Peripheral nerve blocks are widely used in regional anesthesia and correct prediction and early determination of block success are important for the anesthesiologist. Presently, determination of block success is based on both objective and subjective methods, for example, testing for cold sensation, pinprick, muscle force, etc.1,2 These examinations require some anatomical knowledge by the examiner as well as a cooperating patient and may be time-consuming.
Previous studies show that brachial plexus blocks at different levels lead to increased skin temperature in the hand and forearm.3–5 The thermographic patterns after individual peripheral blocking of the major nerves to the hand and forearm have also been described in detail.6 Our preliminary experience using infrared thermographic imaging after an ultrasound (US)-guided lateral infraclavicular block showed characteristic thermographic patterns and increases in skin temperature that were fastest and most pronounced on the digits’ tips. The lateral technique refers to a needle insertion point at the antero-inferior border of the clavicle and with a needle direction 0°–45° posterior, sagittal, and slightly lateral.
In this study, we hypothesized that a temperature difference between the blocked and the nonblocked hand, simply registered by touching the skin 30 minutes after injection, may be a useful diagnostic test for predicting a successful block defined by sensory and motor block of all 4 major nerves (musculocutaneous, radial, ulnar, and median nerves).
The aim of this study was (1) to investigate whether a blinded observer would be able to predict lateral infraclavicular block success based on sensed skin temperature differences between the distal phalanges of the corresponding 2nd and 5th digits of the blocked hand and the nonblocked hands and (2) to evaluate interobserver agreement in assessing temperature differences.
After approval (H-C-2008–047) from the Committees on Biomedical Research Ethics of the Capital Region of Denmark and written informed consent from all subjects, we conducted a blinded, prospective, observational study. Forty-five adult patients, scheduled for surgery of the hand or forearm, were included. Inclusion criteria were patients planning to have lateral infraclavicular block before surgery, ASA I–III, age 18–90 years, weight 50–120 kg. Exclusion criteria were age <18 years, international normalized ratio >1.4, platelet count <80 × 109 liter−1, coagulopathy, medication with vitamin K-antagonist/high-dose heparin or fractionated heparins, allergy to local anesthetics, infection at the site of needle insertion, peripheral neurological disease, Raynaud disease, or patient refusal.
Routine monitoring included continuous electrocardiography, pulse oximetry (earlobe), and noninvasive arterial blood pressure on the thigh. An IV catheter was inserted into a superficial vein of the nonblocked cubital fossa.
The patients were in supine position at a room temperature of 21°C. Direct sunlight was avoided, all bandages and clothing were removed from the forearms and hands, and the patients were allowed to acclimatize for 10 minutes with their hands in prone position and resting with the palmar skin on 5°C cold packs (Nexucare™ cold-hot maxi, 3M Health Care, Neuss, Germany). The cooling of both hands was done in an attempt to standardize measurements and to augment the magnitude of potential temperature difference between hands introduced by the lateral infraclavicular block.
Blocking of the Brachial Plexus
With the use of a high-frequency linear US transducer (HFL 38 ×/13–6 MHz, S-ICU™ Ultrasound System, Sonosite Inc., Bothell, WA), we identified the axillary artery and vein and also, whenever possible, the position of the 3 nerve cords. After disinfection with ethanol-chlorhexidine (83% and 0.5%, respectively), a sterile transparent drape was placed over the planned injection site. The attending anesthesiologist performed the lateral infraclavicular block, originally described by Raj et al7 and later modified by Klaastad et al,8–12 as a single-shot injection of 20 mL ropivacaine 7.5 mg/mL (Naropin® 7.5 mg/mL, Astra-Zeneca A/S, Albertslund, Denmark). The needle insertion point was at the antero-inferior edge of the clavicle just medial to the coracoid process and the needle (Stimuplex® D 22 G 50 mm, 15° or Stimuplex® D 22 G 80 mm, 15°; B. Braun Melsungen AG, Melsungen, Germany) was advanced in the intocephalo-caudal direction in the sagittal plane and slightly lateral. We used in-plane US guidance to position the needletip at 7 o’clock in relation to the axillary artery. When correctly positioned, perineural spread of local anesthetic was observed. All blocks were performed by anesthesiologists with varying degrees of experience.
The Diagnostic Test: Temperature Difference Sensed by Touch of the Skin
Before the initiation of the study, a group of 8 observers was trained to examine the patients in a standardized manner. The observers were instructed to first examine the paired 2nd phalanges followed by the paired 5th phalanges. The distal part of the 2 phalanges was gently and simultaneously squeezed and a (possible) temperature difference was sensed. The attending observers were randomly chosen and interchangeable among the 8 observers. The observers were blinded to the assessments of the other observers, to the side of the lateral infraclavicular block, and to the clinical block assessment.
Each patient was evaluated by 3 different observers at baseline (after 10 minutes of acclimatization, just before the block was performed) and 30 minutes after the block was performed. A sheet covered the patient’s arms and shoulders, hiding both the site of local anesthetic injection and the site of the IV catheter. From thermographic imaging in previous studies, we expected skin temperature increases to be largest at the digits,6 and we also expected increased skin temperature of the 2nd and the 5th digits to represent blocking of the lateral and medial cord, respectively. The blinded observer therefore simultaneously touched the skin of the distal phalanges of (1) the right and the left 2nd digits; and (2) the right and the left 5th digits by slightly squeezing the distal phalanges between the pulps of his own 1st and 2nd digits. The observer registered any sensed skin temperature differences in the following categorical manner: (1) no skin temperature difference or (2) skin temperature difference. If a skin temperature difference was sensed, the blinded observer identified whether the right or left distal phalanx was warmest for both the 2nd and the 5th digit. In our diagnostic test, we predicted a lateral infraclavicular block to be successful, when the following criteria were met: (1) the observer should be able to sense a difference in skin temperature between the blocked and the nonblocked corresponding 2nd and 5th distal phalanges and (2) the observer should appoint skin temperature to be warmest on the blocked side for both the 2nd and the 5th distal phalanges. In contrast, we considered a lateral infraclavicular block to be failed when the following criteria were met: The observer was not able to sense a skin temperature difference between 1 or both of the corresponding right and left distal 2nd and 5th phalanges. We only included patients without skin temperature differences at baseline between the 2nd and 5th corresponding digits for the diagnostic test.
Clinical Block Assessment
A group of 6 nurses was trained to clinically assess the performed lateral infraclavicular block. Before and 30 minutes after performing the lateral infraclavicular block and just after the temperature difference sensed by skin touch was evaluated, a blinded nurse (see the above section) from the trained group evaluated sensory and motor function of both upper limbs. Before initiating the study, the nurses underwent theoretical and practical education to ensure that the block was evaluated in a standardized manner. Cold sensation was assessed by applying a cooled glass ampoule (5°C) over predefined skin areas innervated by the 4 major nerves (musculocutaneous, radial, ulnar, and median nerves), respectively. The sensation was recorded as either cold or not cold. Motor function was tested by the ability to voluntarily: (1) flex the elbow joint (musculocutaneous nerve), (2) extend the wrist (radial nerve), (3) oppose the 1st digit (median nerve), and (4) flex the distal interphalangeal joint of the 5th digit (ulnar nerve). Motor block of each nerve was recorded as either normal or compromised compared with the nonblocked arm.
The block was categorized as successful when all 4 nerves were affected in both sensory and motor function as described above. In contrast, the block was categorized as failed when any part of the tested sensory or motor function was still intact. Patients with successful blocks were transferred directly to surgery, whereas patients with unsuccessful blocks were given a supplementary US-guided block (axillary brachial plexus block or peripheral individual musculocutaneous, ulnar, median, or radial nerve blocks) depending on the failure.
Supplementary Block Assessment by Infrared Thermographic Imaging
In a supplementary assessment, we wanted to verify if there were any significant skin temperature changes after performing the lateral infraclavicular block. We recorded an infrared thermographic image (Thermovision A320, FLIR Systems, Danderyd, Sweden) of the dorsal side of both hands and forearms in prone position at baseline (after acclimatization) and at 30 minutes, starting immediately after performing the block. The camera was newly calibrated and fixed in a standardized position approximately 1.30 m vertically above the bed. Thermovision A320 provides a 2-dimensional thermal image with a thermal resolution of <0.07°C, an accuracy of ±2%, and a picture resolution of 320 × 240 pixels. Because the emissive factor of the skin is 0.98, the measured temperature values were evaluated as skin temperature values. We used a specific software package (ThermaCAM™ Researcher 2.9 Pro, FLIR Systems). Outlined areas of interest and skin temperature were automatically calculated as the mean of the total pixels inside the predefined area of interest. We performed calculations for 2 spot skin temperature measurements, defined as circular areas of 2 mm in diameter distally on the 2nd and 5th digits just radial and proximal to the nail root (skin temperature spot 2 and 5, respectively). Identical measurements of the nonblocked hand were used as control in our assessments.
To evaluate the validity of the test, the diagnostic accuracy was described by the sensitivity, specificity, the predictive values of a positive test (PPV) and negative test, and the positive and negative likelihood ratios.
To evaluate the reliability of the test, the Cohen’s kappa values were calculated for the interobserver agreement between the individual observers. Furthermore, the Fleiss κ was calculated for assessing the reliability of agreement between multiple numbers of observers (in this study: 3).13 The Student t test was used for comparing thermographic skin temperature measurements.
Concerning the validity, we expected 85% successful blocks and a sensitivity of 95%. Using α (2-sided) = 5% and β (power) = 80%, a total of 86 cases were needed. Concerning the reliability, for a 2-tailed test we needed 42 patients to detect a κ-value of 0.80, a null value of 0.40, and a power of 80%.14 Data from our previous study6 were used for estimating the sample size to detect a significant difference in skin temperature in our thermographic assessment. Thus, 20 patients were needed when Δskin temperature = 2.5°C, SD = 3.8, α (2-sided) = 5%, and β (power) = 80%. We agreed to include 45 patients representing 135 cases.
A P-value < 0.05 was considered statistically significant. SPSS software package (SPSS Statistics, version 19.0.0, SPSS, Chicago, IL) was used.
Forty of 45 patients completed the study. Nineteen patients were scheduled for hand or forearm surgery because of fractures and 21 patients were scheduled for elective surgery for various reasons. One patient was excluded because of a toxic reaction to local anesthetic, 2 patients were unable to complete the study due to their inability to keep their hands still during the measurement period, and 2 patients were excluded and therefore did not complete the study because of poor quality of the thermographic images. Thirty of the performed lateral infraclavicular blocks were successful and 10 were failures (Table 1).
Prediction of Block Success
Each of the 40 patients was evaluated by 3 different observers, totaling 120 evaluated cases. In our validity assessment, we excluded 19 cases represented by 15 patients because at baseline the blinded observer sensed a difference in skin temperature between the corresponding distal 2nd and 5th phalanges. Among 10 cases where the detected temperature difference did not correctly predict block success, 4 cases represented by 3 patients were false positive and 6 cases represented by 3 patients were false negative. The dichotomous diagnostic accuracy of a sensed skin temperature difference is demonstrated in Table 2. Thus, among patients with sensed skin temperature differences and a unilateral warmer side indicating a positive diagnostic test, 95% (95% CI, 86%–98%) were successfully blocked (PPV). Moreover, of the patients with a successful lateral infraclavicular block, 92% (95% CI, 83%–97%) had a sensed skin temperature difference between the corresponding 2nd and the 5th distal phalanges and a unilateral warm side (sensitivity).
Interobserver Agreement in Assessing Temperature Differences
In our reliability assessment, all 40 patients representing 120 cases were included. We estimated the Cohen’s kappa coefficient to determine the interobserver agreement among the 3 observers of the (possible) skin temperature differences of the corresponding 2nd and 5th distal phalanges of the blocked and the nonblocked hand. Tables 3 and 4 illustrate examples of the interobserver agreement between 2 observers’ assessments of the skin temperature differences of the 2nd distal phalanx. Finally, we estimated the Fleiss κ for multiple observers as a composite measure for agreement among all observers. Fleiss κ was 0.87 (95% CI, 0.73–0.998,) for the 2nd distal phalanx and 0.74 (95% CI, 0.61–0.87) for the 5th distal phalanx, respectively.
Supplementary Block Assessment by Infrared Thermographic Imaging
Figure 1 illustrates the infrared thermographic images before and 30 minutes after performing a lateral infraclavicular block. In the successfully blocked patients, skin temperature spot 2 and skin temperature spot 5 increased by 5.2°C ± 4.0°C (mean ± SD, P < 0.0001) and 6.6°C ± 4.6°C (P< 0.0001), respectively, from baseline to 30 minutes on the blocked side. In the failed blocks, skin temperature spot 2 and skin temperature spot 5 did not change (Δ skin temperature = –0.36°C ± 4.5°C (P = 0.81) and Δ skin temperature = –0.92°C ± 6.4°C (P = 0.66), respectively). On the nonblocked side, skin temperature spot 2 and skin temperature spot 5 decreased by –7.3°C ± 4.8°C (P < 0.0001) and –7.5°C ± 5.1°C (P < 0.0001), respectively. All 10 patients with failed blocks had thermographic imaging that indicated (partial) block failure. Furthermore, all 30 patients with successful blocks had thermographic imaging indicating block success.
The purpose of this study was to test whether a blinded observer would be able to predict lateral infraclavicular block success based on sensed skin temperature differences between the blocked and the nonblocked hand. A sensitivity and a PPV of 92% and 95%, respectively, indicate that this predefined diagnostic test is valid. In addition, the interobserver agreement evaluated by Fleiss κ demonstrated substantial agreement for the 5th digit and almost perfect agreement for the 2nd digit (κ values interpreted by the following categories: 0.01–0.20 Slight agreement; 0.21–0.40 Fair agreement; 0.41–0.60 Moderate agreement; 0.61–0.80 Substantial agreement; 0.and 81–0.99 Almost perfect agreement).15 This indicates that the test is also reliable. Finally, our thermographic measurements demonstrated significantly increased changes in skin temperature of the 2nd and 5th digits of a lateral infraclavicular-blocked hand. These findings support using skin temperature differences in the distal hand to predict lateral infraclavicular block success.
The blinded observers’ ability to predict lateral infraclavicular or other brachial plexus block success using changes in skin temperature is limited. Galvin et al3 concluded that increased skin temperature measured by infrared thermography in the dermatomes supplying the operative site predicted axillary brachial plexus block success with high accuracy. The sympathetic nerves (the thoracolumbar nervous system) originate from below the first thoracic segment (Th1). The nerve fibers ascend to form the inferior, middle, and superior cervical ganglia supplying the upper limb with sympathetic fibers while joining the brachial plexus in a complex manner. Moreover, the current understanding is that the sympathetic nerve fibers travel peripherally along both major nerves and blood vessels.16 Thus, it is difficult to predict the thermographic patterns after brachial plexus blocks performed at different levels, for example, axillary brachial plexus block, lateral infraclavicular block, and interscalene brachial plexus block. However, we know that peripheral blocking of the ulnar and median nerves leads to characteristic increases in skin temperature in the hand, whereas peripheral blocking of the musculocutaneous or radial nerves does not change skin temperature in the hand.6 Furthermore, the increases in skin temperature were substantial, fastest, and most pronounced distally on the palmar side of the digits with the 4th digit representing a thermographically overlapping area between the median and the ulnar nerves. We therefore hypothesized that a successful lateral infraclavicular block could be predicted most reliably by measuring or sensing a simultaneous increase in skin temperature in the distal part of the 2nd and 5th digits. Since the 4th digit is an overlapping area, this digit is not suited for a diagnostic test. The 5th digit has to be included, whereas either the 1st, 2nd, or 3rd digit could probably be included with identical test results. Changes in skin temperature of the 2nd and 5th digit that can predict lateral infraclavicular block success are likely caused by the arrangement of the 3 nerve cords around the axillary artery.17 We speculate that skin temperature will increase in the 2nd and 5th digit if both the lateral and medial nerve cords are anesthetized, and if these 2 cords are anesthetized, the posterior cord is most likely also anesthetized by the US-guided lateral infraclavicular technique. It is important to stress that the present conclusions are only valid for lateral infraclavicular block. Other brachial plexus blocks may or may not have similar thermographic patterns.
The strength of our study was that qualitative distal skin temperature assessment is easy to perform and appears to be highly predictive. However, there are several limitations to this study. We used clinical block assessment as our outcome measure. A better outcome measure may be block success determined by the surgical procedure. A block was defined as successful when there was sensory and motor deficit for each of the 4 nerves. This may be a rigorous definition, which partly explains the large proportion of 25% failed blocks. Moreover, the large proportion of failed blocks may be explained by the varying degrees of experience of the anesthesiologists who performed the blocks.
In 19 (16%) of the 120 cases, the diagnostic test was not suitable for predicting a successful block because baseline skin temperature differed between hands. In addition, in 10 cases, the diagnostic test was wrong and thereby potentially detrimental (e.g., leading to pain on incision or unnecessary rescue blocking of a primary block that would have been successful). Distal skin temperature is subject to major fluctuations due to changes in environmental temperature, sympathetic nervous activity (stress, anxiety), local inflammation, prior trauma, etc. To reduce the impact of these factors and to standardize measurements, we tried to cool both hands by placing them on cooled packs (5°C). This procedure itself may have introduced fluctuations. Moreover, the method did not provide adequate and constant cooling, since the cooled packs warmed over time, resulting in rewarming of the areas after 20 minutes. Furthermore, the blocked hand relaxes after successful blocking and assumes a position with variable skin contact to the cold packs, also introducing rewarming. By cooling the hands, we may have underestimated the rapidity of changes in skin temperature and overestimated the size of changes in skin temperature.
We included both elective and fracture patients to increase the usability of the test in a general surgical population. However, there may have been some sympathetic tone differences between the 2 groups that caused differences in skin temperature.
In a clinical context, a time span of 30 minutes seems too long and the cooling procedure reduces the applicability of the test. Thus, in future assessments, the validity and reliability of the diagnostic test without cooling are needed. In addition, studies should validate our findings of the optimal timing and cutoff value for the predictive value of changes. It may be important to evaluate the minimum cutoff value of the thermographically measured skin temperature differences of the 2 digits that are possible to detect by simple touch of a blinded observer. Furthermore, it has to be evaluated whether the test performs best as a stand-alone test or whether it should be combined with a clinical block assessment. Finally, our findings may indicate that thermographic patterns and objective measurements of temperature differences are better for predicting block success. However, these findings have to be evaluated in future studies.
In conclusion, we found that blinded observers were able to predict lateral infraclavicular block success with high validity and reliability based on sensed differences between the corresponding distal phalanges of the 2nd and 5th digits of the blocked and nonblocked hand.
Name: Semera Asghar, MD.
Contribution: This author helped conception and design, data retrieval, linkage, validation and analysis, interpretation of results, and prepared the manuscript.
Attestation: Semera Asghar approved the final manuscript, and attests to the integrity of the original data and the analysis reported in this manuscript. Semera Asghar is the archival author.
Name: Kai H. W. Lange, MD, DMSc.
Contribution: This author helped conception and design, data retrieval, linkage, validation and analysis, interpretation of results, and prepared the manuscript.
Attestation: Kai H. W. Lange approved the final manuscript, and attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Lars H. Lundstrøm, MD, PhD, BSc.
Contribution: This author helped linkage, validation and analysis, interpretation of results, and prepared the manuscript.
Attestation: Lars H. Lundstrøm approved the final manuscript, and attests to the integrity of the original data and the analysis reported in this manuscript.
This manuscript was handled by: Terese T. Horlocker, MD.
1. Curatolo M, Petersen-Felix S, Arendt-Nielsen L. Assessment of regional analgesia in clinical practice and research. Br Med Bull. 2004;71:61–76
2. Paqueron X, Gentili ME, Willer JC, Coriat P, Riou B. Time sequence of sensory changes after upper extremity block: swelling sensation is an early and accurate predictor of success. Anesthesiology. 2004;101:162–8
3. Galvin EM, Niehof S, Medina HJ, Zijlstra FJ, van Bommel J, Klein J, Verbrugge SJ. Thermographic temperature measurement compared with pinprick and cold sensation in predicting the effectiveness of regional blocks. Anesth Analg. 2006;102:598–604
4. Hermanns H, Braun S, Werdehausen R, Werner A, Lipfert P, Stevens MF. Skin temperature after interscalene brachial plexus blockade. Reg Anesth Pain Med. 2007;32:481–7
5. Minville V, Gendre A, Hirsch J, Silva S, Bourdet B, Barbero C, Fourcade O, Samii K, Bouaziz H. The efficacy of skin temperature for block assessment after infraclavicular brachial plexus block. Anesth Analg. 2009;108:1034–6
6. Lange KHW, Jansen T, Asghar S, Kristensen PL, Skjønnemand M, Nørgaard P. Skin temperature measured by infrared thermography after specific ultrasound-guided blocking of the musculocutaneous, radial, ulnar, and median nerves in the upper extremity. Br J Anaesth. 2011;106:887–95
7. Raj PP, Montgomery SJ, Nettles D, Jenkins MT. Infraclavicular brachial plexus block–a new approach. Anesth Analg. 1973;52:897–904
8. Klaastad O, Lilleås FG, Røtnes JS, Breivik H, Fosse E. A magnetic resonance imaging study of modifications to the infraclavicular brachial plexus block. Anesth Analg. 2000;91:929–33
9. Klaastad Ø, Smith HJ, Smedby O, Winther-Larssen EH, Brodal P, Breivik H, Fosse ET. A novel infraclavicular brachial plexus block: the lateral and sagittal technique, developed by magnetic resonance imaging studies. Anesth Analg. 2004;98:252–6
10. Klaastad O, Lilleås FG, Røtnes JS, Breivik H, Fosse E. A magnetic resonance imaging study of modifications to the infraclavicular brachial plexus block. Anesth Analg. 2000;91:929–33
11. Sandhu NS, Manne JS, Medabalmi PK, Capan LM. Sonographically guided infraclavicular brachial plexus block in adults: a retrospective analysis of 1146 cases. J Ultrasound Med. 2006;25:1555–61
12. Chin KJ, Alakkad H, Adhikary SD, Singh M. Infraclavicular brachial plexus block for regional anaesthesia of the lower arm (Review). Conchrane Database Syst Rev. 2013;8:CD005487
13. Fleiss JL. Measuring nominal scale agreement among many raters. Psychol Bull. 1971;76:378–82
14. Sim J, Wright CC. The kappa statistic in reliability studies: use, interpretation, and sample size requirements. Phys Ther. 2005;85:257–68
15. Viera AJ, Garrett JM. Understanding interobserver agreement: the kappa statistic. Fam Med. 2005;37:360–3
16. Robertson D, Biaggioni I, Burnstock G, Low PA, Paton JFR. Primer on the Autonomic Nervous System. 20123rd Ed San Diego Academic Press
17. Drake RL, Vogl AW, Mitchell AWMSchmitt William. Upper limb. In: Gray’s Anatomy for Students. 20102nd Ed Philadelphia, PA Churchill Livingstone Elsevier:650–784
© 2015 International Anesthesia Research Society
18. Newcombe RG. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med. 1998;17:857–72