802 GOES ET AL. VAMS-tranquilization:Tense/relaxed item.Friedman's ing"in relation to "Before"(p=0.001).The treatment effect ANOVA revealed no significant differences among experi- was not significant (p=0.126). mental phases for groups SO25 (p=0.078)and SO10 (p=0.126),although differences were found for groups SOs CG/MS analysis.The volatile composition of sweet or- (p=0.044),TT(p=0.003),andH20(p=0.003). ange and tea tree essential oils is shown in Tables 4 and 5, respectively. VAMS-mental sedation.Friedman's ANOVA revealed no significant differences among test phases for all of the Discussion groups[SO2.5(p=0.327):S05(p=0.985:SO1o(p=0.060):TT (p=0.832):andH20(p=0.369)l. The aim of this work was to evaluate the potential anxi- olytic effect of sweet orange (C.sinensis)essential oil in VAMS-physical sedation.Friedman's ANOVA revealed healthy volunteers submitted to an anxiogenic situation.The no significant differences among test phases for the groups experimental model of anxiety,as employed here,was S02.5(p=0.615),S05(p=0.403),TT(p=0.138),andH20 shown to be valid,since the subjects from the control groups (p=0.271).However,differences were found for the group behaved as expected,significantly increasing their anxiety SOo(p=0.009).Separate analyses of the individual items of signs/symptoms during the Stroop task. this category,for this group,showed differences among test The obtained results showed the anxiolytic properties of phases for the lethargic/energetic item(p=0.008). sweet orange essential oil through the lack of a significant increase in state-anxiety and tension,and the lack of a sig- VAMS-other feelings and attitudes.Friedman's ANOVA nificant decrease in tranquility,presented by the individuals exposed to the aroma.Of the two doses(2.5 and 10 drops) revealed no significant differences for the groups SO2.5 (p=0.145),SOs(p=0.814),TT(p=0.110),andH20 that demonstrated anxiolytic activity,the higher one seemed (p=0.618).However,differences were found for the group to be better at preventing anxious symptoms,as its results SOo(p=0.027).Separate analyses of the individual items of were further from reaching statistical significance and were this category,for this group,showed differences among test observed in a greater number of parameters.It is reasonable phases for the withdrawn/gregarious item(p=0.002). to question the fact that the intermediate dose failed to produce an effect.However,this activity profile was ob- Physiologic measurements served before,when the same aroma was tested in animals.9 This lack of a dose/effect relation is common when dealing Data are presented as mean and standard error of the with a mixture of compounds instead of a pure substance. mean in Table 3. Interestingly,the observed anxiolytic effects were not followed by sedative/hypnotic effects.On the contrary,at Heart rate.The ANOVA revealed that the interaction the highest dose,sweet orange oil made the volunteers feel between test phase and treatment was not significant more energetic while performing the test.They also felt more (p=0.641);therefore the two main effects were analyzed.The introverted.It is tempting to speculate that these two test phase effect was significant (p<0.001),in that the heart symptoms together mean that the subjects were more con- rate was greater "During"in relation to "Before"(p<0.001). centrated on the task,which would be a welcome side-effect, The treatment effect was not significant(p=0.525). especially as tranquilizers tend to cause concentration loss. Nevertheless,this should be further investigated. Gastrocnemius electromyogram.The ANOVA revealed The GC/MS analysis determined that limonene corre- that the interaction between test phase and treatment was sponded to 54.48%of the volatile components of our sweet not significant (p=0.653);therefore the two main effects orange oil sample.Previous studies,performed with rodents, were analyzed.The test phase effect was significant using essential oils that were practically pure limonene (p<0.001),in that the muscular tension was greater "Dur- (>97%),9,20 have shown very clear anxiolytic effects, TABLE 3.SUMMARY OF PHYSIOLOGIC RESULTS Experimental phase Parameters Treatment Pretreatment Before During After Heart rate (beats/min) S025 76.7±3.5 70.5±3.2 108.8±9.4 77.7±3.7 SOs 71.5±45 69.9±3.6 95.7±7.6 72.1±5.1 S010 76.5±5.2 74.9±5.3 107.1±10.8 74.5±5.0 TT 80.3±7.0 79.5±7.4 127.1±20.8 75.1±4.9 H2O 69.5±2.9 69.9±3.9 103.1±10.3 70.6±2.5 Electromyogram of the S025 1.1±0.3 0.9±0.2 2.3±0.5 0.9±0.3 gastrocnemius muscle (uV) SO5 3.1±1.0 2.1±0.6 3.7±0.8 1.7±0.7 S010 1.5±0.4 0.8±0.2 3.6±1.5 1.0±0.3 TT 1.0±0.2 1.7±0.5 2.2±0.6 1.6±0.5 H2O 2.7±0.7 1.8±0.4 4.0±0.6 2.5±0.5 Data are presented as means+standard error of the mean. SO2,SOs,SOo represent essential oil of sweet orange 2.5,5,or 10 drops.TT,essential oil of tea tree.VAMS–tranquilization: Tense/relaxed item. Friedman’s ANOVA revealed no significant differences among experi￾mental phases for groups SO2.5 ( p = 0.078) and SO10 ( p = 0.126), although differences were found for groups SO5 ( p = 0.044), TT ( p = 0.003), and H2O ( p = 0.003). VAMS–mental sedation. Friedman’s ANOVA revealed no significant differences among test phases for all of the groups [SO2.5 ( p = 0.327); SO5 ( p = 0.985); SO10 ( p = 0.060); TT (p = 0.832); and H2O ( p = 0.369)]. VAMS–physical sedation. Friedman’s ANOVA revealed no significant differences among test phases for the groups SO2.5 ( p = 0.615), SO5 ( p = 0.403), TT ( p = 0.138), and H2O ( p = 0.271). However, differences were found for the group SO10 ( p = 0.009). Separate analyses of the individual items of this category, for this group, showed differences among test phases for the lethargic/energetic item ( p = 0.008). VAMS–other feelings and attitudes. Friedman’s ANOVA revealed no significant differences for the groups SO2.5 ( p = 0.145), SO5 ( p = 0.814), TT ( p = 0.110), and H2O ( p = 0.618). However, differences were found for the group SO10 ( p = 0.027). Separate analyses of the individual items of this category, for this group, showed differences among test phases for the withdrawn/gregarious item ( p = 0.002). Physiologic measurements Data are presented as mean and standard error of the mean in Table 3. Heart rate. The ANOVA revealed that the interaction between test phase and treatment was not significant ( p = 0.641); therefore the two main effects were analyzed. The test phase effect was significant ( p < 0.001), in that the heart rate was greater ‘‘During’’ in relation to ‘‘Before’’ ( p < 0.001). The treatment effect was not significant ( p = 0.525). Gastrocnemius electromyogram. The ANOVA revealed that the interaction between test phase and treatment was not significant ( p = 0.653); therefore the two main effects were analyzed. The test phase effect was significant ( p < 0.001), in that the muscular tension was greater ‘‘Dur￾ing’’ in relation to ‘‘Before’’ ( p = 0.001). The treatment effect was not significant ( p = 0.126). CG/MS analysis. The volatile composition of sweet or￾ange and tea tree essential oils is shown in Tables 4 and 5, respectively. Discussion The aim of this work was to evaluate the potential anxi￾olytic effect of sweet orange (C. sinensis) essential oil in healthy volunteers submitted to an anxiogenic situation. The experimental model of anxiety, as employed here, was shown to be valid, since the subjects from the control groups behaved as expected,11 significantly increasing their anxiety signs/symptoms during the Stroop task. The obtained results showed the anxiolytic properties of sweet orange essential oil through the lack of a significant increase in state-anxiety and tension, and the lack of a sig￾nificant decrease in tranquility, presented by the individuals exposed to the aroma. Of the two doses (2.5 and 10 drops) that demonstrated anxiolytic activity, the higher one seemed to be better at preventing anxious symptoms, as its results were further from reaching statistical significance and were observed in a greater number of parameters. It is reasonable to question the fact that the intermediate dose failed to produce an effect. However, this activity profile was ob￾served before, when the same aroma was tested in animals.9 This lack of a dose/effect relation is common when dealing with a mixture of compounds instead of a pure substance. Interestingly, the observed anxiolytic effects were not followed by sedative/hypnotic effects. On the contrary, at the highest dose, sweet orange oil made the volunteers feel more energetic while performing the test. They also felt more introverted. It is tempting to speculate that these two symptoms together mean that the subjects were more con￾centrated on the task, which would be a welcome side-effect, especially as tranquilizers tend to cause concentration loss. Nevertheless, this should be further investigated. The GC/MS analysis determined that limonene corre￾sponded to 54.48% of the volatile components of our sweet orange oil sample. Previous studies, performed with rodents, using essential oils that were practically pure limonene ( > 97%),9,20 have shown very clear anxiolytic effects, Table 3. Summary of Physiologic Results Experimental phase Parameters Treatment Pretreatment Before During After Heart rate (beats/min) SO2.5 76.7 – 3.5 70.5 – 3.2 108.8 – 9.4 77.7 – 3.7 SO5 71.5 – 4.5 69.9 – 3.6 95.7 – 7.6 72.1 – 5.1 SO10 76.5 – 5.2 74.9 – 5.3 107.1 – 10.8 74.5 – 5.0 TT 80.3 – 7.0 79.5 – 7.4 127.1 – 20.8 75.1 – 4.9 H2O 69.5 – 2.9 69.9 – 3.9 103.1 – 10.3 70.6 – 2.5 Electromyogram of the gastrocnemius muscle (lV) SO2.5 1.1 – 0.3 0.9 – 0.2 2.3 – 0.5 0.9 – 0.3 SO5 3.1 – 1.0 2.1 – 0.6 3.7 – 0.8 1.7 – 0.7 SO10 1.5 – 0.4 0.8 – 0.2 3.6 – 1.5 1.0 – 0.3 TT 1.0 – 0.2 1.7 – 0.5 2.2 – 0.6 1.6 – 0.5 H2O 2.7 – 0.7 1.8 – 0.4 4.0 – 0.6 2.5 – 0.5 Data are presented as means – standard error of the mean. SO2.5, SO5, SO10 represent essential oil of sweet orange 2.5, 5, or 10 drops. TT, essential oil of tea tree. 802 GOES ET AL