• 3230 catheters were placed in the subclavian vein, 10,958 in the
IJ and 3,188 in the femoral vein for a total of 113,652 catheter days

  
• 2 RCTs and 8 cohort studies
  
• All RCTs are conducted on ICU patients for non-emergent lines with full sterile technique
  
• CRBI
 density was 2.5 per 1,000 catheter days (range 0.6-7.2)
  
• No significant difference in the risk of CRBIs between femoral lines and subclavian/IJ lines in each randomized controlled trial (i.e., no level 1A evidence).
  
• For all studies, the authors report no overall risk of CRBI between the femoral and subclavian site (RR 1.75; 95% CI 0.80-3.8, p = .16).
  
• The studies were homogenous when 2 studies (Nagashima et al. and Lorente et al.) were removed because they were considered statistical outliers. Funnel plot demonstrated bias of these two studies, with these earlier studies favoring the subclavian site.
  
• Overall, the IJ site had a significantly lower risk of CRBI when compared to the femoral site (risk ratio 1.90; 95% CI 1.21-2.97, p = .005, I2 = 35%). When the two outlier studies (see above) were removed, there was no statistical difference.
  
• Risk of DVT was assessed in the two RTCs, both with divergent results. The studies were heterogenous. Meta-analysis of the data from both studies reveals no significant difference in risk of DVT between femoral and IJ/Subclavian.

Central line in place at admission to ICU, central line within 15 days of admission, emergent central line, coagulopathy, thrombocytopenia, anatomic defect, skin lesion or recent surgery at insertion site, BMI >35 (men) or >30 (women)

• Femoral catheterization was associated with:
  
• Higher rate of overall infectious complications (19.8% vs 4.5%; P<.001; incidence density of 20 vs 3.7 per 1000 catheter-days)
  
• Higher rate of major infectious complications (clinical sepsis with or without bloodstream infection, 4.4% vs 1.5%; Not clinically significant)
  
• Higher rate of thrombotic complications (21.5% vs 1.9%; P<.001) and complete thrombosis of the vessel (6% vs 0%; P = .01)
  
• Rates of overall and major mechanical complications were similar between the 2 groups (17.3% vs 18.8 %; P = .74 and 1.4% vs 2.8%; P = .44, respectively).
  
• The only factor associated with infectious complications was femoral catheterization
  
• Femoral catheterization was the only risk factor for thrombotic complications (OR, 14.42; 95% CI, 3.33-62.57; P<.001).

• IJ catheters were more difficult to insert and required more time. More hematomas occurred in the jugular group than in the femoral group (13/366 patients [3.6%] vs 4/370 patients [1.1%], respectively; P=.03).
  
• CRBI occurred in 3 of 324 patients (0.9%) with femoral catheters (incidence per 1000 catheter-days, 1.5; 95% CI, 0.1-6.4) and in 5 of 313 patients (1.6%) with jugular catheters (incidence per 1000 catheter-days, 2.3; 95% CI, 0.3-7.7). This difference was not significant (P = .42).
• The risk of catheter colonization at removal did not differ significantly between the femoral and jugular groups (incidence of 40.8 vs 35.7 per 1000 catheter-days; hazard ratio [HR], 0.85; 95% confidence interval [CI], 0.62-1.16; P=.31).
  
• Jugular catheterization significantly increased incidence of catheter colonization vs femoral catheterization in patients with low BMI, whereas jugular catheterization significantly decreased this incidence in patients with high BMI.