This study investigated the effectiveness of ethanol as an early indicator of resorption during transcervical endometrium resection (TCRE). Ethanol was added as a marker and exhaled alcohol concentrations were measured with an Alcomed test. Ninety-three patients underwent endometrial ablation under general anaesthesia and exhaled ethanol levels and blood sodium concentrations were monitored simultaneously every 3 minutes. In 23% of the cases the Alcomed test was the first indicator of absorption. Relevant changes in blood sodium concentrations appeared later in 9% of the cases and gives a positive Alcomed test. The predictive exhaled ethanol value for preventing the transurethral prostate resection (TUR) syndrome is 1.5 promille. A marked difference in fluid resorption during hysteroscopic myomectomy was observed (400-1000 mi during TCRE and myomectomy vs. 100-300 mi TCRE only). Of 83 TCRE patients, 23 had a positive Alcomed test. Ten patients underwent a hysteroscopic myomectomy. Sensitivity and specificity were determined for a decrease of more than 5 mmol l(-1) in blood sodium and for several ethanol levels. The sensitivity was 100% for an ethanol level of at least 0.05 promille. The specificity was 90.9% and the positive predictive value 41.7%. All patients with a decrease of more than 5 mmol l(-1) in blood sodium concentration were detected using a threshold of 0.05 promille exhaled ethanol and nine out of 10 patients with a blood sodium decrease of at least 5 mmol l(-1) also had an ethanol level of at least 0.05 promille. Unfortunately a positive Alcomed test has a poor prognostic value.