Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

Ultrasonic locating devices for central venous cannulation: meta-analysis.

Hind D, Calvert N, McWilliams R, et al.

Br Med J. 2003;327(7411):361 [abstract]

Reviewed by Eric Harry MD, Children's Hospital and Regional Medical Center, Seattle, WA

Review posted October 5, 2005

---

I. Are the results of the study valid?

A. Primary questions:

1. Did the overview address a focused clinical question?

Yes. The focused clinical question addressed in this meta-analysis was to assess the evidence for the clinical effectiveness of ultrasound guided central venous cannulation. Comparison was made between using real time two-dimensional (2-D) ultrasonography or Doppler needle probes and the anatomic landmark method of cannulation.

2. Were the criteria used to select articles for inclusion appropriate?

As a rapid review to support clinical decision making, the criteria for selecting articles was appropriate, though only English language articles were included in the review.

Inclusion criteria for the meta-analysis were:

• Randomized controlled study design
• Clinical effectiveness of 2-D ultrasound guidance or Doppler ultrasound guidance for the placement of central venous lines.
• Comparison of ultrasonography with the landmark method or the surgical cut-down procedure.
• Inclusion of at least one of five outcomes of interest: risk of failed catheter placement (primary outcome), risk of complications from catheter placement, risk of failure of catheter placement on the first attempt, number of attempts for successful catheterization or time (in seconds) to successful catheterization.

Using the search criteria listed below (I.B.3), 27 trials were identified for review. Five were excluded because their methods of randomization were either unclear or inadequate. Two studies were rejected because veins were located by ultrasound followed by blind venipuncture and 2 more studies were rejected because they were presented in abstract form only. This left 18 studies for inclusion in the review.

B. Secondary questions:

3. Is it unlikely that important, relevant studies were missed?

Maybe. Although fifteen electronic databases were searched from inception to October 2001, all non-English language articles were excluded from the review. References from relevant articles and submissions for sponsorship to the National Institute for Clinical Excellence were also hand searched. The authors did not mention if any attempt was made to identify unpublished studies. The conference proceedings and abstracts could have been searched. The authors of two abstracts could have been contacted instead of simple exclusion. Authors published in the field could have been contacted to identify any the published studies that might have been missed (including studies that are in press or not yet indexed or referenced) and to identify unpublished studies. Without these measures, there was likely to be a publication bias. The full search strategy used to identify articles is published in a separate article (2).

4. Was the validity of the included studies appraised?

Yes. The authors assessed methodological quality descriptively on both the studies that were included and excluded. Studies were appraised primarily on generation of allocation sequence, allocation concealment and adequacy of intention-to-treat analysis. The randomization method was not reported in half of the included studies. None of the studies reported any allocation concealment. This could increase the possibility of "selection bias" in those studies. There was only minimal dropout or withdrawal after randomization so "attrition bias" was not an issue.

Five studies were excluded based on unclear or inadequate randomization. Of the 18 studies included for review, one study was withdrawn from the 'seconds to successful cannulation' outcome based on the inclusion of machine set-up time within the study. No other studies were excluded from relevant outcomes analysis.

5. Were assessments of studies reproducible?

Unclear. The authors did not mention who conducted the literature search, selected the studies, and judged quality of the studies. They only mentioned that two independent "coworkers" extracted data, and discrepancies resolved by consensus. Since professional background and experience could be potential source of biases in any steps of a systematic review, identity of the individuals who conduct the all procedures is very important. Figure 1 tracks all of the studies identified and screened for retrieval. Reasons for exclusion of identified studies were specifically listed within the review and are included above (I.A.2 and I.B.3.).

6. Were the results similar from study to study?

Yes. The studies included for meta-analysis were quite similar in terms of patients, interventions, and outcome measures. However, the physicians who placed the central venous catheters varied from inexperienced residents to experienced cardiac anesthesiologists. This difference could have a significant impact on the success rates of the procedures in each study. There were considerable differences in sample size in the studies included in the meta-analysis (range n=29 to n=160), which could bias the results of the combined analysis towards the results of the studies with larger sample sizes.

In term of outcome determination, the primary outcomes revealed no heterogeneity within 7 studies for failed internal jugular vein catheter placement in adults and in 3 studies for failed internal jugular vein placement in infants (p = 0.33 and 0.51 respectively). This means the included studies seem to measure the same underlying magnitude of effect, hence the results could be combined in meta-analysis.

II. What are the results?

1. What are the overall results of the review?

Results for 2-D ultrasound guidance for internal jugular vein cannulation in adults and infants are summarized in the two tables below (taken from Table 2 of the study).

Summary 2-D Ultrasound Guidance Compared with Landmark Method: Internal Jugular Vein (Adults)
	# Subjects*	Effect Size (95% CI)	P value
Relative Risk
Failed Catheter Placement	608	0.14 (0.06 to 0.33)	<0.0001
Complication with Placement	579	0.43 (0.22 to 0.87)	0.02
Failure on 1st attempt	341	0.59 (0.39 to 0.88)	0.009
Mean #
Attempts for Successful Cannulation	267	-1.50 (-2.53 to -0.47)	0.004
Seconds to Successful Cannulation	372	-69.33 (-92.36 to 46.31)	<0.0001

*Combined number between 2-D ultrasound guidance & landmark method.

Summary 2-D Ultrasound Guidance Compared with Landmark Method: Internal Jugular Vein (Infants)
	# Subjects*	Effect Size (95% CI)	P value
Relative Risk
Failed Catheter Placement	167	0.15 (0.03 to 0.64)	0.01
Complication with Placement	167	0.27 (0.08 to 0.91)	0.03
Failure on 1st attempt	-	-	-
Mean #
Attempts for Successful Cannulation	95	-2.00 (-2.78 to -1.22)	<0.0001
Seconds to Successful Cannulation	127	-349.38 (-801.89 to 103.13)	0.13

*Combined number between 2-D ultrasound guidance & landmark method.

Only 1 study included in the review looked at using 2-D ultrasound guidance for subclavian vein cannulation in adults (5). Though their results showed a relative risk reduction of 86% for failed catheter placement favoring use of ultrasound (RR 0.14, P = 0.006), operators in the study were physicians with limited line placement experience (< 30 previous central line placements).

Likewise, while there was a benefit for use of 2-D ultrasound guidance for femoral line placement in adults in terms of attempts to successful cannulation (mean decrease of 2.7 attempts, 95% CI -5.26 to -0.14), results were again drawn from only 1 study. A trend for a decrease in risk of failed femoral line placement was present (RR 0.29, 95% CI 0.07 to 1.21), though this trend was not statistically significant.

No data were included on use of 2-D ultrasound guidance for placement of subclavian or femoral lines in infants or children.

Use of Doppler ultrasound was associated with a decreased risk of failed internal jugular vein cannulation in adults (RR 0.39, 95% CI 0.17 to 0.92) but an increased risk of failed subclavian vein cannulation (RR 1.48, 95% CI 1.03 to 2.14). Opposite of adults, Doppler ultrasound was associated with a trend towards increased risk of failed internal jugular vein cannulation in infants (RR 1.23, 95% CI 0.3 to 5.11). No data was given using Doppler ultrasound for femoral vein cannulation in infants or adults or for subclavian vein cannulation in infants.

2. How precise were the results?

95% confidence intervals were provided for all outcomes (Table 2-4 and Figure 2) and are included in the results above (II.1). On the whole, the confidence intervals were not terribly precise, but the upper limit of the 95% CI was still quite far from one. For example, the upper limit for the risk of failed IJ placement in infants is 0.64. Although the range of 0.03 to 0.64 is wide, 0.64 is still a meaningful relative risk.

III. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

Clear evidence supports use of 2-D ultrasound guidance for internal jugular vein catheterization in all ages of patients, infant to adult. The beneficial effects of using 2-D ultrasound guidance for internal jugular vein cannulation in adults (significant reductions in the relative risks of failed catheter placement, complications, failure on the 1st attempt and mean number of attempts to successful cannulation) shown in this study are directly applicable to the care of teenagers and adolescents in pediatric intensive care units.

Three studies included in the review were specifically limited to infants and showed similar beneficial results (7,8,9). All three studies involved experienced operators (cardiac anaesthetist or pediatric anesthesia fellows) and all showed a similar benefit favoring use of 2-D ultrasound guidance for internal jugular vein catheterization (significant reductions in the relative risks of failed internal jugular vein cannulation and complications with a trend towards reduction in time to successful cannulation).

In terms of patient safety, the reduction in relative risk of complications during internal jugular vein cannulation shown with use of 2-D ultrasound guidance in adults and infants may also be directly applicable to training inexperienced operators in the PICU.

Limited data was included in the study regarding use of 2-D ultrasound guidance for subclavian or femoral line placement in adults, with results from only 1 study included for both. From these results it is too early to conclude that 2-D ultrasound would be beneficial for either procedure. No information was included on subclavian or femoral vein cannulation in infants.

Finally, real time 2-D ultrasound guidance was superior to Doppler ultrasound for cannulation of the internal jugular vein in adults and infants and for cannulation of the subclavian vein in adults. While Doppler ultrasound has some potential benefit in internal jugular vein cannulation in adults, it is inferior to 2-D ultrasound guidance, did not show a benefit for internal jugular vein cannulation in infants and was associated with a higher risk of failed subclavian vein cannulation in adults.

2. Were all clinically important outcomes considered?

All major clinical outcomes were included in the study. The 5 clinical outcomes considered were:

• Risk of failed catheter placement (primary outcome)
• Risk of complications from catheter placement
• Risk of failure of catheter placement on the first attempt
• Number of attempts for successful catheterization
• Time (in seconds) to successful catheterization.

3. Are the benefits worth the harms and costs?

Given that 2-D ultrasound guidance reduced the relative risks of complications related to internal jugular vein cannulation by 57% in adults and 73% in infants there is a clear potential for a patient safety benefit in all ages of patients.

A cost-benefit analysis performed for the National Health Service (UK) (2) and summarized in the Hind article showed potential cost savings of $3249 (£2000) per 1,000 procedures for using ultrasound guidance for central venous cannulation. This analysis included the cost of purchasing ultrasound machines and training staff in their use.

The only potential downside for use of routine ultrasound guidance for internal jugular vein cannulation mentioned by the authors is the potential for "deskilling" providers in use of the landmark method in emergency situations when ultrasound may not be available. Whether or not this would be true in practice is open to debate.

References:

1. Schultz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995; 273:408-12. [abstract]
2. Calvert N, Hind D, McWilliams RG, Thomas SM, Beverly C, Davidson A. The effectiveness and cost-effectiveness of ultrasound locating devices for central venous access: a systemic review. Health Technol Assess 2003:7(12).[abstract]
3. Randolph AG, Cook D, Gonzales CA, Pribble CG. Ultrasound guidance for placement of central venous catheters: a meta-analysis of the literature. Crit Care Med 1996; 24(12): 2053-58.[abstract]
4. Abboud PAC, Kendall JK. Ultrasound guidance for vascular access. Emergency Medicine Clinics of North America 2004; 22(3): 749-73.
5. Gualtieri E, Deppe S, Sipperly ME, Thompson DR. Subclavian venous catheterization: greater success rate for less experience operators using ultrasound guidance. Critical Care Medicine 1995; 23(4): 692-7.[abstract]
6. Altman DG, Bland JM. Interaction revisited: the difference between two estimates. BMJ 2003; 326:219.[abstract]
7. Verghese ST, McGill WA, Patel RI, Sell JE, Midgley FM, Ruttimann UE. Comparison of three techniques for internal jugular vein cannulation in infants. Paediatric Anaesthesia 2000; 10: 505-11.[abstract]
8. Verghese ST, McGill WA, Patel R, Sell J, Midgley F, Ruttimann U. Ultrasound-guided internal jugular venous cannulation in infants: a prospective comparison with the traditional palpation method. Anesthesiology 1999; 91(1): 71-77.[abstract]
9. Alderson PJ, Burrows FA, Stemp LI, Holtby HM. Use of ultrasound to evaluate internal jugular vein anatomy and to facilitate central venous cannulation in paediatric patients. British Journal of Anaesthesia 1993; 70: 145-8.[abstract]

Comments

Back to the EB Journal Club Index