Is Routine Chest X-ray After Ultrasound-Guided Central Venous Catheter Insertion Choosing Wisely? A Population-Based Retrospective Study of 6875 Patients

Chui J, Saeed R, Jakobowski L et al.

Chest. 2018 Feb 28. pii: S0012-3692(18)30341-6.

DOI: 10.1016/j.chest.2018.02.017

Aim of Study

• to determine the incidence of pneumothorax and CVC misplacement in the perioperative setting
• Cost analysis to examine the economic impact of using a routine post-procedural CXR as a screening test to rule out CVC-related mechanical complications.
• Risk factors associated with higher complication rates of pneumothorax and catheter misplacement that may warrant routine screening.

Design

Large-scale population-based retrospective cohort study

Methodology

Retrospective review of the records of all adult patients who received a central line in three University-affiliated hospitals in London, Ontario, Canada between July 1, 2008 – December 31, 2015

• Catheter misplacement was defined as either
  • the catheter tip was not positioned at the confluence of SVC and RA
  • the catheter was coiled or looped along its course
• If Pt identified to have a CVC-related mechanical complication:
  • further review of the records of their entire hospital stay to review the management of CVC-related mechanical complications

Statistics

• Univariate logistic regression analysis: The association between the complications [pneumothorax (yes or no) or catheter misplacement (yes or no)] and the study variables were examined by logistic regression. Odds ratios with associated 95% confidence intervals and P values were calculated.
• Multivariate logistic regression analysis: to measure the association between the complications and the study variables.

Results

• Incidence of pneumothorax:
  • 29% (17 of 5796 patients) with RIJ
  • 0% (0 of 440 patients) with LIJ
  • 71% (5 of 292 patients) with LSCV
  • 39% (1 of 259 patients) with RSCV
• Management of pneumothorax
  • Of 23 patients with confirmed pneumothorax, (18 were right side and 5 left side)
    • 15 successfully treated with conservative management.
    • 8 (0.12% of 6875) required a chest drain to be inserted
      • 5 immediately after the pneumothorax identified
      • 2 on post-procedural day 1
      • 1 on post-procedural day 2
    • Incidence of misplacement:
      • More common than pneumothorax
      • 51% (88 of 5796 patients) with RIJ
      • 86% (17 of 440 patients) with LIJ
      • 82% (17 of 292 patients) with RSCV
      • 47% (9 of 259 patients) with LSCV
    • Cost-Analysis
      • Total number of chest X-rays performed in this study was 6875
      • The total cost of CXR: $790K-1,375K USD (over 7.5 years) or $105K-$183K/y
      • Cost to diagnose one case of pneumothorax was $34,375-$59,783
      • Cost associated with one chest drain insertion was $98,828-$171,875 USD
      • Number Needed to Test (NNT) to detect one case of pneumothorax was 299
      • The NNT to detect one case where chest drainage was necessary was 860

Conclusions/Discussions

• Complication rates of CVCs, inserted under US guidance, were exceedingly low (only 0.33% pneumothorax and 1.91% misplacement)
  • When these complications occur, most pts did not require any intervention.
  • No pts with catheter misplacement developed any related complications.
• The cost analysis showed that post-procedural CXR was a very expensive screening test to rule out CVC-related mechanical complications.
• Authors identified the site of catheterization as the major determinant of pneumothorax and catheter misplacement, rather than patient or procedure.
• Point-of-care US has recently been suggested as a clinically superior alternative technique to diagnose pneumothorax and confirm catheter misplacement
• Enacting a change in practice is difficult and requires a cultural shift, physician awareness, and a redesign of patient management processes and guidelines.

Stated Limitations from the Study

Strengths:

• By far the largest cohort in the literature that evaluates CVC-related mechanical complications under US
• Study adds important additional information to the evidence base as it:
  • confirms the rarity of CVC-related mechanical complications
  • identified that the site of catheterization was the most important factor predicting pneumothorax and catheter misplacement which may help to develop selective screening strategies in this subset of high-risk patients,
  • allowed for a cost analysis of CXR.

Limitations:

• Retrospective design therefore information bias.
  • To minimize the information bias, authors evaluated both the electronic CXR film and the radiologist’s report from each patient to verify any CVC-related mechanical complications. Authors performed data verification by a second assessor.
• Lack of data regarding the level of expertise and experience of the operators, or the number and duration of CVC insertions for each procedure.

Discussion from Journal Club Meeting (?Change of Practice)

Discussion regarding potential cost in our local trust

Discussion regarding how likely to implement USS as point-of-care to verify CVC placement and any complications

Discussion regarding UK guidelines which are national, issued by NICE and/or the Royal College of Radiologists in this case and change needs to be implemented nationally.

Discussion regarding difficulty of enacting change in clinical practice

Summary by Dr R Hreiche. Journal Club Meeting 08 March 2018.

 

Addendum:

Blog post by Dr Josh Farkas adding further information/opinion regarding this topic.

https://emcrit.org/pulmcrit/does-central-line-position-matter-can-we-use-ultrasonography-to-confirm-line-position/