Perineural versus intravenous administration ofadjuvants to ropivacaine in brachial plexus blockSystematic review and quality of reporting assessment

(1)

Örebro university

School of Medical Sciences Degree project, 30 ECTS 2nd June 2017

Perineural versus intravenous administration of

adjuvants to ropivacaine in brachial plexus block

Systematic review and quality of reporting assessment

Version 2

Author: Masi Koskinen Supervisor: Maija-Liisa Kalliomäki, MD, PhD Tampere, Finland

(2)

Abstract

Introduction Brachial plexus block (BPB) is commonly employed in shoulder surgery where it offers effective but short-lived postoperative analgesia. Perineu-ral and intravenous adjuvants have been studied in search of ways to prolong the duration of analgesia. The effectiveness of adjuvants to ropivacaine and their op-timal route of administration are currently unknown, and therefore the main focus of this review.

Health care recommendations and policies are widely based on results from ran-domised clinical trials (RCTs). However, the quality of reporting of trials is shown to be far from optimal. Hence our secondary focus was to assess the reporting quality of our study material.

Methods We performed a systematic multi-database search of placebo-controlled RCTs where adjuvants to ropivacaine in BPB are administered both perineurally and intravenously. Trial findings concerning analgesia and adverse effects were extracted for comparison. Further, the quality of trial reporting was assessed as adherence to the CONSORT statement and quantified as the total number of items reported in accordance with the statement.

Results Eight trials on four adjuvants were included. Dexamethasone gave in-consistent results. Dexmedetomidine improved the postoperative analgesia per-ineurally and intravenously but was observed to cause intraoperative bradycardia and hypotension. Ketamine produced psychotomimetic adverse effects without enhancing the analgesia. Perineural, but not intravenous, parecoxib improved the pain relief without any evident adverse effects. The mean CONSORT scores was 21.3 (of a maximum of 37).

Conclusions Dexmedetomidine seemed the most promising adjuvant via both routes of administration. Overall, the reporting quality was suboptimal.

(3)

Abbreviations

BPB . . . Brachial plexus block

CENTRAL . . . Cochrane Central Register of Controlled Trials

CINAHL . . . Cumulative Index to Nursing and Allied Health Litera-ture

CONSORT . . . Consolidated Standards of Reporting Trials COX . . . Cyclooxygenase

GRADE . . . Grading of Recommendations Assessment, Development, and Evaluation

MBS . . . Motor Block Score

MEDLINE . . . Medical Literature Analysis and Retrieval System On-line

NRS . . . Numerical Rating Scale VAS . . . Visual Analogue Scale

PICO . . . Patient, Intervention, Comparison, and Outcome PONV . . . Postoperative Nausea and Vomiting

(4)

1. Introduction

1.1 Postsurgical pain and analgesia

Pain after surgery is a major cause of patient discomfort, both physical and cogni-tive-emotional [1]. According to a survey conducted in Uppsala university hos-pital (Sweden), up to a half of patients in the surgery division had suffered from severe pain during the past 24 hours, indicating inadequate pain management [2]. Another survey from Örebro (Sweden) showed that 41 % of patients who under-went orthopaedic surgery experienced moderate-to-severe pain during the early convalescence [3].

Acute postoperative pain has been shown to delay the recovery after surgery [4], and is one of the known risk factors for persistent pain [5]. In a Swedish register study, diagnoses related to persistent pain correlated with a large socio-economic burden amounting to about a tenth of the Swedish gross domestic product in 2007 [6].

The modern approach to postsurgical pain relief is multimodal, i.e. different anal-gesic agents and techniques (e.g. regional anaesthesia) are combined to gain an additive or synergistic effect. One of the central aims of multimodal analgesia is to reduce the need of opioids and avoiding their side-effects (e.g. nausea, vomit-ing, apnea, and obstipation) [7]. Concerns have been raised that morphine might promote tumour angiogenesis, as seen in studies on human breast cancer tissue in vitro and mice in vivo [8]. Furthermore, opioids might even worsen the pain through a phenomenon known as opioid-induced hyperalgesia [9].

(6)

The shoulder (glenohumeral) joint is particularly prone to injuries due to its great mobility and relatively low stability [10]. Surgery on the shoulder and upper arm is common: 22335 operations were made in Sweden in 2015 [11]. Shoul-der surgery, like other orthopaedic surgery, often causes significant postoperative pain which can result in a heavy use of opioid analgesics [12].

Brachial plexus block (BPB) is a regional anaesthetic technique commonly used in shoulder surgery. In this blockade, a single bolus or a continuous infusion of a local anaesthetic agent is injected near the brachial plexus (a bundle of cervical and thoracic nerves, mainly C5–T1). [12] A continuous infusion provides e ffec-tive analgesia but requires placing a perineural catheter which increases the risk of complications [13]. A single injection is less prone to complications but its analgesic effect is short-lived [12]. Recently, a multitude of randomised clinical trials, and systematic reviews thereof, have studied whether adding an adjuvant to a single-shot nerve block prolongs the duration of analgesia.

The choosing of ropivacaine as a focus for this review stems from a practical interest: ropivacaine is the first-line local anaesthetic used in regional nerve blocks in Tampere university hospital (Finland), where it has largely replaced previously used agents, bupivacaine and chirocaine, due to a lower cost and a more beneficial side-effect profile (Maija-Liisa Kalliomäki, personal communication, 13th April 2017). In particular, ropivacaine has less toxic effect on the heart and brain than bupivacaine [14].

1.2 Adjuvants

The spectrum of potential adjuvants is broad and comprises several mechanisms of action: clonidine and dexmedetomidine (alpha-2 adrenergic agonists), dexam-ethasone (corticosteroid), midazolam (benzodiazepine), adrenaline, sodium bicar-bonate, magnesium, opioids, ketamine, and tramadol [15].

Perineural clonidine prolongs the duration of analgesia when added to ropiva-caine, according to a meta-analysis [16]. Dexmedetomidine, being more selective than clonidine, might even be superior to clonidine as indicated by a study which

(7)

compares them as adjuvants to bupivacaine [17].

Three recent systematic reviews study dexmedetomidine mixed with various local anaesthetics in brachial plexus block. Dexmedetomidine is shown to prolong the duration of analgesia but also to cause intraoperative bradycardia and hypoten-sion as possible adverse effects. [18–20] Findings from a volunteer study on ul-nar nerve block favour perineural to intravenous dexmedetomidine as adjuvant to ropivacaine [21].

As for dexamethasone, most trials comprise local anaesthetics other than ropiva-caine. Recent reviews of these trials find the perineural dexamethasone effective [22, 23], and possibly even superior to the intravenous administration [24] (avail-able as of March 2017). However, the last-mentioned review includes both upper and lower limb surgery, and therefore the true effects in the subset of brachial plexus blocks might be different.

As for adrenaline, tramadol and magnesium, a clinical review from 2014 does not find any beneficial effect when co-administered perineurally with ropivacaine [15] .

In order to piece together the current panorama of adjuvants to ropivacaine in brachial plexus blocks, this study includes placebo-controlled trials on any sub-stance tested as an adjuvant both perineurally and intravenously in a single trial setting.

1.3 The CONSORT statement

Reporting of randomised clinical trials (RCTs) is often incomplete. Inadequate reporting can hide poor trial design and lead to biased estimates of intervention effects. This is problematic since RCTs are generally viewed as the most reliable method for evaluating interventions, and since health care recommendations and policies are based on them. [25]

The CONSORT (Consolidated Standards of Reporting Trials) statement was de-veloped as a source of guidance for authors of clinical trials as to how the quality

(8)

of reporting can be improved. The statement comprises a 25-item checklist and a flow diagram. [25]

The flow diagram shows the stepwise progress of participants through a trial, and the checklist (Table A.1) describes which items are essential to include in a trial report. The contents of the list are exemplified in an explanatory article from the authors of the statement. [25]

When trial findings are collected for a systematic review, it is important to con-sider the quality of reporting as part of a wider assessment of the quality of evi-dence for a treatment effect [25,26]. CONSORT was not intended to be used as an instrument of quality assessment [25], but attempts to apply CONSORT for such a purpose have recently been made [27, 28].

1.4 Aims of this study

This review summarises the current knowledge about those substances which have been tested as both perineural and intravenous adjuvants to ropivacaine in brachial plexus block. The primary aim is to find out

- which of the adjuvants enhance the postoperative analgesia,

- whether the perineural or intravenous administration is to be preferred, and - what adverse effects from the adjuvants have been observed.

The secondary aim is to evaluate the quality of reporting of the included trials as adherence to the CONSORT statement.

(9)

2. Methods

Study details and results were extracted and tabulated for comparison. No quan-titative meta-analysis was undertaken. Instead, a descriptive synthesis of the ma-terial was composed. The included trials were also examined for their quality of reporting. No ethical approval was necessary as this is a systematic review of existing literature.

In this review, the Population, Interventions, Comparison and Outcomes (PICO) were

(P) Adults undergoing upper limb surgery under a single-shot brachial plexus block with ropivacaine;

(I) Perineural adjuvant and intravenous placebo, and Intravenous adjuvant and perineural placebo; (C) Placebo perineurally and intravenously;

(O) Duration of analgesia, postoperative analgesic consumption, and adverse effects.

The perineural and intravenous dosing of the adjuvant was required to be equal. Trials where other local anaesthetics than ropivacaine were co-administered were excluded. Duration of sensory block was used as a surrogate measure for the duration of analgesia when the latter was not available.

(10)

2.1 Search of literature

The search involved prospective, randomised, placebo-controlled, three-arm trials which met the PICO. The following databases were used: the Cochrane Central Register of Controlled Trials (CENTRAL), Cumulative Index to Nursing and Al-lied Health Literature (CINAHL), Pubmed (including MEDLINE and PubMed Central), and Scopus. No clinical trials register was searched for unpublished trials.

A search phrase was built around keywords describing the population, interven-tion, and outcomes: ropivacaine, brachial plexus, interscalene, supraclavicular, infraclavicular, axillary, block, blockade, analgesia, duration, sensory, motor, intravenous (including abbreviations iv and i.v.) and intravenously. A combina-tion of the keywords was assumed to be found in the title or the abstract of every relevant trial, and therefore the search was restricted to those fields. The search phrase was structured by using the logical operators AND and OR (Figure 3.1). No restrictions or filters were set for publication type, date, or language.

The search results were imported to a reference manager (Mendeley), and dupli-cates were removed. Titles and abstracts of the remaining citations were manu-ally screened for inclusion. Finmanu-ally, only full-text articles in English electronicmanu-ally available via university libraries in Örebro (Sweden) or Tampere (Finland) were included. Bibliographies of the retrieved publications and recent reviews were looked through to identify additional trials. The search was completed on the 6t h of March 2017.

2.2 Quality of reporting assessment

The CONSORT 2010 checklist (Table A.1) was used as a set of criteria for ade-quate reporting. The article by Moher et al. [25] was consulted for examples and explanations.

(11)

was considered that publishers may have established formats for the summary which authors need to follow.

For every other item on the checklist, an assessment was made to decide whether the reporting in the included manuscripts was in line with the CONSORT recom-mendations, and a binary value of yes or no was recorded. The total number of positive answers per trial report was added up (with a maximum of 36 includ-ing the sub-items). This score was raised by one if a study flow diagram was presented, giving a final maximum score of 37.

(12)

3. Results

3.1 Retrieved literature

The multi-database search yielded 82 citations when duplicates were removed (Figure 3.1). After a manual screening of the titles and abstracts, ten trials met the inclusion criteria. Two of them [29, 30] could not be accessed in full-text, leaving eight trials [31–38] for a further analysis.

The included trials were conducted on four adjuvants: dexmedetomidine, dexam-ethasone, parecoxib, and ketamine. The dosage of ropivacaine and the adjuvants varied across the trials, and no two interventions were exactly similar. Basic char-acteristics of the studies were collected to Table 3.1.

Duration of analgesia was defined as the time from block induction to the first request of analgesic [32–34, 37], except for Abdallah et al. [31] who measured the time to first report of pain at the surgical site. Discomfort from the site of incision was reported by Lee et al. [35] and included in the table as a surrogate measure for the duration on analgesia. Likewise, two other trials [36, 38] only measured the duration of sensory block which was used as a substitute for the duration of analgesia.

(13)

Ropivacaine intravenous OR intravenously OR iv OR i.v. brachial plexus OR interscalene OR supraclavicular OR infraclavicular OR axillary sensory OR motor OR duration

AND block ORblockade OR analgesia AND 132 records: CENTRAL=50 CINAHL=3 PubMed=25 Scopus= 54 0 records identified through other sources 82 records after removing duplicates 82 records screened 72 records excluded: 18 reviews

11 compare local anaesth. 9 only perineural adjuvant 2 only intravenous adjuvant 1 oral adjuvant

5 no adjuvant

2 no non-active placebo 6 compares different anaesthetic techniques 5 no brachial plexus block 4 case reports

2 conference abstracts 1 veterinary study 6 other non-RCT studies

10 articles assessed

8 trials included

2 records excluded: 2 full-text not available

SEARCH PHRASE PRISMA FLOW DIAGRAM

Figure 3.1: Search phrase was built around keywords from the study population, interventions and outcomes, and structured using the logical operators AND and OR. Search results were merged together and duplicates removed. After a manual screening, eight articles were included for review (adapted from PRISMA flow

(14)

Table 3.1: Details of the included trials concerning the study populations (type of surgery and nerve block), interventions (trial arms) and outcomes were extracted for comparison. First author (publ. year) Nr of patients Type of surgery GA Type of brachial plexus block Amount of ropivacaine Trial arms

Outcome, primary Outcome, secondary

Abdallah (2016)

99 Arthroscopic shoulder surgery

Yes Interscalene Ropivacaine 0.5 % (15 ml/75 mg) (1) 0.5 ug/kg dexmedetomidine (2) 0.5 ug/kg IV dexmedetomidine (3) saline (1) Duration of analgesia (2) Cumulative 24-h analgesic consumption (1) Motor blockade (2) Incidence of bradycardia (3) Incidence of hypotension (4) Incidence of postop. neurologic symptoms Desmet (2013) 144 Arthroscopic shoulder surgery

Yes Interscalene Ropivacaine 0.5 % (30 ml/150 mg) (1) 10 mg dexamethasone (2) 10 mg IV dexamethasone (3) saline

Duration of analgesia (1) Pain scores (2) MBS (3) Analgesic need (4) Sleep disturbance (5) Overall satisfaction Kathuria (2015) 60 Upper limb surgery No Supraclavicular Ropivacaine 0.5 % (30 ml/150 mg) (1) 50 ug dexmedetomidine (2) 50 ug IV dexmedetomidine (3) saline

* Sensory and motor block - onset-time

- duration

* Duration of analgesia

* PONV, skin rash, tachycardia, bradycardia, hypotension, hypertension, hypoxemia, sedation, any other side effect Kawanishi (2014) 34 Arthroscopic shoulder surgery No Interscalene Ropivacaine 0.75 % (20 ml/150 mg) (1) 4 mg dexamethasone (2) 4 mg IV dexamethasone (3) none reported

Duration of analgesia (1) NRS morning after surgery (2) Analgesic need (3) Sleep disturbance (4) Overall satisfaction Lee (2002) 51 Forearm and hand surgery No Interscalene Ropivacaine 0.5 % (30 ml/150 mg) (1) 30 mg ketamine (2) 30 mg IV ketamine (3) saline

* Sensory and motor block - onset-time - duration * Adverse effects Liu (2013) 135 Forearm surgery No Axillary Ropivacaine 0.25 % (40 ml/100 mg) (1) 20 mg parecoxib (2) 20 mg IV parecoxib (3) saline

* Sensory and motor block - duration

* Most severe pain score during 24-h postoperative period Rosenfeld (2016) 120 Shoulder surgery

Yes Interscalene Ropivacaine 0.5 % (28 ml/140 mg) (1) 8 mg dexamethasone (2) 8 mg IV dexamethasone (3) saline

Sensory block

- duration

(1) Cumul. 24-h opioid cons. (2) Duration of analgesia (3) Pain score (VAS) (4) Nausea

(5) Admin. of anti-emetic (24-h) (6) Patient satisfaction (7) Adverse event (first 7 days) Sakae

(2017)

60 Arthroscopic shoulder

Yes Interscalene Ropivacaine 0.75 % (20 ml/150 mg) (1) 4 mg dexamethasone

Sensory block

- onset-time

(1) Pain score (VAS) (2) Postoperative nausea and

(15)

3.2 E

ffects of the adjuvants

Both intravenous and perineural dexmedetomidine were found to prolong the du-ration of analgesia, and reduce the cumulative analgesic consumption during the first 24 hours after surgery (Table 3.2). The results for the two routes of adminis-tration did not differ significantly from each other. Abdallah et al. found no dif-ference in the incidence of prespecified adverse effects (bradycardia, hypotension, neurologic symptoms), while Kathuria et al. reported sporadic cases of bradycar-dia, hypotension, and skin rash among patients in the intervention groups. [31,33] The results on dexamethasone diverged from one trial to another. In the study of Desmet et al. the duration of analgesia was prolonged, and the consumption of analgesics was reduced, regardless of the route of administration. As an adverse effect, participants receiving dexamethasone (perineurally or intravenously) had higher blood glucose levels postoperatively than those on placebo. In contrast, the placebo group had a higher incidence of sleep disturbances due to pain during the first night after surgery (a finding also made by Kawanishi et al. [34]). [32] Kawanishi et al. found no significant change in the consumption of analgesics, and observed a prolonged duration of analgesia only after perineural dexametha-sone [34]. Further, Rosenfeld and colleagues saw no significant changes in the duration of analgesia or the incidence of adverse effects but found a reduction in the cumulative consumption of opioids after both administration routes [37]. Sakae et al. found dexamethasone to prolong the duration of the sensory block. This duration was also the longest in absolute terms observed among these trials. Sakae et al. did not measure the cumulative use of analgetics but reported ratios of participants who needed versus did not need an opioid within 24 hours post-operatively, and found a significant difference between the perineural and control groups. [38]

Perineural, but not intravenous, parecoxib was found to prolong the duration of sensory block. The authors reported no evident adverse effects (e.g. haematoma formation). [36]

(16)

Table 3.2: Effects of the adjuvants on the duration of analgesia and the need of analgesics during early convalescence were extracted for comparison. Quality of trial reporting was evaluated as the number of adequately reported CONSORT checklist items, and the presence of a study flow diagram.

Author, first (year)

Adjuvant

Duration of analgesia Need for analgesics during early convalescence

CONSORT total score

(of max 37)

Time (h),

error bar (SD or IQR)

Interventions with significant prolongation As proportion to consumption in control group Interventions with significant reductionº

PN IV C PN IV C Significant Cumulative analgesic

consumption @ 24 h Significant CONSORT, flowchart Desmet (2013) Dexamethasone Both Both (oral paracetamol iv diclofenac) 25 23.4 21.3 12.6 0.6842105263157890.6052631578947371 PN and IV Peri: 2.6 / 59 IV: 2.3 / 56 C: 3.8 / 101 Mean paracetamol (g) / diclofenac i.v. (mg) @ 48 h PN and IV yes Kawanishi (2014) Perineural Neither (loxoprofen) 20 18 14 11.2 59/101 56/101 1 PN Peri: 12.5 IV: 17.5 C: 27.0 Mean loxoprofen (g) @ 20 h None no Rosenfeld (2016) Neither Both (iv morphine equivalent) 22 8.4 9.2 8.4 0.4629629629629630.6481481481481481 None Peri: 12.2 (9.3) IV: 17.1 (15.9) C: 24.1 (14.3)

i.v. morphine equivalent (mg)

PN and IV yes Sakae (2017) Perineural* - - 19 38.7 27.4 28.8 0.5062240663900410.709543568464731 PN - - no Abdallah (2016) Dexmedetomidine Both Both (oral morphine equivalent) 33 10.9 9.8 6.7 0.780219780219780.8083028083028081 PN and IV Peri: 63.9 (58.8 to 69.0) IV: 66.2 (60.6 to 71.8) C: 81.9 (75.0 to 88.9)

Oral morphine equivalent (mg)

PN and IV yes

Kathuria

(2015) Both

Both

(im diclofenac) 13 16.1 16.2 8.9 0.4691666666666670.5 1 PN and IV

Peri: 56.3 (33.3) IV: 60.0 (39.2) C: 120.0 (56.6) Diclofenac i.m. (mg) PN and IV no Lee

(2002) Ketamine Neither* - - 16 9.7 10.7 9.9 None - - no

Liu

(2013) Parecoxib Perineural* - - 22 9.1 7.6 7.3 PN - - yes

PN IV C 0 8.75 17.5 26.25 35 PN IV C 0 6.25 12.5 18.75 25 PN IV C 0 3.75 7.5 11.25 15 PN IV C 0 13.75 27.5 41.25 55 PN IV C 0 5.5 11 16.5 22 PN IV C 0 3 6 9 12 PN IV C 0 4 8 12 16 PN IV C 0 3 6 9 12 PN IV C 0 0.25 0.5 0.75 1 PN IV C 0 0.25 0.5 0.75 1 PN IV C 0 0.25 0.5 0.75 1 PN IV C 0 0.25 0.5 0.75 1 PN IV C 0 0.25 0.5 0.75 1

PN, perineurial. IV, intravenous. C, control. SD, standard deviation. IQR, interquartile range. * Duration of sensory block as surrogate for duration of analgesia

º In parenthesis : Analgesic agent or morphine equivalent as reported in the manuscript

(17)

Ketamine did not significantly extend the time to discomfort from the site of surgery, but adverse effects (mostly from the central nervous system, e.g. unpleas-ant feeling, hallucinations, bad dreams, and drowsiness) occured with a higher incidence after perineural (44 %), and intravenous (94 %) administration of ke-tamine than in the placebo group (with only one case of headache) [35].

Non-inferiority or superiority of one route of administration to the other was tested in two trials. Abdallah et al. showed that intravenous dexmedetomidine was non-inferior to perineural. Sakae et al. found perineural dexamethasone to be superior to intravenous in a post-hoc analysis.

3.3 Quality of reporting

The trial reports were assessed using the CONSORT 2010 checklist (Figure A.1). The total scores inclusive of the presence of a study flow diagram are shown in Table 3.2. In the text below, keywords from the cited items are highlighted to improve the legibility.

The publication from Abdallah et al. [31] showed the most complete adherence to the CONSORT guidelines. This was the only manuscript to describe where the trial took place (item 4b), to discuss the generalisability of the findings (item 21), and to report from where the full trial protocol is available (item 24). The manu-script was checked against the protocol (available through www.clinicaltrials.gov), and no important changes to methods (item 3b) or outcomes (item 6b) were iden-tified. In addition, their subgroup analysis was prespecified as recommended by the guidelines (item 18).

Seven items (2a, 2b, 3a, 5, 15, 19, and 22) were rated as adequately reported in all of the included trials. In contrast, none of the included trials reported com-prehensively on who generated the allocation sequence, enrolled participants, and assigned them to the interventions (item 10).

Besides Abdallah et al., only two trials [32, 38] declared the registration number and the site (item 23) in their reports, but the register site of [32] was not available

(18)

in English nor freely accessible.

Interimanalyses (item 7b) were not mentioned in any of the articles. Additional post-hoc analyses (item 12b) were reported in two trials [33,38] but without giving any reason or explanation. Item 17b was irrelevant in this set of trials as none of the primary outcomes was binary. The rest of the checklist was covered in sufficient detail by a median of six trials per item. A flow diagram was included in four manuscripts [31, 32, 36, 37].

(19)

4. Discussion

This systematic review was undertaken to find out which perineurally and intra-venously injectable adjuvants to ropivacaine enhance the postoperative analgesia from the brachial plexus block, and which of the routes of administration should be preferred. The findings were put into perspective with possible adverse ef-fects. Further, the quality of reporting of the included trials was assessed using the CONSORT statement.

Trials on four adjuvants were identified. Ketamine did not enhance the analge-sia but was reported to cause unpleasant adverse effects. Perineural parecoxib prolonged the sensory block without any evident adverse effects. Dexmedetomi-dine, both perineural and intravenous, improved the postoperative analgesia but sporadic cases of intraoperative bradycardia and hypotension were reported. Re-sults from dexamethasone were inconsistent. The quality of reporting was mostly suboptimal.

The findings concerning ketamine’s adverse effects are not surprising. Ketamine is an effective anaesthetic, but unpleasant psychotomimetic effects are known to fol-low when ketamine enters the systemic bloodstream [39]. In order to reduce these systemic effects, it seems logical to experiment ketamine perineurally. Indeed, at least three trials [30,35,40] have studied perineural ketamine in the brachial plexus block. However, all these trials report an increased incidence of adverse effects without any analgesic advantage, rendering the use of ketamine as an adjuvant unpromising.

A PubMed search on cyclooxygenase (COX) inhibitors in nerve blocks suggests that Liu et al. [36] are unique in studying parecoxib (or any other COX inhibitor)

(20)

as a perineural adjuvant in peripheral nerve blocks. Previously, an intravenous infusion of parecoxib is shown to reduce spinal hyperalgesia in humans [41] and neuraxial administration to be safe in a rat model [42]. The absence of further adjuvant studies on COX inhibitors might nevertheless be due to safety concerns. Our findings concerning effects of perineural dexmedetomidine are in line with re-sults from several recent meta-analyses [18–20]. Studies on intravenous dexmede-tomidine as adjuvant are few and mostly conducted in the setting of spinal anaes-thesia with local anaesthetics other than ropivacaine [43]. Several adequately re-ported trials are needed before an evidence-based recommendation about the use of dexmedetomidine as an adjuvant to ropivacaine can be made.

As for dexamethasone, the varying results presented in our review cannot be ex-plained by differences in the type of surgery, the site of injection or the amount of ropivacaine administered, since they are reported similar in the included trials. It has previously been hypothesised that a high intravenous dose of dexamethasone (e.g. 8 mg or higher) might directly impact peripheral nerves, thereby narrowing the difference in analgesia from intravenous versus perineural administration [44]. This dose-response relationship might explain why, in our study, the highest dose of dexamethasone (10 mg) performed well both perineurally and intravenously while 4 mg only perineurally. However, the negative result on analgesia reported by Rosenfeld et al. [37] contradicts this hypothesis.

The CONSORT total score ranged from 13 to 33 (of max 37), with a mean of 21.3 (57 %). Our result lies between those of two previous CONSORT-based quality assessments: A review [27] of fifteen trials, which study postoperative outcomes after a total knee arthroplasty, shows a mean total score of 79 % (17 of 23 items on the CONSORT 2001 checklist). Another more recent review [28] on 120 surgical RCTs reports a mean overall score of 44 % (13 of 29 items on a modified 2010 checklist).

Seven of the 8 trials in our review were published in 2013 or later. Hence the newest CONSORT guidelines have most likely been available for the authors by the time of writing the manuscripts. The relatively low adherence to the guidelines indicates a need for further propagation of the CONSORT statement. Inadequate

(21)

reporting does not automatically mean that the results are biased but makes it difficult for the reader to properly assess the validity of the findings [25].

Our study comes with limitations: The CONSORT statement is not designed to be used as a tool for quality assessments, but instead, as guidance for authors of trial reports [25]. An instrument specifically developed for evaluating the quality of evidence, including an assessment of reporting bias, is GRADE (Grading of Recommendations Assessment, Development and Evaluation) [26]. The GRADE approach is outcome-centered and requires a quantitative meta-analysis as a ba-sis [26], which is why it could not be chosen for this review. Our analyba-sis of the adjuvants is not likely to have benefitted from a comprehensive quantitative assessment as the number of identified trials on any particular adjuvant was so small.

In our CONSORT-based assessment, judging in a dichotomous manner whether the recommendations were followed was not straightforward. The challenge lay in deciding what level of reporting should be rated as adequate since no clear criteria exist. The inevitable subjectivity of our assessment may have introduced a bias. We are also aware that our phrasing of the research question, which requires both perineural and intravenous administration in the same trial setting, excludes ad-juvants which are not suitable for one or the other route of administration. This framing was, however, deliberate since the different routes of administration may give additional information about whether the effect is mostly central or periph-eral. Unfortunately, no such conclusion can be made based on the results from the currently available trials.

Our study is meant to be part of an on-going effort to improve postoperative pain management by enhancing an existing analgesic technique. The current knowl-edge about the efficacy and safety of adjuvants to ropivacaine is clearly insufficient to allow for a clinician to make an evidence-based choice. Even negative findings, such as those on ketamine, are important since they can lead to do not do recom-mendations which are nowadays provided and promoted by health care institutes in e.g. the United Kingdom [45] and Finland [46].

(22)

Conclusions

In conclusion, both the perineural and intravenous dexmedetomidine seem equally promising adjuvants to ropivacaine in brachial plexus block, but the clinician should be vigilant about the risk of intraoperative bradycardia and hypotension. Our finding of the suboptimal adherence to the CONSORT guidelines calls for attention of authors of future trial reports.

(23)

5. Bibliography

[1] Parsons B, Schaefer C, Mann R, Sadosky A, Daniel S, Nalamachu S, et al. Economic and humanistic burden of post-trauma and post-surgical neuro-pathic pain among adults in the United States. Journal of Pain Research. 2013;6:459–469.

[2] Wadensten B, Fröjd C, Swenne CL, Gordh T, Gunningberg L. Why is pain still not being assessed adequately? Results of a pain prevalence study in a university hospital in Sweden. Journal of Clinical Nursing. 2011;20(5-6):624–634.

[3] Rawal N, Hylander J, Nydahl PA, Olofsson I, Gupta A. Survey of post-operative analgesia following ambulatory surgery. Acta Anaesthesiologica Scandinavica. 1997;41:1017–1022.

[4] Pavlin DJ, Chen C, Penaloza Da, Polissar NL, Buckley FP. Pain as a factor complicating recovery and discharge after ambulatory surgery. Anesthesia and analgesia. 2002;95(3):627–634, table of contents.

[5] Kehlet H, Jensen TS, Woolf CJ. Persistent postsurgical pain: risk factors and prevention. Lancet. 2006 may;367(9522):1618–25.

[6] Gustavsson A, Bjorkman J, Ljungcrantz C, Rhodin A, Rivano-Fischer M, Sjolund KF, et al. Socio-economic burden of patients with a diagnosis re-lated to chronic pain - Register data of 840,000 Swedish patients. European Journal of Pain. 2012;6:289–299.

[7] Wu CL, Raja SN. Treatment of acute postoperative pain. Lancet. 2011 jun;377(9784):2215–25.

(24)

[8] Bimonte S, Barbieri A, Rea D, Palma G, Luciano A, Cuomo A, et al. Mor-phine Promotes Tumor Angiogenesis and Increases Breast Cancer Progres-sion. BioMed Research International. 2014;2015.

[9] Angst MS, Clark JD. Opioid-induced Hyperalgesia. Anesthesiology. 2006;104(3):570–587.

[10] Paulsen F, Waschke J, editors. Sobotta Atlas of Human Anatomy: Volume 3, 15th ed., English/Latin. 15th ed. Munich: Elsevier GmbH; 2011.

[11] Socialstyrelsen. Statistik för operationer i öppen vård (dagkirurgi) [Internet]; 2017. [Accessed 28th March 2017]. Available from: www.socialstyrelsen.se/statistik/statistikdatabas/

operationerioppenvarddagkirurgi.

[12] Julien RE, Williams Ba. Regional anesthesia procedures for outpatient shoulder surgery. International anesthesiology clinics. 2005;43(3):167–75. [13] Aguirre J, Del Moral A, Cobo I, Borgeat A, Blumenthal S. The role of

continuous peripheral nerve blocks. Anesthesiology Research and Practice. 2012;2012.

[14] Simpson D, Curran MP, Oldfield V, Keating GM. Ropivacaine: a review of its use in regional anaesthesia and acute pain management. Drugs. 2005;65(18):2675–717.

[15] Bailard NSNS, Ortiz J, Flores RARA. Additives to local anesthetics for pe-ripheral nerve blocks: Evidence, limitations, and recommendations. Ameri-can Journal of Health-System Pharmacy. 2014;71(5):373–385.

[16] Pöpping DM, Elia N, Marret E, Wenk M, Tramer MR. Clonidine as an Ad-juvant to Local Anesthetics for Peripheral Nerve and Plexus Blocks. Anes-thesiology. 2009;111(2):406–415.

[17] Swami SS, Keniya VM, Ladi SD, Rao R. Comparison of dexmedetomidine and clonidine (alpha2 agonist drugs) as an adjuvant to local anaesthesia in supraclavicular brachial plexus block: A randomised double-blind prospec-tive study. Indian Journal of Anaesthesia. 2012;56(3):243–249.

(25)

[18] Ping Y, Ye Q, Wang W, Ye P, You Z. Dexmedetomidine as an adjuvant to local anesthetics in brachial plexus blocks A meta-analysis of randomized controlled trials. Medicine. 2017;.

[19] Vorobeichik L, Brull R, Abdallah FW. Evidence basis for using perineural dexmedetomidine to enhance the quality of brachial plexus nerve blocks: a systematic review and meta-analysis of randomized controlled trials. British Journal of Anaesthesia. 2017;118(2):167–181.

[20] Hussain N, Grzywacz VP, Ferreri CA, Atrey A, Banfield L, Shaparin N, et al. Investigating the Efficacy of Dexmedetomidine as an Adjuvant to Local Anesthesia in Brachial Plexus Block. Regional Anesthesia and Pain Medicine. 2017;42(2):1.

[21] Marhofer D, Kettner SC, Marhofer P, Pils S, Weber M, Zeitlinger M. Dexmedetomidine as an adjuvant to ropivacaine prolongs peripheral nerve block: A volunteer study. British Journal of Anaesthesia. 2013;110(3):438– 442.

[22] Choi S, Rodseth R, McCartney CJL, Hardman JG. Effects of dexametha-sone as a local anaesthetic adjuvant for brachial plexus block: A systematic review and meta-analysis of randomized trials. British Journal of Anaesthe-sia. 2014;112(3):427–439.

[23] Knezevic NN, Anantamongkol U, Candido KD. Perineural dexamethasone added to local anesthesia for brachial plexus block improves pain but delays block onset and motor blockade recovery. Pain physician. 2015;18(1):1–14. [24] Chong MA, Berbenetz NM, Lin C, Singh S. Perineural Versus Intravenous Dexamethasone as an Adjuvant for Peripheral Nerve Blocks. Regional Anes-thesia and Pain Medicine. 2017;42(3):1.

[25] Moher D, Hopewell S, Schulz KF, Montori V, Gøtzsche PC, Devereaux PJ, et al. CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. BMJ (Clinical research ed). 2010 jan;340:c869.

(26)

[26] Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck-Ytter Y, Schünemann HJ. What is "quality of evidence" and why is it important to clinicians? BMJ (Clinical research ed). 2008 may;336(7651):995–8.

[27] Crowley C, Dowsey MM, Quinn C, Barrington M, Choong PFM. Impact of regional and local anaesthetics on length of stay in knee arthroplasty. ANZ Journal of Surgery. 2012;82(4):207–214.

[28] Yu J, Li X, Li Y, Sun X. Quality of reporting in surgical randomized clinical trials. British Journal of Surgery. 2017 feb;104(3):296–303.

[29] Aref AAEAM. Effects of adding tramadol to 0.5% ropivacaine for interscalene brachial plexus block. Egyptian Journal of Anaesthesia. 2006;22(2):140–143.

[30] Kim GN, Lee IO, Kong MH, Lee MK, Lim SH, Choi YS. The Effect of Ke-tamine Added to a Ropivacaine Supuraclavicular Brachial Plexus Blockade. Korean Journal of Anesthesiology. 2004;46(1):46–50.

[31] Abdallah FW, Dwyer T, Chan VWS, Niazi AU, Ogilvie-Harris DJ, Oldfield S, et al. IV and Perineural Dexmedetomidine Similarly Prolong the Dura-tion of Analgesia after Interscalene Brachial Plexus Block: A Randomized, Three-arm, Triple-masked, Placebo-controlled Trial. Anesthesiology. 2016 mar;124(3):683–695.

[32] Desmet M, Braems H, Reynvoet M, Plasschaert S, Van Cauwelaert J, Pottel H, et al. I.V. and perineural dexamethasone are equivalent in increasing the analgesic duration of a single-shot interscalene block with ropivacaine for shoulder surgery: A prospective, randomized, placebo-controlled study. British Journal of Anaesthesia. 2013 sep;111(3):445–452.

[33] Kathuria S, Gupta S, Dhawan I. Dexmedetomidine as an adjuvant to ropiva-caine in supraclavicular brachial plexus block. Saudi Journal of Anaesthesia. 2015;9(2):148.

[34] Kawanishi R, Yamamoto K, Tobetto Y, Nomura K, Kato M, Go R, et al. Perineural but not systemic low-dose dexamethasone prolongs the duration

(27)

of interscalene block with ropivacaine: a prospective randomized trial. Local and regional anesthesia. 2014;7:5–9.

[35] Lee IO, Kim WK, Kong MH, Lee MK, Kim NS, Choi YS, et al. No en-hancement of sensory and motor blockade by ketamine added to ropivacaine interscalene brachial plexus blockade. Acta anaesthesiologica Scandinavica. 2002 aug;46(7):821–826.

[36] Liu X, Zhao X, Lou J, Wang Y, Shen X. Parecoxib added to ropivacaine prolongs duration of axillary brachial plexus blockade and relieves post-operative pain general. Clinical Orthopaedics and Related Research. 2013 feb;471(2):562–568.

[37] Rosenfeld DM, Ivancic MG, Hattrup SJ, Renfree KJ, Watkins AR, Hentz JG, et al. Perineural versus intravenous dexamethasone as adjuncts to local anaesthetic brachial plexus block for shoulder surgery. Anaesthesia. 2016 apr;71(4):380–388.

[38] Sakae TM, Marchioro P, Schuelter-Trevisol F, Trevisol DJ. Dexamethasone as a ropivacaine adjuvant for ultrasound-guided interscalene brachial plexus block: A randomized, double-blinded clinical trial. Journal of Clinical Anes-thesia. 2017;38.

[39] Durrani Z, Winnie AP, Zsigmond EK, Burnett ML. Ketamine for intravenous regional anesthesia. Anesthesia & Analgesia. 1989;68(3):328–332.

[40] Senel AC, Ukinc O, Timurkaynak A. Does the addition of tramadol and ketamine to ropivacaine prolong the axillary brachial plexus block? BioMed Research International. 2014;2014.

[41] Koppert W, Wehrfritz A, Körber N, Sittl R, Albrecht S, Schüttler J, et al. The cyclooxygenase isozyme inhibitors parecoxib and paracetamol reduce central hyperalgesia in humans. Pain. 2004;108(1-2):148–153.

[42] Kim YH, Lee PB, Park J, Lim YJ, Kim YC, Lee SC, et al. The neurological safety of epidural parecoxib in rats. NeuroToxicology. 2011;32(6):864–870.

(28)

[43] Abdallah FW, Abrishami A, Brull R. The facilitatory effects of intravenous dexmedetomidine on the duration of spinal anesthesia: A systematic review and meta-analysis. Anesthesia and Analgesia. 2013;117(1):271–278. [44] Bolin ED, Wilson S. Perineural Versus Systemic Dexamethasone. Regional

Anesthesia and Pain Medicine. 2015;40(4):393–394.

[45] National Institute for Health and Care Excellence. NICE Guidance: Savings and productivity [Internet]; 2017. [Accessed 12th April 2017]. Available from: www.nice.org.uk/savingsandproductivity/collection. [46] The Finnish Medical Society Duodecim. Vältä viisaasti -suositukset

[Inter-net]; 2017. [Accessed 12th April 2017]. Available from: www.kaypahoito. fi/web/kh/valta-viisaasti-suositukset.

[47] Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ (Clin-ical research ed). 2009 jan;339(jul21_1):b2535.

(29)

A. Appendix

Table A.1: The CONSORT 2010 checklist of topics and items to be included in a clinical trial report according to the CONSORT statement. (Adapted from Moher et al. [25])

Topic Checklist item

Title and abstract

1a Identification as a randomised trial in the title

1b Structured summary of trial design, methods, results, and conclu-sions

Introduction

Background and objectives 2a Scientific background and explanation of rationale 2b Specific objectives or hypotheses

Methods

Trial design 3a Description of trial design (such as parallel, factorial) including allocation ratio

3b Important changes to methods after trial commencement (such as eligibility criteria), with reasons

Participants 4a Eligibility criteria for participants

4b Settings and locations where the data were collected

Interventions 5 The interventions for each group with sufficient details to allow replication, including how and when they were actually adminis-tered

Outcomes 6a Completely defined pre-specified primary and secondary outcome

measures, including how and when they were assessed

6b Any changes to trial outcomes after the trial commenced, with rea-sons

(30)

Table A.1: Continued

7b When applicable, explanation of any interim analyses and stopping guidelines

Randomisation:

Sequence generation 8a Method used to generate the random allocation sequence

8b Type of randomisation; details of any restriction (such as blocking and block size)

Allocation concealment

mechanism

9 Mechanism used to implement the random allocation sequence (such as sequentially numbered containers), describing any steps taken to conceal the sequence until interventions were assigned Implementation 10 Who generated the random allocation sequence, who enrolled

par-ticipants, and who assigned participants to interventions

Blinding 11a If done, who was blinded after assignment to interventions (for

example, participants, care providers, those assessing outcomes) and how

11b If relevant, description of the similarity of interventions

Statistical methods 12a Statistical methods used to compare groups for primary and sec-ondary outcomes

12b Methods for additional analyses, such as subgroup analyses and adjusted analyses

Results

Participant flow 13a For each group, the numbers of participants who were randomly assigned, received intended treatment, and were analysed for the primary outcome

13b For each group, losses and exclusions after randomisation, together with reasons

Recruitment 14a Dates defining the periods of recruitment and follow-up

14b Why the trial ended or was stopped

Baseline data 15 A table showing baseline demographic and clinical characteristics for each group

Numbers analysed 16 For each group, number of participants (denominator) included in each analysis and whether the analysis was by original assigned groups

Outcomes and estimation 17a For each primary and secondary outcome, results for each group, and the estimated effect size and its precision (such as 95 % confi-dence interval)

17b For binary outcomes, presentation of both absolute and relative ef-fect sizes is recommended

(31)

Table A.1: Continued

Ancillary analyses 18 Results of any other analyses performed, including subgroup anal-yses and adjusted analanal-yses, distinguishing pre-specified from ex-ploratory

Harms 19 All important harms or unintended effects in each group (for

spe-cific guidance see CONSORT for harms) Discussion

Limitations 20 Trial limitations, addressing sources of potential bias, imprecision, and, if relevant, multiplicity of analyses

Generalisability 21 Generalisability (external validity, applicability) of the trial find-ings

Interpretation 22 Interpretation consistent with results, balancing benefits and

harms, and considering other relevant evidence Other information

Registration 23 Registration number and name of trial registry

Protocol 24 Where the full trial protocol can be accessed, if available

Funding 25 Sources of funding and other support (such as supply of drugs),

Perineural versus intravenous administration ofadjuvants to ropivacaine in brachial plexus blockSystematic review and quality of reporting assessment