BACKGROUND: Monitoring of PETCO2 in the patients during regional anesthesia may be no less important than under general anesthesia, but will aid in early detection of potentially catastrophic events. However, the utility and accuracy of capnography in non-intubated patients has received little attention. We examined correlation between PETCO2 measured via nasal cannula and PaCO2 values in the sedated spontaneously breathing patients during spinal anesthesia. METHODS: Thirty adult patients who underwent elective surgery were administered optimal doses of tetracaine and epinephrine mixture in their site of operation, length, weight and age. Thereafter, we sampled expired gas by 175ml/min and administered oxygen by 3L/min using oxygen delivery CO2 sampling nasal cannula. End tidal carbon dioxide tension, heart rate, blood pressure and respiratory rate were measured before and 20 min after 0.02 mg/kg midazolam i.v.. And arterial blood gases were once measured 20 min after 0.02 mg/kg midazolam i.v.. RESULTS: The patients, sedation state was asleep or calm in awake. End tidal carbon dioxide tension was significantly increased after midazolam injection (p<0.01), but another values were not different after midazolam injection. Linear regression analysis of arterial carbon dioxide tension and end tidal carbon dioxide tension after midazolam injection yielded y = 0.77x + 4.82 and r2 = 0.76 (p<0.01). CONCLUSIONS: End tidal carbon dioxide tension using oxygen delivery CO2 sampling nasal cannula in the sedated spontaneously breathing patients with midazolam during spinal anesthesia were significantly related with arterial carbon dioxide tension. Therefore, we conclude that monitoring of PETCO2 via nasal cannula is a reliable means during spinal anesthesia.