Br.J. Anaesth. (1976), 48, 651
CYANIDE AND THIOCYANATE CONCENTRATIONS FOLLOWING SODIUM NITROPRUSSIDE INFUSION IN MAN C. J. VESEY, P. V. COLE AND P. J. SIMPSON SUMMARY
Hydrogen cyanide (HCN) has been detected in animal and human tissues following poisoning with sodium nitroprusside (SNP) (Hermann, 1886; Johnson, 1929; Annotation, 1936, 1948; LazarusBarlow and Norman, 1941). Hill (1942) also found HCN in a fatal case of SNP poisoning and was able to demonstrate cyanide formation in whole blood or tissue incubated with nitroprusside. This has been confirmed by others (Page et al., 1955; Vesey, Cole and Simpson, 1976). Following the introduction of SNP to clinical practice (Page, 1951; Page et al., 1955), toxic effects were reported, and these were attributed to increased plasma thiocyanate (SCN) concentrations following prolonged dosage with the drug (Nourok et al., 1964). Johnson (1929) was the first to demonstrate the hypotensive action of SNP in man and he was convinced that the liberation of cyanide was not the cause of any toxic effects. However, it has been suggested (Vesey and Cole, 1975) that HCN is the cause of metabolic acidosis and fatalities following hypotensive anaesthesia with SNP (Jack, 1974; Merrifield and Blundell, 1974; Davies et al., 1975). In our previous studies on patients infused with SNP during surgery, we found that plasma cyanide concentrations were increased fourfold whilst plasma thiocyanate (the major end-product of cyanide detoxication) was increased by only small amounts, if at all. In one patient who received an infusion for 5 days, however, plasma SCN and HCN concenCYRIL J. VESEY, M.SC.; PETER V. COLE, F.F.A.R.C.S. ; PETER J. SIMPSON, F.F.A.R.C.S.; Department of Anaesthesia, St
Bartholomew's Hospital, London EC1A 7BE.
nations both increased by a factor of 10 and the red cells-to-plasma HCN ratio was 200 : 1 (Vesey et al., 1974). These observations have now been extended in a further series of patients. In addition, the HCN concentrations in expired air have been measured to evaluate the possible use of this procedure as an index of plasma cyanide concentrations during SNP infusions. PATIENTS AND METHODS
Nine male and 17 female patients undergoing major orthopaedic surgery were studied (table I). In our practice hypotensive anaesthesia is not contemplated for anyone with a history of either cardiovascular or cerebrovascular disease. One hour after premedication, the patients were anaesthetized with thiopentone. Suxamethonium was given to facilitate orotracheal intubation. The electrocardiogram, and arterial pressure, measured from an indwelling radial artery cannula, were monitored continuously. Anaesthesia was maintained with a minimum of 33% oxygen in nitrous oxide. Patients lying supine or in the lateral position were allowed to breathe spontaneously. IPPV was considered necessary for patients in the prone position, and in these cases tubocurarine was administered. Minimal concentrations of trichloroethylene or halothane were used when necessary. Blood loss during operation was measured by swab weighing and suction volume, and blood transfusion was seldom needed during the hypotensive period. Reversal of neuromuscular blockade was obtained in all cases following return of the arterial pressure to the pre-hypotensive
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
Twenty-six patients, receiving an infusion of sodium nitroprusside (SNP) during surgery, had considerable increases in both red cell and plasma cyanide concentration, but only small changes in plasma thiocyanate concentration. There was a linear relationship between both plasma and RBC cyanide concentrations and the total dose of SNP. The expired cyanide concentration followed the changes in the plasma. We believe that the development of metabolic acidosis, and the recent fatalities involving SNP, are attributable to histotoxic hypoxia as a result of excessive plasma concentrations of cyanide. On the basis of our results, we recommend that the total dose of SNP should not exceed 1.5 mg/kg during short-term infusions and that the plasma cyanide should not exceed 300 nmol%. Plasma thiocyanate concentrations are, in general, an unreliable indication of extent of exposure to cyanide, although they may become important during long-term infusions.
BRITISH JOURNAL OF ANAESTHESIA
652
TABLE I. Details of patients and SNP dose rates
(yr)
Sex
Smokers
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
57 29 55 41 63 52 53 28 46 33 38 52 41 15 77 42 27 50 54 75 15 58 17 32 42 50
M F F M F M M M M M F M F F F F M F F F F F F F F F
+ — + + — — + + + + — — + — — — + — + — _ — — _ _ —
55 90 95 80 80 118 168 115 65 95 120 30 85 57 85 108 65 38 120 90 141 95 54 80 82 69
value, and the patients were observed for 1-2 h in an adjacent recovery room. SNP in physiological saline (200 fig/ml) was introduced to the existing i.v. infusion (Hartmann's solution B.P.), as near to the cannula as possible to minimize deadspace. Administration of the drug through a drip controller (IVAC 200) was commenced immediately before surgery at dose rates sufficient to reduce the systolic arterial pressure to about 60 mm Hg (mean dose rates and total doses are given in table I). Baseline samples of expired air and arterial blood were obtained immediately before the infusion was commenced. The air was collected into a plastic Douglas bag over a timed period (usually 3-5 min). Blood was withdrawn into a pre-cooled heparinized syringe, sealed and placed in a mixture of ice and water. The second samples of air and blood were collected immediately following the administration of nitroprusside. Subsequent blood samples were taken 1 and 24 h later. Control expired air samples were also collected from 15 patients who were comparable in age, sex, weight and anaesthetic technique with those receiving SNP.
SNP dose (nig)
20 16 53 35 8 27 51 36 23 34 67.5
4 26 15 12 24 19 3 26 10 95.4
27 12 42 16 7.6
M e a n dose rate (nmol/kg/min) 11.7 10.4 29.3 17.8
5.1 8.4 12.0 10.4 15.6 18.4 24.7
6.1 24.0 20.9
7.4 15.5 16.2
5.1 10.3
6.6 44.9 14.6 15.4 25.3
9.6 7.5
Isolation of cyanide (a) Plasma cyanide. Immediately following collection, the blood samples were centrifuged at 4 °C and an aliquot of the separated plasma (5-10 ml) was assayed for cyanide. The remainder of the plasma was stored at — 20 °C, some for later determination of thiocyanate, and some, together with an aliquot of the red cells, for eventual assay of vitamin B 12 (the subject of a separate communication). Plasma HCN was isolated by the method of Boxer and Rickards (1951). Scrubbed oxygen-free nitrogen was passed at a rate of 1 litre/min for 1 \ h through a mixture of 1 vol plasma and 3 vol 10% trichloroacetic acid (TCA). HCN was removed from the gas stream by passage through two tubes in series, each containing sodium hydroxide 0.2 mol/litre. (The volume of NaOH used was adjusted to suit the amount of HCN expected.) (b) Red cell cyanide. The isolated red cells were washed six times with normal saline to remove any traces of plasma thiocyanate (contamination of the cells with SCN results in artefactual formation of HCN on assay (Vesey and Wilson, in preparation)) and stored at 4 °C overnight. After the determination
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
Patient
Infusion time (min)
Age
CYANIDE, THIOCYANATE AND NITROPRUSSIDE
Assay of cyanide and thiocyanate Both plasma thiocyanate and separated cyanide were determined by the spectrophotometric method of Aldridge (1945). Potassium thiocyanate solutions of known concentration were used as standards. Plasma thiocyanate was determined on duplicate 1-ml aliquots of the supernatant obtained by centrifuging a mixture of 1 ml of plasma and 9 ml of 10% TCA. One millilitre of NaOH solution containing HCN from (a) (b) or (c) above, or 1 ml of the protein-free TCA extract of the plasma, was mixed with 0.5 ml of HC1 1 mol/litre and two drops of saturated bromine water. Excess bromine was removed by adding three drops of a 2% arsenic trioxide solution and mixing vigorously in a stoppered tube. Ten minutes after the addition of 1.8 ml of a pyridine-benzidine mixture, the absorbance was measured at 532 nm. The pyridine-benzidine mixture was prepared from 1 vol of a solution of 5% benzidine hydrochloride in hydrochloric acid 1 mol/litre and 5 vol of a pyridine-HCl-water solution (6 : 1: 4 by vol respectively).
RESULTS AND DISCUSSION
The results confirm that there is a significant increase in plasma HCN following the infusion of SNP (Vesey et al., 1974). In addition they indicate that for short periods of infusion (2 h approximately) there is a close correlation between the increase in plasma cyanide, measured immediately after the infusion, and the total dose of SNP (r = 0.94, P< 0.001, n = 24 (fig. 1)). The correlation of plasma cyanide
with the mean rate of SNP infusion is poorer, but statistically significant (r = 0.843, P< 0.001, n = 24). When cyanide is introduced into blood, in vitro or in vivo, most of it enters the red cells (Vesey and Wilson, in preparation). In this study 98.4% (SD ±0.83%) of the blood cyanide, immediately after the infusion, was found in the red cells. As with plasma, the red cell cyanide concentrations were more I 200,
1J %
1 /
o —
t.
. 1.0
2.0
3.0 4.0 5.0 //mol SXP/kg boil; \\l
6.0
FIG. 1. Relationship between the increases in plasma cyanide, immediately following infusionj and the total dose of SNP (r = 0.94, P< 0.001, n = 24). The irregular result was for patient 21 who received the highest dose of the series. This result was probably influenced by the dilution of the patient's plasma with the transfusion of 800 ml of blood and 500 ml of saline during surgery. It was not included in our calculations.
3 C-"g20
1
2
3
4
5
•
6
/imolSNP/kgbodywt
FIG. 2. Correlation between the increases in red cell cyanide concentrations immediately following infusion and total SNP dose (r = 0.924, P < 0.001, n = 26).
closely related to the total dose of SNP than to the rate of infusion (r = 0.924, P< 0.001, n = 26; r = 0.849, P< 0.001, n = 26 respectively, fig. 2).
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
of the packed cell volume of the washed cells an aliquot (1-5 ml) was mixed, in an aeration tube, with an equal volume of water. Three volumes of 10% TCA were added to the haemolysate and the liberated HCN was transferred to sodium hydroxide in a stream of nitrogen, as for plasma. (c) Expired air cyanide was isolated and concentrated by the method of Boxer and Rickards (1952). Each air sample was drawn through a solution of silver sulphate 0.02 mol/litre in sulphuric acid 0.1 mol/litre in a Drechsel bottle with a sintered inlet. Any residual anaesthetic gases together with carbon dioxide were removed from the solution by the passage of nitrogen 1 litre/min for 20 min. Following the addition of sulphuric acid (11 mol/litre) to the silver sulphate solution to increase the concentration to 3.5 mol/litre, HCN, set free, was transferred to sodium hydroxide as in (a) and (b) above.
653
BRITISH JOURNAL OF ANAESTHESIA
654
TABLE II. Plasma and RBC cyanide and plasma thwcyanateconcentrationsbefore and after infusion of SNP
Plasma
Red blood cell
Cyanide nmol%
Thiocyanate l±mol%
Pre
Post
lh post
24 h post
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
2 0 15 — 13 14 8 2 4 18 20 9 30 6 20 2 5 2 10 2 0 5 2 0 14 4
5 32 204 79 38 29 94 30 34 143 205 18 200 42 40 102 39 23 80 6 190 75 18 155 29 12
— — 194 — — 0 35 5 13 24 105 6 34 27 26 68 11 — 14 3 129 45 3 81 0 0
3 5 45 3 20 — 8 5 6 19 10 6 0 27 18 0 3 — 16 2 0 5 11 3 — —
Pre
Post
lh ]post
24 h post
3.8 5.2
3.9 6.0
4.0 6.3
3.8 4.4
12.4
13.8 11.8
13.8 12.4
15.3
3.2 4.3
3.7 4.5
13.6 10.5 13.8 13.0
17.2 10.9 14.7 13.6
16.3 11.7 15.3 14.8
12.5
3.6 2.0
4.6 1.8
5.4 1.5
6.6 0.5
10.4
14.0
14.8
14.5
2.6 2.3 3.3 4.8 1.8 3.3 4.7 4.8
3.7 2.3 3.8 5.2 2.2 4.3 5.2 7.8
3.2 2.3 4.6
3.0 1.8 4.0 5.8
— 4.8 5.2 8.9
4.6 4.5 8.8
14.2
15.9
16.3
14.0
4.8 4.9 3.3 4.2
4.5 6.2 4.3 3.8
4.4 7.1 4.5 3.7
3.7 5.9 — —
9.8 2.5 3.3
Both mean red cell and mean plasma cyanide concentrations, (nmol HCN/100 ml)/((xmol SNP/kg), decreased to less than a half of the post-infusion values, 1 h after infusion had ceased (fig. 3). However, where the post-infusion concentrations were high, the decrease in cyanide concentration was much slower (table II). There are obvious differences in the rate at which HCN is metabolized (table I I : patient 8 compared with patient 9). In some cases there was a decrease to less than the pre-infusion values after 24 h (fig. 3 and table II). This may indicate an enhanced detoxication in response to exposure to HCN or, more probably, an increased urinary excretion as a result of the i.v. infusion. Boxer and Rickards (1952) showed that minute amounts of HCN are normally excreted in expired air and are related to the metabolic rate. James (1938) found that the expired HCN concentration increased after the administration of cyanide. Following SNP infusion there were definite increases in expired HCN in the majority of patients (fig. 4). In control patients undergoing a similar type of surgery and anaesthesia,
Z3
9.7 2.8
7.2
13.3 11.7
Pre
Post
0.024 0.028 0.113 0.038 0.063 0.222 0.073 0.017 0.097 0.053 0.014 0.080 0.068 0.003 0.207 0.047 0.080 0.160 0.040 0.017 0.029 0.038 0.019 0.008 0.222 0.291
0.155 1.860 11.70 5.880 0.875 2.200 8.240 2.500 2.920 3.700 18.97 0.680 8.230 4.102 1.220 6.710 2.270 2.520 5.420 0.100 21.20 5.300 1.750 8.000 2.200 0.665
lh post
24 h post
0.042 0.335 8.50 1.639 0.172 0.513 3.450 2.140 0.472 0.393 17.35 0.066 0.548 1.143 0.375 4.024 0.688
0.015 0.055 0.248 0.116 0.051 0.108 0.084 0.138 0.166 0.263 0.035 0.188 0.007 0.130 0.105 0.116
—
—
0.436 0.050 16.53 1.328 0.273 5.400 0.513 0.220
0.063 0.030 0.210 0.118 0.072 0.126 — —
60
e- 40
5«5 oc •5. a. 30
I
I? 20 |
rh
1 10
11-25
of
T
11
X
H-21
»"20
•
• £
^ ^ 3 ?:" 2
T
ii
H -26
11-24
post
1 h post
— 11-22
24 h post
FIG. 3. Change in mean plasma and red cell cyanide concentrations. The difference between values after and before infusion expressed as (nmol HCN/100 ml)/ SNP/kg) together with SD immediately following, 1 h following, and 24 h following infusion.
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
Patient
Cyanide nmol%
655
CYANIDE, THIOCYANATE AND NITROPRUSSIDE
CONTROL PATIENTS
SNP PATIENTS
o CO
o C 4
z: U 3 I Q 2 U4
a: 1 E X
w
50
100
Smokers
150 0
50 100 150 Non-smokers
50 Smokers
0
50 100 Non-smokers
OPERATION TIME (min) FIG. 4. Changes in expired cyanide (expressed as nmol HCN/kg body wt/h) for patients receiving SNP and patients undergoing similar procedures without SNP. Results in smokers and non-smokers are presented separately. Dose rates (nmol SNP/kg/min) are given at the end of each line.
expired HCN concentrations showed little change, except in some who were smokers, but were in most cases less at the end of the operation. In three of the patients who received only small amounts of SNP there was also a decrease in the amount of HCN expired. The initially high concentration of HCN in the expired gas in some smokers probably results indirectly from the HCN present in tobacco smoke. The increases in the expired HCN concentrations above pre-infusion values correlate with the increases in plasma cyanide (r = 0.874, P< 0.001, n = 19; fig. 5). The fact that log10 of the increase in expired HCN correlates with the plasma cyanide concentration (r = 0.843, P< 0.001, n = 19) suggests that the expired cyanide concentration may increase exponentially with increasing plasma cyanide. The relationship to the rate of infusion of SNP or to the total amount infused was less marked (r = 0.665, P< 0.001; r = 0.57, 73 0.001; r = 0.596, P = 0.001 respectively; n = 24). This may reflect the relatively slow rate of the detoxication of cyanide. Similarly, the difference between the mean plasma thiocyanate concentrations before and after infusion were significant only at 1 h after SNP in both smokers and non-smokers {t = 2.567, 0.05>P>0.02; t = 2.06, P 0.02 for smokers; t = 2.06, Ps^O.05 for non-smokers).
•S'40
2
= 20
values, or less, 24 h later. Increased thiocyanate concentrations normally decrease slowly; for example in smokers the return to normal values after the cessation of smoking takes 14 days. However, it is known that increases in the intake of both fluid and chloride ions enhance the rate of SCN excretion (Stoa, 1957). Although the smokers in our study received SNP at a higher mean rate than that given to the non-smokers (18.21 + 6.62; 13.95 +9.99 nmol/kg/ min respectively; n = 1 and 18), the difference is not statistically significant (t = 1.0357, P>0.1). Sensitivity to SNP is very variable but would appear to increase with age. The highest mean dose rate amongst our patients was required by patient 21, a 15-year-old girl (44.9 nmol SNP/kg/min). Mean dose rates plotted against age are shown in figure 7. Although the correlation is poor (r = —0.553) it is statistically significant (0.01 >P^0.001). However, there appears to be no sex difference in the response to SNP.
t • 10
20
30
US
50
» 60
70
Age(>r)
FIG. 7. Correlation between the rate of infusion of SNP and the ages of the patients (r = —0.553, n = 26, 0.01 > P > 0.001). GENERAL DISCUSSION AND CONCLUSIONS
Although 98% of HCN in the blood, following an infusion of SNP, is located in the red cells, its fate is unknown. In the rat, the enzyme /J-mercaptopyruvate sulphur transferase, which converts HCN to thiocyanate, is present in the red cells (Sorbo, 1957), but this enzyme is believed to show only low activity in the red cells of man (Van den Hamer, Morell and Scheinberg, 1967). The inference that the red cell cyanide content has no apparent toxicological significance is suggested by
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
-a ir- 2
657
CYANIDE, THIOCYANATE AND NITROPRUSSIDE
occurred as a result of the absorption of 24 mg of HCN (table III, case 19). Thus 25 mg of HCN could reasonably be expected to be close to the lethal i.v. dose. Assuming a blood volume of about 70 ml/kg in a 70-kg man (Airman and Dittmer, 1971), this would result in a blood cyanide concentration of 19 (xmol%, or 19 |i.mol/litre plasma. We assume that, as we have confirmed in animal studies, 90% of the blood HCN is in the red cells. Similarly, minimum whole blood concentrations recorded in the literature (table III), would represent plasma concentrations of about 10 (imol HCN/litre and, where death has occurred from inhalation of HCN (cases M5 and M6), the calculated value would be even less. Where death has been delayed (cases 19 and M7), a situation similar to SNP overdose, the calculated plasma concentrations approach 1 (xmol%. Thus the fatal cyanide concentration would appear to be between 10 and 20 fxmol/ litre plasma. Mehaffey (1942) showed that the lethal oral dose of SNP for guineapigs was about 15 mg and, on this basis, McDowall, and others (1974) suggested that 250 mg of SNP (about 3.6mg/kg or 12 jxmol/kg) might be the lethal dose in man. Nourok and colleagues (1964) found that the hypotensive effects of SNP by mouth were ho greater than those produced by the thiocyanate arising from it. This suggests that
TABLE I I I . Whole blood cyanide concentrations in man following poisoning, together with postulated plasma HCN concentrations
Reference Gettler and Baine (1938)
Sunshine and Finkle (1964)
Bonnischen and Maehly (1966)
Shanahan (1973)
Case no.
Blood cyanide (fimol/litre)
6 19
119 96
15 48 (M) 54 (M) 33 (F) 86 (M) 40 (M) P5 M5 M6 M7
152
102 154 92\ 70/ 130
3 4
156 122
Relevant details Oral cyanide Oral cyanide; died after 3h Inhaled HCN
43] 112 1 82 f
6lJ
Oral cyanide Inhaled HCN Oral cyanide Inhaled HCN KCN absorbed through skin; died within 4 h — —
Postulated plasma HCN concns. (nmol/litre)* 12 9.6 15 4 11 8 6 10 15 9 7 3* 16 12
Only the lower concentrations have been selected since, obviously in most cases of poisoning, doses much larger than the lethal dose will have been taken. * Plasma cyanide concentrations are postulated to be one-tenth those of whole blood, assuming 90% of the total HCN to be present in the red cells. In case M7, where absorption would be slow and similar to the situation with SNP, 98% of blood cyanide is likely to be present in the red cells.
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
a comparison of the highest concentrations found after SNP infusion (12, 19 and 21 jamoF/o (table I I ) ; 32 (xmol% (Vesey et al., 1974)) with some of the lower total blood cyanide values reported in cases of cyanide poisoning (table III). It has been suggested that between 10 and 12 (imol% is the minimum whole blood cyanide concentration which will cause lethal effects in an otherwise healthy man (Gettler and Baine, 1938; Sunshine and Finkle, 1964). Comparison of these figures would suggest also that HCN in the plasma, rather than in the red cells, is the critical factor in cyanide intoxication. Since the tissues are readily permeable to lipidsoluble HCN, the amount passing into them to cause toxic symptoms, will be closely related to the plasma concentration. It should be possible to suggest an upper limit for plasma HCN and, on the basis of our results (fig. 1), to suggest a maximum total dose of SNP (Vesey, Cole and Simpson, 1975). The minimum lethal oral dose of hydrogen cyanide is variously suggested as 40 mg (Rentoul and Smith, 1973), 0.5mg/kg (Hunt, 1923; Gettler and Baine, 1938) and 37 mg (Gonzales, Vance and Helpern, 1940). Only part of an oral dose would be absorbed and some would be detoxicated in the liver before entry into the general circulation (Gettler and Baine, 1938). The last authors calculated that one death
658
BRITISH JOURNAL OF ANAESTHESIA TABLE IV. Some SNP doses and the estimated resulting plasma cyanide concentrations Reference
Dose of SNP
Details
McDowall and others (1974)
250 mg (3.6 mg/kg or 12|imol/kg) 320 mg/h (4.6 mg/kg/h or 15.3 (imol/kg/h)
Suggested lethal dose for man
250 mg (4.2 mg/kg or 14 (imol/kg)
Merrifield and Blundell (1974)
750 mg (2 mg/kg/h or 20 nmol/kg in 3 h)
A. B. Davies (pers. comm.) Jack (1974)
800 after 1-3 h infusion 1020 after 1 h infusion
Smallest toxic infused dose for man calculated from baboon studies 42-year-old female infused over 930 90 min; severe reversible acidosis 20-year-old male infused over 1300 after 3 h 5 h. Fatal infusion (?) 2200 after 5 h infusion 14-year-old male infused over 2200 80 min. Fatal 39-year-old male infused over 2400 30 min. Fatal
400 mg (10 mg/kg or 33.6 nmol/kg) 750 mg (approx. 11 mg/kg or 37 nmol/kg) Editorial (1975) 3 mg/kg/h (30 (i.mol/kg Recommended rate in3h) Vesey, Cole and 1.5 mg/kg Recommended total dose Simpson (1975) (5 nmol/kg)
2000 after 3 h infusion 330 after 1-3 h infusion
* The estimate of plasma cyanide is based on figure 1 and assumes that the relationship between plasma cyanide and SNP dose is still linear above the concentrations that we have measured.
SNP, taken orally, decomposes in the stomach or liver, and only its breakdown product, HCN, and detoxication product, SCN, enter the general circulation. Therefore the lethal i.v. dose of SNP would be expected to be less than 250 mg. Assuming that the linear relationship between the total dose of SNP and plasma HCN applies at concentrations higher than those we have measured, 250 mg of SNP would cause a plasma concentration of 0.8 \irao\ HCN/100 ml if infused over 2-3 h (table IV). This is close to the lower lethal value for plasma cyanide (1 (xmol%) suggested above. SNP doses which produced metabolic acidoses and, in some cases, death, together with estimated plasma HCN concentrations, are compared with maximum dose rates recommended previously (Editorial, 1975) (table IV). Thus it would appear that, for short-term infusions, a total dose of SNP resulting in plasma HCN concentrations of 1 fxmol% (15 nmol SNP/kg; 4.5 mg/kg (fig. 1)) would be very near to the toxic dose. HCN present in the plasma enters the tissues and is converted to thiocyanate by the mitochondrial enzyme rhodanese. As judged by the increase in plasma cyanide and the slow increase in plasma thiocyanate, the rate of cyanide detoxication in our patients was slower than that of the release of HCN from SNP.
The rate of detoxication of HCN is limited by the availability and low permeability of thiosulphate (Sorbo, 1962). The infusion of SNP, together with the slow release of HCN, will maintain continuous exposure of the tissues to cyanide. Detoxication mechanisms may thus become exhausted at blood concentrations which would not normally cause problems in acute HCN poisoning. This may explain why the toxic effects of SNP, in animals, compared with those of cyanide, although similar, are delayed (MehafFey, 1942), as may occur also in clinical practice (McDowall et al., 1974; Merrifield and Blundell, 1974). On present evidence, we suggest that the total dose of SNP should be limited to 1.5 mg/kg over the duration of an average surgical operation (Vesey, Cole and Simpson, 1975). High doses seem to be unnecessary (Adams, 1975) and, in our hands, the addition of small amounts of halothane and, occasionally, jS-blockade has enabled us to limit the total dose to less than 100 mg and the dose rate to less than 45 nmol/kg/min (0.8 mg/kg/h). Where the total dose is likely to exceed the suggested value, or during long-term administration, we recommend monitoring of plasma cyanide concentrations. Since log10 expired air cyanide concentration is linearly related to the
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
MacRae and Owen (1974)
Estimated plasma cyanide (nmol%)*
CYANIDE, THIOCYANATE AND NITROPRUSSIDE
ACKNOWLEDGEMENTS
The authors would like to thank the Joint Research Board of St Bartholomew's Hospital for invaluable financial assistance and the staff of the Anaesthetic Research Laboratory, St Bartholomew's Hospital, for their expert technical help. REFERENCES
Adams, A. P. (1975). Techniques of vascular control for deliberate hypotension during anaesthesia. Br.J. Anaesth., 47, 777. Aldridge, W. N. (1945). The estimation of microquantities of cyanide and thiocyanate. Analyst, 70, 474. Altman, P. L., and Dittmer, D. S. (1971). Blood and Other Body Fluids, p. 1. Bethesda: Federation of American Societies for Experimental Biology. Annotation (1936). Pharm.J., 82, 125. (1948). Pharm.J., 160, 287. Bonnischen, R., and Maehly, A. C. (1966). Poisoning by volatile compounds. I l l : Hydrocyanic acid. J. Forens. Set., 11, 516.
Boxer, G. E., and Rickards, J. C. (1951). Chemical determination of vitamin B 12 . I I : The quantitative isolation and colorimetric determination of millimicrogram quantities of cyanide. Arch. Biochem. Biophys., 30, 372. (1952). Determination of traces of hydrogen cyanide in respiratory air. Arch. Biochem. Biophys, 39, 287. Davies, D. W., Kadar, D., Steward, D. J., and Munro, I. R. (1975). A sudden death associated with the use of sodium nitroprusside for induction of hypotension during anaesthesia. Can. Anaesth. Soc. J., 22, 547. Editorial (1975). Sodium nitroprusside in anaesthesia. Br. Med. J., 2, 524. Gettler, A. O., and Baine, J. O. (1938). The toxicology of cyanide. Am. J. Med. Sci., 195, 182. Gonzales, T. A., Vance, M., and Helpern, M. (1940). Legal Medicine and Toxicology, 2nd edn, p. 542. New York and London: D. Appleton-Century. Hermann, L. (1886). Ueber die Wirkung des Nitroprussidnatriums. Arch. Physiol., 39, 419. Hill, H. E. (1942). A contribution to the toxicology of sodium nitroprusside. I: Decomposition and determination of sodium nitroprusside. Aust. Chem. Inst. J. Proc, 9,89. Hunt, R. (1923). Cyanwasserstoff, Nitroglukoside, Nitrile5 Rhodanwasserstoff, Isocyanide; in Handbuch der Experimentellen Pharmakologie (ed. A. Heffter), vol. 1, p. 773. Berlin: Julius Springer. Jack, R. D. (1974). Toxicity of sodium nitroprusside. Br.J. Anaesth., 46, 952. James, G. V. (1938). The excretion of sodium cyanide when administered intravenously in small doses. Analyst, 63, 99. Johnson, C. C. (1929). The actions and toxicity of sodium nitroprusside. Arch. Int. Pharmacodyn. Ther., 35, 380. Lazarus-Barlow, P., and Norman, G. M. (1941). A fatal case of poisoning with sodium nitroprusside. Br. Med. J., 2, 407. McDowall, D. G., Keaney, N. P., Turner, J. M., Lane, J. R., and Okuda, Y. (1974). The toxicity of sodium nitroprusside. Br. J. Anaesth., 46, 327. Macrae, W. R., and Owen, M. (1974). Severe metabolic acidosis following hypotension induced with sodium nitroprusside. Br. J. Anaesth., 46, 795. Mehaffey, L. W. (1942). Toxicity of sodium nitroprusside for guinea pigs. Aust. Inst. J. Proc, 9, 93. Merrifield, A. J., and Blundell, M. D. (1974). Toxicity of sodium nitroprusside. Br. J. Anaesth., 46, 324. Moraca, P. P., Bitte, E. M., Hale, D. E., Wasmuth, C. E., and Poutasse, E. F. (1962). Clinical evaluation of sodium nitroprusside as a hypotensive agent. Anesthesiology, 23, 193. Nourok, D. S., Glassock, R. J., Solomon, D. H., and Maxwell, M. H. (1964). Hypothyroidism following prolonged sodium nitroprusside therapy. Am. J. Med. Sci., 248, 129. Osbome, J. S., Adamek, S., and Hobbs, M. E. (1956). Some components of the gas phase of cigarette-smoke. Analyt. Chem., 28, 211. Osuntokun, B. O. (1972). Chronic cyanide neurotoxicity and neuropathy in Nigerians. PI. Fds. Hum. Nutr., 2, 215.
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
plasma cyanide concentration (fig. 5), the possibility of this alternative monitoring technique during SNP infusion should be considered. Moraca and others (1962) and Taylor, Styles, and Lamming (1970) suggested that the sensitivity to SNP increases with age. Our findings confirm this (fig. 7). Thus SNP overdosage is more likely to occur in younger people, in whom, in fact, the reported deaths have occurred. Although only small changes occur during shortterm use, during prolonged SNP infusion plasma thiocyanate concentrations reach high values (fig. 7; Nourok et al., 1964; Vesey et al., 1974). Thus Merrifield and Blundell (1974) and Davies et al., 1975), using short infusions, were unable to find any significant increase of plasma SCN in their patients who died from SNP overdose. Since the changes observed are less than the normal differences between smokers and non-smokers (fig. 6), plasma SCN would seem to be an unsuitable measurement for the assessment of short-term exposure to HCN. On the other hand, with long infusions of SNP, plasma thiocyanate may reach high concentrations and affect thyroid function (Nourok et al., 1964), and measurement of SCN concentrations may be necessary. In conclusion, we believe SNP to be a valuable hypotensive agent if reasonable doses are employed. Although we are unable to recommend a safe upper limit for plasma cyanide or a dose rate for patients undergoing long-term SNP infusion, we note that Osuntokun, Aladetoyinbo and Adeuja (1970) and Osuntokun (1972) reported mean plasma HCN concentrations of 108 nmol% in patients with tropical neuropathy attributed to chronic cyanide exposure from their staple diet of cassava.
659
660
BRITISH JOURNAL OF ANAESTHESIA
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
Osuntokun, B. O., Aladetoyinbo, A., and Adeuja, developpement de l'acidose metabolique et les recents cas A. O. G. (1970). Free cyanide levels in tropical ataxic ayant eu une issue fatale mettant en cause le SNP peuvent etre attribues a une hypoxie histotoxique resultant des neuropathy. Lancet, 2, 372. Page, I. H. (1951). Treatment of essential and malignant concentrations excessive de cyanure dans le plasma. Sur la base de nos resultats, nous recommandons de ne pas hypertension. J.A.M.A., 147, 1311. Corcoran, A. C , Dustan, H. P., and Koppanyi, T. depasser la dose totale de 1,5 mg/kg de SNP, pendant les (1955). Cardiovascular actions of sodium nitroprusside in infusions a court terme et de faire en sorte que le cyanure se trouvant dans le plasma n'excede pas 300 nmol%. Les animals and hypotensive patients. Circulation, 11, 188. Rentoul, E., and Smith, H. (1973). Glaister's Medical concentrations de sulfocyanate dans le plasma ne sont en Jurisprudence and Toxicology, 13th edn, p. 621. London: general pas des indications fiables de l'importance de l'exposition au cyanure, bien que celles-ci puissent devenir Churchill-Livingstone. Shanahan, R. (1973). The determination of submicrogram importantes lors des infusions a long terme. quantities of cyanide in biological materials. J. Forens. Set., 18, 25. CYANID- UND THIOCYANATSorbo, B. (1957). Enzymic transfer of sulphur from KONZENTRATIONEN NACH VERABREICHUNG mercaptopyruvate to sulfite or sulfinates. Biochem. VON NATRIUMNITROPRUSSID BEIM MENSCHEN Biophys. Acta, 24, 324. ZUSAMMENFASSUNG (1962). Enzymic conversion of cyanide to thiocyanate; Sechsundzwanzig patienten zeigten nach Verabreichung in Proceedings of the First International Pharmacological Meeting (ed. B. Brodie), vol. 6, pp. 121, 128. Oxford, von Natriumnitroprussid (NNP) wahrend einer Operation wesentliche Anstiege der Cyanidkonzentrationen sowohl in London, New York, Paris: Pergamon Press. Stoa, K. (1957). Studies on thiocyanate in serum with some den roten Zellen als auch im Plasma, aber nur geringe supplementary investigations in saliva, urine and Veranderungen der Thiocyanatkonzentration im Plasma. cerebrospinal fluid. Vniversitet i Bergen Arbok. Medisinsk Es bestand eine lineare Beziehung zwischen den Cyanidkonzentrationen in Plasma und den roten Zellen und rekke, Nr 2. Sunshine, I., and Finkle, B. (1964). The necessity for zwischen der Gesamtdosis von NNP. Die ausgeatmeten tissue studies in fatal cyanide poisonings. Int. Archiv. Cyanidkonzentrationen glichen den Plasma veranderungen. Wir glauben, dass die Entwicklung metabolischer Azidosen Arbeitsmed., 20, 558. Taylor, T. H., Styles, M., and Lamming, A. J. (1970). und die kiirzlichen Todesfalle bei NNP-Verwendung einer Sodium nitroprusside as a hypotensive agent in general histotoxischen Hypoxie zu zuschreiben sind, die sich als Resultat ubermassiger Cyanidkonzentrationen im Plasma anaesthesia. Br. J. Anaesth., 42, 859. Van den Hamer, C. J. A., Morell, A. G., and Scheinberg, ergibt. Aufgrund unserer Ergebnisse empfehlen wir eine I. H. (1967). A study of the copper content of p-mercapto- Gesamtdosis von NNP von nicht iiber 1,5 mg/kg bei kurzfristigen Verabreichungen, und die Plasma-Cyanidpyruvate trans-sulfurase. J. Biol. Chem., 242, 2514. konzentration sollte nicht mehr als 300 nmol% betragen. Vesey, C. J., and Cole, P. V. (1975). Nitroprusside and Die Plasma-Thiocyanatkonzentrationen sind im allgemcyanide. Br. J. Anaesth., 47, 1115. einen kein verlassliches Zeichen fiir das Ausmass von Linnell, J. C , and Wilson, J. (1974). Some Cyanid im Korper, sie konnen aber bei langer dauernder metabolic effects of sodium nitroprusside in man. Br. Verabreichung von Bedeutung werden. Med. J., 2, 140. Simpson, P. (1975). Sodium nitroprusside in CONCENTRACIONES DE CIANURO Y anaesthesia. Br. Med. J., 3, 229. TIOCIANATO TRAS INFUSION DE • (1976). Changes in cyanide concentrations NITROPRUSURO SODICO EN EL HOMBRE induced by sodium nitroprusside (SNP). Br. J. Anaesth., SUMARIO 48, 268. Wilson, J., and Matthews, D. M. (1966). Metabolic inter- Veintiseis pacientes, sometidos a infusion de nitroprusuro relationships between cyanide, thiocyanate and vitamin sodico (NPS) durante cirugxa, mostraron aumentos conBi 2 in smokers and non-smokers. Clin. Sci., 31, 1. siderables en la concentrati6n de cianuro en los eritrocitos y en el plasma, pero solamente cambios leves en la conCONCENTRATIONS DE CYANURE ET DE centration plasmatica de tiocianato. Habia una relacion SULFOCYANATE A LA SUITE D'UNE lineal entre las concentraciones plasmatica y eritrocitaria de INFUSION DE NITROPRUSSIATE DE SOUDE cianuro y la dosis total de NPS. La concentraci6n de CHEZ L'HOMME cianuro exhalado sigui6 los cambios en el plasma. Creemos RESUME que el desarrollo de acidosis metabolicas, y las recientes Vingt-six patients recevant une infusion de nitroprussiate muertes relacionadas con NPS son atribuibles a hipoxia de soude (SNP) pendant une intervention chirurgicale ont histotoxica como resultado de excesivas concentraciones montri des augmentations considerables de concentrations plasmaticas de cianuro. Basandonos en nuestros resultados, de cyanure aussi bien dans les globules rouges que dans le recomendamos que la dosis total de NPS no debera exceder plasma, mais seulement de petites variations dans les de 1,5 mg/kg durante las infusiones de breve duracion, y concentrations de sulfocyanate dans le plasma. II y a eu que el cianuro plasmatico no sobrepase de 300 nmol%. une relation lineaire entre les concentrations de cyanure En general, las concentraciones de tiocianato plasmatico dans le plasma et dans les globules rouges du sang, et la son poco connables como indicative del grado de exposici6n dose totale de SNP. La concentration de cyanure exhale a al cianuro, aunque pudieran adquirir importancia durante suivi les variations dans le plasma. Nous estimons que le las infusiones de larga duracion.
CYANIDE AND THIOCYANATE CONCENTRATIONS FOLLOWING SODIUM NITROPRUSSIDE INFUSION IN MAN C. J. VESEY, P. V. COLE AND P. J. SIMPSON SUMMARY
Hydrogen cyanide (HCN) has been detected in animal and human tissues following poisoning with sodium nitroprusside (SNP) (Hermann, 1886; Johnson, 1929; Annotation, 1936, 1948; LazarusBarlow and Norman, 1941). Hill (1942) also found HCN in a fatal case of SNP poisoning and was able to demonstrate cyanide formation in whole blood or tissue incubated with nitroprusside. This has been confirmed by others (Page et al., 1955; Vesey, Cole and Simpson, 1976). Following the introduction of SNP to clinical practice (Page, 1951; Page et al., 1955), toxic effects were reported, and these were attributed to increased plasma thiocyanate (SCN) concentrations following prolonged dosage with the drug (Nourok et al., 1964). Johnson (1929) was the first to demonstrate the hypotensive action of SNP in man and he was convinced that the liberation of cyanide was not the cause of any toxic effects. However, it has been suggested (Vesey and Cole, 1975) that HCN is the cause of metabolic acidosis and fatalities following hypotensive anaesthesia with SNP (Jack, 1974; Merrifield and Blundell, 1974; Davies et al., 1975). In our previous studies on patients infused with SNP during surgery, we found that plasma cyanide concentrations were increased fourfold whilst plasma thiocyanate (the major end-product of cyanide detoxication) was increased by only small amounts, if at all. In one patient who received an infusion for 5 days, however, plasma SCN and HCN concenCYRIL J. VESEY, M.SC.; PETER V. COLE, F.F.A.R.C.S. ; PETER J. SIMPSON, F.F.A.R.C.S.; Department of Anaesthesia, St
Bartholomew's Hospital, London EC1A 7BE.
nations both increased by a factor of 10 and the red cells-to-plasma HCN ratio was 200 : 1 (Vesey et al., 1974). These observations have now been extended in a further series of patients. In addition, the HCN concentrations in expired air have been measured to evaluate the possible use of this procedure as an index of plasma cyanide concentrations during SNP infusions. PATIENTS AND METHODS
Nine male and 17 female patients undergoing major orthopaedic surgery were studied (table I). In our practice hypotensive anaesthesia is not contemplated for anyone with a history of either cardiovascular or cerebrovascular disease. One hour after premedication, the patients were anaesthetized with thiopentone. Suxamethonium was given to facilitate orotracheal intubation. The electrocardiogram, and arterial pressure, measured from an indwelling radial artery cannula, were monitored continuously. Anaesthesia was maintained with a minimum of 33% oxygen in nitrous oxide. Patients lying supine or in the lateral position were allowed to breathe spontaneously. IPPV was considered necessary for patients in the prone position, and in these cases tubocurarine was administered. Minimal concentrations of trichloroethylene or halothane were used when necessary. Blood loss during operation was measured by swab weighing and suction volume, and blood transfusion was seldom needed during the hypotensive period. Reversal of neuromuscular blockade was obtained in all cases following return of the arterial pressure to the pre-hypotensive
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
Twenty-six patients, receiving an infusion of sodium nitroprusside (SNP) during surgery, had considerable increases in both red cell and plasma cyanide concentration, but only small changes in plasma thiocyanate concentration. There was a linear relationship between both plasma and RBC cyanide concentrations and the total dose of SNP. The expired cyanide concentration followed the changes in the plasma. We believe that the development of metabolic acidosis, and the recent fatalities involving SNP, are attributable to histotoxic hypoxia as a result of excessive plasma concentrations of cyanide. On the basis of our results, we recommend that the total dose of SNP should not exceed 1.5 mg/kg during short-term infusions and that the plasma cyanide should not exceed 300 nmol%. Plasma thiocyanate concentrations are, in general, an unreliable indication of extent of exposure to cyanide, although they may become important during long-term infusions.
BRITISH JOURNAL OF ANAESTHESIA
652
TABLE I. Details of patients and SNP dose rates
(yr)
Sex
Smokers
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
57 29 55 41 63 52 53 28 46 33 38 52 41 15 77 42 27 50 54 75 15 58 17 32 42 50
M F F M F M M M M M F M F F F F M F F F F F F F F F
+ — + + — — + + + + — — + — — — + — + — _ — — _ _ —
55 90 95 80 80 118 168 115 65 95 120 30 85 57 85 108 65 38 120 90 141 95 54 80 82 69
value, and the patients were observed for 1-2 h in an adjacent recovery room. SNP in physiological saline (200 fig/ml) was introduced to the existing i.v. infusion (Hartmann's solution B.P.), as near to the cannula as possible to minimize deadspace. Administration of the drug through a drip controller (IVAC 200) was commenced immediately before surgery at dose rates sufficient to reduce the systolic arterial pressure to about 60 mm Hg (mean dose rates and total doses are given in table I). Baseline samples of expired air and arterial blood were obtained immediately before the infusion was commenced. The air was collected into a plastic Douglas bag over a timed period (usually 3-5 min). Blood was withdrawn into a pre-cooled heparinized syringe, sealed and placed in a mixture of ice and water. The second samples of air and blood were collected immediately following the administration of nitroprusside. Subsequent blood samples were taken 1 and 24 h later. Control expired air samples were also collected from 15 patients who were comparable in age, sex, weight and anaesthetic technique with those receiving SNP.
SNP dose (nig)
20 16 53 35 8 27 51 36 23 34 67.5
4 26 15 12 24 19 3 26 10 95.4
27 12 42 16 7.6
M e a n dose rate (nmol/kg/min) 11.7 10.4 29.3 17.8
5.1 8.4 12.0 10.4 15.6 18.4 24.7
6.1 24.0 20.9
7.4 15.5 16.2
5.1 10.3
6.6 44.9 14.6 15.4 25.3
9.6 7.5
Isolation of cyanide (a) Plasma cyanide. Immediately following collection, the blood samples were centrifuged at 4 °C and an aliquot of the separated plasma (5-10 ml) was assayed for cyanide. The remainder of the plasma was stored at — 20 °C, some for later determination of thiocyanate, and some, together with an aliquot of the red cells, for eventual assay of vitamin B 12 (the subject of a separate communication). Plasma HCN was isolated by the method of Boxer and Rickards (1951). Scrubbed oxygen-free nitrogen was passed at a rate of 1 litre/min for 1 \ h through a mixture of 1 vol plasma and 3 vol 10% trichloroacetic acid (TCA). HCN was removed from the gas stream by passage through two tubes in series, each containing sodium hydroxide 0.2 mol/litre. (The volume of NaOH used was adjusted to suit the amount of HCN expected.) (b) Red cell cyanide. The isolated red cells were washed six times with normal saline to remove any traces of plasma thiocyanate (contamination of the cells with SCN results in artefactual formation of HCN on assay (Vesey and Wilson, in preparation)) and stored at 4 °C overnight. After the determination
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
Patient
Infusion time (min)
Age
CYANIDE, THIOCYANATE AND NITROPRUSSIDE
Assay of cyanide and thiocyanate Both plasma thiocyanate and separated cyanide were determined by the spectrophotometric method of Aldridge (1945). Potassium thiocyanate solutions of known concentration were used as standards. Plasma thiocyanate was determined on duplicate 1-ml aliquots of the supernatant obtained by centrifuging a mixture of 1 ml of plasma and 9 ml of 10% TCA. One millilitre of NaOH solution containing HCN from (a) (b) or (c) above, or 1 ml of the protein-free TCA extract of the plasma, was mixed with 0.5 ml of HC1 1 mol/litre and two drops of saturated bromine water. Excess bromine was removed by adding three drops of a 2% arsenic trioxide solution and mixing vigorously in a stoppered tube. Ten minutes after the addition of 1.8 ml of a pyridine-benzidine mixture, the absorbance was measured at 532 nm. The pyridine-benzidine mixture was prepared from 1 vol of a solution of 5% benzidine hydrochloride in hydrochloric acid 1 mol/litre and 5 vol of a pyridine-HCl-water solution (6 : 1: 4 by vol respectively).
RESULTS AND DISCUSSION
The results confirm that there is a significant increase in plasma HCN following the infusion of SNP (Vesey et al., 1974). In addition they indicate that for short periods of infusion (2 h approximately) there is a close correlation between the increase in plasma cyanide, measured immediately after the infusion, and the total dose of SNP (r = 0.94, P< 0.001, n = 24 (fig. 1)). The correlation of plasma cyanide
with the mean rate of SNP infusion is poorer, but statistically significant (r = 0.843, P< 0.001, n = 24). When cyanide is introduced into blood, in vitro or in vivo, most of it enters the red cells (Vesey and Wilson, in preparation). In this study 98.4% (SD ±0.83%) of the blood cyanide, immediately after the infusion, was found in the red cells. As with plasma, the red cell cyanide concentrations were more I 200,
1J %
1 /
o —
t.
. 1.0
2.0
3.0 4.0 5.0 //mol SXP/kg boil; \\l
6.0
FIG. 1. Relationship between the increases in plasma cyanide, immediately following infusionj and the total dose of SNP (r = 0.94, P< 0.001, n = 24). The irregular result was for patient 21 who received the highest dose of the series. This result was probably influenced by the dilution of the patient's plasma with the transfusion of 800 ml of blood and 500 ml of saline during surgery. It was not included in our calculations.
3 C-"g20
1
2
3
4
5
•
6
/imolSNP/kgbodywt
FIG. 2. Correlation between the increases in red cell cyanide concentrations immediately following infusion and total SNP dose (r = 0.924, P < 0.001, n = 26).
closely related to the total dose of SNP than to the rate of infusion (r = 0.924, P< 0.001, n = 26; r = 0.849, P< 0.001, n = 26 respectively, fig. 2).
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
of the packed cell volume of the washed cells an aliquot (1-5 ml) was mixed, in an aeration tube, with an equal volume of water. Three volumes of 10% TCA were added to the haemolysate and the liberated HCN was transferred to sodium hydroxide in a stream of nitrogen, as for plasma. (c) Expired air cyanide was isolated and concentrated by the method of Boxer and Rickards (1952). Each air sample was drawn through a solution of silver sulphate 0.02 mol/litre in sulphuric acid 0.1 mol/litre in a Drechsel bottle with a sintered inlet. Any residual anaesthetic gases together with carbon dioxide were removed from the solution by the passage of nitrogen 1 litre/min for 20 min. Following the addition of sulphuric acid (11 mol/litre) to the silver sulphate solution to increase the concentration to 3.5 mol/litre, HCN, set free, was transferred to sodium hydroxide as in (a) and (b) above.
653
BRITISH JOURNAL OF ANAESTHESIA
654
TABLE II. Plasma and RBC cyanide and plasma thwcyanateconcentrationsbefore and after infusion of SNP
Plasma
Red blood cell
Cyanide nmol%
Thiocyanate l±mol%
Pre
Post
lh post
24 h post
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
2 0 15 — 13 14 8 2 4 18 20 9 30 6 20 2 5 2 10 2 0 5 2 0 14 4
5 32 204 79 38 29 94 30 34 143 205 18 200 42 40 102 39 23 80 6 190 75 18 155 29 12
— — 194 — — 0 35 5 13 24 105 6 34 27 26 68 11 — 14 3 129 45 3 81 0 0
3 5 45 3 20 — 8 5 6 19 10 6 0 27 18 0 3 — 16 2 0 5 11 3 — —
Pre
Post
lh ]post
24 h post
3.8 5.2
3.9 6.0
4.0 6.3
3.8 4.4
12.4
13.8 11.8
13.8 12.4
15.3
3.2 4.3
3.7 4.5
13.6 10.5 13.8 13.0
17.2 10.9 14.7 13.6
16.3 11.7 15.3 14.8
12.5
3.6 2.0
4.6 1.8
5.4 1.5
6.6 0.5
10.4
14.0
14.8
14.5
2.6 2.3 3.3 4.8 1.8 3.3 4.7 4.8
3.7 2.3 3.8 5.2 2.2 4.3 5.2 7.8
3.2 2.3 4.6
3.0 1.8 4.0 5.8
— 4.8 5.2 8.9
4.6 4.5 8.8
14.2
15.9
16.3
14.0
4.8 4.9 3.3 4.2
4.5 6.2 4.3 3.8
4.4 7.1 4.5 3.7
3.7 5.9 — —
9.8 2.5 3.3
Both mean red cell and mean plasma cyanide concentrations, (nmol HCN/100 ml)/((xmol SNP/kg), decreased to less than a half of the post-infusion values, 1 h after infusion had ceased (fig. 3). However, where the post-infusion concentrations were high, the decrease in cyanide concentration was much slower (table II). There are obvious differences in the rate at which HCN is metabolized (table I I : patient 8 compared with patient 9). In some cases there was a decrease to less than the pre-infusion values after 24 h (fig. 3 and table II). This may indicate an enhanced detoxication in response to exposure to HCN or, more probably, an increased urinary excretion as a result of the i.v. infusion. Boxer and Rickards (1952) showed that minute amounts of HCN are normally excreted in expired air and are related to the metabolic rate. James (1938) found that the expired HCN concentration increased after the administration of cyanide. Following SNP infusion there were definite increases in expired HCN in the majority of patients (fig. 4). In control patients undergoing a similar type of surgery and anaesthesia,
Z3
9.7 2.8
7.2
13.3 11.7
Pre
Post
0.024 0.028 0.113 0.038 0.063 0.222 0.073 0.017 0.097 0.053 0.014 0.080 0.068 0.003 0.207 0.047 0.080 0.160 0.040 0.017 0.029 0.038 0.019 0.008 0.222 0.291
0.155 1.860 11.70 5.880 0.875 2.200 8.240 2.500 2.920 3.700 18.97 0.680 8.230 4.102 1.220 6.710 2.270 2.520 5.420 0.100 21.20 5.300 1.750 8.000 2.200 0.665
lh post
24 h post
0.042 0.335 8.50 1.639 0.172 0.513 3.450 2.140 0.472 0.393 17.35 0.066 0.548 1.143 0.375 4.024 0.688
0.015 0.055 0.248 0.116 0.051 0.108 0.084 0.138 0.166 0.263 0.035 0.188 0.007 0.130 0.105 0.116
—
—
0.436 0.050 16.53 1.328 0.273 5.400 0.513 0.220
0.063 0.030 0.210 0.118 0.072 0.126 — —
60
e- 40
5«5 oc •5. a. 30
I
I? 20 |
rh
1 10
11-25
of
T
11
X
H-21
»"20
•
• £
^ ^ 3 ?:" 2
T
ii
H -26
11-24
post
1 h post
— 11-22
24 h post
FIG. 3. Change in mean plasma and red cell cyanide concentrations. The difference between values after and before infusion expressed as (nmol HCN/100 ml)/ SNP/kg) together with SD immediately following, 1 h following, and 24 h following infusion.
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
Patient
Cyanide nmol%
655
CYANIDE, THIOCYANATE AND NITROPRUSSIDE
CONTROL PATIENTS
SNP PATIENTS
o CO
o C 4
z: U 3 I Q 2 U4
a: 1 E X
w
50
100
Smokers
150 0
50 100 150 Non-smokers
50 Smokers
0
50 100 Non-smokers
OPERATION TIME (min) FIG. 4. Changes in expired cyanide (expressed as nmol HCN/kg body wt/h) for patients receiving SNP and patients undergoing similar procedures without SNP. Results in smokers and non-smokers are presented separately. Dose rates (nmol SNP/kg/min) are given at the end of each line.
expired HCN concentrations showed little change, except in some who were smokers, but were in most cases less at the end of the operation. In three of the patients who received only small amounts of SNP there was also a decrease in the amount of HCN expired. The initially high concentration of HCN in the expired gas in some smokers probably results indirectly from the HCN present in tobacco smoke. The increases in the expired HCN concentrations above pre-infusion values correlate with the increases in plasma cyanide (r = 0.874, P< 0.001, n = 19; fig. 5). The fact that log10 of the increase in expired HCN correlates with the plasma cyanide concentration (r = 0.843, P< 0.001, n = 19) suggests that the expired cyanide concentration may increase exponentially with increasing plasma cyanide. The relationship to the rate of infusion of SNP or to the total amount infused was less marked (r = 0.665, P< 0.001; r = 0.57, 73 0.001; r = 0.596, P = 0.001 respectively; n = 24). This may reflect the relatively slow rate of the detoxication of cyanide. Similarly, the difference between the mean plasma thiocyanate concentrations before and after infusion were significant only at 1 h after SNP in both smokers and non-smokers {t = 2.567, 0.05>P>0.02; t = 2.06, P 0.02 for smokers; t = 2.06, Ps^O.05 for non-smokers).
•S'40
2
= 20
values, or less, 24 h later. Increased thiocyanate concentrations normally decrease slowly; for example in smokers the return to normal values after the cessation of smoking takes 14 days. However, it is known that increases in the intake of both fluid and chloride ions enhance the rate of SCN excretion (Stoa, 1957). Although the smokers in our study received SNP at a higher mean rate than that given to the non-smokers (18.21 + 6.62; 13.95 +9.99 nmol/kg/ min respectively; n = 1 and 18), the difference is not statistically significant (t = 1.0357, P>0.1). Sensitivity to SNP is very variable but would appear to increase with age. The highest mean dose rate amongst our patients was required by patient 21, a 15-year-old girl (44.9 nmol SNP/kg/min). Mean dose rates plotted against age are shown in figure 7. Although the correlation is poor (r = —0.553) it is statistically significant (0.01 >P^0.001). However, there appears to be no sex difference in the response to SNP.
t • 10
20
30
US
50
» 60
70
Age(>r)
FIG. 7. Correlation between the rate of infusion of SNP and the ages of the patients (r = —0.553, n = 26, 0.01 > P > 0.001). GENERAL DISCUSSION AND CONCLUSIONS
Although 98% of HCN in the blood, following an infusion of SNP, is located in the red cells, its fate is unknown. In the rat, the enzyme /J-mercaptopyruvate sulphur transferase, which converts HCN to thiocyanate, is present in the red cells (Sorbo, 1957), but this enzyme is believed to show only low activity in the red cells of man (Van den Hamer, Morell and Scheinberg, 1967). The inference that the red cell cyanide content has no apparent toxicological significance is suggested by
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
-a ir- 2
657
CYANIDE, THIOCYANATE AND NITROPRUSSIDE
occurred as a result of the absorption of 24 mg of HCN (table III, case 19). Thus 25 mg of HCN could reasonably be expected to be close to the lethal i.v. dose. Assuming a blood volume of about 70 ml/kg in a 70-kg man (Airman and Dittmer, 1971), this would result in a blood cyanide concentration of 19 (xmol%, or 19 |i.mol/litre plasma. We assume that, as we have confirmed in animal studies, 90% of the blood HCN is in the red cells. Similarly, minimum whole blood concentrations recorded in the literature (table III), would represent plasma concentrations of about 10 (imol HCN/litre and, where death has occurred from inhalation of HCN (cases M5 and M6), the calculated value would be even less. Where death has been delayed (cases 19 and M7), a situation similar to SNP overdose, the calculated plasma concentrations approach 1 (xmol%. Thus the fatal cyanide concentration would appear to be between 10 and 20 fxmol/ litre plasma. Mehaffey (1942) showed that the lethal oral dose of SNP for guineapigs was about 15 mg and, on this basis, McDowall, and others (1974) suggested that 250 mg of SNP (about 3.6mg/kg or 12 jxmol/kg) might be the lethal dose in man. Nourok and colleagues (1964) found that the hypotensive effects of SNP by mouth were ho greater than those produced by the thiocyanate arising from it. This suggests that
TABLE I I I . Whole blood cyanide concentrations in man following poisoning, together with postulated plasma HCN concentrations
Reference Gettler and Baine (1938)
Sunshine and Finkle (1964)
Bonnischen and Maehly (1966)
Shanahan (1973)
Case no.
Blood cyanide (fimol/litre)
6 19
119 96
15 48 (M) 54 (M) 33 (F) 86 (M) 40 (M) P5 M5 M6 M7
152
102 154 92\ 70/ 130
3 4
156 122
Relevant details Oral cyanide Oral cyanide; died after 3h Inhaled HCN
43] 112 1 82 f
6lJ
Oral cyanide Inhaled HCN Oral cyanide Inhaled HCN KCN absorbed through skin; died within 4 h — —
Postulated plasma HCN concns. (nmol/litre)* 12 9.6 15 4 11 8 6 10 15 9 7 3* 16 12
Only the lower concentrations have been selected since, obviously in most cases of poisoning, doses much larger than the lethal dose will have been taken. * Plasma cyanide concentrations are postulated to be one-tenth those of whole blood, assuming 90% of the total HCN to be present in the red cells. In case M7, where absorption would be slow and similar to the situation with SNP, 98% of blood cyanide is likely to be present in the red cells.
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
a comparison of the highest concentrations found after SNP infusion (12, 19 and 21 jamoF/o (table I I ) ; 32 (xmol% (Vesey et al., 1974)) with some of the lower total blood cyanide values reported in cases of cyanide poisoning (table III). It has been suggested that between 10 and 12 (imol% is the minimum whole blood cyanide concentration which will cause lethal effects in an otherwise healthy man (Gettler and Baine, 1938; Sunshine and Finkle, 1964). Comparison of these figures would suggest also that HCN in the plasma, rather than in the red cells, is the critical factor in cyanide intoxication. Since the tissues are readily permeable to lipidsoluble HCN, the amount passing into them to cause toxic symptoms, will be closely related to the plasma concentration. It should be possible to suggest an upper limit for plasma HCN and, on the basis of our results (fig. 1), to suggest a maximum total dose of SNP (Vesey, Cole and Simpson, 1975). The minimum lethal oral dose of hydrogen cyanide is variously suggested as 40 mg (Rentoul and Smith, 1973), 0.5mg/kg (Hunt, 1923; Gettler and Baine, 1938) and 37 mg (Gonzales, Vance and Helpern, 1940). Only part of an oral dose would be absorbed and some would be detoxicated in the liver before entry into the general circulation (Gettler and Baine, 1938). The last authors calculated that one death
658
BRITISH JOURNAL OF ANAESTHESIA TABLE IV. Some SNP doses and the estimated resulting plasma cyanide concentrations Reference
Dose of SNP
Details
McDowall and others (1974)
250 mg (3.6 mg/kg or 12|imol/kg) 320 mg/h (4.6 mg/kg/h or 15.3 (imol/kg/h)
Suggested lethal dose for man
250 mg (4.2 mg/kg or 14 (imol/kg)
Merrifield and Blundell (1974)
750 mg (2 mg/kg/h or 20 nmol/kg in 3 h)
A. B. Davies (pers. comm.) Jack (1974)
800 after 1-3 h infusion 1020 after 1 h infusion
Smallest toxic infused dose for man calculated from baboon studies 42-year-old female infused over 930 90 min; severe reversible acidosis 20-year-old male infused over 1300 after 3 h 5 h. Fatal infusion (?) 2200 after 5 h infusion 14-year-old male infused over 2200 80 min. Fatal 39-year-old male infused over 2400 30 min. Fatal
400 mg (10 mg/kg or 33.6 nmol/kg) 750 mg (approx. 11 mg/kg or 37 nmol/kg) Editorial (1975) 3 mg/kg/h (30 (i.mol/kg Recommended rate in3h) Vesey, Cole and 1.5 mg/kg Recommended total dose Simpson (1975) (5 nmol/kg)
2000 after 3 h infusion 330 after 1-3 h infusion
* The estimate of plasma cyanide is based on figure 1 and assumes that the relationship between plasma cyanide and SNP dose is still linear above the concentrations that we have measured.
SNP, taken orally, decomposes in the stomach or liver, and only its breakdown product, HCN, and detoxication product, SCN, enter the general circulation. Therefore the lethal i.v. dose of SNP would be expected to be less than 250 mg. Assuming that the linear relationship between the total dose of SNP and plasma HCN applies at concentrations higher than those we have measured, 250 mg of SNP would cause a plasma concentration of 0.8 \irao\ HCN/100 ml if infused over 2-3 h (table IV). This is close to the lower lethal value for plasma cyanide (1 (xmol%) suggested above. SNP doses which produced metabolic acidoses and, in some cases, death, together with estimated plasma HCN concentrations, are compared with maximum dose rates recommended previously (Editorial, 1975) (table IV). Thus it would appear that, for short-term infusions, a total dose of SNP resulting in plasma HCN concentrations of 1 fxmol% (15 nmol SNP/kg; 4.5 mg/kg (fig. 1)) would be very near to the toxic dose. HCN present in the plasma enters the tissues and is converted to thiocyanate by the mitochondrial enzyme rhodanese. As judged by the increase in plasma cyanide and the slow increase in plasma thiocyanate, the rate of cyanide detoxication in our patients was slower than that of the release of HCN from SNP.
The rate of detoxication of HCN is limited by the availability and low permeability of thiosulphate (Sorbo, 1962). The infusion of SNP, together with the slow release of HCN, will maintain continuous exposure of the tissues to cyanide. Detoxication mechanisms may thus become exhausted at blood concentrations which would not normally cause problems in acute HCN poisoning. This may explain why the toxic effects of SNP, in animals, compared with those of cyanide, although similar, are delayed (MehafFey, 1942), as may occur also in clinical practice (McDowall et al., 1974; Merrifield and Blundell, 1974). On present evidence, we suggest that the total dose of SNP should be limited to 1.5 mg/kg over the duration of an average surgical operation (Vesey, Cole and Simpson, 1975). High doses seem to be unnecessary (Adams, 1975) and, in our hands, the addition of small amounts of halothane and, occasionally, jS-blockade has enabled us to limit the total dose to less than 100 mg and the dose rate to less than 45 nmol/kg/min (0.8 mg/kg/h). Where the total dose is likely to exceed the suggested value, or during long-term administration, we recommend monitoring of plasma cyanide concentrations. Since log10 expired air cyanide concentration is linearly related to the
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
MacRae and Owen (1974)
Estimated plasma cyanide (nmol%)*
CYANIDE, THIOCYANATE AND NITROPRUSSIDE
ACKNOWLEDGEMENTS
The authors would like to thank the Joint Research Board of St Bartholomew's Hospital for invaluable financial assistance and the staff of the Anaesthetic Research Laboratory, St Bartholomew's Hospital, for their expert technical help. REFERENCES
Adams, A. P. (1975). Techniques of vascular control for deliberate hypotension during anaesthesia. Br.J. Anaesth., 47, 777. Aldridge, W. N. (1945). The estimation of microquantities of cyanide and thiocyanate. Analyst, 70, 474. Altman, P. L., and Dittmer, D. S. (1971). Blood and Other Body Fluids, p. 1. Bethesda: Federation of American Societies for Experimental Biology. Annotation (1936). Pharm.J., 82, 125. (1948). Pharm.J., 160, 287. Bonnischen, R., and Maehly, A. C. (1966). Poisoning by volatile compounds. I l l : Hydrocyanic acid. J. Forens. Set., 11, 516.
Boxer, G. E., and Rickards, J. C. (1951). Chemical determination of vitamin B 12 . I I : The quantitative isolation and colorimetric determination of millimicrogram quantities of cyanide. Arch. Biochem. Biophys., 30, 372. (1952). Determination of traces of hydrogen cyanide in respiratory air. Arch. Biochem. Biophys, 39, 287. Davies, D. W., Kadar, D., Steward, D. J., and Munro, I. R. (1975). A sudden death associated with the use of sodium nitroprusside for induction of hypotension during anaesthesia. Can. Anaesth. Soc. J., 22, 547. Editorial (1975). Sodium nitroprusside in anaesthesia. Br. Med. J., 2, 524. Gettler, A. O., and Baine, J. O. (1938). The toxicology of cyanide. Am. J. Med. Sci., 195, 182. Gonzales, T. A., Vance, M., and Helpern, M. (1940). Legal Medicine and Toxicology, 2nd edn, p. 542. New York and London: D. Appleton-Century. Hermann, L. (1886). Ueber die Wirkung des Nitroprussidnatriums. Arch. Physiol., 39, 419. Hill, H. E. (1942). A contribution to the toxicology of sodium nitroprusside. I: Decomposition and determination of sodium nitroprusside. Aust. Chem. Inst. J. Proc, 9,89. Hunt, R. (1923). Cyanwasserstoff, Nitroglukoside, Nitrile5 Rhodanwasserstoff, Isocyanide; in Handbuch der Experimentellen Pharmakologie (ed. A. Heffter), vol. 1, p. 773. Berlin: Julius Springer. Jack, R. D. (1974). Toxicity of sodium nitroprusside. Br.J. Anaesth., 46, 952. James, G. V. (1938). The excretion of sodium cyanide when administered intravenously in small doses. Analyst, 63, 99. Johnson, C. C. (1929). The actions and toxicity of sodium nitroprusside. Arch. Int. Pharmacodyn. Ther., 35, 380. Lazarus-Barlow, P., and Norman, G. M. (1941). A fatal case of poisoning with sodium nitroprusside. Br. Med. J., 2, 407. McDowall, D. G., Keaney, N. P., Turner, J. M., Lane, J. R., and Okuda, Y. (1974). The toxicity of sodium nitroprusside. Br. J. Anaesth., 46, 327. Macrae, W. R., and Owen, M. (1974). Severe metabolic acidosis following hypotension induced with sodium nitroprusside. Br. J. Anaesth., 46, 795. Mehaffey, L. W. (1942). Toxicity of sodium nitroprusside for guinea pigs. Aust. Inst. J. Proc, 9, 93. Merrifield, A. J., and Blundell, M. D. (1974). Toxicity of sodium nitroprusside. Br. J. Anaesth., 46, 324. Moraca, P. P., Bitte, E. M., Hale, D. E., Wasmuth, C. E., and Poutasse, E. F. (1962). Clinical evaluation of sodium nitroprusside as a hypotensive agent. Anesthesiology, 23, 193. Nourok, D. S., Glassock, R. J., Solomon, D. H., and Maxwell, M. H. (1964). Hypothyroidism following prolonged sodium nitroprusside therapy. Am. J. Med. Sci., 248, 129. Osbome, J. S., Adamek, S., and Hobbs, M. E. (1956). Some components of the gas phase of cigarette-smoke. Analyt. Chem., 28, 211. Osuntokun, B. O. (1972). Chronic cyanide neurotoxicity and neuropathy in Nigerians. PI. Fds. Hum. Nutr., 2, 215.
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
plasma cyanide concentration (fig. 5), the possibility of this alternative monitoring technique during SNP infusion should be considered. Moraca and others (1962) and Taylor, Styles, and Lamming (1970) suggested that the sensitivity to SNP increases with age. Our findings confirm this (fig. 7). Thus SNP overdosage is more likely to occur in younger people, in whom, in fact, the reported deaths have occurred. Although only small changes occur during shortterm use, during prolonged SNP infusion plasma thiocyanate concentrations reach high values (fig. 7; Nourok et al., 1964; Vesey et al., 1974). Thus Merrifield and Blundell (1974) and Davies et al., 1975), using short infusions, were unable to find any significant increase of plasma SCN in their patients who died from SNP overdose. Since the changes observed are less than the normal differences between smokers and non-smokers (fig. 6), plasma SCN would seem to be an unsuitable measurement for the assessment of short-term exposure to HCN. On the other hand, with long infusions of SNP, plasma thiocyanate may reach high concentrations and affect thyroid function (Nourok et al., 1964), and measurement of SCN concentrations may be necessary. In conclusion, we believe SNP to be a valuable hypotensive agent if reasonable doses are employed. Although we are unable to recommend a safe upper limit for plasma cyanide or a dose rate for patients undergoing long-term SNP infusion, we note that Osuntokun, Aladetoyinbo and Adeuja (1970) and Osuntokun (1972) reported mean plasma HCN concentrations of 108 nmol% in patients with tropical neuropathy attributed to chronic cyanide exposure from their staple diet of cassava.
659
660
BRITISH JOURNAL OF ANAESTHESIA
Downloaded from http://bja.oxfordjournals.org/ at University of California, San Francisco on January 27, 2015
Osuntokun, B. O., Aladetoyinbo, A., and Adeuja, developpement de l'acidose metabolique et les recents cas A. O. G. (1970). Free cyanide levels in tropical ataxic ayant eu une issue fatale mettant en cause le SNP peuvent etre attribues a une hypoxie histotoxique resultant des neuropathy. Lancet, 2, 372. Page, I. H. (1951). Treatment of essential and malignant concentrations excessive de cyanure dans le plasma. Sur la base de nos resultats, nous recommandons de ne pas hypertension. J.A.M.A., 147, 1311. Corcoran, A. C , Dustan, H. P., and Koppanyi, T. depasser la dose totale de 1,5 mg/kg de SNP, pendant les (1955). Cardiovascular actions of sodium nitroprusside in infusions a court terme et de faire en sorte que le cyanure se trouvant dans le plasma n'excede pas 300 nmol%. Les animals and hypotensive patients. Circulation, 11, 188. Rentoul, E., and Smith, H. (1973). Glaister's Medical concentrations de sulfocyanate dans le plasma ne sont en Jurisprudence and Toxicology, 13th edn, p. 621. London: general pas des indications fiables de l'importance de l'exposition au cyanure, bien que celles-ci puissent devenir Churchill-Livingstone. Shanahan, R. (1973). The determination of submicrogram importantes lors des infusions a long terme. quantities of cyanide in biological materials. J. Forens. Set., 18, 25. CYANID- UND THIOCYANATSorbo, B. (1957). Enzymic transfer of sulphur from KONZENTRATIONEN NACH VERABREICHUNG mercaptopyruvate to sulfite or sulfinates. Biochem. VON NATRIUMNITROPRUSSID BEIM MENSCHEN Biophys. Acta, 24, 324. ZUSAMMENFASSUNG (1962). Enzymic conversion of cyanide to thiocyanate; Sechsundzwanzig patienten zeigten nach Verabreichung in Proceedings of the First International Pharmacological Meeting (ed. B. Brodie), vol. 6, pp. 121, 128. Oxford, von Natriumnitroprussid (NNP) wahrend einer Operation wesentliche Anstiege der Cyanidkonzentrationen sowohl in London, New York, Paris: Pergamon Press. Stoa, K. (1957). Studies on thiocyanate in serum with some den roten Zellen als auch im Plasma, aber nur geringe supplementary investigations in saliva, urine and Veranderungen der Thiocyanatkonzentration im Plasma. cerebrospinal fluid. Vniversitet i Bergen Arbok. Medisinsk Es bestand eine lineare Beziehung zwischen den Cyanidkonzentrationen in Plasma und den roten Zellen und rekke, Nr 2. Sunshine, I., and Finkle, B. (1964). The necessity for zwischen der Gesamtdosis von NNP. Die ausgeatmeten tissue studies in fatal cyanide poisonings. Int. Archiv. Cyanidkonzentrationen glichen den Plasma veranderungen. Wir glauben, dass die Entwicklung metabolischer Azidosen Arbeitsmed., 20, 558. Taylor, T. H., Styles, M., and Lamming, A. J. (1970). und die kiirzlichen Todesfalle bei NNP-Verwendung einer Sodium nitroprusside as a hypotensive agent in general histotoxischen Hypoxie zu zuschreiben sind, die sich als Resultat ubermassiger Cyanidkonzentrationen im Plasma anaesthesia. Br. J. Anaesth., 42, 859. Van den Hamer, C. J. A., Morell, A. G., and Scheinberg, ergibt. Aufgrund unserer Ergebnisse empfehlen wir eine I. H. (1967). A study of the copper content of p-mercapto- Gesamtdosis von NNP von nicht iiber 1,5 mg/kg bei kurzfristigen Verabreichungen, und die Plasma-Cyanidpyruvate trans-sulfurase. J. Biol. Chem., 242, 2514. konzentration sollte nicht mehr als 300 nmol% betragen. Vesey, C. J., and Cole, P. V. (1975). Nitroprusside and Die Plasma-Thiocyanatkonzentrationen sind im allgemcyanide. Br. J. Anaesth., 47, 1115. einen kein verlassliches Zeichen fiir das Ausmass von Linnell, J. C , and Wilson, J. (1974). Some Cyanid im Korper, sie konnen aber bei langer dauernder metabolic effects of sodium nitroprusside in man. Br. Verabreichung von Bedeutung werden. Med. J., 2, 140. Simpson, P. (1975). Sodium nitroprusside in CONCENTRACIONES DE CIANURO Y anaesthesia. Br. Med. J., 3, 229. TIOCIANATO TRAS INFUSION DE • (1976). Changes in cyanide concentrations NITROPRUSURO SODICO EN EL HOMBRE induced by sodium nitroprusside (SNP). Br. J. Anaesth., SUMARIO 48, 268. Wilson, J., and Matthews, D. M. (1966). Metabolic inter- Veintiseis pacientes, sometidos a infusion de nitroprusuro relationships between cyanide, thiocyanate and vitamin sodico (NPS) durante cirugxa, mostraron aumentos conBi 2 in smokers and non-smokers. Clin. Sci., 31, 1. siderables en la concentrati6n de cianuro en los eritrocitos y en el plasma, pero solamente cambios leves en la conCONCENTRATIONS DE CYANURE ET DE centration plasmatica de tiocianato. Habia una relacion SULFOCYANATE A LA SUITE D'UNE lineal entre las concentraciones plasmatica y eritrocitaria de INFUSION DE NITROPRUSSIATE DE SOUDE cianuro y la dosis total de NPS. La concentraci6n de CHEZ L'HOMME cianuro exhalado sigui6 los cambios en el plasma. Creemos RESUME que el desarrollo de acidosis metabolicas, y las recientes Vingt-six patients recevant une infusion de nitroprussiate muertes relacionadas con NPS son atribuibles a hipoxia de soude (SNP) pendant une intervention chirurgicale ont histotoxica como resultado de excesivas concentraciones montri des augmentations considerables de concentrations plasmaticas de cianuro. Basandonos en nuestros resultados, de cyanure aussi bien dans les globules rouges que dans le recomendamos que la dosis total de NPS no debera exceder plasma, mais seulement de petites variations dans les de 1,5 mg/kg durante las infusiones de breve duracion, y concentrations de sulfocyanate dans le plasma. II y a eu que el cianuro plasmatico no sobrepase de 300 nmol%. une relation lineaire entre les concentrations de cyanure En general, las concentraciones de tiocianato plasmatico dans le plasma et dans les globules rouges du sang, et la son poco connables como indicative del grado de exposici6n dose totale de SNP. La concentration de cyanure exhale a al cianuro, aunque pudieran adquirir importancia durante suivi les variations dans le plasma. Nous estimons que le las infusiones de larga duracion.
