<1cm) or major (ie. >1cm or enough to affect the success of the arterial stab). Whether the needle was repositioned subcutaneously in order to obtain a sample and whether or not the attempt was successful were also collected. We recorded the volume (mls) of blood obtained, whether patients received any systemically acting analgesia (paracetamol or opioids) by any route within the preceding hour and the results of arterial blood analysis. After the procedure, participants within the control group were asked whether they would have preferred to have had local anaesthesia.
Data Analysis
Data were prospectively input into a password protected SPSS v20 (IBM) database. The distribution of reported pain was negatively skewed and therefore medians (with interquartile ranges, IQR) are presented and compared using the Mann-Whitney U-test or Kruskall-Wallis as appropriate. Other continuous variables were normally distributed and so represented by their means (and standard deviations, SD) and compared using independent samples t-tests. We compared pain to other continuous variables using Spearmans correlation co-efficient. Confidence intervals were generated at the 95% level and significance set at 5%. Intention-To-Treat analysis was performed.
Sample size
Given the lack of published data we estimated a minimum clinically relevant difference in pain between groups to be 10% (10 mm on the pain VAS) ADDIN EN.CITE Ruyssen-Witrand20111344(11)1344134417Ruyssen-Witrand, A.,Tubach, F.,Ravaud, P.Systematic review reveals heterogeneity in definition of a clinically relevant difference in painJ Clin EpidemiolJ Clin Epidemiol463-4706542011( HYPERLINK \l "_ENREF_11" \o "Ruyssen-Witrand, 2011 #1344" 11). Assuming equal standard deviations of 10% between groups, we required 32 subjects (16 per group) for 80% power, and 44 subjects (22 per group) for 90% power, at a 5% significance level.
Ethics and Research Governance
This protocol was approved by the Hertfordshire National Research Ethics Service Committee of the East of England (reference: 11/EE/0426) and given a favourable opinion by the Medicines and Healthcare products Regulatory Agency (MHRA). Participants gave informed written consent and no identifiable information was gathered.
Results
During the study period, 55 patients met the inclusion criteria. Twelve patients were excluded before randomisation due to the urgency of need for results (n=8) or inability to consent (n=4). Therefore, 43 patients consented and were randomised. Two patients were excluded post-randomisation (one withdrew consent once allocated to the control group and the other declined local anaesthesia once allocated to the intervention group), leaving 41 participants for comparative analysis (Figure 1). Participant characteristics are shown in Table 1.
Table 2 shows the outcomes of arterial puncture. The perceived pain of arterial puncture was objectively low and moreover, not significantly different between groups (Figure 2). There were no significant differences in the biochemical results between groups. There was no significant correlation between patients reported level of pain and the volume of blood obtained or any other biochemical variables.
Table 3 shows the univariate analysis for reported pain. Those who had a smaller needle puncture their radial artery appeared to perceive less pain, albeit not statistically significant (Figure 4). The amount of wrist movement during the procedure and success of the stab were non-significantly related with the amount of reported pain. After the ABG was taken, six participants in the control group said they would have preferred local anaesthesia and this preference was associated with a significantly higher reported level of pain (Figure 5).
Discussion
This is the first robust randomised trial to investigate the effect of subcutaneous local anaesthesia on the perceived pain of radial artery puncture. Previous well-executed studies have shown topical anaesthetic agents to be ineffective in reducing the pain of arterial puncture and our data is in agreement ADDIN EN.CITE Olday200291(9, 10)919117Olday, S.J.,Walpole, R.,Wang, J.Y.Y.Radial artery cannulation: topical amethocaine versus lidocaine infiltrationBritish Journal of AnaesthesiaBritish Journal of Anaesthesia580-5828842002Aaron200392929217Aaron, S.D.,Vandemheen, K.L.,Naftel, S.A.,Lewis, M.J.,Rodger, M.A.Topical tetracaine prior to arterial puncture: a randomized, placebo-controlled clinical trialRespiratory MedicineRespiratory Medicine1195-1199972003( HYPERLINK \l "_ENREF_9" \o "Olday, 2002 #91" 9, HYPERLINK \l "_ENREF_10" \o "Aaron, 2003 #92" 10), suggesting that subcutaneously injected lidocaine provides no reliable reduction in the perceived pain of radial artery puncture.
Anecdotally, we have found that clinicians who avoid the use of local anaesthesia for radial artery puncture suggest that subcutaneousy injected lidocaine increases the pain of the procedure, due to the incompatible pH and temperature of the solution, as well as tissue distension and the 2nd stab which follows. Our results cannot confirm or deny these claims but challenge the concept (Figure 2) and as we found no differences between groups. We hypothesise that there may be value in further research into four particular situations where we found an apparent reductions in pain (potentially representing Type 2 errors) and discuss these results. Firstly, patients with no intravenous analgesia on board appeared to experience higher levels of pain (Figure 3). This finding is intuitive and in context, patients requiring an arterial blood gas analysis are usually unwell and warrant intravenous access. Therefore, we suggest that such patients receive intravenous analgesia (which would likely have been given for their presenting complaint anyway) in line with best medical practice. Secondly, there seemed to be a relationship between perceived pain and the number of attempts made to puncture the radial artery. Again, this finding is expected and we suggest that junior clinicians receive adequate simulated training prior to performing the procedure on patients. The use of a smaller bore of needle to puncture the radial artery appeared to reduce the pain of the procedure (Figure 4), which is again expected ADDIN EN.CITE Gill20071333(12)1333133317Gill, S.G.,Prausnitz, M.RDoes Needle Size Matter?Journal of Diabetes Science and TechnologyJournal of Diabetes Science and Technology725-729152007( HYPERLINK \l "_ENREF_12" \o "Gill, 2007 #1333" 12). Anecdotally, some clinicians avoid using small needles as they believe that smaller luminal diameter may impede the acquisition of the sample and therefore render it inadequate for analysis (eg. due to haemolysis or coagulation). However, the first author routinely uses a non-stock 25-gauge needle in preference to the larger needle supplied and has not experienced such problems. Finally, patients who reported greater levels of pain moved their wrist more during the procedure, which may be related to needle displacement into bone, nerve and muscle or procedural anxiety ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_1" \o "Lightower, 1977 #86" 1-3). We are unable to explan this finding and as with previous points, certainly welcome further controlled research on this topic to clarify such uncertainties.
By avoiding the routine use of injected lidocaine anaesthesia for radial artery puncture, time and cost savings could be made. In our Trust, a vial of 10mls 1% lidocaine solution costs approximately 0.87. The cost of the additional equipment needed to deliver this agent (eg. a 5ml syringe and 25-gauge needle or insulin syringe and gauze) increases the cost to approximately 1. However, this price does not consider the indirect costs of transportation of medicines, staff to manage the stock, staff time to re-stock cupboards/trollies and sharps/glass disposal. Our Trust performs over 800 arterial punctures and blood gas analyses per month so by avoiding the use of local anaesthesia in most cases, we believe that our Trust could make savings many thousands of pounds per year. These savings are theoretically transferable to any Hospital in the world and may help health services redirect funds. Further, physician time is saved as it takes additional time to prepare and deliver the local anaesthesia, as well as wait for its effect.
Our study has limitations which are important to consider. We achieved only 80% power for our primary outcome analysis. This lower sample size which is likely to have impacted on our ability to detect small differences between groups was due to the planned trial cessation once RGW left the department. Future trials should aim to recruit larger samples if it is considered important to determine if there is a clinically important smaller difference to detect. We were unable to blind either party to the intervention; however, this should not have confounded our results because most patients are unlikely to have specific preconceived ideas about the technique for obtaining an arterial blood sample and all procedures were performed by a single clinician. We suggest that future trials consider a placebo arm to minimise researcher bias, as well as examining patients ideas of local anaesthesia as this may be a more useful marker of patient preference.
Conclusion
This trial found no evidence of a difference in pain experienced by those with or without subcutaneously injected local anaesthesia prior to radial arterial blood gas sampling.
References ADDIN EN.REFLIST
1. Lightower JVJ, Elliott MW. Local anaesthetic infiltration prior to arterial puncture for blood gas analysis: a survey of current practice and a randomized double blind placebo controlled trial. J R Coll Physicians Lond. 1977; 31: 645-6.
2. Giner J, Casan P, Belda J, al. e. Pain during arterial puncture. Chest 1996; 110: 1143-5.
3. Scott DB. Topical anaesthesia of intact skin. Br J Par Ther. 1986; 7: 134-5.
4. Sado DM, Deakin CD. Local anaesthesia for venous cannulation and arterial blood gas sampling: are doctors using it? J R Soc Med. 2005; 98: 158-60.
5. Hudon TL, Dukes FS, Reilly K. Use of Local Anesthesia for Arterial Punctures. Am J Crit Care. 2006; 15: 595-9.
6. Giner J, Casan P, Belda J, Litvan H, Sanchis J. [Use of the anesthetic cream EMLA in arterial punction]. Rev Esp Anestesiol Reanim. 2000; 47: 63-6.
7. Joly LM, Spaulding C, Monchi M, Ali OS, Weber S, Benhamou D. Topical lidocaine-prilocaine cream (EMLA) versus local infiltration anesthesia for radial artery cannulation. Anesth Analg. 1998; 8: 403-6.
8. Kim JY, Yoon J, Yoo BS, Lee SH, Choe KH. The effect of a eutectic mixture of local anesthetic cream on wrist pain during transradial coronary procedures. J Invasive Cardiol. 2007; 19: 6-9.
9. Olday SJ, Walpole R, Wang JYY. Radial artery cannulation: topical amethocaine versus lidocaine infiltration. British Journal of Anaesthesia. 2002; 88: 580-2.
10. Aaron SD, Vandemheen KL, Naftel SA, Lewis MJ, Rodger MA. Topical tetracaine prior to arterial puncture: a randomized, placebo-controlled clinical trial. Respiratory Medicine. 2003; 97: 1195-9.
11. Ruyssen-Witrand A, Tubach F, Ravaud P. Systematic review reveals heterogeneity in definition of a clinically relevant difference in pain. J Clin Epidemiol. 2011; 65: 463-70.
12. Gill SG, Prausnitz MR. Does Needle Size Matter? Journal of Diabetes Science and Technology. 2007; 1: 725-9.
Figures
Figure 1. Participant flow-chart.
Figure 2. A box-plot of reported pain (median, IQR) between groups, compared with the Mann-Whitney U-Test.
Figure 3. A box-plot of reported pain (median, IQR) and the presence of systemically active intravenous analgesia, compared with the Mann-Whitney U-Test.
Figure 4. A box-plot of reported pain (median, IQR) and the calibre of needle used to puncture the radial artery, compared with the Mann-Whitney U-Test.
Figure 5. A box-plot of reported pain (median, IQR) and the post-hoc preference for local anaesthesia, compared with the Mann-Whitney U-Test.
Tables
Table 1. Baseline Characteristics
Control (N=21)Intervention (N=20)Age (SD)54.8 (20.5)53.2 (21.5)Gender (%)Male12 (57%)13 (65%)Female9 (43%)7 (35%)Intravenous Analgesia On-Board (%)6 (29%)4 (20%)Quality of Radial Pulse (%)Weak2 (10%)1 (5%)Normal19 (90%)18 (90%)Bounding0 (0%)1 (5%)
Table 2. Outcomes
Control (N=21)Intervention (N=20)p-valuePerceived pain (Median, IQR)1.8 (0.4, 5.4)1.6 (0.7, 3.5)0.938Needle delivering local Anaesthesia (%)31 gauge (insulin) -18 (90%)/25 gauge (orange) -2 (10%)Needle puncturing the Radial Artery (%)25 gauge (orange)18 (86%)19 (95%)0.60622 gauge (stock)3 (9%)1 (5%)Wrist Movement During Attempt at Arterial Puncture (%)None18 (86%)20 (100%)0.356Minor1 (5%)0 (0%)Major2 (10%)0 (0%)Success of Arterial Puncture (%)Direct puncture on 1st advance13 (62%)11 (55%)0.304Subcutaneous repositioning6 (29%)9 (45%)Failed (2nd attempt)2 (10%)0 (0%)Mean Volume of blood obtained (SD)1.69 (1.16)1.70 (0.71)0.975Would have preferred local anaesthetic (%)6 (28.6)--
Table 3. Univariate Analysis
NReported PainMedianIQRp-valueGenderMale251.800.50, 4.300.751Female161.400.15, 5.30Age (years)<50130.950.40, 2.300.14050-6463.200.70, 8.20>64221.150.30, 5.40Intravenous Analgesia On-BoardYes100.600.30, 2.200.105No312.300.50, 6.10Quality of Radial PulseWeak300, 4.300.416Normal371.800.50, 5.40Bounding11.301.30, 1.30Needle delivering local Anaesthesia31 gauge (insulin)191.600.70, 3.700.94725 gauge (orange)11.301.30, 1.30Needle puncturing the Radial Artery 25 gauge (orange)371.400.40, 4.300.15022 gauge (stock)44.352.30, 6.95Wrist Movement During Attempt at Arterial PunctureNone381.350.40, 3.700.087Minor17.007.00, 7.00Major27.706.90, 8.50Success of Arterial PunctureDirect puncture on 1st advance240.600.30, 3.500.094Subcutaneous repositioning152.201.30, 5.20Failed (2nd attempt)26.155.40, 6.90Post-hoc preference for local anaestheticYes66.955.20, 7.800.006No150.500.30, 3.30
e f s a h i l
ɼه|oU|D hc h# OJ QJ ^J mHsH 2h,
h# B*OJ QJ ^J fH ph""" q
h,
h# OJ QJ ^J h# H*OJ QJ ^J h&