OBJECTIVETo compare the procedural and in-hospital outcomes in a large series of diabetic and non-diabetic patients undergoing selective percutaneous coronary intervention (PCI) and to evaluate the influence of diabetes mellitus on the procedural and in-hospital outcomes.

METHODS1294 consecutive patients underwent selective PCI from January to December 2002 in this institution were analyzed retrospectively. Baseline clinical, in-lab and in-hospital outcome information were recorded. Rates of procedural success, device success and clinical success were analyzed and logistic regression was performed to model the association between diabetes status and outcomes.

RESULTSTwo hundred and sixty-nine patients (20.8%) complicated with diabetes. Type C lesion, double and triple vessel diseases were more prevalent in diabetics than those in non-diabetics. The pre-PCI diameter stenosis of diabetics was significantly more severe than that of non-diabetics (91.00 +/- 6.62 vs 89.81 +/- 6.64, P < 0.01). The balloon length, maximum balloon diameter and maximum balloon inflation pressure, maximum inflation duration were larger in diabetics than those in non-diabetics [(17.07 +/- 6.31) mm vs (16.07 +/- 7.28) mm, (2.30 +/- 1.11) mm vs (2.12 +/- 0.94) mm, (9.86 +/- 4.40) atm vs (9.05 +/- 4.75) atm, (20.94 +/- 14.69) s vs (18.26 +/- 14.65) s, respectively, P < 0.05]. The stent diameter was smaller in diabetics than that in non-diabetics [(3.15 +/- 0.47) mm vs (3.23 +/- 0.43) mm, P < 0.05]. The procedural success rate showed no significant difference between two groups (89.6% vs 90.3%, P > 0.05). But a higher incidence of acute/subacute stent thrombosis was observed in diabetics compared with that in non-diabetics (1.9% vs 0.5%, P < 0.05). The rate of clinical success was similar between diabetics and non-diabetics (99.3% vs 99.2%, P > 0.05). Diabetes was not an independent predictor of acute outcomes in the regression model.

CONCLUSIONSA higher incidence of acute/subacute stent thrombosis was observed in diabetics. The incidence of procedural and in-hospital major adverse cardiac events and the rate of clinical success were similar between diabetics and non-diabetics. Diabetes was not an independent predictor of in-hospital outcomes after selective PCI.