Statistical analyses were performed with SAS 9.1.3 (SAS Institute, Cary, NC) and Statistica 7.1 (StatSoft, Tulsa, OK). Age and level of education were comparable in patients and healthy volunteers; the number of males was higher in the patients group (90 versus 50%; χ2 3.8, P = 0.05). Six patients were classed as Child A and 4 as Child B; average MELD 5-Fluoracil nmr was 9.1 (2.2). None of the patients or the healthy volunteers had abnormal BMI. Four (40%) patients and two (20%) healthy volunteers had initial
muscle mass depletion. Quality of life was impaired in patients compared to healthy volunteers [SF36-Physical: 41.1 (9.6) versus 50.4 (7.7), P = 0.02; Table 1]. No significant differences were observed in diurnal preference/daytime sleepiness between the two study groups. As expected based on patients selection, subjective sleep quality (PSQI questionnaire) was also comparable, whereas actigraphy documented significantly lower sleep efficiency in the patients [68.5 (15.9) versus 81.0 (9.0)%, P = 0.05; Table 1]. On average, habitual sleep timing was delayed in patients compared to healthy volunteers, although the differences were not significant (Table 1). Both groups exhibited
fluctuations in subjective sleepiness in the course of the day, with a peak in the early afternoon (Fig. 2). All healthy volunteers and patients had normal PHES and Scan test performances at baseline. However, patients scored worse on both measures (Table 2). All healthy volunteers and nine patients had a normal wake EEG; one patient had grade I EEG slowing according BGJ398 nmr to Amodio et al.28 On average, patients had a significantly slower EEG than healthy volunteers (Table 2). Eight healthy volunteers and eight patients reached consolidated non-REM sleep during the nap opportunity in both experimental conditions, thus nap EEG analysis was limited to these subjects. Patients slept significantly longer than
healthy volunteers [51.0 (14.5) versus 30.4 (15.6) minutes; P = 0.02; Table 2]. However, the length of the solid blocks of non-REM sleep selected for spectral analysis was comparable in the two groups (Table 2). Arachidonate 15-lipoxygenase Power spectra (1.5-25 Hz) of the baseline nap EEG were comparable in patients and healthy volunteers [factor group (patients versus healthy volunteers) not significant; Fig. 3]. Six (60%) patients and five (50%) healthy volunteers were randomized to receive AAC on study day 4, whereas the remainder received it on study day 8. The AAC was well tolerated, although three healthy volunteers and two patients complained of nausea. Patients had significantly higher baseline ammonia levels compared to healthy volunteers [202 (61) versus 147 (45) μg/dL, P = 0.03]. The AAC resulted in a significant increase in ammonia levels in both groups, with highest values at approximately 4 hours after the AAC (Fig. 2).