1

Corresponding author: Anne Holmberg Department of anesthesiology, Oslo University Hospital, P.O. Box 4956 Nydalen, 0424 Oslo, Norway. Ph: +47-481 27 279. Fax: +47 22 11 98 56. E-mail address: anneholmberg77@hotmail.com

Microcirculation and haemodynamics after infraclavicular brachial plexus blocks using adrenaline as an adjuvant to lidocaine; a randomised, double-blind, crossover study in healthy volunteers A. Holmberg,1 A. V. T. Ho,2 D. Fernand,3 K. Toska,4,5 T. Wester ,6 Ø. Klaastad,7 T. Drægni8 and A. R. Sauter1,9 Institutional affiliations: 1 Consultant, Department of Anaesthesiology, Oslo University Hospital, Oslo, Norway 2 MD, Ph D student, Faculty of medicine, Section for physiology, University of Oslo, Oslo Norway 3 Student, Faculty of medicine, Section for physiology, University of Oslo, Oslo Norway 4 Professor, Faculty of medicine, Section for physiology, University of Oslo, Oslo Norway 5 Consultant, Department of Medical Biochemistry, Oslo University Hospital, Oslo, Norway 6 Consultant, Department of Surgery, Section for plastic and recontructive surgery, Østfold Hospital Trust, Moss, Norway 7 Consultant, Department of Anaesthesiology, University Hospital of North Norway, Tromsø, Norway

8Research nurse, Department of Research and Development, Oslo University Hospital, Oslo, Norway

9 Consultant, Department of Anaesthesiology and Department of Research and Development, Division of Emergencies and Critical Care, Oslo University hospital, Oslo, Norway, Twitter: @ax_sauter Correspondence to: Dr A. Holmberg, anneholmberg77@hotmail.com Short title: Circulatory changes after brachial plexus block Key words: brachial plexus block, microcirculation, adrenaline, heart rate, hemodynamics E-mail adresses for all authors: Anne Holmberg: anneholmberg77@hotmail.com Ai Van Thuy Ho: ai_van_11@me.com Dmitry Fernand: dmitryfernand@gmail.com Karin Toska: karin.toska@gmail.com Torjus Wester: Torjus.Wester@so-hf.no Øivind Klaastad: oe.klaastad@gmail.com Tomas Drægni: TOMDRA@ous-hf.no Axel R. Sauter: axsa@online.no We are sorry, but most of the authors do not use twitter. We have added twitter handles only for the auhor who use the social networking service.

2

Summary

We evaluated the effect of adrenaline when added to lidocaine in infraclavicular brachial plexus

blocks on human skin microcirculation (nutritive and subpapillary) and systemic cardiovascular

variables. Twelve healthy, non-smoking, male volunteers were included, each attending two study

sessions two weeks apart, with a crossover design. In both sessions, they received an ultrasound-

guided infraclavicular brachial plexus block in the non-dominant arm with 0.4 ml.kg-1 lidocaine 15

mg.ml-1 with or without adrenaline 5 µg.ml-1. Microcirculation was assessed by laser Doppler

fluxmetri (subpapillary blood flow), capillary video microscopy (nutritive blood flow), and continuous

temperature measurements. Heart rate and arterial pressure were recorded continuously and non-

invasively. Median (IQR [range]) subpapillary blood flow increased substantially 30 minutes after the

brachial plexus block, from 8.5 (4.4-13.5[2.9-28.2]) to 162.7(111.0-197.8[9.5-206.7]) arbitrary units

with adrenaline (p=0.017), and from 6.9(5.3-28.5[1.8-42.1] to 133.7(16.5-216.7[1.0-445.0] arbitrary

units without adrenaline (p=0.036). Nutritive blood flow (functional capillary density, capillaries.mm-

2, measured at the dorsal side of the hand) decreased in the blocked extremity when adrenaline was

used as adjuvant, from median (IQR [range]) 45(36-52[26-59]) to 38(29-41[26-42]) (p=0.028),

whereas no significant change occurred without adrenaline. Median (IQR [range]) finger skin

temperature (degrees Celsius) increased by median 44% (data pooled) with no significant differences

between the groups. No significant changes were found in the systemic cardiovascular variables with

or without adrenaline. We conclude that lidocain infraclavicular brachial plexus blocks cause an

increase in skin subpapillary blood flow. The addition of adrenaline gives more strong and long-

lasting blocks, but decrease the nutritive blood flow.

3

Introduction

Brachial plexus blocks are often used to provide analgesia and anaesthesia for upper limb surgery. A

block of the sympathetic fibers following the brachial plexus block will also increase peripheral blood

flow and skin temperature [1-4]. Brachial blocks are therefore used with the intention to improve

peripheral circulation after microvascular procedures, reconstruction surgery after traumatic

amputation, in treatment of vasospasm induced by Raynaud disease, and in treatment of peripheral

embolism [5,6].

The sympathetic blockade, any systemic uptake of local anaesthetics or adjuvants may induce

changes in blood pressure and cardiac output. Adrenaline is commonly added as an adjuvant to local

anaesthetics to prolong the duration of anaesthesia and reduce the peak plasma level of local

anaesthetics [7,8]. Systemic effects of adrenaline include increased heart rate, contractility and

arterial blood pressure. Locally, adrenaline causes vasoconstriction in the skin through an alfa-

adrenergic effect, whereas binding to beta-adrenergic receptors in arterioles in skeletal muscle

induce vasodilation [9].

Comparing brachial plexus blocks with and without adrenaline, McGregor et al. found a greater

increase in blood flow to the arm and a higher temperature rise when adding adrenaline to an

axillary brachial plexus block [10]. The authors suggested that these findings were caused by the

systemic effect of adrenaline inducing a rise in cardiac output. However, peripheral blood flow is

known to consist of two different components, which differ in their importance on perfusion of

tissue cells: the subpapillary blood flow which is controlled by synchronous opening or closing of

arteriovenous anastomosis for thermoregulation, and the nutritive blood flow in capillaries to cells in

skin and muscle[11]. A blockade of vasoconstrictive sympathetic nervous impulses to blood vessels,

for instance with adrenaline added to the block, is expected to increase both local nutritive blood

flow to skin and muscle, as well as increase blood flow through arteriovenous anastosmoses[12].

4

However, the effect of adrenaline adjuvant on these two different vascular entities have not been

studied in detail, to our knowledge, so far.

The aim of our study was to investigate the effect of adrenaline when added to lidocaine in

infraclavicular brachial plexus block on human skin microcirculation, both subpapillary blood flow

and nutritive blood flow, as well as the general systemic cardiovascular variables [13,14]. For our

primary aim, we tested the hypothesis that adrenaline added to lidocaine in brachial plexus block

cause alterations in both subpapillary blood flow and nutritive blood flow. Secondarily we tested if

the addition of adrenaline may cause changes in systemic cardiovascular variables. At last, we

wanted to characterize the time course of changes in blood flow to the blocked arm and control arm

as well as the characteristics of block with or without adrenaline adjuvant.

5

Methods

The protocol of this randomized, double blind, prospective, crossover study in healthy volunteers

was reviewed and approved by the Regional Committee for Medical Research Ethics in South-Eastern

Norway (2012/2189b) and the Norwegian Medicines Agency. The trial was registered in the EU

Clinical trials Register (EudraCT number 2012-005651-17) prior to enrolment of volunteers started. It

was conducted in adherence to guidelines for Good Clinical Practice. An external monitor from

Akershus University hospital monitored the study. This manuscript adheres to the applicable

CONSORT guidelines.

Twelve healthy, non-smoking male volunteers, age 18 – 60, American Society of Anesthesiologists

(ASA) physical status 1 or 2 with body mass index 18-35 kg.m-2 were recruited by poster

advertisements, and included after informed written consent. The subjects fasted at least 6 hours,

and were instructed not to consume coffee, black tea or garlic 12 hours prior to the experiments. No

medication was allowed during the last seven days before the experiments. Exclusion criteria were:

known nerve injury or nerve compression syndrome of the upper limb; diabetes mellitus; known

reduced immune response; any regular medication; and contraindications to lidocaine or adrenaline.

Each volunteer attended two study sessions two weeks apart, following an AB/BA crossover setup. In

both sessions, the participants received an ultrasound guided infraclavicular brachial plexus block in

the non-dominant arm. A volume of 0.4 ml.kg-1 lidocaine 15 mg.ml-1 was injected, either with or

without adrenaline 5 µg.ml-1. The sequence of blocks, i.e. if adrenaline was added in the first or the

second session, was randomized by computer-generated codes and packed in sealed envelopes by a

person not involved in the experiments. In each experiment, the local anaesthetic solution (with or

without adrenaline) was provided in two 20 ml syringes by an investigator not taking part in the

conduction of the experiment, thus the volunteers and all investigators present at the experimental

sessions were blinded to group allocation.

6

The volunteers were placed in the supine position in an operating theatre with controlled room

temperature of 21°C for 20 minutes. In order to keep the subjects relaxed, the investigators were as

quiet as possible during the experiments and soothing music was played in the room. An intravenous

line was established in the contralateral arm to the block. Standard monitoring including ECG, pulse

oximetry, and upper arm non-invasive blood pressure was applied.

Lateral sagittal infraclavicular block

An ultrasound-guided lateral sagittal infraclavicular brachial plexus block was performed in a

standardized way by one of the authors (ARS) as described in previous publications [15,16]. A Phillips

HD 11 XE ultrasound unit (Philips Medical Systems, Bothell, WA, USA) with a C8-5, 5 to 8 MHz, curved

linear transducer was used. A 22-gauge, 80mm Sono Plex Stim cannula (Pajunk GmbH, Geisingen,

Germany) was placed under ultrasound guidance, in-plane in a central position to the plexus, cranio-

posterior and close to the axillary artery. Start of block was defined as the time when the injecting

needle was withdrawn (‘needle out’).

The effect of the blocks was assessed after all measurements were completed, 60 min after local

anaesthetics injection. Sensory testing included the musculocutaneous, radial, median, ulnar, and

medial antebrachial cutaneous nerve. Blocks were assessed by temperature testing, using an ice bag

repeatedly touching the skin on predefined positions in the sensory area of the nerves. A sensory

block was considered successful when the volunteer had either no sensation for cold or touch

(anaesthesia) or no sensation for cold (analgesia). Motor testing included the musculocutaneous,

radial, median and ulnar nerve. A motor block was considered successful when the subjects had less

than 50% of normal muscle power as estimated by the examiner. To estimate block duration, the

volunteers were asked to report the first time after the block when they were able to hold and drink

from a cup with the blocked arm.

7

Laser Doppler measurements

Laser Doppler Fluxmetry (LDF) (MBF3D, Moor Instruments, Devon, UK) was used to measure

continuous skin blood flow from the left and right hypothenar, expressed in arbitrary units (AU). A

low-energy monochromatic laser beam is reflected from red blood cells 1 – 2 mm below the skin

surface [17,18]. This permits measurement of blood flow mainly in the subpapillary vessels, including

the arteriovenous anastomoses, where the blood flow is primarily serving a temperature-regulating

purpose[11]. The laser Doppler probes were fixed to the skin with double-sided tape. The noise-

limiting filter of the instrument was set at its highest level (21 kHz), and the emitted wavelength was

820 mm. The flux output signal was filtered with a time constant of 0.1 second and sent to the

recording computer. The sampling frequency was 2 Hz.

Capillary video microscopy

Capillary video microscopy with a few micrometres penetration depth was used to assess the

nutritive circulation of the skin. This permits assessments of both the functional density and the flow

velocity of visible capillaries. It was performed twice with a handheld digital microscope (GT700 UV,

Firefly, Belmont, MA, USA), first after 20 minutes of acclimatization and again 30 minutes after the

brachial plexus block, by one of the investigators (AH). Baby skin care oil (Sebamed®, Boppard,

Germany) was applied on the skin before examination. The microscope was then placed between the

first and second metacarp on the dorsal side of both hands. The hand of the side to be blocked was

examined first. Five video sequences were recorded for each examination, each video lasting 15

seconds. The recorded sequences were subsequently analysed using the software QuickTime Player

10.4 for MAC (Apple Inc., Cupertino, CA, USA) using a ME32C colour display unit (Samsung

electronics, Surrey, UK). Xscope software for MAC (the Iconfactory, Greeensboro, NC, USA) was used

to create a grid dividing the screen into 16 equal squares. Capillaries were counted in the four central

squares. When artefact occurred in any of the selected squares, four alternative adjacent squares

8

were randomly chosen. Average numbers from the capillary counts of two investigators (AH, ARS)

were used for analyses of functional capillary density (average number of counted papillary loops per

mm2).

A third investigator (TW) with broad experience in capillary flow assessment examined the flow

velocity. The flow was classified into five different categories: 0) No flow; 1) Sluggish flow; 2)

Continous low flow; 3) Continous high flow; 4) Brisk flow [19].

Temperature measurements

Thermocouples were attached to the digital pulp of finger 1, 2, 4, and 5 on both hands, excluding the

middle finger with the blood pressure cuff. The temperatures were continuously sampled at 1-

minute intervals with a multi-channel Hydra Series II data logger (Fluke, Everett, WA, USA) and

transferred for storage to a recording computer. Due to technical failure we had to replace the Hydra

Series II data logger and used a C28 K type thermometer (Comark, Norwich, UK), with temperature

measurements every 5 minutes on two volunteers on day two. A Hydra Series III (Fluke, Everett, WA,

USA) data logger was used for one volunteer on day two.

Non-invasive continuous hemodynamic monitoring

Heart rate (HR) was calculated over each R-R interval of a three lead electrocardiogram (ECG). Finger

arterial blood pressure was recorded continuously and non-invasively from the third finger of both

hands, using two Nexfin measurement devices (Nexfin®, BMEYE B.V., Amsterdam, the Netherlands).

Blood pressure measurements were performed on both sides to have comparable situations on the

blocked extremity and on the control arm. Only the recording from the dominant hand, which was

not used for plexus block, is reported. Both hands were placed on the arm boards of the operation

table in the height of the right atrium. The instantaneous blood pressure curve was transferred to a

9

recording computer where beat-to-beat mean arterial pressure (MAP) was calculated by integration

over each RR-interval.

The automatic self-calibration of the Nexfin disturbed minor parts of the recordings. After the

experiments, the MAP values measured during these calibration periods (3-5 heart beats) were

replaced by values obtained by interpolation from the MAP values on the beats immediately before

and after calibration. This was done with a custom-made computer program coded in MATLAB

(MathWorks inc, Natick, MD, USA).

Statistical methods and data analysis:

The main hypothesis was that adding adrenaline to a brachial plexus block cause alterations in both

components of skin blood flow. Based on pilot measurements (with a comparable setup as we used

in our study) we used the Doppler flow for statistical power analyses. We expected average Doppler

flow measurements around 40 arbitrary units (AU) and a SD of 20 AU. Based on previous studies and

our own pilot measurements, we expected an increase in blood flow of at least 50% [10]. A sample

size of 10 volunteers would have 80% power to detect a difference in means of 50% assuming

average laser Doppler flow rates of 40 AU (mean1) and 60 AU (mean2) with a common standard

deviation of 20 AU (with 0.5 covariance), using a 2-sided one-sample t-test with alpha 0.05. To

compensate for a lower efficacy of a non-parametric (Wilcoxon rank-sum) test we increased the

number by 15% to a sample size of 12 volunteers.

In order to calculate an average time course of the cardiovascular responses to the brachial block,

the median value from all subjects was calculated in each separate time bin of 0.1 second for the

cardiovascular variables. The mean temperature of the four fingers on each hand was calculated for

each time bin of 1 minute, and the median response from all subjects was calculated.

10

Data were entered into an Access CRF database and further analysed by IBM SPSS, version 25.0 (SPSS

Inc., Chicago, IL, USA). As a result of the small sample size, non-parametric methods were used and

data are presented as median, interquartile range and range. Wilcoxon rank-sum test for paired

samples are used for all data except for evaluating block success, when Pearson Chi Square test with

Yates correction is used. P-values smaller than 0.05 are considered statistically significant.

Considering the exploratory nature of the study, no adjustment for multiple testing/estimation was

performed for the primary or secondary outcomes.

11

Results:

Twelve male volunteers were included after given written informed consent. An incorrect group

allocation occurred in one participant who mistakenly received treatment without adrenaline twice.

The data set of this volunteer was therefore excluded. Furthermore, 3 more subjects were excluded

from the cardiovascular analyses, due to technical errors during the recordings. Because of functional

failure with the microscope, functional capillary density is missing for one of the subjects after

brachial plexus block with adrenaline.

The median (IQR [range]) age, weight, height, and BMI were 22(21-27[19-53]) years, 77(66-86[61-

95]) kg, 180(174-186[169-200]) cm, and 22.5(21.5-23.8[20.5-29.3]) kg.m-2, respectively. Median (IQR

[range]) injected volume of lidocaine in the blocks was 31(26-34[24-38]) ml, either with or without

adrenaline 5 µg.ml-1.

Block assessment

During the experiments, all volunteers reported a subjective feeling of effective blocks. At block

assessment one hour after initiation of the blocks, block success according to our strict predefined

criteria was significantly higher in the blocks with adrenaline. Complete sensory blockade of all distal

nerves (musculocutaneous, radial, median, ulnar, and medial antebrachial cutaneous nerve)

combined with complete motor blockade of all distal nerves (musculocutaneous, radial, median and

ulnar nerve) was seen in 10 (91%) of the blocks with adrenaline and 2(18%) of the blocks without

adrenaline, p=0.003. Median (IQR [range]) block duration (tested by obtained motor control) was

203(125-230[113-285]) min in the blocks with adrenaline and only 67(62-72[30-93]) min in the blocks

without adrenaline (p=0.003).

Laser Doppler measurements

12

A substantial increase in blood flow was observed in all subjects in the blocked arm, both in blocks

with adrenaline and blocks without adrenaline (figure 1, table 1, table 2). There was no statistically

significant difference in blood flow between block with and without adrenaline in the blocked arm

(p=0.889) (figure 1).

Capillary video microscopy

Functional capillary density decreased 30 min after brachial plexus block with adrenaline compared

with pre-block measurements in the blocked arm (figure 2, table 1, table 2). After the blocks without

adrenaline, no significant decrease was found (table 1, table 2). However, functional capillary

density was always significantly lower in the blocked arm compared to the control arm 30 min after

block (p=0.005 with adrenaline, p=0.033 without adrenaline). We did not find a statistically

significant difference in functional capillary density when comparing brachial plexus block with and

without adrenaline (p=0.074).

No significant changes were found for the capillary flow patterns and flow velocity.

Temperature measurements

Thirty min after brachial plexus block, we found increased temperature in the blocked arms

compared to the pre-block measurements in both the blocks with and without adrenaline (table 1).

The temperature changes were also significantly higher in the blocked arm compared to the control

arm in both types of block (table 2). Adrenaline delayed the onset of the temperature change in the

fingers. Time to half temperature rise in minutes (median (IQR [range])) was 11.0(9.4-15.5[6.6-26.0])

minutes with adrenaline and 7.6(3.7-11.2[0.0-14.4]) minutes without adrenaline, p= 0.013. Thirty

minutes after the block, no statistically significant temperature differences were found when

comparing brachial plexus block with and without adrenaline (p=0.074).

13

Non-invasive continuous hemodynamic monitoring

After brachial block, no substantial changes in MAP were seen, neither with nor without adrenaline

(figure 1, table 1). We did not find a statistically significant difference in HR and MAP when

comparing brachial plexus block with and without adrenaline (p=0.674 and p=0.575).

14

Discussion:

Our study shows an increase in hypothenar subpapillary skin blood flow after infraclavicular brachial

block in the anaesthetized extremity. Skin temperature increased correspondingly. In contrast,

nutritive blood flow to the skin on the dorsal side of the hand decreased when adrenaline was used

as an adjuvant. Compared with the control arm, nutritive blood flow to the skin was lower after

blocks with and without adrenaline. The use of adrenaline had no significant effect on cardiovascular

variables, but resulted in stronger nerve blocks.

Peripheral nerve blocks decrease the intensity of sympathetic nervous signals transmitted by

noradrenaline to blood vessels in the arm, both in muscles and skin. A block of the sympathetic nerve

fibres innervating the arteriovenous anastomosis in the glabrous (non-hairy) skin increases blood

flow in the palms and fingers [4]. In our study a substantial increase in blood flow in the blocked arm

was observed after infraclavicular brachial plexus blocks. However, the sympathetic block did not

increase blood flow to the nutritive arterioles in the non-glabrous skin on the dorsal side of the hand

were arteriovenous anastomosis are not present. On the contrary, we found lower functional

capillary density after infraclavicular blocks compared with the control arm, indicating a decreased

skin microcirculation in this area. This observation could be explained by a steal effect caused by the

increase in blood flow to the arteriovenous anastomosis in the palms and fingers [20]. Circulating

adrenaline may also cause direct vasoconstriction in nutritive blood flow to the skin. The

vasoconstrictive effect might increase when there is a steal effect in the arteriovenous anastomosis

due to the sympathetic blockade.

When we evaluated the blocks after about 60 minutes, we found significantly better blocks and

higher block success in the adrenaline group compared with the non-adrenaline group. We evaluated

all five distal nerves, and considered all partly failed blocks as failed blocks. Another definition of

block success, as suggested by Asghar et al, use increase in skin temperature as a valid and reliable

15

measure for predicting successful infraclavicular blocks [21]. According to this definition, all blocks

would be considered successful during the first 30 min. Several participants reported a subjective

feeling of an early block resolution after the no-adrenaline blocks before the scheduled block

evaluation after 60 minutes, which indicates that the blocks were about to wear out at block

evaluation.

Similar observations as ours on block quality were reported by Dogru et al when comparing the

effect of high doses of adrenaline with no adrenaline for axillary plexus blocks with lidocaine [22].

Adrenaline prolonged the duration of the blocks due to vasoconstriction at the injection site and

slower diffusion of local anaesthetics into the nerve fibres. The longer duration and better quality of

the adrenaline blocks in our study can explain the greater changes in subpapillary and nutritive blood

flow. Better block quality after addition of adrenaline might also cause a stronger steal effect to the

nutritive skin perfusion compared with the blocks without adrenaline.

Using continuous non-invasive monitoring we did not observe an effect of adrenaline on systemic

cardiovascular variables like heart rate or blood pressure. We can therefore presume that adrenaline

had little or no systemic effect that can explain the results of our study. The systemic effect of

adrenaline might depend on the volume and the injection site. A previous study by Dogru et al.

showed that low doses of adrenaline offered more stable hemodynamics (blood pressure and heart

rate), but similar block quality compared with high doses [22]. A recent study on paracervical blocks

showed a highly significant increase in blood pressure, heart rate and cardiac output when

adrenaline was used as an adjuvant to bupivacaine [23].

Laser Doppler was used in several studies to investigate changes in skin perfusion after brachial

plexus blocks, all of them reporting alterations in circulations. Arterial vasodilatation, increase in

blood flow velocity, and increase in blood flow through the ipsilateral brachial artery have been

16

shown after axillary brachial plexus blocks [24]. Landsverk et al observed alterations in the oscillatory

components of the flowmetry signal in the blocked arm and the contra-lateral arm. However, blood

flow was not significantly increased in this study [2]. Interestingly, the group measured skin perfusion

on the volar side of the hand where arteriovenous anastomosis are not present. Lethipalo and

colleagues observed increased skin perfusion in the index finger after interscalene brachial plexus

block [25].

Our findings may have important implications for the clinical setting: Brachial plexus blocks increase

subpapillary blood flow in glabrous skin. On the other hand, the nutritive blood flow to the non-

glabrous skin decreases after the blockade with adrenaline adjuvant, which enhance the block

strength and duration. From these findings it may be speculated that adrenaline adjuvant is to be

recommended when a strong block for surgery or pain relief is required, whereas a block without

adrenaline may be recommended if the purpose of the block is to increase the microcirculation and

oxygenation of peripheral cells. Both aspects should be considered when brachial plexus blocks are

performed to enhance vascular microcirculation.

Our study has several limitations. All our participants were healthy and non-smoking. In a clinically

setting, this is not always the case. We did not include women because female reproductive

hormones may influence on the non-noradrenergic mechanisms of vasoconstriction [26]. Our

participants were predominantly young, thus the results may not apply in older patients because of

age related alterations in skin blood flow. Further, an acclimatisation period of 20 minutes might

have been to short and may have resulted in changes in the measurements in the control arm during

the course of our registrations. Our results only applies to infraclavicular blocks with lidocain 15

mg.ml-1 using adrenaline 5 µg.ml-1 as an adjuvant with an injection volume of 0.4 ml.kg-1. Because of

a small sample size in our study, we might have type 2 errors in our statistical calculations.

17

In conclusion, subpapillary skin blood flow increase after infraclavicular brachial plexus blocks,

whereas nutritive blood flow to the non-glabrous skin decreases. Adrenaline as an additive did not

have a significant effect on cardiovascular variables, but resulted in stronger nerve blocks and

possibly a more pronounced decrease in nutritive blood flow.

Competing interests:

Oslo University Hospital founded the complete work. In addition AH were 50% employed by a

scholarship from the South-Eastern Norway Regional Health Authority.

Acknowledgements:

Johan Raeder has kindly read through the paper and given constructive input to the manuscript

preparation.

18

References:

1. Iskandar H, Wakim N, Benard A, et al. The effects of interscalene brachial plexus

block on humeral arterial blood flow: a Doppler ultrasound study. Anesthesia and

Analgesia 2005; 101: 279-81, table of contents.

2. Landsverk SA, Kvandal P, Kjelstrup T, et al. Human skin microcirculation after

brachial plexus block evaluated by wavelet transform of the laser Doppler flowmetry

signal. Anesthesiology 2006; 105: 478-84.

3. Lange KH, Jansen T, Asghar S, Kristensen PL, Skjonnemand M, Norgaard 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.

4. Hermanns H, Werdehausen R, Hollmann MW, Stevens MF Assessment of skin

temperature during regional anaesthesia-What the anaesthesiologist should know. Acta

Anaesthesiol Scand 2018; 62: 1280-9.

5. Berger A, Tizian C, Zenz M Continuous plexus blockade for improved circulation in

microvascular surgery. Ann Plast Surg 1985; 14: 16-9.

6. Aguirre J, Del Moral A, Cobo I, Borgeat A, Blumenthal S The role of continuous

peripheral nerve blocks. Anesthesiol Res Pract 2012; 2012: 560879.

7. Sinnott CJ, Cogswell IL, Johnson A, Strichartz GR On the mechanism by which

epinephrine potentiates lidocaine's peripheral nerve block. Anesthesiology 2003; 98:

181-8.

8. Newton DJ, McLeod GA, Khan F, Belch JJ Mechanisms influencing the vasoactive

effects of lidocaine in human skin. Anaesthesia 2007; 62: 146-50.

9. Rang HP. Rang and Dale`s pharmacology, 8th ed. edn. London, UK, Elsevier

Churchill Livingstone, 2016.

10. McGregor AD, Jones WK, Perlman D Blood flow in the arm under brachial plexus

anaesthesia. J Hand Surg Br 1985; 10: 21-4.

11. Gschwandtner ME, Ambrozy E, Fasching S, et al. Microcirculation in venous ulcers

and the surrounding skin: findings with capillary microscopy and a laser Doppler

imager. Eur J Clin Invest 1999; 29: 708-16.

12. Robertson D, Biaggioni I. Primer on the autonomic nervous system, 3rd edn.

Amsterdam ; Boston, Elsevier/AP, 2012.

13. Ambrozy E, Waczulikova I, Willfort-Ehringer A, Ehringer H, Koppensteiner R,

Gschwandtner ME Microcirculation in mixed arterial/venous ulcers and the

surrounding skin: clinical study using a laser Doppler perfusion imager and capillary

microscopy. Wound Repair Regen 2009; 17: 19-24.

14. Szili-Torok T, Paprika D, Peto Z, et al. Effect of axillary brachial plexus blockade on

baroreflex-induced skin vasomotor responses: assessing the effectiveness of

sympathetic blockade. Acta Anaesthesiol Scand 2002; 46: 815-20.

15. Sauter AR, Dodgson MS, Stubhaug A, Halstensen AM, Klaastad Ø Electrical nerve

stimulation or ultrasound guidance for lateral sagittal infraclavicular blocks: a

randomized, controlled, observer-blinded, comparative study. Anesth Analg 2008;

106: 1910-5.

16. Sauter AR, Smith HJ, Stubhaug A, Dodgson MS, Klaastad Ø Use of magnetic

resonance imaging to define the anatomical location closest to all three cords of the

infraclavicular brachial plexus. Anesth Analg 2006; 103: 1574-6.

17. Bergersen T A study of arteriovenous anastomoses in human skin with special

reference to their response to local temperature (thesis) 1999.

19

18. Lossius K, Eriksen M Spontaneous flow waves detected by laser Doppler in human

skin. Microvasc Res 1995; 50: 94-104.

19. Wester T, Haggblad E, Awan ZA, et al. Assessments of skin and tongue

microcirculation reveals major changes in porcine sepsis. Clin Physiol Funct Imaging

2011; 31: 151-8.

20. Walloe L Arterio-venous anastomoses in the human skin and their role in temperature

control. Temperature (Austin) 2016; 3: 92-103.

21. Asghar S, Lange KH, Lundstrom LH Blinded observer evaluation of distal skin

temperature for predicting lateral infraclavicular block success. Anesthesia and

Analgesia 2015; 120: 246-51.

22. Dogru K, Duygulu F, Yildiz K, Kotanoglu MS, Madenoglu H, Boyaci A

Hemodynamic and blockade effects of high/low epinephrine doses during axillary

brachial plexus blockade with lidocaine 1.5%: A randomized, double-blinded study.

Reg Anesth Pain Med 2003; 28: 401-5.

23. Sjoeen GH, Falk RS, Hauge TH, Langesaeter E Haemodynamic stability after

paracervical block: A randomized, controlled, double-blind study comparing

bupivacaine-adrenaline with bupivacaine. Acta Anaesthesiol Scand 2018.

24. Li J, Karmakar MK, Li X, Kwok WH, Ngan Kee WD Regional hemodynamic

changes after an axillary brachial plexus block: a pulsed-wave Doppler ultrasound

study. Reg Anesth Pain Med 2012; 37: 111-8.

25. Lehtipalo S, Winso O, Koskinen LO, Johansson G, Biber B Cutaneous sympathetic

vasoconstrictor reflexes for the evaluation of interscalene brachial plexus block. Acta

Anaesthesiol Scand 2000; 44: 946-52.

26. Stephens DP, Aoki K, Kosiba WA, Johnson JM Nonnoradrenergic mechanism of

reflex cutaneous vasoconstriction in men. Am J Physiol Heart Circ Physiol 2001; 280:

H1496-504.

20

Table 1: Comparison of measurements before and 30 minutes after infraclavicular plexus block.

Blood flow (Laser Doppler), functional capillary density (FDC), and skin temperature was registered in

the blocked arm and the control arm. Heart rate and mean arterial pressure was measured

simultaneously. Lidocaine with and without adrenaline as an additive was used for the blocks. Values

are median (IQR [range]).

Adrenaline Side N Before block 30 min after block P-value

Blood flow

(AU)

Yes Block 8 8.5(4.4-13.5[2.9-28.2]) 162.7(111.0-197.8[9.5-206.7]) 0.017

Control 8 -0.8(-0.9-4.1[-0.9-5.4]) -0.8(-0.9-13.8[-0.9-18.2]) 0.036

No Block 8 6.9(5.3-28.5[1.8-42.1] 133.7(16.5-216.7[1.0-445.0] 0.036

Control 8 4.9 (-0.8-11.4 [-0.9-16.7]) 9.8(-0.7-21.9[-0.9-37.5]) 0.069

FDC

(Capillaries.mm-

2)

Yes Block 10 45(36-52[26-59]) 38(29-41[26-42]) 0.028

Control 10 40(34-53[31-65]) 48(42-58[40-70]) 0.005

No Block 11 39(35-49[29-51]) 40(33-44[30-56]) 0.594

Control 11 45(38-50[30-53]) 47(40-51[35-71]) 0.477

Temperature

(oC)

Yes Block 11 22.1(21.5-22.9[20.5-25.4]) 34.1(33.6-34.6[25.9-35.1]) 0.003

Control 11 22.0(21.7-26.3[20.4-28.9]) 21.5(21.0-22.1[20.1-23.4]) 0.006

No Block 11 22.4(21.5-25.7[20.8-29.6]) 33.9(31.4-34.5[20.8-34.8]) 0.004

Control 11 22.5(21.9-27.3[20.6-30.6]) 21.7(20.9-22.6[20.7-27.3]) 0.004

HR

(beats.min-2)

Yes 8 56.2(48.4-64.7[43.5-77.1]) 62.6(55.6-70.9[51.5-72.6]) 0.263

No 8 52.9(49.9-59.7[46.8-79.0]) 63.5(54.1-68.7[50.7-70.0]) 0.208

MAP

(mmHg)

Yes 8 95.8(91.8-111.8[91.3-122.0]) 101.6(88.0-114.6[70.3-125.2] 0.779

No 8 90.1(76.7-96.6[71.4-102.7]) 90.7(89.8-101.0[72.4-124.2]) 0.161

21

Table 2: Comparison of changes in blood flow (Laser Doppler), functional capillary density (FDC), and

skin temperature before block to 30 minutes after block in the blocked arm compared with the

unblocked (control) arm. Lidocaine with and without adrenaline as additive was used for the blocks.

Values are median (IQR [range]).

N Adrenaline Blocked arm Control arm P-value

∆ Blood flow

(AU)

8 Yes 142.0(107.4-189.2[-1.2-200.0]) 0.0(0.0-9.8[-0.0-13.5]) 0.017

8 No 120.5(-0.3-211.1[-0.9-413.9]) 1.9(0.0-11.3[-1.8-29.6]) 0.050

∆ FCD

(Capillaries.mm-2)

10 Yes -8.6(-15.2-1.4[-21.3-3.9]) 5.7(3.5-14.8[0.4-16.7]) 0.007

11 No -4.5(-8.1-6.7[-15.7-14.8]) 1.6(-4.1-12.5[-8.8-28.5]) 0.182

∆ Temperature

(oC)

11 Yes 11.8 (9.3-12.5 [4.4-14.2]) -0.7(-4.1-(-0.3) [-5.9-0.2]) 0.003

11 No 10.3(4.9-12.0 [-0.7-12.8]) -1.2(-3.4-(-0.2) [-5.2-0.1]) 0.004

22

Figure legends:

Figure 1:

A: Changes in subpapillary blood flow: Skin blood flow measured by laser Doppler in the blocked arm

and the non-blocked arm (control) after brachial block with and without adrenaline. B: Average

temperature in the four fingers in the blocked arm and non-blocked arm (control) after blocks with

and without adrenaline. C: Average beat-to-beat recordings of MAP. D: Average beat-to-beat

recordings of HR.

Figure 2:

Changes in nutritive blood flow: Picture A shows functional capillary density between the first and

second metacarp on the dorsal side of the hand before infraclavicular block. Figure B shows

functional capillary density in the same place in the same patient 30 minutes after infraclavicular

block with adrenaline.