Beruflich Dokumente
Kultur Dokumente
Original Contribution
a r t i c l e i n f o a b s t r a c t
Article history: Study objective and background: The role of the programmed intermittent bolus (PIB) technique for infusion of
Received 29 July 2017 local anesthetics in continuous peripheral nerve blockade (CPNB) remains to be elucidated. Randomized
Received in revised form 13 August 2017 controlled trials (RCTs) on PIB versus continuous infusion for CPNB have demonstrated conicting results and
Accepted 15 August 2017 no systematic review or meta-analysis currently exists. We aimed to delineate via systematic review with
Available online xxxx
meta-analysis if there is any analgesic benet to performing PIB versus continuous infusion for CPNB.
Design: We conducted a systematic review and random-effects meta-analysis of RCTs.
Keywords:
Anesthetics, local
Data sources: We searched Medline, Embase, and the Cochrane Library without language restriction from
Regional anesthesia inception to 2-May-2017.
Eligibility criteria: Included RCTs had to compare PIB to continuous infusion in adult surgical patients receiving
any upper or lower limb CPNB for postoperative analgesia. VAS pain scores were the primary outcome. The
Cochrane Risk of Bias Tool with GRADE methodology was utilized to assess evidence quality.
Results: Nine RCTs (448 patients) met the inclusion criteria. Two studies performed upper limb blocks and
the rest lower limb blocks. Five RCTs activated the CPNB with long-acting local anesthetic and only ve
used multi-modal analgesia. PIB modestly reduced VAS pain scores at 6 h ( 14.2 mm; 95%CI 23.5 mm to
5.0 mm; I2 = 82.5%; p = 0.003) and 12 h ( 9.9 mm; 95%CI 14.4 mm to 5.4 mm; I2 = 12.4%; p b
0.001), but not at later time points. There were no other meaningful differences in the rest of the outcomes,
apart from more residual motor block with PIB (OR 4.27; 95% CI 1.0816.9; p = 0.04; NNTH = 8). GRADE scoring
ranged from low to very low.
Conclusions: The existing evidence demonstrates that PIB does not meaningfully reduce VAS pain scores in CPNB.
This systematic review provides important information about the limitations of existing studies. Future studies
should reect contemporary practice and focus on more painful operations.
2017 Elsevier Inc. All rights reserved.
1. Introduction
http://dx.doi.org/10.1016/j.jclinane.2017.08.018
0952-8180/ 2017 Elsevier Inc. All rights reserved.
70 M.A. Chong et al. / Journal of Clinical Anesthesia 42 (2017) 6976
continuous infusion in the setting of continuous peripheral nerve block- 2.3. Article screening and data extraction
ade (CPNB) for postoperative analgesia [4,5]. The elucidation of whether
or not PIB is of benet is important: on one hand, PIB may result in bet- All titles, abstracts, and full texts (where required to assess the study
ter patient outcomes, as suggested by the benets in the labor analgesia for inclusion) were reviewed in duplicate by MC and CL. Any disagree-
literature; on the other hand, PIB requires specialized pumps that are ments were resolved by consensus with SD. Data from included studies
capable of giving the automated boluses and, therefore, may be costly were extracted independently onto standardized forms by MC, YW, or
to adopt [6]. Finally, we are not aware of any existing systematic review CL. Extracted data included important baseline demographic informa-
and meta-analysis that synthesizes the evidence for PIB in regional tion of each study, information regarding assessment of the risk of
anesthesia. Therefore, given this gap in the literature and clinical equi- bias of the study, and the pre-specied outcomes. To facilitate meta-
poise, we designed and conducted a systematic review and meta-anal- analysis, medians, IQR, and range values were approximated into
ysis of RCTs to elucidate the effect of PIB on patient centered outcomes means and their corresponding standard deviation using methods
in CPNB for postoperative analgesia. Based on the benets in the labor suggested by the Cochrane Library [8]. Where necessary (e.g. data
analgesia literature, we hypothesized that PIB would result in better values not reported in text and only within graphs), numerical data
analgesia with less local anesthetic and rescue opioid consumption. were extracted from graphs by digital measurement. We attempted to
contact principal investigators of included studies for additional infor-
mation, where necessary.
2. Methods
2.4. Risk of bias evaluation
This systematic review and meta-analysis complies with the PRISMA
statement [7]. Our institutional research ethics board does not require The Cochrane Risk of Bias Tool was utilized to appraise each included
approval for systematic reviews and meta-analyses, as no data is being study's risk of bias by MC and CL and all discrepancies were resolved by
collected from patients. The clinical question, study inclusion criteria, consensus [9]. We considered studies to be at low risk of bias if they
outcomes, and analysis plan were dened a priori. scored 3 or higher on these criteria: (1) appropriately generated
the randomization sequence, (2) appropriate allocation concealment,
2.1. Literature search (3) blinded study personnel and participants, and (4) blinded outcome
assessors, and (5) reported data completely. Furthermore, the study had
Ovid Medline, Embase, and the Cochrane Library were searched to demonstrate no signicant selective reporting bias or other source of
from inception to 2 May 2017 without language restriction for RCTs bias [9]. Finally, GRADE methodology was utilized to provide an overall
meeting the following inclusion criteria. appraisal of the quality of evidence underlying each outcome [10].
Fig. 1. PRISMA owsheet describing the study selection process. Three databases (MEDLINE, EMBASE, and Cochrane) were searched for relevant articles and screened against our inclusion
criteria.
were relied upon to administer the intermittent boluses (e.g. the inter- patient-controlled analgesia [PCA]) [11,1518]. Finally, only 5 studies
vention boluses in that study were not programmed) [13]. reported administration of multi-modal analgesia [4,5,1517]. We con-
sidered 6 of 9 studies to be at low risk of bias (Fig. 2, Risk of Bias Assess-
3.2. Study characteristics ment) [4,5,11,1517].
All studies could be pooled for meta-analysis except for one, where
The baseline characteristics of the included studies are listed in Table the authors did not report the number of patients assigned to each
1 and outcomes reported by each study are shown in a supplementary arm [14]. This small study of 55 patients scheduled for hallux valgus sur-
table (Supplemental Digital Content 2). For block placement, ultrasound gery found marginally decreased local anesthetic consumption favoring
guidance [5,1416] and neurostimulation [4,11,17,18] were used in 4 PIB, but otherwise no other differences in pain scores, patient satisfac-
studies each, and placement technique was not reported by the last tion, or side effects [14].
study [19]. Seven studies activated the nerve block with a bolus dose
of local anesthetic: two studies used mepivacaine [4,17] and 5 utilized 3.3. Primary outcome: VAS pain scores
a long-acting local anesthetic (Table 1) [5,11,15,16,18]. In two studies,
the block activation solution was not reported [14,19]. All studies uti- Patients receiving PIB had lower pain scores in the early postopera-
lized ropivacaine or levobupivacaine for the maintenance dosing of tive period (Fig. 3). Notably, the p-value threshold for this analysis
the CPNB. Certain studies focused on less painful surgeriessuch as hal- (given the repeated comparisons) was set via Bonferroni correction at
lux valgus surgerywith a full listing of details available in Table 1 [4,14, p b 0.007 (initial = 0.05 divided by 7 comparisons). Specically, PIB
17]. Interestingly, only 4 studies employed PCRA boluses [4,5,14,19], patients experienced less pain at 6 h ( 14.2 mm; 95% CI 23.5 mm
with the rest utilizing rescue analgesia via other means (e.g. IV opioid to 5.0 mm; p = 0.003; I2 = 82.5%), 12 h ( 9.9 mm; 95% CI
72
Table 1
Baseline characteristics of included studies.
De Meyer (Abstract) 2016 55 Hallux valgus surgery Subparaneural U/S Levobupivacaine NR 5 mL per hour (9.8 mL PCRA (6 mL per 30 min) NR NR
[14] sciatic (0.125%) per 2 h)
Hamdani [5] 2014 101 Major shoulder surgery Interscalene U/S Ropivacaine (0.2%) 20 mL of ropivacaine 4 mL per hour (4 mL PCRA (5 mL per 30 min) Single Yes
0.25% and lidocaine 0.5% per hour)
Hillegass [18] 2013 45 Total knee arthroplasty Femoral Both Ropivacaine (0.2%) 20 mL of ropivacaine 0.5% 10.1 mL per hour (5 mL IV PCA Single NR
per 30 min)
Mezzatesta [11] 1997 20 Upper limb reconstructive Axillary NS Bupivacaine (0.25%) 2 mg/kg of bupivacaine 0.25 mg/kg per hour IM opioids Multiple NR
microsurgery 0.5% with epinephrine (0.25 mg/kg per hour)
Short (Abstract) [19] 2016 23 Major foot and ankle Popliteal sciatic NR Ropivacaine (0.2%) NR 5 mL per hour (10 mL PCRA (volume NR) NR NR
surgery per 2 h)
Taboada [17] 2008 44 Hallux valgus surgery Popliteal sciatic NS Levobupivacaine 30 mL of mepivacaine 1.5% 5 mL per hour (5 mL Oral opioids Single Yes
(0.125%) per hour)
Taboada [4] 2009 50 Hallux valgus surgery Popliteal sciatic NS Levobupivacaine 30 mL of mepivacaine 1.5% 5 mL per hour (5 mL PCRA (3 mL per 15 min, Single Yes
(0.125%) per hour) max 6 mL of PCRA
per hour)
Thapa [15] 2017 50 ACL reconstruction Adductor canal U/S Ropivacaine (0.5%) 15 mL of ropivacaine 0.5% 2.5 mL per hour IV PCA NR Yes
(15 mL per 6 h)
Wang [16] 2016 60 Total hip arthroplasty Fascia iliaca U/S Ropivacaine (0.2%) 40 mL of ropivacaine 0.2% 10 mL per hour Oral opioids Unclear Yes
(10 mL per hour)
U/S = ultrasound guided; NS = neurostimulation guided; IV = intravenous; IM = intramuscular; PCA = patient-controlled analgesia; NR = not reported (or insufcient information provided to make a determination); ACL = anterior cruciate
ligament; PCRA = patient-controlled regional anesthesia (on-demand boluses of local anesthetic given through the nerve block catheter); MMA = multi-modal analgesia. Conference abstracts are denoted with (Abstract) after the principal
investigator's last name in the trial column. The rest of the papers were published in peer-reviewed journals.
M.A. Chong et al. / Journal of Clinical Anesthesia 42 (2017) 6976 73
Fig. 3. Forest plot of VAS pain scores over time. Data are presented by time point with a Bonferroni correction applied for multiple comparisons.
reduction would not be considered by most to be clinically meaningful data (I2 = 82.5%), and the fact that not all studies reported pain scores
and there were no differences in VAS pain scores at other time points for each time point (Fig. 3). Finally, nearly half the studies utilized
[20,21]. Other caveats include the small number of studies and patients PCRA boluses, which may have obscured any difference between PIB
included in some comparisons, high statistical heterogeneity in the 6 h versus continuous infusion (Table 1). Taken altogether, we do not
Table 2
On-going trials listed on clinicaltrials.gov.
ClinicalTrials.gov identier Type of block Type of surgery Study statusa Estimated date of completiona
Table 3
Summary of ndings table.
Outcomes Illustrative comparative risks (95% CI) Relative effect (95% CI) Number of subjects Quality of the
(number of studies) evidence (GRADE)
Assumed risk (standard care group) Corresponding risk reduction in PIB group
(95% CI)
VAS pain scores The mean VAS pain scores ranged VAS pain scores were 14.2 mm 14.2 mm (23.5 mm to 227 (5)
at 6 h from 10 to 30 in the control group (23.5 mm to 5.0 mm) lower in the 5.0 mm) Very lowa,b
PIB group
VAS pain scores The mean VAS pain scores ranged VAS pain scores were 9.9 mm 9.9 mm (14.4 mm to 236 (5)
at 12 h from 10 to 30 in the control group (14.4 mm to 5.4 mm) lower in the 5.4 mm) Very lowa,b
PIB group
Prolonged motor 53 per 1000 patients 140 (4 to 432) events avoided per 1000 OR 4.27 (1.0816.9) 117 (3)
blockade patients Lowa
a
High risk of bias due to unclear method of allocation concealment for the majority of studies and deciencies in the reporting of randomization sequence generation technique.
b
Very high statistical heterogeneity.
believe that the current evidence supports the PIB technique to have 4.3. Conclusion
any meaningful analgesic benet in CPNB.
The secondary outcomes also reected the lack of analgesic benet Although PIB has seen much success for labor analgesia, our system-
for PIB, with no differences in side effects, opioid consumption, or rescue atic review and meta-analysis did not demonstrate clinically meaning-
analgesia requirement. The only exception was increased rates of resid- ful analgesic benet in CPNB across the current evidence base [3].
ual motor blockade with PIB (OR 4.27; 95% CI 1.0816.9; p = 0.04; Unfortunately, as we have delineated, the published RCTs of PIB are
NNTH = 8). Although no study reported permanent sequelae from fraught with limitations and have poor external validity. Nonetheless,
this residual motor blockade, it is possible that such prolonged motor given the premium cost of PIB-capable infusion pumps, the lack of
block may hinder postoperative mobilization and impact patient satis- meaningful benet in the existing literature serves as a word of caution
faction. Finally, despite our initial hypothesis that PIB decreased local for widespread adoption of the PIB technique for CPNB [6].
anesthetic consumption, this was only demonstrated in two studies, Our systematic review provides important information to inform the
with the remainder showing no meaningful difference [5,11]. conduct of future studies, given that we have identied several repeated
limitations in the existing trials of PIB versus continuous infusion in
CPNB. We suggest that future studies involving PIB be adequately
4.2. Study limitations powered to detect meaningful differences in important patient
outcomes, such as VAS pain scores or patient satisfaction. Furthermore,
The 9 RCTs that were included in this review only comprised 448 future trials should reect contemporary anesthetic practice and utilize
patients. Given this small overall sample size, the majority of our pre- techniques shown to improve analgesia, such as multi-modal analgesia
specied subgroup analyses were underpowered. We await with great and PCRA boluses [1,27]. Finally, given the very limited pain score
interest the completion and publication of several on-going studies in- reduction demonstrated herein, it may be most worthwhile to focus
vestigating PIB in regional anesthesia (NCT02707874, NCT02589288, any further trials on only the most painful of surgeries.
and NCT02539628; Table 2). Although these trials do overlap with the Supplementary data to this article can be found online at http://dx.
existing literature in terms of patient population and type of nerve doi.org/10.1016/j.jclinane.2017.08.018.
block being studied, the additional sample size will be informative
given the small size and number of existing studies.
Regarding the heterogeneity of the included trials, signicant clinical Acknowledgements
differences exist between existing studies. For example, although we
had hoped to present subgroup analysis by type of block and block loca- Dr. Chong is grateful to the Medical Evidence, Decision Integrity,
tion to account for some of this heterogeneity, there were insufcient Clinical Impact (MEDICI) Centre at Western University (London, Ontario,
studies to do so. Such subgroup analyses would be of high interest, Canada) for providing access to the statistical analysis software and to Dr.
given that optimal CPNB dosing is known to vary by these anatomic fac- Nicolas M. Berbenetz for reviewing the nal manuscript. This research
tors [22,23]. Additionally, PCRA and IV PCA as rescue analgesia are did not receive any specic grant from funding agencies in the public,
known to provide different analgesic proles [24]. Taken together, commercial, or not-for-prot sectors.
these variations among the selected trials make the interpretation of
the pooled results more challenging. References
To comment on block placement, nearly half of the studies utilized [1] Singelyn FJ, Seguy S, Gouverneur JM. Interscalene brachial plexus analgesia after
neurostimulation, which some authors suggest leads to a slightly higher open shoulder surgery: continuous versus patient-controlled infusion. Anesth
failure rate compared to ultrasound guidance [25]. Fortunately, this did Analg 1999;89(5):121620.
[2] Wong CA, Ratliff JT, Sullivan JT, Scavone BM, Toledo P, McCarthy RJ. A randomized
not result in a signicant amount of failed blocks among the studies comparison of programmed intermittent epidural bolus with continuous epidural
reviewed. Nonetheless, block and catheter placement method remains infusion for labor analgesia. Anesth Analg 2006;102(3):9049.
an important consideration, as certain studies demonstrate that block [3] George RB, Allen TK, Habib AS. Intermittent epidural bolus compared with continu-
ous epidural infusions for labor analgesia: a systematic review and meta-analysis.
quality is improved when ultrasound guidance is utilized [26]. Unfortu-
Anesth Analg 2013;116(1):13344.
nately, due to the paucity of studies, our pre-specied subgroup analysis [4] Taboada M, Rodriguez J, Bermudez M, Amor M, Ulloa B, Aneiros F, et al. Comparison
by block placement method was not possible. of continuous infusion versus automated bolus for postoperative patient-controlled
analgesia with popliteal sciatic nerve catheters. Anesthesiology 2009;110:1504.
Finally, with regard to the generalizability of the results, only 5 of 9
[5] Hamdani M, Chassot O, Fournier R. Ultrasound-guided continuous interscalene
studies utilized multi-modal analgesia [4,5,1517]. Given that multi- block: the inuence of local anesthetic background delivery method on postopera-
modal analgesia has been demonstrated to decrease pain intensity, tive analgesia after shoulder surgery: a randomized trial. Reg Anesth Pain Med
we suspect that its absence from some trials likely exaggerated the 2014;39(5):38793.
[6] McKenzie CP, Cobb B, Riley ET, Carvalho B. Programmed intermittent epidural bolus-
effect of the PIB on pain scoreswhich was already very modest in our es for maintenance of labor analgesia: an impact study. Int J Obstet Anesth 2016;26:
analysis [1,27]. 328.
76 M.A. Chong et al. / Journal of Clinical Anesthesia 42 (2017) 6976
[7] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The [18] Hillegass MG, Field LC, Stewart SR, Borckardt JJ, Dong L, Kotlowski PE, et al. The ef-
PRISMA statement for reporting systematic reviews and meta-analyses of studies cacy of automated intermittent boluses for continuous femoral nerve block: a pro-
that evaluate health care interventions: explanation and elaboration. J Clin spective, randomized comparison to continuous infusions. J Clin Anesth 2013;25:
Epidemiol 2009;62(10):e134. 2818.
[8] Higgins J, Green S. Cochrane handbook for systematic reviews of interventions. The [19] Short A, Chin KJ, Ghosh SM, Jin R, Chan V. A comparison of programmed intermittent
Cochrane Collaboration; 2011. bolus versus continuous infusion local anesthetic regimens for continuous popliteal
[9] Corbett MS, Higgins JP, Woolacott NF. Assessing baseline imbalance in randomised nerve block. Regional Anesthesia and Pain Medicine. Conference: 41st Annual
trials: implications for the Cochrane risk of bias tool. Res Synth Methods 2014; Regional Anesthesiology and Acute Pain Medicine Meeting of the American Society
5(1):7985. of Regional Anesthesia and Pain Medicine, ASRA, 41. ; 2016.
[10] Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, et al. GRADE: [20] Farrar JT, Berlin JA, Strom BL. Clinically important changes in acute pain outcome
an emerging consensus on rating quality of evidence and strength of recommenda- measures: a validation study. J Pain Symptom Manag 2003;25(5):40611.
tions. BMJ 2008;336(7650):9246. [21] Farrar JT, Portenoy RK, Berlin JA, Kinman JL, Strom BL. Dening the clinically important
[11] Mezzatesta JP, Scott DA, Schweitzer SA, Selander DE. Continuous axillary brachial difference in pain outcome measures. Pain 2000;88(3):28794.
plexus block for postoperative pain relief. Intermittent bolus versus continuous infu- [22] Madison SJ, Monahan AM, Agarwal RR, Furnish TJ, Mascha EJ, Xu Z, et al. A random-
sion. Reg Anesth 1997;22:35762. ized, triple-masked, active-controlled investigation of the relative effects of dose,
[12] Friedrich JO, Adhikari NK, Beyene J. Inclusion of zero total event trials in meta-anal- concentration, and infusion rate for continuous popliteal-sciatic nerve blocks in
yses maintains analytic consistency and incorporates all available data. BMC Med volunteers. Br J Anaesth 2015;114(1):1219.
Res Methodol 2007;7:5. [23] Ilfeld BM, Le LT, Ramjohn J, Loland VJ, Wadhwa AN, Gerancher JC, et al. The effects of
[13] Lee HM, Choi WJ, Choi KT. Comparison of intermittent versus continuous local anesthetic concentration and dose on continuous infraclavicular nerve blocks:
interscalene brachial plexus block for postoperative analgesia. Anesth Pain Med a multicenter, randomized, observer-masked, controlled study. Anesth Analg 2009;
2010;5:1114. 108(1):34550.
[14] Meyer A, Breebaart M, Schepens T, Slappendel R. Continuous infusion versus auto- [24] Borgeat A, Perschak H, Bird P, Hodler J, Gerber C. Patient-controlled interscalene an-
mated intermittent bolus of levobupivacaine 0,125% in ultrasound guided algesia with ropivacaine 0.2% versus patient-controlled intravenous analgesia after
subparaneural sciatic nerve catheters. a prospective double blind randomised study. major shoulder surgery: effects on diaphragmatic and respiratory function. Anesthe-
Regional Anesthesia and Pain Medicine. Conference: 35th Annual European Society siology 2000;92(1):1028.
of Regional Anaesthesia and Pain Therapy Congress, ESRA, 41. ; 2016. p. e4950. [25] Abrahams MS, Aziz MF, Fu RF, Horn JL. Ultrasound guidance compared with electri-
[15] Thapa D, Ahuja V, Verma P, Gombar S, Gupta R, Dhiman D. Post-operative analgesia cal neurostimulation for peripheral nerve block: a systematic review and meta-
using intermittent vs. continuous adductor canal block technique: a randomized analysis of randomized controlled trials. Br J Anaesth 2009;102(3):40817.
controlled trial. Acta Anaesthesiol Scand 2017;60:137985. [26] Williams SR, Chouinard P, Arcand G, Harris P, Ruel M, Boudreault D, et al. Ultrasound
[16] Wang N, Li M, Geng J, Chen X, Guo X. A clinical study of the efcacy of automated guidance speeds execution and improves the quality of supraclavicular block.
intermittent boluses for continuous fascia iliaca block. [Chinese]. Nat Med J China Anesth Analg 2003;97(5):151823.
2016;96:17504. [27] Elia N, Lysakowski C, Tramer MR. Does multimodal analgesia with acetaminophen,
[17] Taboada M, Rodriguez J, Bermudez M, Valino C, Ulloa B, Aneiros F, et al. A new au- nonsteroidal antiinammatory drugs, or selective cyclooxygenase-2 inhibitors and
tomated bolus technique for continuous popliteal block: a prospective, randomized patient-controlled analgesia morphine offer advantages over morphine alone?
comparison with a continuous infusion technique. Anesth Analg 2008;107(4): Meta-analyses of randomized trials. Anesthesiology 2005;103(6):1296304.
14337.