| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Nitrous oxide is widely used in anesthesia and critical care medicine. The effect of nitrous oxide on diaphragm contractility in humans is unknown. We evaluated the effect of a 50% nitrous oxide50% oxygen mixture on diaphragm contractility in healthy adult volunteers. The sniff transdiaphragmatic pressure (Sn Pdi) and the twitch transdiaphragmatic pressure (Tw Pdi) elicited by bilateral supramaximal phrenic nerve stimulation were measured before during and after inhalation of a mixture of 50% nitrous oxide and 50% oxygen. Sn Pdi decreased by 15.4% during nitrous oxide inhalation, with a value of 136 ± 21 cm H2O before nitrous oxide and a value of 115 ± 27 cm H2O during nitrous oxide inhalation (P = 0.03). Similarly, Tw Pdi decreased from 21.2 ± 1.8 cm H2O before nitrous oxide inhalation to 16.9 ± 4.1 cm H2O during nitrous oxide inhalation (P = 0.03). The effect of nitrous oxide was totally abolished 20 min after its discontinuation. Nitrous oxide has a short-acting suppressant effect on the pressure generating capacity of the diaphragm in healthy humans. IMPLICATIONS: We investigated whether nitrous oxide (a common component of gas anesthesia) reduces diaphragm strength in humans. Diaphragm strength is reduced by nitrous oxide but the effect wears off within 20 min of administration. Caution is advised when using nitrous oxide without anesthesiologist supervision in patients at risk of ventilatory failure
Nitrous oxide (N2O) is an anesthetic gas that has proved its efficacy and safety in various minor procedures such as venous cannulation, fracture reductions, lumbar punctures, bone marrow aspirations, laceration repairs, and dental care, both in children and in adults (13). In adults, N2O is widely used during dental care and labor (4), and it is also a common component of the inhaled gas mixtures used to maintain general anesthesia. N2O is traditionally (but not always) (5) considered safe, yet there are few data concerning the isolated effect of N2O on diaphragm function in humans. In patients with critical illness, acquired neuromuscular abnormalities contribute to weaning failure in over 60% of patients (6). Our evaluation of these patients includes the assessment of phrenic nerve and diaphragm function by measuring the twitch transdiaphragmatic pressure (Tw Pdi) elicited by magnetic stimulation of the phrenic nerves (7,8); further details are given by Polkey and Moxham (9). In some patients, passage of the esophageal and gastric catheters necessary to make this measurement can be difficult to tolerate and we reasoned that the use of a short-acting sedative could facilitate this process. N2O is safe and is short-acting. Therefore N2O seemed to be suitable for this purpose. Accordingly, the aim of the present study was to evaluate the in vivo effect of N2O (mixed 50% N2O and 50% oxygen) in humans. Although we had a specific rationale for performing the study, we assumed that this study would be of general interest to anesthesiologists and critical care physicians, given that N2O is widely used.
Six healthy volunteers (3 men and 3 women), age 2942 yr (mean, 34 ± 5 yr), participated in the study. The study was approved by the Ethics Committee of the Royal Brompton Hospital and all the subjects gave their informed consent. Esophageal (Pes) and gastric (Pgas) pressures were recorded using conventionally placed balloon-tip catheters 110 cm in length. The catheters were connected to differential pressure transducers, carrier amplifiers, a 12-bit NB-MIO-16 analog-digital board, and a Macintosh Quadra Centris 650 personal computer running customized Labview 2.0software (National Instruments, Austin, TX). Transdiaphragmatic pressure (Pdi) was obtained online by subtraction of Pes from Pgas. Pressure signals were sampled at 100 Hz. Diaphragm contractility was assessed using two methods, the maximal sniff pressure and the Tw Pdi elicited by magnetic stimulation of the phrenic nerves in the neck. Although the sniff is the classical method of assessing diaphragm function, it relies on the subject making a maximal voluntary effort that cannot be assumed during N2O administration. Therefore our primary outcome measure was Tw Pdi, which is independent of subject effort and suitable for use in the operating room (8). Maximal sniff efforts were performed from functional residual capacity (FRC) in the sitting position without a nose-clip (10). The monitor screen was visible to subjects who were asked to perform short maximal sniffs. Sniffs were recorded until no further increase in sniff Pdi could be achieved. The maximal sniff Pdi was used for analysis. Tw Pdi was measured after bilateral anterior magnetic phrenic nerve stimulation (BAMPS) with two figure-of-eight coils each powered by a Magstim 200stimulator (Magstim, Whitland, Dyfed, Wales) (7). Stimuli were given with the subject seated at relaxed end-expiration after a rest period of 20 min to avoid twitch potentiation (11). To optimize the position of the coils, several anterolateral stimulations were performed at varying positions and coil orientations until the point was determined at which the maximum response could be elicited. Having determined the optimal stimulation position, a ramp of increasing intensity of magnetic stimulation was performed in each case to assess supramaximality. The ramp consisted of three bilateral stimuli at 80%, 85%, 90%, 95%, and 100% of maximal output given in random order. Supramaximal stimulation was indicated by the leveling off or "plateauing" of Tw Pdi in response to increasing stimulus intensity. Subjects were only permitted to proceed with the study if supramaximal stimulation could be obtained. In fact, as is usual with BAMPS (8,12), supramaximal stimulation was obtained in all subjects. All subsequent stimulations were performed in optimal position and at maximum magnetic stimulation output. During the study the subjects were asked to breathe gently, to avoid deep inspirations, and to remain silent. They breathed through a mouthpiece and each wore a nose-clip. Because of the need to use a mouthpiece to administer N2O, subjects could not close their mouths before the magnetic stimulation. Therefore, we used a custom-built occlusion valve routinely used in our intensive care unit (13). This valve was inserted in the ventilatory circuit adjacent to the subjects airway and was electronically linked to the magnetic stimulators, so that the airway was occluded at the time of the phrenic stimulation. The pressure tracings were visible to the operator who stimulated the phrenic nerves until five satisfactory stimulations had been obtained. The tracings were subsequently assessed against the criteria listed below. The pressures elicited by phrenic nerve stimulation (Tw Pes, Tw Pgas, and Tw Pdi) were defined as the difference between the baseline pressures immediately before the twitch and the peak pressure after the twitch. Twitches were only accepted for analysis if performed at relaxed end-expiration, as judged by Pes, and when baseline Pdi pressures were similar to those seen at end-expiration during normal breathing, indicating relaxation of the diaphragm. For twitch pressures, the mean pressure of those twitches considered acceptable for analysis is presented. N2O-oxygen gas mixture (50% N2O, 50% O2) was administered from an anesthetic machine and circle system connected to a mouthpiece. Subjects wore a nose clip to prevent gas leakage. Total gas flow from the anesthetic machine was maintained at 12 L/min throughout the study. A gas monitor was used to continuously monitor inspired oxygen concentration and end-tidal N2O and end-tidal CO2 concentration. In addition electrocardiogram, respiratory rate, and pulse oximetry were monitored continuously. Magnetic stimulation was started after the end-tidal N2O concentration had reached a plateau (minimum 3 min) and the difference between end-tidal and inspired N2O was 2% or less. The subjects continued to breathe the N2OO2 mixture throughout the protocol. A consultant anesthesiologist was present throughout the studies, which were performed in an anesthetic room fully equipped for airway management and resuscitation. Subjects fasted for a minimum of 4 h before the study. All the subjects were familiar with magnetic stimulation and the sniff maneuver and performed a training session of N2O inhalation before the study protocol to confirm that they could maintain full relaxation. The protocol started with the maximal sniff maneuver (Pre N2O sniff Pdi). After a 20-min rest, a ramp was performed to determine baseline Tw Pdi (Pre N2O Tw Pdi). Subjects were then exposed to 3 min of N2O inhalation, and then a second set of BAMPS were performed to establish Tw Pdi during N2O inhalation (N2O Tw Pdi). Immediately after these BAMPS, the N2O inhalation was stopped and the subjects were asked to perform at least 5 sniffs without otherwise inhaling room air. Then, after a second period of 20 min rest breathing room air, a final set of BAMPS and sniffs were performed (Post N2O Tw Pdi, Post N2O sniff Pdi). To assess the between-occasion variability of magnetic stimulation, studies were repeated in three subjects at intervals of 12 wk. One subject was assessed on three occasions, on a 50% N2O50% oxygen mixture on two occasions and a 30% N2O70% oxygen mixture on one occasion. Data are given as mean ± SD. Wilcoxons signed rank test was used to assess whether there was a significant difference between baseline Tw Pdi and sniff Pdi and the same variables during N2O inhalation and 20 min after N2O discontinuation. A P value below 0.05 was considered as the limit of significance.
Anthropometric data are shown in Table 1. The sniff Pdi decreased during N2O inhalation, with a value of 136 ± 21 cm H2O before, and 115 ± 27 cm H2O during N2O inhalation (P = 0.03).
In all subjects, N2O showed a suppressive effect on Tw Pdi (Fig. 1). Tw Pdi decreased from 21.2 ± 1.8 cm H2O before N2O inhalation to 16.9 ± 4.1 cm H2O during N2O inhalation, P = 0.03. The greatest suppressant effect was observed in women (Fig. 1). The post N2O Tw Pdi measurements were similar to the pre N2O Tw Pdi measurements (Fig. 1), with a mean Post Tw Pdi of 21.8 ± 2.4 cm H2O after N2O inhalation, P = 0.8. There was no difference in Tw Pdi time to peak pressure between the Pre N2O, N2O, and Post N2O measurement. The investigators noted subjectively that the same concentration of N2O induced varying degrees of sedation in different subjects. Although attempts to quantify this proved fruitless, the more intense the depth of sedation, the larger was the decrease in Tw Pdi.
In one female subject (who had the smallest body mass index [BMI] and the largest decrease in Tw Pdi with N2O), a dose-dependent suppressive effect of N2O on Tw Pdi was observed (Fig. 2). With a similar Tw Pdi before and after N2O inhalation on both occasions, Tw Pdi decreased to 11.3 ± 2.1 cm H2O during 50% N2O and 15.3 ± 2.2 cm H2O during 30% N2O.
Partitioning of the Tw Pdi during the three conditions showed that there was no significant change in the esophageal and gastric contribution to Pdi (Fig. 3). The Pes/Pgas ratio was 1.36 ± 0.61, 1.41 ± 0.97 and 1.41 ± 0.70 during the Pre N2O twitch, the N2O twitch, and the Post N2O twitch, respectively, P = 0.8.
The percent decrease in Tw Pdi during N2O inhalation was inversely correlated to BMI (r2 = 0.742, P = 0.03) (Fig. 4).
This study shows that the inhalation of N2O has a significant suppressant effect on the pressure generating capacity of the diaphragm in healthy humans. This observation was confirmed by magnetic stimulation of the phrenic nerves and therefore shows the phenomenon to be largely a result of the effect of N2O on peripheral nerve or muscle. However, the effect of N2O was totally abolished 20 minutes after its discontinuation. Theoretically Tw Pdi could have decreased because of submaximal intensity of phrenic nerve stimulation. Although this possibility could also have been excluded by measuring the diaphragm action potential, we did not use this approach because of current concerns regarding the validity of surface electromyelography (EMG) signals (14,15). However, because all subjects were initially supramaximal and because the Tw Pdi returned exactly to baseline in all subjects, we doubt that submaximal stimulation could explain our data. In addition, because N2O is sedative in nature it was easily possible to maintain coil position during N2O inhalation in all subjects. Tw Pdi can decrease because of diaphragm lengthening, which might theoretically have occurred despite the tendency of anesthesia to decrease FRC (16). To address this point we have presented in Figure 3 data concerning the two different components, Tw Pgas and Tw Pes, contributing to the Tw Pdi. It is well established that increasing lung volume decreases Tw Pdi by reducing Tw Pes (12). However in the present study N2O reduced Tw Pgas and Tw Pes equally making lung volume considerations an unlikely explanation for our findings. Previous investigators (17) have assumed that the depressive effects of N2O on respiration in humans are mediated via central drive. However, unlike traditional measures of respiratory muscle function, the Tw Pdi evoked by peripheral nerve stimulation is not influenced by central drive or the subjects aptitude or motivation. Therefore our conclusion that N2O directly diminishes diaphragm contractility is sustainable. As discussed above, the diaphragm action potential was not measured in this study and therefore a transmission defect cannot be excluded on our data. However, previous studies in this area have failed to identify a neuromuscular transmission defect (18). Even if our data could be explained by an N2O induced transmission defect, the clinical significance of our data would be unchanged. The sample size for this study was relatively small (n = 6), but the data clearly show that all subjects had a reduction in Tw Pdi during N2O inhalation, although the magnitude varied between subjects according to their BMI. Because the study required the administration of an anesthetic on at least two occasions to healthy humans, we sought to minimize the number of study sessions and we only studied the 50% N2O50% oxygen mixture, which is the usual mixture used in clinical practice (3). This also explains why a dose-response study was only performed in one subject. The sample size used in the present study is comparable to that used in previous investigations in human subjects in this area (17) Animal studies have shown that N2O has a dose-dependent suppressive effect on the amplitude of motor-evoked potentials (MEP) elicited by electrical stimulation of the motor cortex (19). The MEP was recorded in peripheral muscles and from the epidural space of the thoracic cord in response to electrical stimulation of the motor cortex breathing an air-oxygen mixture and at increasing N2O concentrations (10% to 70%). With 50% N2O, EMG amplitudes were suppressed to 46% of the baseline value in the fore leg, whereas latencies did not change significantly, indicating that N2O either effects the muscle or, conceivably has a selective effect on some of the motor axons. However, N2O does not alter neuromuscular transmission (18), and sensory pathway function up to the primary sensory receiving area (judged by the latency and amplitude of compound muscle action potentials recorded from the thenar muscle after median nerve stimulation) remains unchanged by N2O in previous studies (20,21). We accept that depression of the central pathways documented in this fashion could explain the reduction in rib cage displacement and tidal volume observed by Warner et al. (17). However, as noted above, this mechanism cannot explain the present data suggesting direct impairment of in respiratory muscle contractility as an additional and complimentary explanation for Warner et al.s observation. We observed a larger suppressant effect of N2O on Tw Pdi in women than in men, with a 32% decrease and a 10% decrease, respectively. This suppressant effect was highly correlated to the BMI (Fig. 2). This point has not been reported previously. The reason for this observation was not addressed in the present study. One possibility is that people with a higher BMI have more fat, which absorbs N2O, thus "sparing" skeletal muscle. The results of this study have some important clinical implications. First, the assessment of Tw Pdi is not reliable during N2O inhalation. Thus, this anesthetic gas cannot be used to reduce the potential discomfort associated with magnetic stimulation of the phrenic nerves. Second, the depressant effects of N2O were absent 20 minutes after inhalation. Thus, this supports the use of N2O to avoid the discomfort associated with the insertion of esophageal and gastric balloons. A "washout" period is then required to avoid the confounding effects of this gas on diaphragmatic contractility, but this is also required to avoid potentiation (11). Clinical experience shows that the effect of N2O is rapidly reversible with discontinuation of administration, and it is conceivable that a shorter time period after N2O discontinuation could be used in clinical practice, but further studies are required to verify this point. Third, we have shown that N2O can have a suppressant effect on the pressure-generating capacity of the diaphragm in healthy humans. This finding could be of clinical importance in situations of "natural" and "pathological" diaphragm fatigue. Bellemare and Grassino (22) have demonstrated experimentally that there is a threshold above which diaphragm fatigue occurs. Healthy people never approach this threshold except during extreme loading (23). Patients with chronic obstructive pulmonary disease or neuromuscular disease may breathe close to the threshold at rest and cross-over with acute exacerbations. We therefore recommend that N2O be used with caution in such patients. In summary, N2O causes a dose-dependent depression of diaphragm contractility. Thus, although N2O may be used to facilitate the passage of measurement catheters it cannot be used during magnetic stimulation to assess diaphragm function. The clinical risk posed by N2O was not determined by our study, but we suggest that when N2O is used without supervision of an anesthesiologist (as in the emergency room or during practical procedures) caution be exercised in patients susceptible to ventilatory failure because of preexisting neurological or respiratory disease.
Dr. Fauroux was supported by a grant from the Société de Pneumologie de Langue Française. Other workers were supported by institutional funding.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|