3 In the systematic review by Balk et al.,2 published after the three meta-analyses, the authors reviewed all uncontrolled and controlled data in total. The authors identified 2 RCTs, 8 comparative studies and 25 cohort studies and found that when considering all evidence there was a better BP reduction (8 mmHg) in the angioplasty versus medical treatment arm. However, the studies were uncontrolled buy SB431542 and non-randomized so many methodological issues existed in the majority and in particular, there was the suggestion that the ‘intensive

medical therapy’ was not equal between the groups. In addition, the combined adverse event rates included death by 30 days which was 3% with the other complications of transient deterioration in kidney function

of up to 13%, renal artery injury of 5% and peri-procedural cardiovascular system (CVS) events of 3%. Thus, one can conclude that the review does not favour one treatment modality, that there is weak evidence for similar CVS outcomes and the small improvement in BP (mainly in bilateral renal disease) is likely outweighed by the morbidity. Leertouwer et al.9 performed a meta-analysis of renal arterial stent placement in comparison with renal angioplasty in patients with RAS, including studies published up to August 1998. This systematic review did not report on the quality of the studies as did Balk et al.2 and included uncontrolled BKM120 VAV2 studies. It suggested that stents are better but is very weak in the quality of its conclusions because of the uncontrolled nature of the data it surveyed. Despite achieving changes in arterial patency,

none of the four studies mentioned above has shown significant advantage in slowing renal progression through renal angioplasty over and above conventional medical therapy. Interpretation is limited by the fact that each of these studies has focused on patients with hypertension rather than those with documented progressive renal impairment. In the ASTRAL study the rate of progression of renal impairment (as shown by the slope of the reciprocal of the serum creatinine level) was −0.07 × 10−3 L/µmol per year in the revascularization group, compared with −0.13 × 10−3 L/µmol per year in the medical therapy group, a nonsignificant difference favouring revascularization of 0.06 × 10−3 L/µmol per year (95% confidence interval, −0.002–0.13; P = 0.06).3 This nonsignificant trend is weakened by the fact that the number of patients able to be reported on at 5 years was 72 (revascularization) versus 61 (medical).