Although the mu-opioid agonist morphine affects ventilatory control in men and women in different ways, no data exist regarding the influence of sex on the ventilatory effects of inhalational anesthetics. The authors compared the effect of sevoflurane on the ventilatory response to isocapnic hypoxia in healthy young men and women.
Breath-to-breath ventilatory responses to hypoxic steps (number of hypoxic steps, four-six; duration, 3 min; end-tidal oxygen tension, approximately 50 mmHg; end-tidal carbon dioxide tension clamped at approximately 4 mmHg above resting values) were assessed in nine men and nine women without and with low-dose sevoflurane (end-tidal concentration, 0.25%). The bispectral index of the electroencephalogram was measured concomitantly.
Sevoflurane reduced the hypoxic ventilatory sensitivity significantly in both sexes (men: control, 0.62 +/- 0.17 vs. sevoflurane, 0.38 +/- 0.19 l x min(-1) x %(-1); women: control, 0.52 +/- 0.30 vs. sevoflurane, 0.34 +/- 0.15 l x min(-1) x %(-1)). Sevoflurane-induced reductions of the hypoxic responses were not different in the men and women. During sevoflurane inhalation, the bispectral index values decreased equally in men and women.
In contrast to morphine, the influence of a low dose of the inhalational anesthetic sevoflurane on the ventilatory response to hypoxia is independent of sex.
RESPIRATORY depression is a serious side effect of anesthetics and opioids. [1,2] We showed previously that the [micro sign]-opioid receptor agonist morphine produces respiratory depression that differs quantitatively and qualitatively between men and women.  With respect to the ventilatory response to hypoxia, men showed a shift of response to lower baseline minute ventilation (Vi), with no change in the increase in ventilation caused by a 3-min isocapnic hypoxic challenge (the acute hypoxic response). Conversely, women showed, apart from a similar shift to lower baseline values of Vi, a 50% reduction of acute hypoxic response. We relate these differences in respiratory depression to sex-dependent genetic factors. Animal and human studies indicate that the occurrence of sex-related differences in opioid-mediated behavior further involves analgesia, motor behavior, memory, and addition. [3,4]
Whether the sex differences we previously observed are a unique feature of [micro sign]-receptor agonists or they also apply to respiratory depressants that do not act at opioid receptors is not known. Suggestions for sex-related differences in the influence of anesthetics on ventilation and the control of breathing are found in the literature: Warner et al.  showed possible sex differences in the pattern of respiratory muscle use in persons anesthetized with halothane during quiet breathing; and ethanol-induced depression of the hypoxic drive seems to depend on sex. 
In this study, we compared the influence of the volatile anesthetic sevoflurane, at a subanesthetic concentration, on the ventilatory response to 3-min isocapnic hypoxic pulses in men and women. Because sevoflurane, administered at an end-tidal concentration of 0.25%, causes respiratory depression and sedation,  we measured the bispectral index (BIS) of the electroencephalogram (EEG) to determine whether possible sex differences in the ventilatory responses could be related to the differences in the effect of sevoflurane on the central nervous system arousal state.
Participants and Apparatus
We recruited 20 healthy young persons (10 men and 10 women) to participate in this experimental protocol after we obtained approval from the Leiden University Medical Center Ethics Committee. All volunteers participated in previous studies and were familiar with the experimental procedures.  All the women reported normal menstrual cycles. Experiments were performed exclusively in the follicular phase of their cycles. When required, a pregnancy test was performed on the morning of the study. The participants were asked to refrain from using stimulants and depressants for at least 12 h before the study and were instructed not to eat or drink for at least 6 h before the study.
Before data collection was started, an intravenous catheter was inserted and electrodes for EEG measurement were placed. Subsequently, the volunteers rested for 30 min. Next, a mask (Vital Signs, Totowa, NJ) was fitted over the nose and mouth (a nose clip was not used). The inspired and expired gas flows were measured using a pneumotachograph connected to a pressure transducer and electronically integrated to yield a volume signal. Corrections were made for the alterations in gas viscosity resulting from changes in the oxygen concentration of the inhaled gas mixtures. The pneumotachograph was connected to a T-piece. One arm of the T-piece received a gas mixture from a gas-mixing system consisting of three mass-flow controllers (Bronkhorst High-Tec, Veenendaal, The Netherlands). A DEC PDP 11/23 micro-computer (Digital Equipment Co., Maynard, MA) provided control signals to the mass-flow controllers so the composition of the inspired gas mixtures could be adjusted to force the end-tidal pressures of oxygen and carbon dioxide (PETO(2) and PETCO(2), respectively) to follow a specified pattern in time. The oxygen and carbon dioxide concentrations of inspired and expired gases and the arterial hemoglobin oxygen saturation (Sp (O)(2)) were measured using a Datex Multicap gas monitor and Datex Satelite Plus pulse oximeter, respectively (Datex-Engstrom, Helsinki, Finland).
Electrodes (Zipprep; Aspect Medical Systems, Framingham, MA) were placed according to the 10/20 system for electrode placement at At1-FP(1) and AT2-FP2for bipolar recordings of the EEG. The EEG was recorded using an Aspect A-1000 EEG monitor. The monitor computed the BIS over 4-s epochs. We averaged the values during 1-min intervals and used the data points during the 1-min period before, and the third minute of hypoxia for further analysis.
The study began with the assessment of resting PETCO(2) levels during a 10- to 15-min period in which the participants breathed a normoxic gas mixture. Thereafter, the PETCO(2) was increased by approximately 4 mmHg. This value was maintained in the control and sevoflurane studies. The PETO(2) waveform was as follows:(1) min 0–3, 110 mmHg;(2) a stepwise decrease to 50 mmHg; and (3) min 3–6, 50 mmHg. This sequence was repeated three to five times (Figure 1 and Figure 2). One series of hypoxic pulses was performed without the inhalation of sevoflurane, and one series was performed while the participants inhaled 0.25% end-tidal sevoflurane (after reaching a 0.25% end-tidal concentration, a 12-min equilibration period preceded the sevoflurane studies). The sequence of the study (control-sevoflurane, sevoflurane-control) was randomized among the participants. Between control and sevoflurane studies, a 20-min resting period was allowed.
Data Analysis and Inclusion-Exclusion Criteria
The studies were evaluated by taking mean values of the last 10 breaths before hypoxia was induced and at the end of the 3 min of hypoxia. The difference between the hypoxic and preceding normoxic data point was determined to calculate the hypoxic ventilatory sensitivity or Delta Vi/DeltaSpO(2). Thus, we obtained four to six values of Delta Vi/DeltaSpO(2) per participant per treatment.
The volunteers were instructed to keep their eyes closed during the control and sevoflurane studies. Data inclusion criteria were based on the central nervous system arousal state of the participants, as determined by the EEG:(1) the state of wakefulness during control studies (BIS >or= to 94) and (2) a BIS level less than 94 during the sevoflurane studies. These values are based on the results of previous studies. 
Sevoflurane versus Control. To detect the significance of differences between the sevoflurane and control studies, a two-way analysis of variance (factors: subject and treatment) was performed on relevant variables and parameters (see Table 1). Separate analyses were performed in men and women.
Men versus Women. To detect any differences between men and women, a three-way analysis of variance (factors: subject, sex, and treatment) was performed on relevant variables and parameters (Table 1). A significant sex difference was assumed when the factor sex or the interactive term sex X treatment were significant.  Finally, a Student t test was used to analyze the male and female sevoflurane acute hypoxic response, expressed as a percentage of control and calculated from the mean values obtained from each participant.
Probability values < 0.05 were considered significant. All values reported are the mean of the participants +/- SD.
One woman did not complete the study because she experienced nausea. One man showed continuous irregular breathing during the control studies. Data for both volunteers were discarded. The anthropometric data of the remaining 18 participants showed no difference in age between men and women (men [n = 9], 26.6 +/- 6.9 yr vs. women [n = 9], 23.7 +/- 3.7 yr). However, men were taller and heavier (men, 77.7 +/- 8.7 kg and 182.6 +/- 6.0 cm vs. women, 65.3 +/- 8.8 kg [P < 0.05 by Student t test], and 168 +/- 5.7 cm [P < 0.05]).
In men, 52 control and 46 sevoflurane responses to hypoxic pulses were obtained. Six sevoflurane responses were discarded because of an increase in BIS more than 94. In women, 47 control and 46 sevoflurane responses were acquired. Four sevoflurane responses were omitted from the analysis because of arousal. Figure 1shows an example of an increase in BIS above 94 that occurred during the sevoflurane study of a male participant. In control and sevoflurane studies, the magnitude of the hypoxic response was time (or pulse number) independent (by analysis of variance). There was no significant effect of the sequence of studies (control-sevoflurane vs. sevofluran-control) on the magnitude of the hypoxic responses. Hypoxic arterial hemoglobin- oxygen saturation derived from the pulse oximeter showed a significant effect of sex (Table 1). The SpO(2) in women during sevoflurane inhalation differed by approximately 3% from women and men controls and men receiving sevoflurane. Sevoflurane reduced the BIS and hypoxic ventilatory sensitivities in men and women (Table 1, Figure 3). Figure 2shows examples of typical responses of a female participant. The hypoxic ventilatory sensitivities during sevoflurane inhalation were 63 +/- 25% of control in men, compared with 64 +/- 23% of control in women (difference not significant). The 95% confidence interval (CI) of the male hypoxic sensitivities during sevoflurane inhalation (defined as the ratio of the sevoflurane response over the control response) overlapped with the corresponding value for the women: 95% CI for men = 43–83% of control; 95% CI for women = 46–81% of control. Linear regression of the sevoflurane hypoxic sensitivities against the sevoflurane BIS failed to show a significant correlation (men: r =-0.35, P = 0.34; women: r =-0.40, P = 0.26).
A subanesthetic concentration of sevoflurane produced (1) an identical central nervous system arousal state in men and women as judged by the value of the BIS, (2) a reduction of the hypoxic ventilatory sensitivity of a similar magnitude in men and women, and (3) lower hypoxic saturations at similar PETO(2) and PETCO(2) levels in women compared with men.
Our decision to discard hypoxic responses when the BIS level was more than 94 during sevoflurane inhalation could be debated. It is our experience, and that of others, that BIS values more than 94 correspond with a state of wakefulness in which persons are alert and respond to commands. [2,8] In general, when BIS values are less than 94, participants are lethargic and need auditory stimuli to respond to commands. [2,8] Increases in BIS seen in the sevoflurane studies were often short-lived and related to an increase in ventilatory response to hypoxia (Figure 1) but not always to leg, arm, or head movement. It is noteworthy that the increase in breathing movement during hypoxia did not affect the participants' central nervous system arousal state.
In humans, low concentrations of inhalational anesthetics reduce the magnitude of the ventilatory response to isocapnic hypoxia by an effect within the peripheral chemoreflex loop, most likely at the carotid bodies. [9,10] The biologic mechanisms involved remain unknown. A small effect of low-dose anesthetics on the acute hypoxic response caused by a general depression of the central nervous system is not excluded.  Our finding of a 30–40% depression of the acute hypoxic response by sevoflurane, 0.25%, corresponds with results of a previous study from our laboratory.  In the current study, we could not demonstrate sex differences in the ability of sevoflurane to reduce the acute hypoxic ventilatory response. This indicates that the influence of sevoflurane on ventilatory control is not controlled by sex-dependent genetic factors. The possibility of the involvement of non-sex-related genetic determinants in the depression of ventilatory responses by anesthetics remains unknown and merits further study.
Our findings are in contrast with the observation that sex differences exist in the effect of morphine on ventilatory responses originating at the carotid bodies. [1,11] We relate the difference in outcome of this study using sevoflurane and our previous studies using morphine to the specific mechanisms by which opioids and anesthetics cause respiratory depression. Our respiratory studies then suggest the existence of sexual dimorphism in [micro sign]-opioid receptors (for example, sex differences in number, distribution, density, affinity, response to activation, interaction with other receptors, or all of these) in specific regions in the peripheral and central nervous system involved in the ventilatory control loop, but not of receptors through which volatile anesthetics exert their various effects, such as nicotinic acetylcholine or [Greek small letter gamma]-aminobutyric acid A receptors, [12,13] at least not in regions of the nervous system involved in the control of breathing.
A consistent finding in our study with strict control of PETO(2) and PETCO(2) was the lower SpO(2) values in women compared with men during the inhalation of sevoflurane (Table 1). To the best of our knowledge, this is the first time this observation has been reported. More studies are necessary to determine whether our findings can be reproduced, and, if they can, whether they are the result of a sex difference in the end-tidal-arterial oxygen tension gradient or in the arterial hemoglobin-oxygen saturation at identical arterial oxygen tensions. The former may be related to the observation in isolated sheep lungs of a sex difference in vasoconstrictive response to reduced inspired oxygen pressure (an attenuated response was observed in the lungs of postpubertal female sheep that was mediated by estradiol). [14,15] The latter suggests sex differences in the influence of anesthetics on the oxyhemoglobin-dissociation curve.