Transcutaneous oximetry has been used to define the level of amputation in arteritic patients, with discrepant results. We have studied preoperative statix oximetry in 33 arteritic patients at the Leriche-Fontaine Stage IV, who underwent 36 amputations (thighs = 6, Legs+Symes = 14, transmetatarsal = 7, toes = 9). Oximetry included the measurement, at the level of amputation, of the transcutaneous partial oxygen pressure (tc pO2), of the tissue oxygenation ratio (TOR), preferably with a precordial electrode, of the tc pO2 gain after oxygen inhalation and of the gain ratio with the reference electrode. Two patients died postoperatively, the amputation stump did not heal in another 8 patients. The tc pO2 value was 36.6 +/- 16.2 mm Hg in the healed group and 21.1 +/- 19.9 mm Hg in the non healed group. The TOR respectively was 71.2 +/- 32.7% and 38.6 +/- 29.9% in these two groups. These differences were statistically significant. The difference between the two groups in the measurements made after the oxygenation test was not statistically significant. With a tc pO2 threshold at 26 mm Hg, the sensitivity was 73% and the specificity 75%, the positive predictive value 90%, the negative predictive value 46% and the value of the test 73.5%. The thresholds calculated to be 56% for TOR, 11 mm Hg for the gain and 32% for the gain ratio, did not improve the performances of oximetry. In our study, the tc pO2 was the parameter that best allowed predicting healing was obtained with smaller values than the threshold. Other elements such as the general condition and diabetes have played a role in the prognosis.