- Email: [email protected]
Influence of volatile anaesthetics on hypercapnoeic ventilatory responses in mice with blunted respiratory drive†
British Journal of Anaesthesia 92 (5): 697±703 (2004)
DOI: 10.1093/bja/aeh124
Advance Access publication March 5, 2004
LABORATORY INVESTIGATIONS In¯uence of volatile anaesthetics on hypercapnoeic ventilatory responses in mice with blunted respiratory drive² H. Groeben1,2*, S. Meier2, C. G. Tankersley1, W. Mitzner1 and R. H. Brown1 1
Department of Environmental Health Sciences/Division of Physiology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, USA. 2University of Essen, Essen, Germany *Corresponding author: Department of Anesthesiology and Critical Care Medicine, University of Essen, Hufelandstrasse 55, 45122 Essen, Germany. E-mail: [email protected]
Methods. Using whole body plethysmography, we assessed respiratory rate (RR) and pressure amplitude in 11 male C3 mice at rest, during anaesthesia with iso¯urane, sevo¯urane or des¯urane, and during recovery. To test respiratory drive, mice were exposed to 8% carbon dioxide. Data were analysed by two-way-analysis of variance with post hoc tests and Bonferroni correction. Results. RR was unaffected during sevo¯urane anaesthesia up to 1.0 MAC. Likewise, sevo¯urane at 1.5 MAC affected RR less than either iso¯urane (P=0.0014) or des¯urane (P=0.0048). The increased RR to a carbon dioxide challenge was blocked by all three anaesthetics even at the lowest concentration, and remained depressed during recovery (P<0.0001). Tidal volume was unaffected by all three anaesthetics. Conclusions. In C3 mice, spontaneous ventilation was less affected during sevo¯urane compared with either iso¯urane or des¯urane anaesthesia. However, the RR response to hypercapnoeia was abolished at 0.5 MAC for all the anaesthetic agents and remained depressed even at the end of recovery. Our data suggest that different volatile anaesthetics have varying effects on the control of breathing frequency but all block the respiratory response to carbon dioxide. Therefore, a genetic predisposition to a blunted carbon dioxide response represents a susceptibility factor that interacts with hypercapnoeic hypoventilation during maintenance of anaesthesia and in the emergence from anaesthesia, regardless of the agent used. Br J Anaesth 2004; 92: 697±703 Keywords: anaesthetics volatile; complications, hypercapnia Accepted for publication: December 23, 2003
It is well established that volatile anaesthetic agents directly affect the neural control of breathing. At clinically relevant concentrations, volatile anaesthetics alter the magnitude and pattern of breathing.1±5 In addition, subanaesthetic concentrations of volatile anaesthetics, as low as 0.1 MAC, may signi®cantly diminish the response to hypoxia and hypercapnoeia.6±9
Clinically, the respiratory depressant effects of volatile anaesthetic agents may contribute to postoperative morbidity and mortality through decreased responsiveness to hypoxia or hypercapnoeia. This is of particular concern in ² Presented in part at the ATS annual meeting, Atlanta, May 18±22, 2002.
Ó The Board of Management and Trustees of the British Journal of Anaesthesia 2004
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
Background. Subanaesthetic concentrations of volatile anaesthetics signi®cantly affect the respiratory response to hypoxia and hypercapnoeia. Individuals with an inherited blunted respiratory drive are more affected than normal individuals. To test the hypothesis that subjects with blunted hypercapnoeic respiratory drive are diversely affected by different anaesthetics, we studied the effects of three volatile anaesthetics on the control of breathing in C3H/HeJ (C3) mice, characterized by a blunted hypercapnoeic respiratory response.
Groeben et al.
Methods Animals Our study protocol was approved by the Johns Hopkins Animal Care and Use Committee. Eleven male C3H/HeJ mice were purchased from Jackson Laboratories (Bar Harbor, ME, USA) and housed in the animal facilities at Johns Hopkins University for 5±6 weeks prior to the ®rst experiments. Water and chow (Agway Pro-Lab RMH 1000) were provided ad libitum. All experiments were performed at the age of 14 weeks (12 days) and a mean weight of 32.8 (3.4) g, at the same time of day for each animal.
Measurements Whole body plethysmograph
Ventilatory function was assessed by whole body plethysmography under unrestrained conditions.15 16 Each animal was permitted to acclimatize in a barometric, acrylic, cylindrical chamber (400 cm3) for at least 30 min before ventilation measurements were obtained. The chamber was placed in an insulated box to control environmental conditions (i.e. quiet and dark surroundings) and to prevent heat loss. The chamber temperature was maintained within the thermoneutral zone for mice of 26±28°C [i.e. 26.9 (0.8)°C] and each measurement was made with a type T thermocouple device. The inspired gas (FIO2=1.0) was humidi®ed (90% relative humidity) and directed through the chamber at a ¯ow rate of 300 ml min±1.
At constant chamber volume, changes in pressure due to inspiration and expiration were measured using a differential pressure transducer (model 8510B-2; Endevco, San Juan Capistrano, CA, USA) and recorded on a strip chart recorder (Grass Polygraph model 7D; Grass Instrument, Quincy, MA, USA). The out¯ow air from the chamber was analysed for oxygen, carbon dioxide and the respective volatile anaesthetic (iso¯urane, sevo¯urane, or des¯urane) using a Capnomac Ultra ULT-V-27±06 (Datex, Helsinki, Finland). Protocol
In randomized order, all animals received all three anaesthetic agents on different days. After the animal became quiescent, the chamber was sealed for 60 s to permit measurements of respiratory rate (RR) and the pressure amplitude (Amp; in mm) resulting from each tidal breath as a correlate of the tidal volume. The product of RR and Amp was calculated as an equivalent of minute ventilation (MV). While the chamber was sealed, metabolically induced changes in carbon dioxide and oxygen were <1% of the target level. Subsequently, an inspired gas mixture (FICO2=0.08; Puritan-Bennett Medical Gases, Overland Park, KS, USA) was administered to the chamber for 6 min. At the end of this carbon dioxide challenge, measurements of RR and Amp and calculation of MV were repeated as described above. After these baseline measurements, iso¯urane, sevo¯urane (Abbott Laboratories, North Chicago, IL, USA) or des¯urane (Baxter Healthcare, Deer®eld, IL, USA) was added to the hyperoxic inspirate via an agent-speci®c vaporizer integrated in the gas ¯ow system. Each anaesthetic was administered on a different day in random order with at least 1 day between study days for each mouse. Each volatile anaesthetic was administered in increasing concentrations of 0.5, 1.0, and 1.5 MAC (minimal alveolar concentration required to produce immobility in response to noxious stimuli) speci®c for C3H/HeJ mice. The MAC for C3H/HeJ mice is 1.5% for iso¯urane, 2.4% for sevo¯urane and 8.6% for des¯urane.18 Sevo¯urane and des¯urane were administered for 20 min at each concentration step, while iso¯urane was administered for 30 min at each concentration to achieve a steady state of anaesthesia. After reaching steady state, measurements of RR and Amp were repeated and a carbon dioxide challenge followed, as described above. Immediately after the carbon dioxide challenge at the highest concentration of the anaesthetic, the anaesthetic agent was turned off and the mice were observed during recovery from anaesthesia for the next 20 min (sevo¯urane and des¯urane) or 30 min (iso¯urane). During recovery, measurements were performed at 5, 10, 20 and 30 min for iso¯urane (RI-IV) and at 5, 10, 15 and 20 min for sevo¯urane and des¯urane (RI-IV). An example of the original recording is presented in Figure 1. At the end of
698
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
individuals with compromised respiratory drive, such as patients with chronic obstructive airway disease and obstructive sleep apnoea.10±14 Several studies indicate that there are possible differences in respiratory depressant effects among the currently available volatile anaesthetic agents.1 4 6 8±11 C3H/HeJ (C3) inbred mice are characterized by a blunted response to hypoxia and hypercapnoeia.15 16 Thus, C3 mice serve as a genetic model of individuals (mice) with blunted chemosensitivity. To elucidate possible differences in the deleterious effects of various volatile anaesthetic agents on respiratory function in individuals with blunted chemosensitivity, we studied the effects of volatile anaesthetics in C3 mice. We have already demonstrated that the genetically determined difference in respiratory drive of the C3 mice compared with mice with an undisturbed respiratory drive was enhanced during recovery from iso¯urane anaesthesia.17 However, the question remains whether shorter-acting volatile anaesthetics can avoid the postanaesthetic depressed response to respiratory stimuli such as hypercapnoeia. Therefore, the present study compared the effects of iso¯urane, sevo¯urane and des¯urane on the magnitude and pattern of spontaneous breathing at rest and during acute hypercapnoeic challenge in C3 mice.
Volatile anaesthetics and respiratory drive
recovery a ®nal carbon dioxide challenge was performed and subsequently the mice were returned to their cages.
Data analysis Data are presented as mean (SD). Measurements of RR, Amp and MV before and after carbon dioxide challenge were analysed by two-way analysis of variance (ANOVA) with post hoc tests and Bonferroni correction for multiple comparisons, controlling for anesthetic agent and concentration. Signi®cance was considered to be P<0.05. We tested the null hypotheses that (i) the effects of the three agents on RR, Amp and MV are the same and (ii) the effects of the three agents on the response to hypercapnoeia are the same.
Results A total of 11 mice were studied. All mice received all three anaesthetic agents and completed the study. At baseline, there was no difference in RR, Amp or MV between the mice on different days.
Fig 2 Respiratory rate (RR) and pressure amplitude (Amp), as an expression of tidal volume, in 11 male C3H/HeJ mice measured at baseline and during iso¯urane, sevo¯urane or des¯urane anaesthesia at 0.5, 1.0 and 1.5 MAC. Data are mean and SD. Four additional measurements were performed during recovery (RI±RIV) at 5, 10, 15 and 20 min for iso¯urane and des¯urane and 5, 10, 20 and 30 min for iso¯urane anaesthesia. During sevo¯urane anaesthesia, RR was preserved up to 1.0 MAC but RR was signi®cantly depressed at 1.0 MAC during anaesthesia with either iso¯urane (P<0.0001) or des¯urane (P<0.0001). The increase in Amp for iso¯urane at 1.5 MAC and RI (P<0.0001) could compensate only partly for the decrease in RR, while there were no changes in Amp for sevo¯urane and des¯urane. *=Signi®cant difference for iso¯urane vs sevo¯urane; #=signi®cant difference for iso¯urane vs des¯urane; §=signi®cant difference for sevo¯urane vs des¯urane (P<0.0167).
Effect of iso¯urane, sevo¯urane and des¯urane on ventilation at rest Iso¯urane
At 1.0 and 1.5 MAC of iso¯urane, RR decreased signi®cantly and dose-dependently compared with RR at baseline and 0.5 MAC. Subsequently, RR increased signi®cantly during recovery from anaesthesia (P<0.0001; Fig. 2). Amp increased signi®cantly at 1.5 MAC and after 5 min of recovery (P<0.0001). Des¯urane
At 1.0 and 1.5 MAC of des¯urane, RR decreased signi®cantly and dose-dependently compared with RR at baseline
699
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
Fig 1 Original recordings of one C3 mouse before and during iso¯urane anaesthesia. Six seconds of each step and one carbon dioxide challenge are shown. Carbon dioxide at baseline led to deepening of the breath and an increase in respiratory rate. Increasing concentrations of iso¯urane (without additional carbon dioxide) affected respiratory rate dramatically.
Groeben et al.
and 0.5 MAC. Subsequently, RR increased signi®cantly during recovery from anaesthesia (P<0.0001; Fig. 2). Amp did not change signi®cantly. Sevo¯urane
Effect of volatile anaesthetics on the response to hypercapnoeia At baseline, the hypercapnoeic challenge caused a signi®cant increase in RR, Amp and MV (P<0.0001; Figs 3 and 4). All three inhalation anaesthetics attenuated the tachypnoeic response to hypercapnoeia (Fig. 4). Even after recovery from all three anaesthetics, the tachypnoeic response was completely blocked (P<0.0001). While there were varying, albeit small, changes in Amp to hypercapnoeia (a slight dose-dependent decrease during des¯urane anaesthesia; Fig. 4) at the end of recovery, there were no signi®cant differences (Fig. 4). In particular, there was no difference in
Fig 3 The product of respiratory rate (RR) and pressure amplitude (Amp), as an equivalent of minute ventilation, of 11 male C3H/HeJ mice measured at baseline and during iso¯urane, sevo¯urane or des¯urane anaesthesia at 0.5, 1.0, and 1.5 MAC. Data are mean and SD. At rest, four additional measurements were performed during recovery (RI±RIV) at 5, 10, 15 and 20 min for iso¯urane and des¯urane and 5, 10, 20 and 30 min for iso¯urane anesthesia (A). The percentage increase in MV in response to exposure to 8% carbon dioxide is presented at baseline, 0.5, 1.0 and 1.5 MAC and at the end of recovery (B). During sevo¯urane anaesthesia, MV was preserved up to 1.0 MAC but was signi®cantly depressed at 1.0 MAC during anaesthesia with iso¯urane (P=0.0008) or des¯urane (P<0.0001). During recovery, MV was higher at the ®rst and second time points during anaesthesia with sevo¯urane and des¯urane compared with iso¯urane. The response to hypercapnoeia was signi®cantly more depressed during des¯urane anaesthesia compared with iso¯urane and sevo¯urane at 0.5 MAC. At the end of recovery the response to hypercapnoeia was completely blocked regardless of the agent used. Overall, sevo¯urane preserved spontaneous breathing more than iso¯urane and des¯urane, but there was no relevant difference between the three agents in the dramatic depression of the response to hypercapnoeia. *=Signi®cant difference for iso¯urane vs sevo¯urane; #=signi®cant difference for iso¯urane vs des¯urane; §=signi®cant difference for sevo¯urane vs des¯urane (P<0.0167).
RR, Amp and MV for all three agents when the response at 1.5 MAC was compared with the end of recovery (Figs 3 and 4). Immediately after the last measurements, the mice were returned to their cage. They were considered awake when they were running and adequately responding to their cage mates. All mice ful®lled these criteria.
700
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
In contrast to results for iso¯urane and des¯urane, during sevo¯urane administration RR was unchanged up to 1.0 MAC and decreased signi®cantly at 1.5 MAC. During recovery from anaesthesia, RR increased signi®cantly (P<0.0001; Fig. 2). Amp did not change signi®cantly. RR was less affected by sevo¯urane compared with iso¯urane and des¯urane at 1.0 (P=0.0036 and P<0.0001 respectively) and 1.5 MAC (P=0.0014 and P=0.0048 respectively; Fig. 2). Moreover, RR recovered more quickly after sevo¯urane anaesthesia than after either iso¯urane or des¯urane. After 5 min of recovery from sevo¯urane anaesthesia, RR was signi®cantly greater than under iso¯urane (P=0.0005; Fig. 2) and at 10 min it was signi®cantly greater than for iso¯urane (P=0.0002) and des¯urane (P<0.0001; Fig. 2). However, by the end of recovery there was no difference in RR among the three anaesthetic agents (P>0.05). Amp was signi®cantly higher during iso¯urane anaesthesia at 1.5 MAC compared with sevo¯urane and des¯urane anaesthesia and at the beginning of the recovery phase (RI) for iso¯urane vs sevo¯urane. There were no statistically signi®cant changes in Amp during sevo¯urane and des¯urane anaesthesia. At the end of recovery (RIIRIV), there were no signi®cant differences in Amp between all three anaesthetic agents. The product of RR and Amp (as an equivalent of MV) strengthens the results seen in RR. During sevo¯urane anaesthesia, MV was maintained at baseline up to 1.0 MAC and decreased signi®cantly during iso¯urane and des¯urane anaesthesia (P=0.0111 and P=0.0008 respectively). Only at 1.5 MAC did the increased Amp during the iso¯urane and des¯urane anaesthesia compensate for the decreased RR, resulting in no difference in MV compared with sevo¯urane. Finally, after the 5 and 10 min time points during recovery (RI, RII), MV increased signi®cantly faster during sevo¯urane and des¯urane anaesthesia than during iso¯urane (P<0.0001; Fig. 3).
Volatile anaesthetics and respiratory drive
Discussion In mice predisposed to a blunted carbon dioxide response, spontaneous ventilation was preserved during sevo¯urane anaesthesia compared with iso¯urane or des¯urane anaesthesia up to 1.0 MAC. However, even at the lowest concentration tested, responses to hypercapnoeia were signi®cantly impaired by all three volatile anaesthetics. Moreover, the hypercapnoeic breathing response remained blocked at the end of the recovery period for all three anaesthetic agents. While hypoxic drive can be altered at very low anaesthetic concentrations in patients with supposedly normal respiratory drive, the response to hypercapnoeia is believed to be more robust.1 Moreover, patients with a blunted respiratory drive are likely to be at greater risk of hypoxia
701
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
Fig 4 Changes in respiratory rate (RR) and pressure amplitude (Amp), as an expression of tidal volume, to 8% carbon dioxide in 11 male C3H/HeJ mice measured at baseline and during iso¯urane, sevo¯urane or des¯urane anaesthesia at 0.5, 1.0 and 1.5 MAC and at the end of recovery. Data are mean and SD. Even at the lowest concentrations of all three anaesthetics, RR was signi®cantly decreased (P<0.0001). The decrease with 0.5 MAC des¯urane was only partly compensated for by an increase in Amp (P=0.0029). In recovery, the response to carbon dioxide was signi®cantly suppressed for all three anaesthetics and did not differ from the blockade during deep anaesthesia at 1.5 MAC (P<0.0001). *=Signi®cant difference for iso¯urane vs sevo¯urane; #=signi®cant difference for iso¯urane vs des¯urane; §=signi®cant difference for sevo¯urane vs des¯urane (P<0.0167).
and hypercapnoeia and more prone to potential sequelae under the in¯uence of volatile anaesthetics than individuals with a full normal response to carbon dioxide.12±14 The evaluation of the effect of different anaesthetics on respiratory drive in humans is complicated by a wide range of interindividual differences.4 20 Inbred mouse strains offer a unique advantage in studying pharmacological effects by eliminating the variance of phenotypes due to genetic in¯uences.19 Furthermore, by restricting our investigation to male mice, we also eliminated possible gender-related in¯uences. Therefore, the present study focused on the effect of hypercapnoeia in an animal model with an established reduced hypercapnoeic response.16 Indeed, we have demonstrated previously that C3 mice show greater susceptibility to iso¯urane than mice with a normal respiratory drive.17 In particular, in the most critical period of recovery from iso¯urane anaesthesia, C3 mice presented a complete lack of response to hypercapnoeia.17 Besides interindividual differences, environmental factors also in¯uence the response to anaesthetic agents. Volunteers showed a normal response to a hypercapnoeic challenge during iso¯urane sedation after visual or auditory stimulation.6 8 9 21 In contrast, when the room was silent and the lights dimmed, the response to the same carbon dioxide challenge was signi®cantly suppressed.6±9 To account for these effects and to minimize environmental stimulation, in the present study the mice were kept in dark and quiet surroundings at a comfortable ambient temperature. Because earlier studies have indicated potential differences between the commonly used volatile anaesthetics, we compared the effects of iso¯urane, sevo¯urane and des¯urane on spontaneous breathing and the response to hypercapnoeia.6 8 However, the results in these earlier studies were unclear, and were at least partly explained by the response variability among the subjects. By using isogenetic mice, we reduced the genetic variability and focused on the physiological differences caused by the three anaesthetics. We observed that spontaneous ventilation was affected differently by the three anaesthetics. While iso¯urane and des¯urane led to a signi®cant decrease to about 50% of baseline at 1.0 MAC, a more dramatic decrease (>80%) occurred at 1.5 MAC. Only during iso¯urane anaesthesia were the changes in RR partially compensated for by an increase in Amp at 1.5 MAC. In contrast, during sevo¯urane anaesthesia the RR was unchanged up to 1.0 MAC and only decreased to 31% of baseline at 1.5 MAC, an effect less dramatic than the decline in RR during either iso¯urane or des¯urane anaesthesia. In the recovery period, within about 10 min the mice reached their baseline RR and showed a brief but signi®cant overshoot, which was probably due to slight hypercapnoeia induced by hypoventilation during deep anaesthesia. Overall, without carbon dioxide blood gas measurements, we can only speculate about the cause of this mild hyperventilation. However, it is important to note that the
Groeben et al.
to which this risk is due to a genetic predisposition remains to be determined.10 11 In conclusion, in an inbred mouse model predisposed to blunted carbon dioxide breathing, spontaneous ventilation was preserved with the use of sevo¯urane up to 1.0 MAC compared with iso¯urane or des¯urane anaesthesia. However, there were no differences among the three volatile anesthetic agents in blocking the response to hypercapnoeia. Even the newer, short-acting anaesthetic agents signi®cantly blunted the response to hypercapnoeia. Of greatest signi®cance, the response to hypercapnoeia remained depressed even after the anaesthetics had been discontinued and the animals appeared fully recovered. Future research to determine the genetic basis for potential susceptibility to anaesthetic agents might help to identify individuals at risk and reduce postoperative morbidity and mortality.
References
702
1 Knill RL, Gelb AW. Ventilatory responses for hypoxia and hypercapnia during halothane sedation and anesthesia in man. Anesthesiology 1978; 49: 244±51 2 SjoÈgren D, Sollevi A, Ebberyd A, Lindahl SGE. Iso¯urane anaesthesia (0.6 MAC) and hypoxic ventilatory responses in humans. Acta Anaesthesiol Scand 1995; 39: 17±22 3 Lockhart SH, Rampil IJ, Yasuda N, Eger EI, Weiskopf RB. Depression of ventilation by des¯urane in humans. Anesthesiology 1991; 74: 484±8 4 Sollevi A, Lindahl SGE. Hypoxic and hypercapnic ventilatory responses during iso¯urane sedation and anaesthesia in women. Acta Anaesthesiol Scand 1995; 39: 931±8 5 Hirshman CA, McCullough RE, Cohen PJ, Weil JV. Depression of hypoxic ventilatory response by halothane, en¯urane and iso¯urane in dogs. Br J Anaesth 1977; 49: 957±62 6 van den Elsen M, Sarton E, Teppema L, Berkenbosch A, Dahan A. In¯uence of 0.1 minimum alveolar concentration of sevo¯urane, des¯urane and iso¯urane on dynamic ventilatory response to hypercapnia in humans. Br J Anaesth 1998; 80: 174±82 7 Sarton E, Dahan A, Teppema L, et al. Acute pain and central nervous system arousal do not restore impaired hypoxic ventilatory response during sevo¯urane sedation. Anesthesiology 1996; 85: 295±303 8 Dahan A, Sarton E, van den Elsen M, van Kleef J, Teppema L, Berkenbosch A. Ventilatory responses to hypoxia in humans. In¯uences of subanesthetic des¯urane. Anesthesiology 1996; 85: 60±8 9 van den Elsen M, Dahan A, Berkenbosch A, DeGoede J, van Kleef JW, Olievier ICW. Does subanesthetic iso¯urane affect the ventilatory response to acute isocapnic hypoxia in healthy volunteers? Anesthesiology 1994; 81: 860±7 10 Rosenberg J, Wildschiùdtz G, Pedersen MH, von Jessen F, Kehlet H. Late postoperative nocturnal episodic hypoxaemia and associated sleep pattern. Br J Anaesth 1994; 72: 145±50 11 Reeder MK, Muir AD, FoeÈx P, Goldman MD, Loh L, Smart D. Postoperative myocardial ischemia: temporal association with nocturnal hypoxaemia. Br J Anaesth 1991; 67: 626±31 12 Loadsman JA, Hillman DR. Anesthesia and sleep apnoea. Br J Anaesth 2001; 86: 254±66 13 Thalhofer S, Dorow P. Central sleep apnea. Respiration 1997; 64: 2±9
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
mice were able to increase their RR to over 300 b.p.m. when challenged with carbon dioxide at baseline. Therefore, even during their overshoot in RR to approximately 230 b.p.m. during the recovery period, the mice retained their ability to increase their RR further. The increase in Amp remained unchanged during sevo¯urane and iso¯urane anaesthesia compared with baseline. In contrast, during des¯urane anaesthesia Amp increased at 0.5 MAC relative to baseline, followed by a dose-dependent decrease at 1.0 and 1.5 MAC. Des¯urane was the only agent tested that caused reductions in both the tachypnoeic response and the tidal breath amplitude to the hypercapnoeic challenge. However, in contrast to the signi®cant difference during spontaneous breathing under anaesthesia at 1.0 MAC, we observed only small differences among the three anaesthetic agents in the response to the hypercapnoeic challenge. When we examined the results of hypercapnoeic challenges in general, the responses were similarly attenuated among the three anesthetic agents, even at the lowest concentration. Most importantly, the anaesthetic effect on hypercapnoeia continued well into recovery when the mice appeared to be awake, and anaesthetic concentrations in the exhaust from the chamber were at the lower level of detection (0.02 MAC for sevo¯urane and 0.01 MAC for des¯urane), i.e. concentrations were considered subanaesthetic. However, even at this time point, the response to hypercapnoeia was almost completely blocked. In surgical patients, this would be the time when they would be transferred from close observation in the recovery room to a quiet normal room on the ¯oor. This observation was not entirely consistent with observations in studies of human subjects. At low doses of iso¯urane and sevo¯urane (0.1 MAC), human responses to hypercapnoeia were minimally affected and des¯urane showed no effect.1 4 6±9 However, these results were obtained in human volunteers who had normal responses to carbon dioxide at rest and who were free of any diseases affecting the control of breathing. In contrast, C3 mice have a blunted response to carbon dioxide and serve as a model of possible pathological responses in humans with reduced respiratory drive. As in animal models, there are limitations and differences compared with human responses. For example, humans respond to hypercapnoeia with a preferential increase in their tidal volume, while mice preferentially increase their RR. Therefore, further studies are necessary to evaluate the effects of low concentrations of volatile agents in patients with pre-existing lung disease or a compromised breathing response to carbon dioxide, such as patients with Pickwick syndrome or patients who are predisposed to central sleep apnoea syndrome.12±14 These are the patients at greatest risk of adverse events after surgery, especially when potent analgesic medications are added in the immediate postoperative period, possibly potentiating the depression of respiratory drive. The extent
Volatile anaesthetics and respiratory drive
14 Connolly LA. Anesthetic management of obstructive sleep apnea patients. J Clin Anesth 1991; 3: 461±9 15 Tankersley CG, Elston RC, Schnell AH. Genetic determinants of acute hypoxic ventilation: patterns of inheritance in mice. J Appl Physiol 2000; 88: 2310±8 16 Tankersley CG, Fitzgerald RS, Kleeberger SR. Differential control of ventilation among inbred strains of mice. Am J Physiol 1994; 267: R1371±7 17 Groeben H, Meier S, Tankersley CG, Mitzner W, Brown RH. Heritable differences in respiratory drive and breathing pattern in mice during anaesthesia and emergence. Br J Anaesth 2003, 91: 541±5
18 Sonner JM, Gong D, Eger EI. Naturally occurring variability in anesthetic potency among inbred mouse strains. Anesth Analg 2000; 91: 720±6 19 Haldane JBS, Waddington CH. Inbreeding and linkage. Genetics 1931; 16: 357±74 20 Hirshman CA, McCullough RE, Weil JV. Normal values for hypoxic and hypercapnic ventilatory drives in man. J Appl Physiol 1975; 38: 1095±98 21 Temp JA, Henson LC, Ward DS. Effects of a subanesthetic minimum alveolar concentration of iso¯urane on two tests of the hypoxic ventilatory response. Anesthesiology 1994; 80: 739±50
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
703
DOI: 10.1093/bja/aeh124
Advance Access publication March 5, 2004
LABORATORY INVESTIGATIONS In¯uence of volatile anaesthetics on hypercapnoeic ventilatory responses in mice with blunted respiratory drive² H. Groeben1,2*, S. Meier2, C. G. Tankersley1, W. Mitzner1 and R. H. Brown1 1
Department of Environmental Health Sciences/Division of Physiology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, USA. 2University of Essen, Essen, Germany *Corresponding author: Department of Anesthesiology and Critical Care Medicine, University of Essen, Hufelandstrasse 55, 45122 Essen, Germany. E-mail: [email protected]
Methods. Using whole body plethysmography, we assessed respiratory rate (RR) and pressure amplitude in 11 male C3 mice at rest, during anaesthesia with iso¯urane, sevo¯urane or des¯urane, and during recovery. To test respiratory drive, mice were exposed to 8% carbon dioxide. Data were analysed by two-way-analysis of variance with post hoc tests and Bonferroni correction. Results. RR was unaffected during sevo¯urane anaesthesia up to 1.0 MAC. Likewise, sevo¯urane at 1.5 MAC affected RR less than either iso¯urane (P=0.0014) or des¯urane (P=0.0048). The increased RR to a carbon dioxide challenge was blocked by all three anaesthetics even at the lowest concentration, and remained depressed during recovery (P<0.0001). Tidal volume was unaffected by all three anaesthetics. Conclusions. In C3 mice, spontaneous ventilation was less affected during sevo¯urane compared with either iso¯urane or des¯urane anaesthesia. However, the RR response to hypercapnoeia was abolished at 0.5 MAC for all the anaesthetic agents and remained depressed even at the end of recovery. Our data suggest that different volatile anaesthetics have varying effects on the control of breathing frequency but all block the respiratory response to carbon dioxide. Therefore, a genetic predisposition to a blunted carbon dioxide response represents a susceptibility factor that interacts with hypercapnoeic hypoventilation during maintenance of anaesthesia and in the emergence from anaesthesia, regardless of the agent used. Br J Anaesth 2004; 92: 697±703 Keywords: anaesthetics volatile; complications, hypercapnia Accepted for publication: December 23, 2003
It is well established that volatile anaesthetic agents directly affect the neural control of breathing. At clinically relevant concentrations, volatile anaesthetics alter the magnitude and pattern of breathing.1±5 In addition, subanaesthetic concentrations of volatile anaesthetics, as low as 0.1 MAC, may signi®cantly diminish the response to hypoxia and hypercapnoeia.6±9
Clinically, the respiratory depressant effects of volatile anaesthetic agents may contribute to postoperative morbidity and mortality through decreased responsiveness to hypoxia or hypercapnoeia. This is of particular concern in ² Presented in part at the ATS annual meeting, Atlanta, May 18±22, 2002.
Ó The Board of Management and Trustees of the British Journal of Anaesthesia 2004
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
Background. Subanaesthetic concentrations of volatile anaesthetics signi®cantly affect the respiratory response to hypoxia and hypercapnoeia. Individuals with an inherited blunted respiratory drive are more affected than normal individuals. To test the hypothesis that subjects with blunted hypercapnoeic respiratory drive are diversely affected by different anaesthetics, we studied the effects of three volatile anaesthetics on the control of breathing in C3H/HeJ (C3) mice, characterized by a blunted hypercapnoeic respiratory response.
Groeben et al.
Methods Animals Our study protocol was approved by the Johns Hopkins Animal Care and Use Committee. Eleven male C3H/HeJ mice were purchased from Jackson Laboratories (Bar Harbor, ME, USA) and housed in the animal facilities at Johns Hopkins University for 5±6 weeks prior to the ®rst experiments. Water and chow (Agway Pro-Lab RMH 1000) were provided ad libitum. All experiments were performed at the age of 14 weeks (12 days) and a mean weight of 32.8 (3.4) g, at the same time of day for each animal.
Measurements Whole body plethysmograph
Ventilatory function was assessed by whole body plethysmography under unrestrained conditions.15 16 Each animal was permitted to acclimatize in a barometric, acrylic, cylindrical chamber (400 cm3) for at least 30 min before ventilation measurements were obtained. The chamber was placed in an insulated box to control environmental conditions (i.e. quiet and dark surroundings) and to prevent heat loss. The chamber temperature was maintained within the thermoneutral zone for mice of 26±28°C [i.e. 26.9 (0.8)°C] and each measurement was made with a type T thermocouple device. The inspired gas (FIO2=1.0) was humidi®ed (90% relative humidity) and directed through the chamber at a ¯ow rate of 300 ml min±1.
At constant chamber volume, changes in pressure due to inspiration and expiration were measured using a differential pressure transducer (model 8510B-2; Endevco, San Juan Capistrano, CA, USA) and recorded on a strip chart recorder (Grass Polygraph model 7D; Grass Instrument, Quincy, MA, USA). The out¯ow air from the chamber was analysed for oxygen, carbon dioxide and the respective volatile anaesthetic (iso¯urane, sevo¯urane, or des¯urane) using a Capnomac Ultra ULT-V-27±06 (Datex, Helsinki, Finland). Protocol
In randomized order, all animals received all three anaesthetic agents on different days. After the animal became quiescent, the chamber was sealed for 60 s to permit measurements of respiratory rate (RR) and the pressure amplitude (Amp; in mm) resulting from each tidal breath as a correlate of the tidal volume. The product of RR and Amp was calculated as an equivalent of minute ventilation (MV). While the chamber was sealed, metabolically induced changes in carbon dioxide and oxygen were <1% of the target level. Subsequently, an inspired gas mixture (FICO2=0.08; Puritan-Bennett Medical Gases, Overland Park, KS, USA) was administered to the chamber for 6 min. At the end of this carbon dioxide challenge, measurements of RR and Amp and calculation of MV were repeated as described above. After these baseline measurements, iso¯urane, sevo¯urane (Abbott Laboratories, North Chicago, IL, USA) or des¯urane (Baxter Healthcare, Deer®eld, IL, USA) was added to the hyperoxic inspirate via an agent-speci®c vaporizer integrated in the gas ¯ow system. Each anaesthetic was administered on a different day in random order with at least 1 day between study days for each mouse. Each volatile anaesthetic was administered in increasing concentrations of 0.5, 1.0, and 1.5 MAC (minimal alveolar concentration required to produce immobility in response to noxious stimuli) speci®c for C3H/HeJ mice. The MAC for C3H/HeJ mice is 1.5% for iso¯urane, 2.4% for sevo¯urane and 8.6% for des¯urane.18 Sevo¯urane and des¯urane were administered for 20 min at each concentration step, while iso¯urane was administered for 30 min at each concentration to achieve a steady state of anaesthesia. After reaching steady state, measurements of RR and Amp were repeated and a carbon dioxide challenge followed, as described above. Immediately after the carbon dioxide challenge at the highest concentration of the anaesthetic, the anaesthetic agent was turned off and the mice were observed during recovery from anaesthesia for the next 20 min (sevo¯urane and des¯urane) or 30 min (iso¯urane). During recovery, measurements were performed at 5, 10, 20 and 30 min for iso¯urane (RI-IV) and at 5, 10, 15 and 20 min for sevo¯urane and des¯urane (RI-IV). An example of the original recording is presented in Figure 1. At the end of
698
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
individuals with compromised respiratory drive, such as patients with chronic obstructive airway disease and obstructive sleep apnoea.10±14 Several studies indicate that there are possible differences in respiratory depressant effects among the currently available volatile anaesthetic agents.1 4 6 8±11 C3H/HeJ (C3) inbred mice are characterized by a blunted response to hypoxia and hypercapnoeia.15 16 Thus, C3 mice serve as a genetic model of individuals (mice) with blunted chemosensitivity. To elucidate possible differences in the deleterious effects of various volatile anaesthetic agents on respiratory function in individuals with blunted chemosensitivity, we studied the effects of volatile anaesthetics in C3 mice. We have already demonstrated that the genetically determined difference in respiratory drive of the C3 mice compared with mice with an undisturbed respiratory drive was enhanced during recovery from iso¯urane anaesthesia.17 However, the question remains whether shorter-acting volatile anaesthetics can avoid the postanaesthetic depressed response to respiratory stimuli such as hypercapnoeia. Therefore, the present study compared the effects of iso¯urane, sevo¯urane and des¯urane on the magnitude and pattern of spontaneous breathing at rest and during acute hypercapnoeic challenge in C3 mice.
Volatile anaesthetics and respiratory drive
recovery a ®nal carbon dioxide challenge was performed and subsequently the mice were returned to their cages.
Data analysis Data are presented as mean (SD). Measurements of RR, Amp and MV before and after carbon dioxide challenge were analysed by two-way analysis of variance (ANOVA) with post hoc tests and Bonferroni correction for multiple comparisons, controlling for anesthetic agent and concentration. Signi®cance was considered to be P<0.05. We tested the null hypotheses that (i) the effects of the three agents on RR, Amp and MV are the same and (ii) the effects of the three agents on the response to hypercapnoeia are the same.
Results A total of 11 mice were studied. All mice received all three anaesthetic agents and completed the study. At baseline, there was no difference in RR, Amp or MV between the mice on different days.
Fig 2 Respiratory rate (RR) and pressure amplitude (Amp), as an expression of tidal volume, in 11 male C3H/HeJ mice measured at baseline and during iso¯urane, sevo¯urane or des¯urane anaesthesia at 0.5, 1.0 and 1.5 MAC. Data are mean and SD. Four additional measurements were performed during recovery (RI±RIV) at 5, 10, 15 and 20 min for iso¯urane and des¯urane and 5, 10, 20 and 30 min for iso¯urane anaesthesia. During sevo¯urane anaesthesia, RR was preserved up to 1.0 MAC but RR was signi®cantly depressed at 1.0 MAC during anaesthesia with either iso¯urane (P<0.0001) or des¯urane (P<0.0001). The increase in Amp for iso¯urane at 1.5 MAC and RI (P<0.0001) could compensate only partly for the decrease in RR, while there were no changes in Amp for sevo¯urane and des¯urane. *=Signi®cant difference for iso¯urane vs sevo¯urane; #=signi®cant difference for iso¯urane vs des¯urane; §=signi®cant difference for sevo¯urane vs des¯urane (P<0.0167).
Effect of iso¯urane, sevo¯urane and des¯urane on ventilation at rest Iso¯urane
At 1.0 and 1.5 MAC of iso¯urane, RR decreased signi®cantly and dose-dependently compared with RR at baseline and 0.5 MAC. Subsequently, RR increased signi®cantly during recovery from anaesthesia (P<0.0001; Fig. 2). Amp increased signi®cantly at 1.5 MAC and after 5 min of recovery (P<0.0001). Des¯urane
At 1.0 and 1.5 MAC of des¯urane, RR decreased signi®cantly and dose-dependently compared with RR at baseline
699
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
Fig 1 Original recordings of one C3 mouse before and during iso¯urane anaesthesia. Six seconds of each step and one carbon dioxide challenge are shown. Carbon dioxide at baseline led to deepening of the breath and an increase in respiratory rate. Increasing concentrations of iso¯urane (without additional carbon dioxide) affected respiratory rate dramatically.
Groeben et al.
and 0.5 MAC. Subsequently, RR increased signi®cantly during recovery from anaesthesia (P<0.0001; Fig. 2). Amp did not change signi®cantly. Sevo¯urane
Effect of volatile anaesthetics on the response to hypercapnoeia At baseline, the hypercapnoeic challenge caused a signi®cant increase in RR, Amp and MV (P<0.0001; Figs 3 and 4). All three inhalation anaesthetics attenuated the tachypnoeic response to hypercapnoeia (Fig. 4). Even after recovery from all three anaesthetics, the tachypnoeic response was completely blocked (P<0.0001). While there were varying, albeit small, changes in Amp to hypercapnoeia (a slight dose-dependent decrease during des¯urane anaesthesia; Fig. 4) at the end of recovery, there were no signi®cant differences (Fig. 4). In particular, there was no difference in
Fig 3 The product of respiratory rate (RR) and pressure amplitude (Amp), as an equivalent of minute ventilation, of 11 male C3H/HeJ mice measured at baseline and during iso¯urane, sevo¯urane or des¯urane anaesthesia at 0.5, 1.0, and 1.5 MAC. Data are mean and SD. At rest, four additional measurements were performed during recovery (RI±RIV) at 5, 10, 15 and 20 min for iso¯urane and des¯urane and 5, 10, 20 and 30 min for iso¯urane anesthesia (A). The percentage increase in MV in response to exposure to 8% carbon dioxide is presented at baseline, 0.5, 1.0 and 1.5 MAC and at the end of recovery (B). During sevo¯urane anaesthesia, MV was preserved up to 1.0 MAC but was signi®cantly depressed at 1.0 MAC during anaesthesia with iso¯urane (P=0.0008) or des¯urane (P<0.0001). During recovery, MV was higher at the ®rst and second time points during anaesthesia with sevo¯urane and des¯urane compared with iso¯urane. The response to hypercapnoeia was signi®cantly more depressed during des¯urane anaesthesia compared with iso¯urane and sevo¯urane at 0.5 MAC. At the end of recovery the response to hypercapnoeia was completely blocked regardless of the agent used. Overall, sevo¯urane preserved spontaneous breathing more than iso¯urane and des¯urane, but there was no relevant difference between the three agents in the dramatic depression of the response to hypercapnoeia. *=Signi®cant difference for iso¯urane vs sevo¯urane; #=signi®cant difference for iso¯urane vs des¯urane; §=signi®cant difference for sevo¯urane vs des¯urane (P<0.0167).
RR, Amp and MV for all three agents when the response at 1.5 MAC was compared with the end of recovery (Figs 3 and 4). Immediately after the last measurements, the mice were returned to their cage. They were considered awake when they were running and adequately responding to their cage mates. All mice ful®lled these criteria.
700
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
In contrast to results for iso¯urane and des¯urane, during sevo¯urane administration RR was unchanged up to 1.0 MAC and decreased signi®cantly at 1.5 MAC. During recovery from anaesthesia, RR increased signi®cantly (P<0.0001; Fig. 2). Amp did not change signi®cantly. RR was less affected by sevo¯urane compared with iso¯urane and des¯urane at 1.0 (P=0.0036 and P<0.0001 respectively) and 1.5 MAC (P=0.0014 and P=0.0048 respectively; Fig. 2). Moreover, RR recovered more quickly after sevo¯urane anaesthesia than after either iso¯urane or des¯urane. After 5 min of recovery from sevo¯urane anaesthesia, RR was signi®cantly greater than under iso¯urane (P=0.0005; Fig. 2) and at 10 min it was signi®cantly greater than for iso¯urane (P=0.0002) and des¯urane (P<0.0001; Fig. 2). However, by the end of recovery there was no difference in RR among the three anaesthetic agents (P>0.05). Amp was signi®cantly higher during iso¯urane anaesthesia at 1.5 MAC compared with sevo¯urane and des¯urane anaesthesia and at the beginning of the recovery phase (RI) for iso¯urane vs sevo¯urane. There were no statistically signi®cant changes in Amp during sevo¯urane and des¯urane anaesthesia. At the end of recovery (RIIRIV), there were no signi®cant differences in Amp between all three anaesthetic agents. The product of RR and Amp (as an equivalent of MV) strengthens the results seen in RR. During sevo¯urane anaesthesia, MV was maintained at baseline up to 1.0 MAC and decreased signi®cantly during iso¯urane and des¯urane anaesthesia (P=0.0111 and P=0.0008 respectively). Only at 1.5 MAC did the increased Amp during the iso¯urane and des¯urane anaesthesia compensate for the decreased RR, resulting in no difference in MV compared with sevo¯urane. Finally, after the 5 and 10 min time points during recovery (RI, RII), MV increased signi®cantly faster during sevo¯urane and des¯urane anaesthesia than during iso¯urane (P<0.0001; Fig. 3).
Volatile anaesthetics and respiratory drive
Discussion In mice predisposed to a blunted carbon dioxide response, spontaneous ventilation was preserved during sevo¯urane anaesthesia compared with iso¯urane or des¯urane anaesthesia up to 1.0 MAC. However, even at the lowest concentration tested, responses to hypercapnoeia were signi®cantly impaired by all three volatile anaesthetics. Moreover, the hypercapnoeic breathing response remained blocked at the end of the recovery period for all three anaesthetic agents. While hypoxic drive can be altered at very low anaesthetic concentrations in patients with supposedly normal respiratory drive, the response to hypercapnoeia is believed to be more robust.1 Moreover, patients with a blunted respiratory drive are likely to be at greater risk of hypoxia
701
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
Fig 4 Changes in respiratory rate (RR) and pressure amplitude (Amp), as an expression of tidal volume, to 8% carbon dioxide in 11 male C3H/HeJ mice measured at baseline and during iso¯urane, sevo¯urane or des¯urane anaesthesia at 0.5, 1.0 and 1.5 MAC and at the end of recovery. Data are mean and SD. Even at the lowest concentrations of all three anaesthetics, RR was signi®cantly decreased (P<0.0001). The decrease with 0.5 MAC des¯urane was only partly compensated for by an increase in Amp (P=0.0029). In recovery, the response to carbon dioxide was signi®cantly suppressed for all three anaesthetics and did not differ from the blockade during deep anaesthesia at 1.5 MAC (P<0.0001). *=Signi®cant difference for iso¯urane vs sevo¯urane; #=signi®cant difference for iso¯urane vs des¯urane; §=signi®cant difference for sevo¯urane vs des¯urane (P<0.0167).
and hypercapnoeia and more prone to potential sequelae under the in¯uence of volatile anaesthetics than individuals with a full normal response to carbon dioxide.12±14 The evaluation of the effect of different anaesthetics on respiratory drive in humans is complicated by a wide range of interindividual differences.4 20 Inbred mouse strains offer a unique advantage in studying pharmacological effects by eliminating the variance of phenotypes due to genetic in¯uences.19 Furthermore, by restricting our investigation to male mice, we also eliminated possible gender-related in¯uences. Therefore, the present study focused on the effect of hypercapnoeia in an animal model with an established reduced hypercapnoeic response.16 Indeed, we have demonstrated previously that C3 mice show greater susceptibility to iso¯urane than mice with a normal respiratory drive.17 In particular, in the most critical period of recovery from iso¯urane anaesthesia, C3 mice presented a complete lack of response to hypercapnoeia.17 Besides interindividual differences, environmental factors also in¯uence the response to anaesthetic agents. Volunteers showed a normal response to a hypercapnoeic challenge during iso¯urane sedation after visual or auditory stimulation.6 8 9 21 In contrast, when the room was silent and the lights dimmed, the response to the same carbon dioxide challenge was signi®cantly suppressed.6±9 To account for these effects and to minimize environmental stimulation, in the present study the mice were kept in dark and quiet surroundings at a comfortable ambient temperature. Because earlier studies have indicated potential differences between the commonly used volatile anaesthetics, we compared the effects of iso¯urane, sevo¯urane and des¯urane on spontaneous breathing and the response to hypercapnoeia.6 8 However, the results in these earlier studies were unclear, and were at least partly explained by the response variability among the subjects. By using isogenetic mice, we reduced the genetic variability and focused on the physiological differences caused by the three anaesthetics. We observed that spontaneous ventilation was affected differently by the three anaesthetics. While iso¯urane and des¯urane led to a signi®cant decrease to about 50% of baseline at 1.0 MAC, a more dramatic decrease (>80%) occurred at 1.5 MAC. Only during iso¯urane anaesthesia were the changes in RR partially compensated for by an increase in Amp at 1.5 MAC. In contrast, during sevo¯urane anaesthesia the RR was unchanged up to 1.0 MAC and only decreased to 31% of baseline at 1.5 MAC, an effect less dramatic than the decline in RR during either iso¯urane or des¯urane anaesthesia. In the recovery period, within about 10 min the mice reached their baseline RR and showed a brief but signi®cant overshoot, which was probably due to slight hypercapnoeia induced by hypoventilation during deep anaesthesia. Overall, without carbon dioxide blood gas measurements, we can only speculate about the cause of this mild hyperventilation. However, it is important to note that the
Groeben et al.
to which this risk is due to a genetic predisposition remains to be determined.10 11 In conclusion, in an inbred mouse model predisposed to blunted carbon dioxide breathing, spontaneous ventilation was preserved with the use of sevo¯urane up to 1.0 MAC compared with iso¯urane or des¯urane anaesthesia. However, there were no differences among the three volatile anesthetic agents in blocking the response to hypercapnoeia. Even the newer, short-acting anaesthetic agents signi®cantly blunted the response to hypercapnoeia. Of greatest signi®cance, the response to hypercapnoeia remained depressed even after the anaesthetics had been discontinued and the animals appeared fully recovered. Future research to determine the genetic basis for potential susceptibility to anaesthetic agents might help to identify individuals at risk and reduce postoperative morbidity and mortality.
References
702
1 Knill RL, Gelb AW. Ventilatory responses for hypoxia and hypercapnia during halothane sedation and anesthesia in man. Anesthesiology 1978; 49: 244±51 2 SjoÈgren D, Sollevi A, Ebberyd A, Lindahl SGE. Iso¯urane anaesthesia (0.6 MAC) and hypoxic ventilatory responses in humans. Acta Anaesthesiol Scand 1995; 39: 17±22 3 Lockhart SH, Rampil IJ, Yasuda N, Eger EI, Weiskopf RB. Depression of ventilation by des¯urane in humans. Anesthesiology 1991; 74: 484±8 4 Sollevi A, Lindahl SGE. Hypoxic and hypercapnic ventilatory responses during iso¯urane sedation and anaesthesia in women. Acta Anaesthesiol Scand 1995; 39: 931±8 5 Hirshman CA, McCullough RE, Cohen PJ, Weil JV. Depression of hypoxic ventilatory response by halothane, en¯urane and iso¯urane in dogs. Br J Anaesth 1977; 49: 957±62 6 van den Elsen M, Sarton E, Teppema L, Berkenbosch A, Dahan A. In¯uence of 0.1 minimum alveolar concentration of sevo¯urane, des¯urane and iso¯urane on dynamic ventilatory response to hypercapnia in humans. Br J Anaesth 1998; 80: 174±82 7 Sarton E, Dahan A, Teppema L, et al. Acute pain and central nervous system arousal do not restore impaired hypoxic ventilatory response during sevo¯urane sedation. Anesthesiology 1996; 85: 295±303 8 Dahan A, Sarton E, van den Elsen M, van Kleef J, Teppema L, Berkenbosch A. Ventilatory responses to hypoxia in humans. In¯uences of subanesthetic des¯urane. Anesthesiology 1996; 85: 60±8 9 van den Elsen M, Dahan A, Berkenbosch A, DeGoede J, van Kleef JW, Olievier ICW. Does subanesthetic iso¯urane affect the ventilatory response to acute isocapnic hypoxia in healthy volunteers? Anesthesiology 1994; 81: 860±7 10 Rosenberg J, Wildschiùdtz G, Pedersen MH, von Jessen F, Kehlet H. Late postoperative nocturnal episodic hypoxaemia and associated sleep pattern. Br J Anaesth 1994; 72: 145±50 11 Reeder MK, Muir AD, FoeÈx P, Goldman MD, Loh L, Smart D. Postoperative myocardial ischemia: temporal association with nocturnal hypoxaemia. Br J Anaesth 1991; 67: 626±31 12 Loadsman JA, Hillman DR. Anesthesia and sleep apnoea. Br J Anaesth 2001; 86: 254±66 13 Thalhofer S, Dorow P. Central sleep apnea. Respiration 1997; 64: 2±9
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
mice were able to increase their RR to over 300 b.p.m. when challenged with carbon dioxide at baseline. Therefore, even during their overshoot in RR to approximately 230 b.p.m. during the recovery period, the mice retained their ability to increase their RR further. The increase in Amp remained unchanged during sevo¯urane and iso¯urane anaesthesia compared with baseline. In contrast, during des¯urane anaesthesia Amp increased at 0.5 MAC relative to baseline, followed by a dose-dependent decrease at 1.0 and 1.5 MAC. Des¯urane was the only agent tested that caused reductions in both the tachypnoeic response and the tidal breath amplitude to the hypercapnoeic challenge. However, in contrast to the signi®cant difference during spontaneous breathing under anaesthesia at 1.0 MAC, we observed only small differences among the three anaesthetic agents in the response to the hypercapnoeic challenge. When we examined the results of hypercapnoeic challenges in general, the responses were similarly attenuated among the three anesthetic agents, even at the lowest concentration. Most importantly, the anaesthetic effect on hypercapnoeia continued well into recovery when the mice appeared to be awake, and anaesthetic concentrations in the exhaust from the chamber were at the lower level of detection (0.02 MAC for sevo¯urane and 0.01 MAC for des¯urane), i.e. concentrations were considered subanaesthetic. However, even at this time point, the response to hypercapnoeia was almost completely blocked. In surgical patients, this would be the time when they would be transferred from close observation in the recovery room to a quiet normal room on the ¯oor. This observation was not entirely consistent with observations in studies of human subjects. At low doses of iso¯urane and sevo¯urane (0.1 MAC), human responses to hypercapnoeia were minimally affected and des¯urane showed no effect.1 4 6±9 However, these results were obtained in human volunteers who had normal responses to carbon dioxide at rest and who were free of any diseases affecting the control of breathing. In contrast, C3 mice have a blunted response to carbon dioxide and serve as a model of possible pathological responses in humans with reduced respiratory drive. As in animal models, there are limitations and differences compared with human responses. For example, humans respond to hypercapnoeia with a preferential increase in their tidal volume, while mice preferentially increase their RR. Therefore, further studies are necessary to evaluate the effects of low concentrations of volatile agents in patients with pre-existing lung disease or a compromised breathing response to carbon dioxide, such as patients with Pickwick syndrome or patients who are predisposed to central sleep apnoea syndrome.12±14 These are the patients at greatest risk of adverse events after surgery, especially when potent analgesic medications are added in the immediate postoperative period, possibly potentiating the depression of respiratory drive. The extent
Volatile anaesthetics and respiratory drive
14 Connolly LA. Anesthetic management of obstructive sleep apnea patients. J Clin Anesth 1991; 3: 461±9 15 Tankersley CG, Elston RC, Schnell AH. Genetic determinants of acute hypoxic ventilation: patterns of inheritance in mice. J Appl Physiol 2000; 88: 2310±8 16 Tankersley CG, Fitzgerald RS, Kleeberger SR. Differential control of ventilation among inbred strains of mice. Am J Physiol 1994; 267: R1371±7 17 Groeben H, Meier S, Tankersley CG, Mitzner W, Brown RH. Heritable differences in respiratory drive and breathing pattern in mice during anaesthesia and emergence. Br J Anaesth 2003, 91: 541±5
18 Sonner JM, Gong D, Eger EI. Naturally occurring variability in anesthetic potency among inbred mouse strains. Anesth Analg 2000; 91: 720±6 19 Haldane JBS, Waddington CH. Inbreeding and linkage. Genetics 1931; 16: 357±74 20 Hirshman CA, McCullough RE, Weil JV. Normal values for hypoxic and hypercapnic ventilatory drives in man. J Appl Physiol 1975; 38: 1095±98 21 Temp JA, Henson LC, Ward DS. Effects of a subanesthetic minimum alveolar concentration of iso¯urane on two tests of the hypoxic ventilatory response. Anesthesiology 1994; 80: 739±50
Downloaded from http://bja.oxfordjournals.org/ at Fresno Pacific University on December 26, 2014
703