a b s t r a c t

Background: Currently, whether long-axis in-plane (LA-IP) is superior to short-axis out-of-plane (SA-OOP) dur- ing ultrasound-guided vascular access remains inconclusive. We, therefore, conducted a meta-analysis of ran- domized controlled trials to compare the effects of LA-IP vs SA-OOP techniques in patients undergoing ultrasound-guided vascular access (USGVA).

Methods: A computer-based literature search of PubMed, Embase, and the Cochrane Library (up to October 2015) was performed to identify randomized controlled trials that evaluated the effects of LA-IP compared with SA-OOP in patients undergoing USGVA. The primary end point was the First-pass success rate. Secondary end points in- cluded mean time to success, mean attempts to success, and incidence of the complication of hematoma. Weight- ed mean differences (WMDs) and Relative risks (RRs) with 95% confidence intervals (CIs) were calculated by random-effects model.

Results: Five eligible studies with a total of 470 patients satisfied the inclusion criteria. There was no significant difference for the first-pass success rate (RR, 1.06; 95% CI, 0.91-1.23; P = .44), mean time to success (WMD, 4.78 seconds; 95% CI, -4.43 to 13.99; P = .31), mean attempts to success (WMD, 0.06 times; 95% CI, -0.23 to 0.35; P = .69), and incidence of the complication of hematoma (RR, 2.86; 95% CI, 0.32-25.42; P = .35) between the LA-IP and SA-OOP groups.

Conclusions: There is insufficient evidence to definitively choose either LA-IP or SA-OOP in patients undergoing USGVA. Further robustly well-designed trials are warranted to investigate the appropriate technique in patients receiving USGVA.

(C) 2016

Introduction

Vascular cannulation as a common invasive procedure is often re- quired in most of clinical settings, particularly in emergency department (ED), intensive care unit, and operating room. In general, vascular can- nulation includes catheterization of vein and artery, which is commonly used as an important aspect of patient care for the administration of fluids and medications [1]. Importantly, it allows continuous monitoring purposes, such as central venous pressure detection, arterial blood gas analysis, and other tests of Blood indicators. Nowadays, the latest

? Disclosure: The authors declare no conflict of interest.

?? Funding: None.

* Corresponding author: Department of Ultrasonography, Qilu Hospital, Shandong Uni- versity, No. 107 Wenhua Xi Road, Jinan, 250000, Shandong, China. Tel./fax: +86 531

82169114.

E-mail address: [email protected] (J. Li).

¹ These authors contributed equally to this work.

guidelines stated that the routine use of ultrasound guidance is recom- mended for vascular cannulation, especially for internal jugular vein cannulation, radial artery catheterization, and percutaneous intrave- nous central catheterization [1]. In addition, numerous studies have shown that ultrasound-guided vascular access (USGVA) can improve the first-pass success rate; reduced number of needle passes and com- plication incidence; and shorten access time for radial artery catheteri- zation [2,3], peripheral intravenous access [4], and nerve block [5,6].

There are 2 basic approaches for USGVA techniques, that is, long-axis in-plane (LA-IP) and short-axis out-of-plane (SA-OOP) [1]. Using an in- animate model, Blaivas et al [7] found that SA approach represented faster than LA approach in terms of vascular access time and LA ap- proach was associated with improved visibility of the needle tip during vessel puncture [8]. Unluckily, it is currently unknown which puncture technique provides the optimal conditions for USGVA. Several random- ized controlled trials (RCTs) were designed to compare the effects of 2 different needling techniques in success rate at the first attempt, cathe- ter insertion time, and number of needle sticks in patients undergoing USGVA [9-13]. However, these studies not only had wide variation in

http://dx.doi.org/10.1016/j.ajem.2015.12.092

0735-6757/(C) 2016

sample size but also conveyed inconclusive results. We, therefore, con- ducted a meta-analysis involving available RCTs to compare the effects of 2 different needling approaches in patients undergoing USGVA.

Methods

Data sources and literature search

Computer-based literature search was performed on PubMed, Embase, and the Cochrane Library from inception through October 2015 for eligible trials with using the following keywords: (“long axis” OR “in plane”) and (“short axis” OR “out of plane”) and “ultrasound.” El- igible trials limited by RCT and written in English were included. In ad- dition, bibliographies of all potential studies, including reference lists, citation searches, the latest guidelines, and relevant systematic reviews, were manually searched.

Study selection

The following selection criteria were included: (1) population: ei- ther adult or pediatric patients undergoing USGVA; (2) intervention vs control: LA-IP vs SA-OOP; (3) outcome measures: the primary end point was the first-pass success rate, and secondary end points included mean time to success and mean attempts to success; and (4) study de- sign: RCT.

Data extraction and outcome measurement

All data were extracted by 2 independent investigators (YBG and JHY). In detail, the data included first author, Publication year, country, study design/Jadad score, number of patients (LA-IP/SA-OOP), weight (LA-IP/SA-OOP), clinical setting, vessel type, Ultrasound device, ultra- sound type, catheters, operator experience, and results. Disagreements among authors were settled by discussion or a third investigator (JMM). Predefined primary end point was the first-pass success rate, and secondary endpoints included mean time to success (defined as the time period between penetration of skin and aspiration of venous blood into the catheter) and mean attempts to success (defined as the number of times the needle was withdrawn and redirected).

Bias assessment“>Quality and risk-of-bias assessment

The quality of RCTs was evaluated according to the Jadad scale [14]. In detail, randomization (0-2 points), blinding (0-2 points), and the

dropouts and withdrawals (0-1 points) were defined in the scale. A score of less than or equal to 2 indicates low quality, whereas a score of greater than or equal to 3 indicates high quality [15]. In addition, the risk of bias was assessed by the Cochrane risk-of-bias tool [16].

Statistical analysis

The present study was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta- Analyses statement [17]. Weighted mean differences (WMDs) for con- tinuous outcomes with 95% confidence intervals (CIs) and relative risks (RRs) for dichotomous outcomes with 95% CIs were calculated by random-effects model [18]. The heterogeneity across studies was tested using the I² statistic, and the I² greater than 50% indicated significant heterogeneity [19]. If so, we would use sensitivity analyses conducted by sequentially excluding each study to identify the potential sources of heterogeneity and investigate the influence of a single study on the overall pooled estimate. Importantly, to check the influence of various factors on the first-pass success rate, we further performed subgroup analyses based on differently clinical setting among studies, such as ul- trasound type (Doppler ultrasound vs real-time 2-dimensional [2D] ul- trasound), vessel type (the radial artery vs the Right internal jugular vein), and sample size (>= 99 vs b 99). Furthermore, publication bias was not assessed because only 5 RCTs (b 10) were included in the pres- ent study. All meta-analyses were performed using RevMan 5.3 (The Cochrane Collaboration, Oxford, UK). A 2-sided P value of less than .05 indicated statistical significance.

Results

Eligible studies and studies of characteristics

A total of 235 relevant articles were identified from the initial search. After reviewing the titles and abstracts and removing duplicates, 212 were excluded for various reasons, and finally, 5 eligible RCTs [9-13] in- volving a total of 470 patients were identified for our meta-analysis (Fig. 1). Next, the main characteristics of the available 5 RCTs are shown in Table 1. In detail, all the 5 RCTs were published in English and involved adult patients. They were published from 2011 to 2014, and the sample sizes of them ranged from 40 to 163. Five RCTs were conducted in 5 different countries including Turkey [9], Ireland [10], the United States [11], China [12], and Egypt [13]. Among the included trials, Doppler ultrasound was applied in 3 RCTs [9,11,12], whereas real-time 2D ultrasound in 2 RCTs [10,13]. With respect to vessel type,

Fig. 1. Search strategy and flow chart of the meta-analysis.

Table 1

Main characteristics of RCTs included in the meta-analysis

Study/country

Study design/ Jadad score

No. of patients (LA- IP/SA-OOP)

Weight (LA-IP/ SA-OOP)

Clinical setting

Vessel type

Ultrasound device

Ultrasound type

Catheters Operator experience

Results

Berk et al [9]/ RCT/3 108 (54/54) Turkey

(76 +- 16)/78 +-

18 (kg)

(84.1 +- 15.7)/

Patients undergoing elective surgery

Patients

The radial artery

The right

Esaote My Lab 30, US

Machine, Florance, Italy

SonoSite,

Doppler ultrasound

Real-time

20-G 1.v.

cannula (Lakhani Medicare Pvt, Ltd, Haryans, India)

Anesthetists who had placed N 50 Arterial lines by using either in plane or out of plane approaches

Anesthetists with experience of

There were shorter Cannulation time and higher first-pass success rate in LA-IP group than SA-OOP group (all P b .05).

The first-pass success rate was higher in SA-OOP group than LA-IP

Chittoodan et al

/Ireland

Mahler et al

Blinded

RCT/4

99 (50/49)

(85.2 +- 13.

5) (kg)

(202 +- 66)/

undergoing

cardiac surgery

Emergency department

internal jugular vein

The

Micromaxx, Bothwell, WA

Sonosite Micromax;

2D

ultrasound

Doppler

Not reported

18- or 20-G

angiocath IV

N 50 US-guided internal jugular cannulations

Attending physicians or nurses with experience of

group (P b .006).

Fewer needle redirections were required in SA-OOP group than LA-IP group (P b .004).

SA-OOP technique required less insertion time than

LA-IP (P = .02), and

/USA RCT/3 40 (20/20)

Quan et al [12]/ RCT/3 163 (82/81) China

Tammam et al RCT/3 60 (30/30) [13]/Egypt

(201 +- 54) (lb)

(72.1 +- 10.5)/

(76.4 +- 12.2) (kg)

Body mass index (kg/m²) (28.15 +- 8.20)/(27.41 +-

7.17)

patients

with difficult IV access

Patients undergoing liver surgery or splenic resection

Critical care and hemodialysis patients

basilic

vein

The radial artery

The right internal jugular vein

Sonosite, Inc, Bothell, WA

Terason2000 +, Terason, Burlington, MA

GE LogiqBook XP; GE, Inc,

Fairfield, CT

ultrasound

Doppler ultrasound

Real-time 2D

ultrasound

catheter

(Medex, Inc., Carlsbad, CA)

22-G arterial cannula (Insyte WTM, Becton Dickinson, Singapore)

16- or 18-G

(Cook, Bloomington, IN)

N 20 US-guided

peripheral or Central lines and workshop attendance Anesthesiologists who had previously cannulated 450 radial arteries and used the US- guided technique for approximately 200 procedures Attending anesthetists, nephrologists, or intensivists who had at least 5 years of experience in performing the internal jugular vein catheterization

success rate was higher in SA-OOP group than LA-IP group (P = .61).

There was higher first-pass success rate in modified SA- OOP group than LA- IP group (P b .0001), but no difference for cannulation time.

No difference was found in SA-OOP and LA-IP groups for average access time and number of attempts (P = .94,

.84).

Pooled analysis of end points”>there were 2 trials with the radial artery [9,12], 2 trials with the right in- ternal jugular vein [10,13], and 1 trial with the basilic vein [11]. Further- more, all puncturing operations were performed by experienced attending physicians, anesthetists, nephrologists, and intensivists or nurses.

Quality and risk-of-bias assessment

Details of quality and risk-of-bias assessment are, respectively, summarized in Table 1 and Fig.2. Two authors agreed on each item of the Jadad scores and the Cochrane risk-of-bias tool (YBG and JHY). As a result, the mean (SD) Jadad score was 3.2 (0.4). The risk-of-bias analysis showed that all trials were followed at low risk in terms of ran- dom sequence generation (selection bias), allocation concealment (se- lection bias), and incomplete outcome data (attrition bias), but only 1 study [10] was followed at low risk in blinding of outcome assessment (detection bias).

Pooled analysis of end points

The pooled results indicated that there was no significant difference for the first-pass success rate (5 RCTs [9-13]; RR, 1.06; 95% CI, 0.91-1.23; P = .44; P = .0006 for heterogeneity; I² = 80%; Fig. 3A), mean time to success (4 RCTs [9,10,12,13]; WMD, 4.78 seconds; 95% CI, - 4.43 to 13.99; P = .31; P = .0003 for heterogeneity; I² = 84%; Fig. 3B), mean at- tempts to success (4 RCTs [9-11,13]; WMD, 0.06 times; 95% CI, -0.23 to 0.35; P = .69; P = .0007 for heterogeneity; I² = 82%; Fig. 3C), and inci- dence of the complication of hematoma (3 RCTs [9,12,13]; RR, 2.86; 95% CI, 0.32-25.42; P = .35; P = .002 for heterogeneity; I² = 85%; Fig. 3D) between the LA-IP and SA-OOP groups. Here, heterogeneity was signif- icant for the first-pass success rate. Further exclusion of any single study did not resolve the heterogeneity and did not change the pooled results. Subsequently, we performed subgroup analyses for the first-pass suc- cess rate to determine the effect of various exclusion criteria according to ultrasound type, vessel type, and sample size (Table 2). These exclu- sion criteria did not obviously alter the overall combined RR of the first-

Fig. 2. Risk-of-bias assessment. A, Risk-of-bias graph, the authors’ judgments about each risk-of-bias item presented as percentages across all included studies. B, Risk-of-bias summary, the authors’ judgments about each risk-of-bias item for the each included studies.

pass success rate, which ranged from 0.92 (95% CI, 0.51-1.67; P = .79) to

1.73 (95% CI, 0.78-1.59; P = .54).

Discussion

In the current meta-analysis, we summarized preliminary evidence in the clinical literature on the effects of LA-IP vs SA-OOP techniques in patients with USGVA. Our results indicated that there was no signifi- cant difference for the first-pass success rate, mean time to success, mean attempts to success, and incidence of the complication of hemato- ma to success between the LA-IP and SA-OOP groups.

Currently, numerous studies have confirmed the role of USGVA in patients from various clinical departments, especially in those with dif- ficult access [20-23]. In general, 2D images display with 2 basic scan- ning axes either LA-IP or SA-OOP for the target vessel. Each has its own advantages and disadvantages in terms of directing the needle on the target vessel [1]. However, relatively few studies have fully investi- gated placement techniques and compared the effects of 2 different needling approaches. To date, there is no acceptable standard scanning axis for USGVA. The purpose of our meta-analysis is to compare the ef- fects of LA-IP vs SA-OOP techniques on the first-pass success rate, mean time to success, mean attempts to success, and incidence of the compli- cation of hematoma to success in patients with USGVA.

Our results provided preliminary evidence that neither LA-IP nor SA-OOP appeared to be better than the other in terms of improving the first-pass success rate, shortening mean time to success, and reducing mean attempts to success. Subsequently, given significant

heterogeneity for the first-pass success rate, we further conducted sen- sitivity analyses subgroup analyses and found that the overall combined RR of the first-pass success rate was not obviously changed. We believe that heterogeneity across studies could be seen as a result of clinical and methodological differences, such as different ultrasound type and de- vice, vessel type, clinical setting, catheters, and operator experience. In terms of safety, we found that both LA-IP and SA-OOP techniques would cause some complications including thrombosis, hematoma, edema, vasospasm, and posterior wall damage. Of them, hematoma was the most common complication, which is also consistent with the literatures [9,23]. The overall results suggested that the 2 techniques did not seem much different in terms of leading to incidence of hemato- ma. Nevertheless, we still believed that the LA-IP technique may be more pro to reduce incidence of hematoma.

We also provided some additionally valuable implications for clinical practice and future research. First of all, considering the obvious hetero- geneity and the potential bias, our results could not be used as strong basis for choosing the suitable operating technique of scanning axis for patients with USGVA, and we believe that the principle of individu- ation may be followed. The use of ultrasound guidance for improving first-pass success rate and reducing complications varies according to vessel type selected [5]. Future research needs to focus on comparing the first-pass success rate among the central venous, radial artery, and peripheral intravenous. Next, only 1 trial was applied with blinding of outcome assessment. We also acknowledge that it may be difficult to blind to investigators and participants during the puncture process. Be- cause blinding can prevent ascertainment bias and protect the sequence

Fig. 3. Forest plot of the meta-analysis of RCTs. A, The first-pass success rate. B, Mean time to success. C, Mean attempts to success. D, Incidence of the complication of hematoma.

after allocation [24], future researchers should focus on applying the ap- propriate blinding such as blind to outcome assessments. Finally, all the

5 RCTs included only adult patients; hence, the effects of the 2

techniques for vascular access in pediatric patients lack of full under- standing. Further studies are, therefore, necessary to pay more attention to this important topic.

Table 2

Subgroup analyses based on various exclusion criteria for the first-pass success rate

Outcome	n (N)	LA-IP	SA-OOP	RR (95% CI)	P	I2 (%)	Heterogeneity P	Model
All included trials [9-13]	470 (5)	187/236	197/234	1.06 (0.98-1.16)	.16	80	.0006	Fixed effects
Trials with Doppler ultrasound [9,11,12]	311 (3)	118/156	119/155	1.00 (0.73-1.36)	.98	85	.002	Random effects
Trials with real-time 2D ultrasound [10,13]	159 (2)	69/80	78/79	1.12 (0.78-1.59)	.54	95	b.0001	Random effects
Sample size >= 99 [9,10,12]	370 (3)	140/186	148/184	1.04 (0.77-1.41)	.79	87	.0004	Random effects
Sample size b 99 [11,13]	100 (2)	47/50	49/50	1.03 (0.90-1.19)	.65	51	.15	Random effects
Vessel type-the radial artery [9,12]	271 (2)	101/136	100/135	0.92 (0.51-1.67)	.79	92	.0004	Random effects
Vessel type-the right IJV [10,13]	159 (2)	69/80	78/79	1.12 (0.78-1.59)	.54	95	b.0001	Random effects

Abbreviations: n, number of patients; N, number of trials; IJV, internal jugular vein.

When interpreting our results, some limitations should be taken into account. First, the present study is based on only 5 RCTs with various clinical settings, different ultrasound device and type, and different nee- dle type and target vessels, which led to heterogeneity and potentially influenced the results. Second, almost all trials were not blinded, espe- cially in those studies with a small sample. This could lead to perfor- mance and detection bias. Third, because of limited data, we failed to assess the effects of the other importantly clinical outcomes, for exam- ple, patient pain and patient satisfaction scores. Finally, those unpub- lished or missing data may lead to selection bias.

Conclusions

In sum, the current preliminary evidence suggests that there is insuf- ficient evidence to definitively choose either LA-IP or SA-OOP in patients undergoing USGVA. Hence, the principle of individuation may appear to be more appropriate. Considering the limitations and the potential bias, further robustly well-designed trials are warranted to investigate the most appropriate technique as well as the optimal populations in pa- tients with USGVA.

Acknowledgments

We would like to acknowledge the authors of the original articles who provided additional data and information about their work beyond what was included in their published articles.

References

1. Troianos CA, Hartman GS, Glas KE, Skubas NJ, Eberhardt RT, Walker JD, et al. Special articles: guidelines for performing ultrasound guided vascular cannulation: recom- mendations of the American Society of Echocardiography and the Society of Cardio- vascular Anesthesiologists. Anesth Analg 2012;114(1):46-72.
2. Gao YB, Yan JH, Gao FQ, Pan L, Wang XZ, Lv CJ. Effects of ultrasound-guided radial artery catheterization: an updated meta-analysis. Am J Emerg Med 2015;33(1): 50-5.
3. Shiloh AL, Savel RH, Paulin LM, Eisen LA. Ultrasound-guided catheterization of the radial artery: a systematic review and meta-analysis of randomized controlled trials. Chest 2011;139(3):524-9.
4. Nelson D, Jeanmonod R, Jeanmonod D. Randomized trial of tourniquet vs blood pres- sure cuff for target vein dilation in ultrasound-guided peripheral intravenous access. Am J Emerg Med 2014;32(7):761-4.
5. Flores S, Herring AA. Ultrasound-guided dorsal penile nerve block for ED paraphimosis reduction. Am J Emerg Med 2015;33(6):863 [e3-5].
6. Cao X, Zhao X, Xu J, Liu Z, Li Q. Ultrasound-guided technology versus neurostimulation for sciatic nerver block: a meta-analysis. Int J Clin Exp Med 2015;8(1):273-80.
7. Blaivas M, Brannam L, Fernandez E. Short-axis versus long-axis approaches for teaching ultrasound-guided vascular access on a new inanimate model. Acad Emerg Med 2003;10(12):1307-11.
8. Stone MB, Moon C, Sutijono D, Blaivas M. Needle tip visualization during ultrasound- guided vascular access: short-axis vs long-axis approach. Am J Emerg Med 2010; 28(3):343-7.
9. Berk D, Gurkan Y, Kus A, Ulugol H, Solak M, Toker K. Ultrasound-guided radial arte- rial cannulation: long axis/in-plane versus short axis/out-of-plane approaches? J Clin Monit Comput 2013;27(3):319-24.
10. Chittoodan S, Breen D, O’Donnell BD, Iohom G. Long versus short axis ultrasound guided approach for internal jugular vein cannulation: a prospective randomised controlled trial. Med Ultrason 2011;13(1):21-5.
11. Mahler SA, Wang H, Lester C, Skinner J, Arnold TC, Conrad SA. Short- vs long-axis ap- proach to ultrasound-guided peripheral intravenous access: a prospective random- ized study. Am J Emerg Med 2011;29(9):1194-7.
12. Quan Z, Tian M, Chi P, Cao Y, Li X, Peng K. Modified short-axis out-of-plane ultra- sound versus conventional long-axis in-plane ultrasound to guide radial artery can- nulation: a randomized controlled trial. Anesth Analg 2014;119(1):163-9.
13. Tammam TF, El-Shafey EM, Tammam HF. Ultrasound-guided internal jugular vein access: comparison between short axis and long axis techniques. J Kidney Dis Transpl 2013;24(4):707-13.
14. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17(1):1-12.
15. Kjaergard LL, Villumsen J, Gluud C. Reported methodologic quality and discrepancies between large and small randomized trials in meta-analyses. Ann Intern Med 2001; 135(11):982-9.
16. Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD, et al. The Cochrane collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011;343: d5928.
17. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 2009; 339:b2700.
18. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7(3):

    177-88.

    Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta- analyses. BMJ 2003;327(7414):557-60.

19. Reusz G, Csomos A. The role of ultrasound guidance for vascular access. Curr Opin Anaesthesiol 2015;28(6):710-6.
20. Lalu MM, Fayad A, Ahmed O, Bryson GL, Fergusson DA, Barron CC, et al. Ultrasound- guided subclavian vein catheterization: a systematic review and meta-analysis. Crit Care Med 2015;43(7):1498-507.
21. Butterfield M, Abdelghani R, Mohamad M, Limsuwat C, Kheir F. Using ultrasound- guided peripheral catheterization of the internal jugular vein in patients with diffi- cult peripheral access. Am J Ther 2015;43(12 Suppl 1):1-3.
22. Gu WJ, Wu XD, Wang F, Ma ZL, Gu XP. Ultrasound guidance facilitates radial artery catheterization: a meta-analysis with trial sequential analysis of randomized con- trolled trials. Chest 2015:1-50.
23. Schulz KF, Grimes DA. Blinding in randomised trials: hiding who got what. Lancet 2002;359(9307):696-700.