Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Vein harvesting techniques compared

Systematic review

Bandolier 119 was intrigued by a paper applying best evidence to surgical interventions [1]. It pulled together evidence about perioperative medicine, performed a systematic review of pilot studies, and suggested mechanisms of implementation. The surgical literature has many good meta-analyses. One, on techniques of vein harvesting for coronary artery bypass surgery [2], examines how different techniques can lead to very different rates of harmful postoperative leg wound infections.

Systematic review

The search used MEDLINE for studies reporting on minimally invasive vein harvesting for CABG, and compared it with conventional harvesting techniques. For inclusion studies had to be properly randomised, though clearly they could not be blind, and care was taken to avoid double-reporting. Minimally invasive methods included several techniques and instrumentation, including endoscopic techniques. Conventional vein harvesting used standard surgical instrumentation and direct visualisation through longitudinal or skin-bridging techniques to harvest the vein.

The outcome of postoperative leg wound infection was defined as drainage of pus from the wound, documented infection with positive culture, or the requirement for additional surgical or medical treatment, like use of antibiotics.


There were 14 included trials with 1,527 patients. Trials were as large as 250 patients, and as small as 30. There were variable rates of conversion from minimally invasive to conventional harvesting, up to 22% in one trial. The follow up was variable, being as short as six days and as long as 42 days after the operation.

Leg wound infection rates varied between 3% and 30% (mean 13%) with conventional techniques, and 0% and 10% (mean 3%) with minimally invasive techniques (Figure 1). The number needed to treat with minimally invasive harvesting to prevent one leg wound infection that would have occurred with conventional harvesting was 10 (95% CI 8 to 14). The same result was found for larger and smaller trials (smaller trials had 65 patients or fewer), and for endoscopic harvesting (Table 1).

Figure 1: Infection rates for conventional and minimally invasive vein harvesting

Table 1: NNTs and sensitivity analysis for leg wound infections, comparing minimally invasive with conventional harvesting

Number of
Included studies
Relative risk
(95% CI)
(95% CI)
All studies
0.24 (0.16 to 0.36)
10 (8 to 14)
Larger studies
0.26 (0.16 to 0.43)
10 (8 to 15)
Smaller studies
0.21 (0.10 to 0.44)
9 (6 to 19)
Endoscopic only
0.24 (0.15 to 0.39)
10 (7 to 14)


Minimally invasive vein harvesting probably costs more and takes longer, but reduces postoperative leg wound infection significantly. Questions needing to be answered include whether it can be done in every institution, and whether it is cost-effective. Given that hospital acquired infection can cost an average of £3,150 and lead to an extra 14 days in hospital, as well as killing some patients, the answer looks obvious. More fun work for our health economists.


  1. H Kehlet, DW Wilmore. Multimodal strategies to improve surgical outcomes. American Journal of Surgery 2002 193: 630-641.
  2. T Athanasiou et al. Leg wound infection after coronary artery bypass grafting: a meta-analysis comparing minimally invasive versus conventional vein harvesting. Ann Thorac Surg 2003 76: 2141-2146.

previous story