NITROUS OXIDE ELIMINATION AND DIFFUSION HYPOXIA DURING NORMO- AND HYPOVENTILATION

Transcription

1 British Journal of Anaesthesia 1993; 71: NITROUS OXIDE ELIMINATION AND DIFFUSION HYPOXIA DURING NORMO- AND HYPOVENTILATION S. EINARSSON, O. STENQVIST, A. BENGTSSON, E. HOULTZ AND J. P. BENGTSON SUMMARY We studied the elimination rate of nitrous oxide in 36 patients undergoing orthopaedic surgery. They were a/located randomly to one of six groups which differed in time of nitrous oxide exposure and mode of ventilation. In order to simulate recovery conditions, nitrous oxide administration was discontinued after 30, 60 or 120 min of exposure. Either normoventilation or hypoventilation was used. The mean excretion rate was 1 litre min- 1 at 1 min, declining to 100 ml min~ 7 at 30 min. with relatively small effects of different modes of ventilation and times of exposure. In spite of an F/ Oj of 0.30, there were significant decreases in Sp Oj during both normo- and hypoventilation. The smallest end-tidal oxygen concentrations were reached at min in the groups with hypoventilation, after 1 or 2 h of nitrous oxide exposure. (Br. J. Anaesth. 1993; 71 : ) KEY WORDS Anaesthetics, gases: nitrous oxide. Hypoxia: diffusion. The time required for the elimination of nitrous oxide from the body during recovery from anaesthesia is predicted to be similar to that required for saturation with the gas [1,2]; there is a close agreement in relative uptake and excretion curves for nitrous oxide [3,4]. There are few quantitative studies of nitrous oxide elimination, but in one study, peak nitrous oxide outflow was 1 litre at 1.5 min [5]. The effect of ventilation on end-tidal concentrations has been studied [6], but it is not clear if ventilation has any effect on body elimination of nitrous oxide. In theory, there is no appreciable effect of increasing ventilation if the anaesthetic is as poorly soluble as nitrous oxide [7]. Arterial oxygen desaturation may develop when nitrous oxide-oxygen anaesthesia is discontinued and the patient breathes room air [8]. Because nitrous oxide is more soluble in blood than nitrogen, the excretion of nitrous oxide into the alveoli is greater than the uptake of nitrogen from the alveoli into the blood. The dilution of alveolar oxygen by nitrous oxide may cause arterial hypoxaemia. This diffusion hypoxia is said not to be clinically significant in healthy patients who maintain normoventilation [9, 10]. During controlled normoventilation of patients undergoing neurosurgery, Pa^t decreased by 2.2 kpa, 10 min after the inspired gas was changed from 79% nitrous oxide in oxygen to air [11]. Normal subjects sedated with 75 % nitrous oxide for 90 s during moderate hyperventilation demonstrated arterial desaturation, caused mainly by apnoea [12]. The aim of this investigation was to study nitrous oxide elimination rates during recovery with normoand hypoventilation after different durations of exposure to nitrous oxide, and to evaluate the influence on end-tidal oxygen concentration and oxygen saturation. PATIENTS AND METHODS The study was approved by the Ethics Committee of the University of Goteborg. A non-rebreathing system with a Servo 900C ventilator was used, with a 1-litre mixing box connected distal to the expiratory valve. Gas for concentration analyses was sampled from the outlet of the mixing box and from the connection between the Y-piece and the tracheal tube. A Capnomac Ultima (Datex Instrumentarium OY, Helsinki, Finland) monitored inspiratory and end-tidal concentrations of nitrous oxide, oxygen, carbon dioxide and isoflurane or enflurane. This gas monitor also intermittently monitored mixed expired gas concentrations. The gas analyses were collected on-line to a computer using a Datex Daisy software program. A Datex Cardiocap was used for pulse oximetry. An Ohmeda 5420 Volume Monitor (Ohmeda, BOC Health Care Division) measured expiratory ventilation volume. This is a turbine vane monitor with a stated tidal volume accuracy of ± 8 % or ± 40 ml, whichever is greater, within the range ml. A heat-moisture exchanger was used for inspiratory gas humidification. Anaesthesia was induced with thiopentone 4 5 mg/kg body weight, tracheal intubation was facilitated with suxamethonium 1 mg kg" 1 and pancuronium mg kg" 1 was used for SvEINN EINARSSON, M.D.; OLA STENQVIST, M.D., PH.D.; ANDERS BENGTSSON, M.D., PH.D.; ERIK HOUI/TZ, M.D. ; JAN PETER BENGTSON, M.D., PH.D. ; Department of Anaesthesia and Intensive Care, Sahlgren Hospital, University of Gdteborg, Goteborg, Sweden. Accepted for Publication: February 8, 1993.

2 190 BRITISH JOURNAL OF ANAESTHESIA TABLE I. Patient data (mean (range)) Group N-30 H-30 N-60 H-60 N-120 H-120 Weight (kg) Height (cm) Age (yr) Sex (M/F) 80 (56-95) 176 ( ) 45 (17-81) 4/2 75 (58-94) 175 ( ) 36 (15-73) 4/2 80 (65-91) 180 ( ) 54 (23-76) 6/0 69 (56-81) 170 ( ) 40 (15-69) 3/3 68 (43-93) 173 ( ) 52 (18-77) 3/3 67 (52-80) 172 ( ) 37 (22-50) 1/ i c E o TO FIG. 1. Mean (SEM) nitrous oxide excretion rates (KN,O) during normoventilation and hypoventilation after nitrous oxide exposure of 30 ( ), 60 ( ) or 120 (O) min. neuromuscular block. Fentanyl (total dose mg kg" 1 ) was used at induction and during anaesthesia, to provide analgesia. Inspiratory concentrations of 30 % oxygen, < 2 % enflurane or isoflurane, and % nitrous oxide were used. Nitrous oxide was discontinued during the anaesthetic in order to simulate postanaesthetic conditions, after the patients had received it for 30, 60 or 120 min. Then air and oxygen were used, with the inspiratory oxygen concentration unchanged at 30%. The inhalation agent vaporizer setting was also unchanged during the subsequent 30-min study period. To substitute for the anaesthetic effect of nitrous oxide, midazolam 0.04 mg kg" 1 was given i.v. 5 min before cessation of nitrous oxide administration. During the period of nitrous oxide elimination, either normoventilation or hypoventilation was maintained: the inspiratory ventilation volume used to maintain an end-tidal carbon dioxide concentration (E' COI ) of 4.0 % was either kept constant or reduced by 50 % 5 min before the discontinuation of nitrous oxide and then kept constant during the following 30-min study period. Ventilatory frequency was unchanged, at 14 b.p.m. If 5p o, decreased to less than 90%, the inspiratory oxygen concentration was increased by 5 % increments until oxygen saturation reached 90 % or more. After obtaining informed consent, we studied 36 ASA I or II patients undergoing orthopaedic surgery with an expected anaesthesia duration of at least 2.5 h. Patients were allocated randomly to one of six groups: Group N-30. Patients were given nitrous oxide (Fi Nt ) for 30 min. The inspiratory ventilation volume used to maintain E' COJ 4.0% at 30 min was kept constant during the following 30- min study period. Group H-30. Patients were given nitrous oxide for 30 min, as in group N-30. The inspiratory ventilation used to maintain E' COJ 4.0 % at 25 min was reduced by 50 % 5 min before the nitrous oxide elimination period. Ventilatory frequency was unchanged, at 14 b.p.m. Group N-60. Nitrous oxide was given for 60 min and the inspiratory ventilation volume maintaining E' CO 4.0 % at 60 min was kept constant during the study period. Group H-60. Nitrous oxide exposure was for 60 min and inspiratory ventilation volume was reduced by 50 % 5 min before discontinuation of nitrous oxide, as in group H-30. Group N-120. Nitrous oxide exposure was for 120 min. Ventilation volumes were as in groups N- 30 and N-60. Group H-120. Nitrous oxide was given for

3 N,O ELIMINATION AND DIFFUSION HYPOXIA min. Ventilation volumes were as in groups H-30 and H-60. Two patients, one in group H-60 and one in group H-120, had Sp o, values less than 90% at 6 and 3 min, respectively. Fl Oi was increased to 0.55 and 0.35, respectively, until adequate Sp o, was obtained. The inspiratory to end-tidal oxygen concentration differences of these patients were extrapolated to Fi Of 0.30, as were the oxygen saturation values. TABLE II. Best-fit curves created for each patient according to the formula VN,0 (ml/70 kg min~ l ) = bt" m (mean (SEM)) b SEM m SEM Group N-30 H-30 N-60 H-60 N-120 H Statistics Results are expressed as mean (SEM) or as mean (range) and compared by one-factor ANOVA repeated measures within groups, followed by Fischer's PLSD test. A one-way ANOVA and Fischer's PLSD test were used for comparison between groups. Best-fit curves for nitrous oxide excretion rates were created according to a power model. Statistical significance analysis was performed through the calculation of best fit curves for each patient. The following formula was used: FN 2 O (ml/70 kg min" 1 ) = br m where b = 1-min elimination rate; t = time after cessation of administration of nitrous oxide (min); m = negative "slope" of the curve. Statistical significance was assumed for P < RESULTS Patient data are presented in table I. The nitrous oxide excretion rates were normalized by weight to ml min~v70 kg (fig. 1). The mean excretion rate was 987 ml min" 1 at 1 min, declining to 103 ml min" 1 at 30 min. Best fit curves gave significantly greater b and m values in groups N-30 and N-60 compared with groups H-60 and H-120 (P < 0.05 in all cases), indicating larger initial excretion rates in groups N- 30 and N-60, but a more prolonged nitrous oxide excretion in groups H-60 and H-120 (table II). The mean expired ventilation volume was 7.13 litre in the three groups with normoventilation and 3.20 litre in the groups with hypoventilation. E' COJ decreased gradually in the groups with normoventilation during the 30-min nitrous oxide elimination period. Mean E' COI values at 30 min were 3.60 (SEM 0.16)% in group N-30, 3.83 (0.11)% in group N-60 and 3.78 (0.17)% in group N-120. Corresponding values during hypoventilation were 6.75 (0.24)%, 7.35 (0.15)% and 7.40 (0.20)% in groups H-30, H-60 and H-120, respectively. Mean end-tidal nitrous oxide concentration declined rapidly from 66 to 70 % at 0 min to 6-9 % at 5 min and to 2 4 % at 30 min during normoventilation. In the groups with hypoventilation, the values were % at 5 min and 5-9 % at 30 min (fig. 2). The maximum inspiratory to end-tidal oxygen concentration difference was 5.7% in group N-30, 6.0% in group N-60 and 6.3% in group N-120, all at 3 min. During hypoventilation, corresponding maximum values were 11.9% at 4 min in group H-30, 14.8 % at 10 min in group H-60 and 14.8 % at 15 min in group H-120 (fig. 3). The curves were significantly different between normo- and hypoventilation (P < 0.05 in all cases), but also between groups H-30 and H-60, and between groups H-30 and H-120 (P < 0.05 in both cases) : ' : 50 End-1tidal N FIG. 2. Mean (SEM) end-tidal nitrous oxide concentration during normoventilation and hypoventilation after nitrous oxide exposure of 30 ( ). 60 ( ) or 120 (O) min.

4 192 BRITISH JOURNAL OF ANAESTHESIA i i i I i FIG. 3. Mean (SEM) inspiratory to end-tidal oxygen difference (i o E' Ot ) during normoventilation and hypoventilation after nitrous oxide exposure of 30 (\j), 60 ( ) or 120 (O) min. Mean Sp Oi was % at the time of cessation of nitrous oxide (fig. 4). The smallest mean values were 96.0 % at 15 min in group N-30, 96.5 % at 1 min in group N-60, 97.2% at 4 min in group N-120, 96.3% at 3 min in group H-30, 93.8% at 5 min in group H-60 and 93.5 % at 4 min in group H-120. The decrease in 5p Ol between 0 min and 15 min was significant in group N-30 (P < 0.05). In the groups with hypoventilation, there were significant decreases in Sp o between 0 and 4 min in group H-30, between 0 and 3-20 min in group H-60 and between 0 and 3-30 min in group H-120 (P < 0.05 in all cases). DISCUSSION This study has demonstrated that nitrous oxide is eliminated at a rate similar to that at which it is taken up. The time of exposure, within the range min, had no significant influence on excretion rates. This is an indication that the output reservoir formed by the body is finite and almost full after 30 min of nitrous oxide administration. The effect of hypoventilation on nitrous oxide elimination was most pronounced during the first few minutes, which can be interpreted as a slower elimination from the lung volume of the patient. The ventilatory volumes had little effect on the excretion rates after 5 min. Thus the greatest impact of augmenting ventilation on nitrous oxide elimination is achieved during the first 5 min. The prerequisites for diffusion hypoxia were present after the first 3-5 min of recovery, which has been suggested previously [7]. The mean inspiratory to end-tidal oxygen concentration difference during hypoventilation, after 1 2 h of exposure to nitrous oxide, reached a maximum of almost 15 % at min. This would correspond to an end-tidal oxygen concentration of 6% during air breathing. The end-tidal gas concentrations after 1 h of nitrous oxide administration can be used to simulate a period of recovery breathing room air (fig. 5). is crucial to maintain adequate alveolar oxygen concentrations during the first 30 min of recovery. had little effect on nitrous oxide elimination after about 5 min; in contrast, the end-tidal nitrous oxide concentration was about doubled in the 5 30 min interval during hypoventilation compared with normoventilation. The endtidal carbon dioxide concentration increased to about 7 % during the first 30 min of hypoventilation during recovery. Thus both nitrous oxide and carbon dioxide dilute alveolar oxygen. With air breathing, the decreasing oxygen tension affects the haemoglobin oxygen saturation on the steep part of the oxygen dissociation curve, with resultant arterial desaturation. In figure 5, we have attempted to calculate the arterial oxygen saturation from the oxygen dissociation curve, using the inspiratory to end-tidal oxygen concentration difference in the study [13]. after anaesthesia without nitrous oxide would decrease the end-tidal oxygen concentration, but to a much smaller extent. A recent review on mechanisms of postoperative hypoxaemia found diffusion hypoxia during nitrous oxide elimination "unimportant" [14]. Our study conflicts with this view. In spite of an inspiratory oxygen concentration of 30 %, we found decreases in Sp 0, during both normo- and hypoventilation. In clinical practice, we recommend that nitrous oxide is discontinued early in the recovery period, so that the patient is monitored during the first elimination phase. As hyperventilation does not influence the elimination of nitrous oxide to a large extent, and increases the risk of a subsequent hypoventilation during transport to the recovery area, hyperventilation should be avoided. During transport from the operation theatre to the recovery

5 N,O ELIMINATION AND DIFFUSION HYPOXIA ; 1 Bfl T T. "Jar 97 ' < e 193 ; ii i o" > Time (min] FIG. 4. Mean (SEM) pulse oximetry oxygen saturation (SpOl) during normoventilation and hypoventilation after nitrous oxide exposure of 30 ( ). 60 ( ) or 120 (O) min lili mm FIG. 5. Calculated arterial oxygen saturation (Sa^ (heavy stippling) and alveolar gas concentrations (ATPD) ( = oxygen; Q = carbon dioxide; 0 = nitrous oxide; S3 = nitrogen) during recovery after 1 h of nitrous oxide anaesthesia with normo- and hypoventilation using room air. room, we strongly recommend the use of oxygen therapy, regardless of the duration of the nitrous oxide anaesthesia. In conclusion, the nitrous oxide elimination rates were close to those predicted previously, but there is a risk of diffusion hypoxia for at least 30 min from cessation of nitrous oxide administration in the presence of hypoventilation. ACKNOWLEDGEMENT Financial support was provided by AGA AB Medical Research Fund. REFERENCES 1. Severinghaus JW. The rate of uptake of nitrous oxide in man. Journal of Clinical Investigation 1954; 33: Mapleson WW. Quantitative prediction of anesthestic concentrations. In: Papper EM, Kitz RJ, eds. Uptake and Distribution of Anesthetic Agents. New York: McGraw-Hill, Frumin MJ, Salanitre E, Rackow H. Excretion of nitrous oxide in anesthetized man. Journal of Applied Physiology 1961; 16: Salanitre E, Rackow H, Greene LT, Klonymus D, Epstein RM. Uptake and excretion of subanesthetic concentrations of nitrous oxide in man. Anesthtsiology 1962; 23: Sheffer L, Steffenson JL, Birch AA. Nitrous-oxide-induced diffusion hypoxia in patients breathing spontaneously. Anesthtsiology 1972; 37: Stoelting RK, Eger El II. The effects of ventilation and anesthetic solubility on recovery from anesthesia: an in vivo and analog analysis before and after equilibration. Anesthesiology 1969; 30: Eger El II. Anesthetic Uptake and Action. Baltimore: Williams and Wilkins, 1974; Fink BR. Diffusion anoxia. Anesthesiology 1955; 16: Rackow H, Salanitre E, Frumin MJ. Dilution of alveolar gases during nitrous oxide excretion in man. Journal of Applied Physiology 1961; 16: Frumin MJ, Edelist G. Diffusion anoxia: a critical reappraisal. Anaesthesia 1969; 31: Sugioka K, Cattermole RW, Sebel P. Arterial oxygen tensions measured continuously in patients breathing 21% oxygen and nitrous oxide or air. British Journal of Anaesthesia 1987; 59: Northwood D, Sapsford DJ, Jones JG, Griffiths D, Wilkins C. Nitrous oxide sedation causes post-hyperventilation apnoea. British Journal of Anaesthesia 1991; 67: Guyton AC. Textbook of Medical Physiology. Philadelphia: Saunders, 1986; Jones JG, Sapsford DJ, Wheatley RG. Postoperative hypoxaemia: mechanisms and time course. Anaesthesia 1990; 45: