Fig 1.
A common reason for declining trial participation was that long travel distances made patients unwilling to return to the hospital for the 1-year follow-up visit. Otosclerosis is a clinical diagnosis, and 4 patients were found at surgery to have middle ear disorders where stapedotomy is not indicated. One patient in the placebo group was excluded in violation of the protocol because of a fracture of the stapes footplate. A total of 5 patients never received the study drug because their operations were postponed after study drugs had been delivered to the ward (but before drug infusion started).
Table 1.
Demographic characteristics, hearing, and surgical factors in study patients.
Fig 2.
Hearing thresholds of the placebo group and the N-Acetyl cysteine group (“Treated group”) at baseline.
Thresholds for air-conducted sound (lower set of points) are measured by standard headphones covering the ears; thresholds for bone-conducted sound (top data set) are determined by placing a vibrator on the mastoid process. Bone-conducted sound was not measured at frequencies higher than 4000 Hz as this is not part of normal clinical routine. Rings represent the mean values; vertical lines are 95% confidence intervals for the mean. Hearing was measured at the same set of frequencies in all patients, but a horizontal offset was added when plotting to improve the clarity of presentation.
Fig 3.
(Upper graph) Change in hearing thresholds 6 weeks after surgery (as compared to measurements before the operation).
Hearing was measured at the same set of frequencies as in Fig. 2. The size of each dot corresponds to the number of patients, obtained from the histogram for each frequency (5 dB bins) and boxes enclose the 95% confidence interval. Horizontal lines show the mean value at each frequency. One patient with complete sensorineural hearing loss after surgery is omitted from the plot (but included in all statistical analyses). (Lower graph). Change in bone-conduction thresholds at 6 weeks. The scaling of the dots is the same in both panels.
Fig 4.
Change in hearing thresholds one year after surgery (as compared to measurements obtained before the operation). The size of each dot corresponds to the number of patients (obtained from the histogram for each frequency; 5 dB bins). Boxes enclose the 95% confidence interval and the horizontal lines give the mean at each frequency. One patient with complete sensorineural hearing loss after surgery is omitted from the plot. (Lower graph). Change in bone-conduction thresholds at 1 year. Dot scaling is identical in both graphs.
Table 2.
Changes in scores for tinnitus, dizziness, and other variables.