To determine if the use of needle enhancing software facilitate injection technique in ultrasound-guided peripheral nerve blocks.
To determine if the use of needle enhancing software facilitate injection technique in ultrasound-guided peripheral nerve blocks.
Prospective, blinded, randomized controlled trial.
Eight hind limbs from canine cadavers.
The limbs were randomly allocated to two groups; software on (group I) and software off (group II). Eight anaesthetists with no previous experience of ultrasound-guided regional anaesthesia were recruited. Thirty-six procedures were carried out (18 per group). After sciatic nerve visualisation via ultrasonography, the anaesthetist introduced a needle guided by ultrasonography. When the tip of the needle was considered by the anaesthetist to be as close as possible to the nerve without touching it, 0.05 mL of methylene blue dye was injected. Parameters evaluated included: number of attempts to visualise the needle with ultrasonography, time spent to perform the technique, subjective evaluation of ease of needle visualisation, proximity of the tip of the needle to the nerve, and, at dissection of the leg, inoculation site of the dye in relation to the nerve.
Significant differences between groups were identified in relation to the number of attempts (group I: median 1, IQR: 1 – 1 attempts versus group II: median 1, IQR: 1 – 4 attempts, p = 0.019), and the relationship between the dye and the nerve during hind limb dissection (72.2% of the nerves were stained in group I versus 16.6% in group II, p = 0.003). No significant difference between groups was observed with respect to the time taken to perform the procedure (group I: median 25.5, IQR: 18.4 – 44.3 seconds versus group II: median 35.7, IQR: 18.6–78.72 seconds, p = 0.31), subjective evaluation of the needle visualization (p = 0.45) or distance between the tip of the needle and the nerve as measured from the ultrasound screen (p = 0.23).
This study identified greater success rate in nerve staining when the needle enhancing software was used. The results suggest that the use of this technique could improve injection technique amongst inexperienced anaesthetists performing ultrasound-guided peripheral nerve blocks in dogs.
Local and regional anaesthesia techniques in dogs have been used increasingly to relieve pain related to medical and surgical procedures (Shilo et al. 2010). The successful nerve block depends primarily on an optimal distribution of local anaesthetic around the nerve (Costa-Farré et al. 2011) at an appropriate drug concentration. Ensuring close proximity of the injection to the nerve can be challenging and a number of methods (anatomical landmarks, paraesthesia, neurostimulation, etc.) have been reported for location and close administration of the anaesthetics drug.
The popularity of real-time ultrasound guidance for nerve blockade has increased dramatically over the last years in human anaesthesia (Luyet et al. 2010) and several studies have already been published in relation to its use in veterinary anaesthesia (Campoy et al. 2010; Echeverry et al. 2010, 2012; Shilo et al. 2010; Costa-Farré et al. 2011; Schroeder et al. 2011; Mahler 2012). Ultrasound demonstrates in real time the relative locations of the needle, the nerves of interest, the structures to be avoided by the needle (i.e. blood vessels) and the distribution of local anaesthetic injected (Klaastad et al. 2009). For these reasons, ultrasound-guided nerve blocks may be advantageous compared to ‘blind’ techniques (including neurostimulation). Higher success rates, improved procedure safety, reduction of local anaesthetic dose, provision of a faster onset, more predictable duration of effect and overall improvement of block quality are some of the main advantages reported (Marhofer et al. 1998, 2005; Sala-Blanch & De Andrés 2004).
Ultrasound-guided regional anaesthesia, however, requires mastering of a number of skills, including the knowledge of basic ultrasound physics, ultrasound machine settings, sonographic anatomy, and the manual dexterity to visualise and advance the needle close to the nerve (Luyet et al. 2010). Information is scarce regarding the details of the learning process and skill development required to conduct safe and effective ultrasound-guided regional anaesthesia (Sites et al. 2004, 2007a). The American and European Society of Regional Anaesthesia and Pain Therapy (Sites et al. 2009a) have suggested recommendations for education and training in ultrasound-guided regional anaesthesia. These recommendations consist of 10 tasks that are helpful in performing an ultrasound-guided nerve block.
Apart from the anatomical and theoretical knowledge required, the use of appropriate equipment may be beneficial in order to maximise the success and efficacy of the technique. enhanced needle visualization (SonoSite, Inc, WA, USA) is software that intensifies the brightness of the needle during ultrasound-guided procedures (Fig. 1) and this may facilitate the efficacy of the technique. The aim of this study was to evaluate the use of this software during ultrasound-guided nerve blocks undertaken by veterinary surgeons with little or no previous experience in ultrasound-guided regional anaesthesia. It was hypothesised this software could improve injection techniques, in relation to the peri-neural distribution of injected dye.
The study was performed by members of Veterinary Clinical Services (VCS) at the Royal Veterinary College after approval from the institutional ethics committee had been obtained (URN 2011 1109).
Eight hind limbs from canine cadavers (originally frozen, but then thawed) were used in this study, four left and four right limbs. Medium to large size dogs were selected in order to obtain good sonographic images of sciatic nerves and permit multiple samples per leg. Each leg was divided from proximal to distal into several 2–3 cm long regions. Parallel lines were drawn on the skin with a sharpie in order to recognize each region. This allowed a number of anaesthetists to perform the injections on different areas of the sciatic nerve of the same leg. Legs were allocated to one of two groups: each group had two right and two left hind limbs. In group I (Software ON) the injections were performed using the enhanced needle visualization Software, and in group II (Software OFF) the software was not used. Four or five different anaesthetists, depending on the length of the limb, performed the injections on a different region of each hind limb. Eight veterinary anaesthetists from the Department of VCS (four senior clinicians and four residents) were recruited to this study. All of them had little or no previous experience of ultrasound-guided regional anaesthesia.
The hair of the lateral aspect of hind limbs was clipped from the greater trochanter of the femur to the stifle joint; the skin was cleaned with 50% surgical spirit and coupling gel was applied. The limbs were scanned using an ultrasound machine (model Sonosite M-Turbo, UK) with a linear array transducer (10–13 MHz). The ultrasonographic appearance and location of the sciatic nerve was shown to the anaesthetists by a board certified radiologist (LB) prior to the anaesthetists’ performed their task. Transverse images of the sciatic nerve were obtained. The task consisted in introducing a 25 mm 22 gauge needle with a 1 mL syringe filled with methylene blue dye (Methylthioninium chloride injection 1% w/v, Martindale Pharmaceuticals, UK) attached, through the skin and advancing it, guided by the ultrasound images. The needle was introduced at an angle of 45° to the limb surface using an in-plane approach (Fig. 2). When the tip of the needle was considered to be placed in the correct position (proximal to the nerve but not inside it), 0.05 mL of methylene blue dye was injected. Each anaesthetist performed the same task four times (one right and one left limb with the software ON and one right and left hind limb with the software OFF). All anaesthetists were unaware as to whether the software was on or off at the time of injection. The order of injections in the hind limbs was randomly allocated using a random number function (Microsoft Office Excel 2007, version 12.0, WA, USA). Once all injections were completed, dissection of the hind limbs was carried out by another investigator (JV) who also was unaware as to whether the software had been used or not. Macroscopic evaluation of the relationship between the dye and the sciatic nerve was performed.
The number of attempts was defined as number of withdrawals of the needle back towards the skin surface in order to amend the angle of approach. The time spent to perform the technique was considered the time interval from introduction of the needle through the skin for the first time until the dye had been injected. After the task was performed, the anaesthetists recorded a subjective evaluation of the clarity of needle visualisation (three categories) (Table 1). The distance between the position of the tip of the needle in relation to the nerve as viewed on the ultrasound screen, was measured with the ultrasound machine callipers before dye injection (three categories) (Table 1). Finally, the sciatic nerve dying was evaluated during hind limb dissection by the same investigator who carried out the dissections (three categories) (Table 1).
|Subjective evaluation of the clarity of observation of the needle|
|Poor (needle cannot be visualised)|
|Good (needle can be visualised but it is hard to see the tip of the needle)|
|Excellent (needle can be visualised and tip of the needle is easy to see)|
|Subjective evaluation of the position of the tip of the needle in relation with the nerve in the ultrasound screen before injecting the dye|
|Tip of the needle cannot be visualised|
|More than 0.5 cm|
|Less than 0.5 cm|
|Relationship between the dye and the nerve during hind limb dissection|
|Nerve is not stained|
|Nerve is slightly stained (less than 2.0 mm)|
|Nerve stained (more than 2.0 mm)|
Statistical analysis of the results was performed using the pasw version 18.0 (Statistical Package for the Social Sciences for Windows, Inc., IL, USA). Categorical data were described with frequency (%) and compared with the Chi-squared and Fisher's exact tests as appropriate. Post hoc pairwise comparisons with the chi squared test were performed where significant differences were detected in the initial analysis. Numerical data were assessed graphically for normality; results were reported with median and interquartile range (IQR) and comparisons between groups undertaken using the non-parametric Mann–Whitney U test. Differences were considered significant when p < 0.05.
Median and interquartile ranges (IQR) for the number of attempts to visualise the needle were 1 (IQR 1 – 1) in group I, and 1 (IQR 1 – 4) in group II and were significantly different (p = 0.019). Most of the anaesthetists needed <4 attempts to visualise the needle. The maximum number of attempts required occurred when the software was off and the anaesthetist needed nine attempts. No differences were detected due to operator seniority (senior/resident, p = 0.572) or hind limb (left/right, p = 0.952).
A significant difference was found between groups I and II in relation to nerve staining during hind limb dissection. Group I showed more stained nerves than group II (13/18 versus 3/18 respectively, p = 0.003). Thirteen injections (72.2%) produced some degree of staining of the nerve in the group I (seven nerves were stained <2 mm in length and six nerves were stained more than 2 mm) compared with three injections (16.6%) in group II (one nerve was stained <2 mm and two nerves were stained more than 2 mm in length) (Table 2).
|Nerve staining, no. of nerves (%)|
|No staining||Less than 2.0 mm||More than 2.0 mm||Total|
|On (Group I)||5 (27.8)||7 (38.9)||6 (33.3)||18 (100)|
|Off (Group II)||15 (83.3)||1 (5.6)||2 (11.1)||18 (100)|
Prior to injecting the dye, anaesthetists classified the perceived distance between needle and nerve as less than 0.5 cm in 88.8% of the procedures in both groups (16/18 in each group). In group I, the tip of the needle was not visualised at all in two cases. In group II, the tip of the needle was believed to be placed more than 0.5 cm away from the nerve in two cases. No significant difference between groups was detected in the distance between the tip of the needle and the nerve as measured on the ultrasound screen (p = 0.236).
In procedures in which anaesthetists reported the needle to be very close to the nerve (<0.5 cm) nerve staining was assessed on dissection (categorised as: stained or not stained). Thirty-two procedures reached this criterion (16 procedures in group I and 16 procedures in group II). Significantly more limbs were classified as stained in group I (11 stained nerves per 16 procedures, 68.7%) compared to group II (two stained nerves per16 procedures, 12.5%, p = 0.002). No differences were observed between nerve staining regarding operator type (senior/resident, p = 0.879) or hind limb blocked (left/right, p = 0.705). The time to perform the procedure was not significantly different between groups (p = 0.311). Group I spent a median of 25.5 seconds (IQR 18.4–44.3) to perform the procedure versus a median of 35.7 seconds (IQR 18.6–78.72) in group II.
In group I, anaesthetists reported poor visualisation of the needle in 16.7% (3/18) of the procedures, good visualisation in 16.7% (3/18) and excellent visualisation in 66.6% (12/18). In group II, poor visualisation of the needle was reported in 33.3% (6/18) of the procedures, good in 22.3% (4/18) and excellent in 44.4% (8/18). No significant difference was observed between both groups (p = 0.45).
This study suggests that the use of this software increased success in nerve staining compared with injections performed without the aid of software. These results support the use of this technology to improve the injection technique and clinical outcome for anaesthetists without limited experience of ultrasound guided peripheral nerve blocks. Thirteen out of 18 injections (72%) performed with the software were placed in proximity to the nerve (staining nerves either less or more than 2 mm). Only three out of 18 injections (16%), performed without use of the software achieved this.
Previous work in veterinary and human anaesthesia supports the use of new technologies to improve the efficacy and reduce complications associated with techniques such as peripheral nerve blocks (Abrahams et al. 2009; Campoy et al. 2010). Neurostimulation has been considered the gold standard for peripheral nerve blocks in human anaesthesia (Abrahams et al. 2009) and has been proved also to be a good tool in veterinary anaesthesia (Campoy et al. 2008). However, it does not provide visualisation of the needle tip, and vessels and other structures may be traumatised during the procedure. For this reason, neurostimulation-guided nerve block is still considered as a ‘blind’ technique (Echeverry et al. 2010). Ultrasound has been reported to provide similar accuracy as neurostimulation (Abrahams et al. 2009) but because needle and anatomical structures are visualised in real time, complications related to needle placement have been reported to be reduced and effective injected local anaesthetic dose increased (Marhofer et al. 1998). Efficacy of ultrasound-guided regional anaesthesia depends critically on image quality of the target organ, the needle, and the ultrasound devices (Maecken et al. 2007).
As with any local or regional anaesthetic technique, training is required to achieve an acceptable level of performance. Luyet et al. (2010) compared the learning curves for peripheral nerve blocks guided by neurostimulation with blocks guided by ultrasonography. That ultrasound-guided technique′s learning curve appeared steeper; anaesthetists requiring 10–15 cases in order to reach a an acceptable level of proficiency in the ultrasound group, compared to 25–30 cases when using neurostimulation. In addition to this, success rate statistically was higher in the ultrasound compared to the neurostimulation group (89% versus 80%). In the current study, the anaesthetists were assisted in the localisation of the sciatic nerve, such that only the injection process would be evaluated. Recognition of the nerve is paramount for ultrasound-guided regional anaesthesia. For this reason, direct comparison between the current and Luyet′s study was not possible. The success rate obtained in the current study using the software (72% of injections) could be considered relatively high given that every anaesthetist performed only four to six procedures overall (two or three with the software turned on) and may not have reached a plateau of proficiency. Therefore, although the learning curve was not evaluated in the current study, these results support the use of this technology to improve the ability to perform this task.
The sciatic nerve was selected in this study for several reasons. Firstly, ultrasonographic anatomy of the sciatic nerve has been widely described and it is reported to be a relatively easily recognised nerve. Benigni et al. (2007) described it as made of two components, the common peroneal nerve and the tibial nerve. The two components of the sciatic nerve appear as hypoechoic tubular structures surrounded by a thin hyperechoic rim: the cranial one, representing the common peroneal nerve, always appears slightly smaller in diameter than the tibial nerve. In frozen-thawed sections, as evaluated in the current study, the nerve is said to appear more uniformly hypoechoic due to the post-mortem degeneration of the nerve structures (Benigni et al. 2007). This may have facilitated the visualization of the nerve. In the current study the same approach was adopted for all limbs (frozen-thawed) and hence comparisons within the study between study groups should remain valid. Secondly, the sciatic nerve is a long nerve so several injections could be performed on the same hind limb. Finally, the sciatic nerve block was studied as it is considered one of the most commonly used peripheral nerve block technique and hence clinically relevant (Vettorato et al. 2012).
A low volume of methylene blue (0.05 mL) was injected in order to evaluate more accurately the anatomical area where the tip of the needle was placed. It was considered likely that a greater volume would result in good or complete staining of the nerve even if the needle tip was not placed particularly close to the nerve on injection. Campoy et al. (2008, 2010) reported a minimum local anaesthetic volume of 0.05 mL kg−1 to perform sciatic nerve block using either neurostimulation or ultrasonography in dogs. In the current study, 16 injections (13 in group I and three in group II) produced some degree of nerve staining. It is therefore possible that lower volumes than previously reported could be sufficient to achieve an effective nerve block when using ultrasound guided methodology, and this should be further investigated. This would agree with several studies performed in human anaesthesia where lower local anaesthetic doses were required for ultrasonographic methods compared with ‘blind’ techniques including neurostimulation (Pawan & Satya 2010; Salinas 2010).
During two dye injections, the dye was seen to completely surround the nerve, in what is reported in the literature as a ‘doughnut’ spreading pattern (Choquet et al. 2012). During limb dissection, these two injections were associated with the greatest area of staining of the sciatic nerve despite the low volume of injection. Ultrasonography allowed visualisation of the nature of local anaesthetic spread around the nerve. This feature has been considered a valuable additional factor for successful nerve blockade (Sites et al. 2009b). In their study, Costa-Farré et al. (2011) repositioned the needle when necessary to obtain a distribution of local anaesthetic around at least 50% of the nerve surface. They reported a 100% success rate when this criterion was applied. Finally, no intraneural injection was visualised macroscopically in any case.
The number of attempts to visualise the needle by the anaesthetists was significantly lower when the software was used. This finding would support the hypothesis that the software provided better visualisation of the needle. However, the subjective evaluation made by the anaesthetists regarding their perception of visualisation of the needle did not show any significant difference between groups. One possible explanation for this was that the anaesthetists may not have consciously realised that they could see the needle better. Sites et al. (2004) stated that the ability of a novice to visualise the needle by maintaining its position within the longitudinal plane of the ultrasound beam represented the most difficult aspect of the procedure. With respect to needle visualisation, Sites et al. (2007b) stated that the goal is to simultaneously minimize refraction and maximize reflection back toward the probe by keeping the needle perpendicular to the ultrasound beam. With deeper nerve targets, the angle of incidence between the beam and needle becomes shallower such that more ultrasound waves are redirected (by refraction and reflection) and fewer waves successfully return to the probe. The end result is that the needle becomes less visible. For this reason, the visibility of a needle in ultrasound-guided percutaneous procedures is often limited by dispersion of the needle's reflections away from the transducer. With this software, a needle enhancement algorithm was developed that maximizes the received reflections by steering the ultrasound beam precisely perpendicular to the needle. The resulting image clearly depicts the needle as a bright line, even if the needle is not perpendicular to the ultrasound beam (Cheung & Rohling 2004).
In our study, anaesthetists performed the task introducing the needle with an angle of 45° using in-plane approach. Another needle angle has been described for sciatic nerve blockade in dogs (Campoy et al. 2010) although our approach has also been described in human anaesthesia and is said to be easy to perform (Marhofer 2010). Takatani et al. (2011) found the needle enhancing software, the same type as used in our study, very helpful even with different needle insertion angles and they concluded this technology could allow safer ultrasound procedures in various approaches. For all these reasons, facts related to needle visibility as selection of type of needle, angle of needle insertion, etc. (Schafhalter-Zoppoth et al. 2004; Maecken et al. 2007) could be avoided with this technology.
In the current study, anaesthetists believed the tip of the needle was very close to the nerve in 88% of the procedures in both groups (16 out of 18 cases in each group). Comparing these results with the results obtained in the hind limb dissection, the anaesthetists working without the software misclassified the position of the tip of the needle more often than when the software was on. Only two out of 16 nerves (12.5%) were stained when anaesthetists thought the tip of the needle was very close to the nerve and were working without the software. Sites et al. (2007a) reported the most frequent error experienced by novices was loss of visualisation of the needle tip. Interestingly, the success rate with the software was much higher (68.7% versus 12.5%) and only in five of 16 procedures where the tip of the needle was considered very close to the nerve (<0.5 cm), the nerve was not stained with methylene blue. These findings would support the conclusion that anaesthetists were able to identify the tip of the needle more effectively when the software was used.
No significant difference was found between groups regarding the time the anaesthetist spent performing the task (25.5 seconds in group I and 35.7 seconds in group II). However, one anaesthetist in group I spent more than 90 seconds performing the task (149.6 seconds) compared with five anaesthetists in group II (204.9, 90, 192.3, 91.3 and 90.8 seconds).
One limitation of this study was the small sample size. A greater number of procedures per anaesthetist would also be needed to be able to study the learning curve. Our results, however, suggest that fewer patients potentially would be needed to learn this local anaesthetic technique because we could observe a significant difference when comparing the nerve staining between both groups.
Finally, our study was performed on cadavers and this fact could make operators more confident to advance the needle as much as they wanted, change position of the needle, etc. During hind limb dissections, no signs of intraneural injection were observed, though we cannot guarantee nerves were not penetrated during some procedures. Although visibility of hypodermic needle would be adequate in order to carry out the ultrasound guided peripheral nerve blockade in a clinical scenario, its bevel would not be the best choice to perform a safe injection close to the nerve. A 14° long-bevel needles have been proven to produce more nerve damage than 45° short-bevel needles (Jeng et al. 2010). Therefore, this fact must be taken into account when nerve block is performed in live animals.
In conclusion, this study demonstrated that inexperienced anaesthetists performing ultrasound-guided regional anaesthesia in cadavers had a higher success rate when working with the needle enhancing software. This finding suggests the use of this equipment could aid performance but further studies are necessary to evaluate these findings in a clinical setting.
The authors wish to thank all anaesthesia team of the Royal Veterinary College for assistance with this study and BCF Technology Ltd. for providing the ultrasound machine.