Unlike femoral access, the trans-radial approach to coronary angiography requires a greater learning curve for the inexperienced interventionist, and demands a greater technical challenge than the femoral approach, which leads to a high incidence of failures and complications, mainly in the initial stages of their formation. Some explanations for procedural failure include inability for radial artery puncture, vessel tortuosity, and radial spasm. Radial spasm has been reported from 5 to 30% in experienced centers, with a decrease to 2.5% when associated with the administration of nitrate added to local anesthesia. Although radial spasm is infrequent and usually occurs once the artery is cannulated or during manipulation of the catheters, it can sometimes occur before cannulation due to multiple unsuccessful attempts to cannulate the artery, which can also produce pain in the patient.

The investigators will conduct an experimental, placebo-controlled, longitudinal, prospective, double-blind, parallel arm clinical trial. Patients will be randomized to receive 1:1 an intervention or placebo. Intervention group will receive a subcutaneous "cocktail" with 0.5 ml of 500 mcg of nitroglycerin + 1 ml of 2% simple lidocaine. Placebo group will receive a subcutaneous injection with 0.5 ml of 0.9% saline solution + 1 ml of 2% simple lidocaine.

The investigators hypothesize that the maneuver (subcutaneous infiltration of local anesthetic together with nitroglycerin) will facilitate radial access in terms of fewer punctures until the insertion of the arterial introducer, shorter access time, lower incidence of radial spasm, and lower crossover rate to femoral access, compared with placebo. In terms of safety, the investigators do not believe that the maneuver represents a risk to the patient in terms of incidence of hypotension or headache. In fact, the investigators believe that the maneuver will provide less pain at the site of puncture referred by the patient due to the less number of unsuccessful punctures.

For the calculation of the required sample, since the incidence of our primary end-point of a compound of crossover to femoral access and/or access site-related vascular complications has not been reported so far, the sample will be calculated with the rate of events obtained in our center at 6 months of recruiting patients. For this calculation, the proportional difference formula with an alpha value of 0,05 and a beta value of 0,20 will be used to have a statistical power of 80%. The investigators estimate that 1,500 patients will be required in a 11-month period.