Twenty-two studies were included in the review (n=1,270 patients, range seven to 292): one randomised controlled trial (RCT), 10 prospective non-randomised studies and 11 retrospective non-randomised studies. One retrospective study (n=183) was excluded from the pooled analyses due to the possibility of an overlap in population with another study.
Overall success rate of percutaneous closure was 92% (95% CI 90.1 to 93.9) with a range of 65% to 100%. Success rate for totally percutaneous closure (where percutaneous closure was attempted and successful without open dissection in either groin) was 79% (95% CI 74.9 to 83.1; 13 studies).
A significantly better success rate with ProGlide (93.7%, 95% CI 91.5 to 95.9; three studies) was found compared with Prostar (90%, 95% CI 88.4 to 91.6; 19 studies). No significant difference was found in success rate between studies that considered all comers and those that excluded patients with femoral artery calcification and groin scarring. Three studies reported a significant difference in success rate with increasing sheath size. Failure was attributed to patients with scarred groins in one study; another study found no association between success rate and previous catheterisation. One prospective study found a significant association with failure and obesity and another demonstrated no correlation of failure to obesity.
Wound complication rates ranged from 2% to 26% and the overall access-related complication rate was 4.4% (95% CI 3.5 to 5.3). P-EVAR was associated with fewer access-related complications compared with patients who underwent standard open groin arterial exposure (RR 0.47, 95% CI 0.28 to 0.78; seven studies); the authors reported that no significant statistical heterogeneity was found. Three studies (one RCT and two retrospective studies) found an increase in access-related complication rate in P-EVAR with sheath size. One retrospective study demonstrated a significant association of complications with large sheath size and obesity.
Mean hospital stay was 2.07 days (seven studies). Hospital stay was shorter in patients who underwent P-EVAR compared with EVAR (weighted mean 2.7 versus 3.5 days; three studies), although only one prospective study found a statistically significant difference.
P-EVAR was associated with significantly less operative time than open groin exposure (weighted mean 106 versus 145 minutes); operative time was reported to be significantly shorter in five of the eight studies.
The RCT found no significant between group differences (P-EVAR, open EVAR and failed P-EVAR) in blood loss. Two prospective studies reported similar blood loss between P-EVAR and open EVAR, but greater blood loss with failed P-EVAR. One other prospective study found significantly less blood loss with P-EVAR compared with open EVAR.
One RCT and one prospective study found ambulation time was shorter with P-EVAR compared with open EVAR. One retrospective study reported ambulation in 81% of patients four to six hours following P-EVAR.