Background Acceleromyography is regularly used as an isolated test to detect residual paralysis. The performance of acceleromyography, however, has not been investigated for the setting where calibration is impossible. This study first evaluated the reliability of a single acceleromyographic train-of-four (TOF) ratio (T4/T1) to detect residual paralysis and compared it with tactile estimation of fade after double-burst stimulation and 100-Hz, 5-s tetanus. The second part of the study investigated whether uncalibrated acceleromyographic TOF ratio can predict time to complete recovery. Methods Anesthesia was induced and maintained with propofol and sufentanil. In the first part of the study (n = 40) neuromuscular blockade was assessed by mechanomyography. After signal stabilization 0.15 mg/kg cisatracurium was given. At the end of surgery a first physician evaluated manual fade after double-burst stimulation, then, in the same patient, a single acceleromyographic TOF ratio was recorded; thereafter a second physician, unaware of the results, assessed fade after a 100-Hz, 5-s tetanus. Sensitivity, specificity, and negative and positive predictive value of the three tests to detect a mechanomyographic TOF > or =0.9 were calculated. In the second part of the study (n = 25) neuromuscular recovery was assessed simultaneously with mechanomyography and uncalibrated acceleromyography (current set manually at 60 mA); the time intervals from acceleromyographic TOF ratios of 0.6, 0.7, 0.8, and 0.9 until complete recovery, i.e., adductor pollicis mechanomyography 0.9 TOF ratios, were determined. Results The sensitivity of double burst stimulation was 29% (95% confidence interval [CI], 13-45%), its specificity was 100%, the negative predictive value was 29% (95% CI, 13-45%), and the positive predictive value was 100%. For a single acceleromyographic TOF ratio the respective values were 70% (95% CI, 54-86%), 88% (95% CI, 67-100%), 47% (95% CI, 23-71%) and 95% (95% CI, 86-100%). The respective values for 100-Hz, 5-s tetanus were 74% (95% CI, 59-89%), 55% (95% CI, 23-88%), 38% (95% CI, 12-64%), and 85% (95% CI, 72 -99%). At an uncalibrated acceleromyographic TOF ratio was 0.6, complete recovery occurred within 16 min (95% CI, 13.5-17.8 min). At acceleromyographic TOF ratios of 0.7, 0.8, and 0.9 this time interval was 12.5 min (95% CI, 10.2-14.8 min), 8 min (95% CI, 6.1-9.9 min), and 4 min (95% CI, 2.7-5.8 min), respectively. Conclusions Acceleromyographic TOF performed better than double-burst stimulation or 100 Hz tetanus, but it did not reliably detect low degrees of residual paralysis when used as an isolated test at the end of surgery. The uncalibrated acceleromyographic TOF ratio, however, did predict the time to complete recovery.