Journal of the Neurological Sciences, 1978, 37:1-29 © Elsevier/North-HollandBiomedicalPress
1
PROFOUND HYPOXIA IN Papio anubis AND M a c a c a mulatta - - PHYSIOLOGICAL AND NEUROPATHOLOGICAL EFFECTS I. Abrupt Exposure following Normoxia II. Abrupt Exposure following Moderate Hypoxia
J. B.
BRIERLEY,PAMELA F. PRIOR, JANE CALVERLEYand A. W. BROWN
Medical Research Coucil Laboratories (JBB, JC and A WB) , Carshalton, Surrey, and The EEG Department (PFP), The London Hospital, Whitechapel, London E1 1BB (Great Britain)
SUMMARY Lightly anaesthetized and spontaneously breathing P. anubis (PA) and M. mulatta (MM) inhaled at ambient pressure 3. 2~ oxygen (= 37,500 ft or 11,430 m)
from air and also after pre-exposure to 14 ~ oxygen ( - 10,000 ft or 3,048 m). The EEG, ECG, respiratory rate, arterial and cerebral venous sinus pressures, end-tidal pO~ and pCOz and body temperature were recorded. Arterial and cerebral venous sinus blood gases, pH and pyruvate and lactate contents were estimated. Before hypoxia, MM showed a relative hyperventilation. Profound hypoxia, from air, ended with the "last breath" at 89-205 sec in PA and at 93-570 sec in MM. Brain damage was restricted to one MM (4 exposures). Profound hypoxia after exposure to 14 ~ oxygen ended with the "last breath" at 87-210 sec in PA and at 120 sec-94 min (including 9 exposures over 5 min) in MM. Brain damage was restricted to one MM ("last breath" at 94 min). In the two MM with brain damage there was evidence of reduction in cerebral perfusion near the end of profound hypoxia. Brain damage in one animal contrasts with the frequent and often severe brain damage in MM after equivalent sub-atmospheric decompressions preceded by exposure to moderate altitude (10,000 ft).
INTRODUCTION At present there is no convincing evidence to show that pure hypoxic hypoxia (i.e. a reduced arterial oxygen tension with normal cerebral blood flow) can cause damage in the brains of intact spontaneously breathing primates including man. Schnei-
2 der (1963) stated that "In hypoxaemia produced by lowering of the pO2 of the inspired air there is a rapid development of a vicious circle when the critical threshold is e~ceeded: a hypoxidosis is then quickly enhanced by a decrease in respiration" (p. 151. Brierley and Meldrum (unpublished data) showed that circulatory and respiratory failure were closely associated. Lightly anaesthetized M. mutatta inhaled pure nitrogen via an endotracheal tube. Respiratory and heart rates slowed, blood pressure fell and the "last breath" occurred within 60 sec. Rapid mechanical ventilation with air restored circulation, respiration and the EEG. The brains of all 7 animals were normal. In view of the evidence for considerable cardio-respiratory impairment at the end of anoxia or profound hypoxia, it was notable that profound hypoxia represented by sub-atmospheric decompression to high altitude (using a decompression chamber) resulted in brain damage (often severe) in Rhesus monkeys (Nicholson, Freeland and Brierley 1970; Blagbrough, Brierley and Nicholson 1973). There was no recorded evidence of circulatory failure and manual ventilation was required in only 2 animals. Brain damage, of ischaemic type, lay along arterial boundary zones of cerebrum and cerebellum, but was variable in the basal ganglia. The decompressions were of squarewave profile. In the 18 monkeys of the first study exposed to an altitude of 37,500 ft (11,430 m) for 10-16 rain there was brain damage in 7. In the 21 of the second study exposed to the same altitude, 10 showed brain damage and 2 required manual ventilation. Decompressions to high altitude always followed an initial exposure to "cabin altitude" representing cabin pressurization for aircraft flying at or above 37,500 ft. This was 10,000 ft (3,048 m) for 10-15 min in all animals. Similar brain damage occurs in man as a consequence of abrupt systemic hypotension (Adams, Brierley, Connor and Treip 1966) and was produced in the normoxaemic Rhesus monkey by a rapid reduction in cerebral perfusion which had to be maintained up to apnoea and during mechanical ventilation (Brierley, Brown, Excell and Meldrum 1969). As the limit of tolerance to profound hypoxia is indicated by respiratory failure it was essential in the present investigation to employ spontaneously breathing and lightly anaesthetized animals in order to record physiological data before and after the "last breath". Pre-exposure to an altitude of 10,000 ft might have deferred the time to "last breath" in Rhesus monkeys exposed to an altitude of 37,500 ft in the decompression chamber. The consequences of exposure to high altitude (profound hypoxia) with and without immediate pre-exposure to the lesser altitude (moderate hypoxia) were, therefore, compared. The altitudes of I0,000 ft and 37,500 ft were converted to equivalent fractions of inspired oxygen (Luft 1964), i.e. 14 and 3.2 ~ oxygen respectively, the remainder being nitrogen. The study was divided into two parts. I. Physiological and neuropathological effects of exposure to profound hypoxia (3.2 ~o oxygen)from air, and II. Physiological and neuropathological effects of exposure to profound hypoxia (3.2~ oxygen) immediately after pre-exposure to moderate hypoxia (14~ oxygen). Papio anubis (PA) as well as Macaca mulatta (MM) were submitted to the two types of exposure in order to assess possible species differences.
MATERIALSAND METHODS Ten adult Papio anubis (weight range 5.7-9.3 kg) and 14 young adult Macaca mulatta (weight range 3.8-11.5 kg) were used. Each experiment was carried out in two stages with a 5-day interval (BrJerley and Excell 1966; Brierley et al. 1969; Brierley, Prior, Calverley and Brown 1977). The animals were fasted overnight and aseptic conditions with antibiotic cover were used for the first stage but not for the second as survival did not exceed 48 h. In MM 1 the second stage involved no surgery and survival was 8 days.
Stage I Under pentobarbitone anaesthesia (45 mg/kg intraperitoneally) an in-dwelling cannula was inserted via a femoral artery. The skull was trephined anterior to the torcula for blood sampling from the superior longitudinal sinus and silver-silver chloride epidural ball electrodes (2 parasagittal rows of 7, each 13 mm from the mid-line with inter-electrode distances of 9 or 10 mm) were connected to a microconnector mounted on the skull to record the EEG. Stage H Anaesthesia was induced with pentobarbitone in 12 MM (but with phencyclidine hydrochloride, Sernylan, Parke Davis: 5.75 mg/kg intramuscularly in MM 2 and ketamine hydrochloride, Vetalar, Parke Davis: 19.6 mg/kg intramuscularly in MM 10). Atropine 0.6 mg was injected intramuscularly. In MM 1, 2 and 10 anaesthesia was maintained by intermittent intravenous injections of pentobarbitone. In all other animals a continuous intravenous infusion of the steroid anaesthetic alphaxolone-alpbadolone (Althesin; veterinary preparation, Saffan, Glaxo) was regulated to maintain the EEG features of light to moderate anaesthesia, i.e. continuous EEG activity or definite but not pronounced burst suppression. At this level small alterations of infusion rate did not affect blood pressure or paCO2. In MM the standard solution of Saffan (12 mg/ml total steroids) was diluted with 6 % dextran in saline and given at a mean rate of 0.02 (range 0.006-0.093) ml/min of the undiluted solution. The baboons required infusion of the undiluted solution at a mean rate of 0.07 (range 0.0058-0.186) ml/min. Occasional supplementary bolus injections (0.03-0.5 ml) led to a rapid tall in mean arterial blood pressure of about 5 mm Hg lasting up to 30 sec. After the start of profound hypoxia no further pentobarbitone was given but the infusion of Saffan was continued at the pre-hypoxic rate. Physiological monitoring (details in Brierley et al. 1977) before, during and after the hypoxic exposures included the epidural EEG, ECG, heart rate (audible signal and digital display), respiratory rate and excursion (thoracic cuff or side-arm in the endotracheal tube), arterial blood pressure, venous sinus pressure and haemoglobin content, oxygen saturation, pOz, pCO2, pH, lactate and pyruvate in simultaneous samples of arterial and cerebral venous sinus blood. End-tidal (ET) pCO2 was recorded in all animals and ET pO2 in 8 PA and 2 MM (Beckman gas analysers). Bipolar recordings of EEG background activity were rated visually on a 6-point scale. A score was given for
4 each minute of recording: 1, continuous activity; 2, 3 and 4, mild, moderate and severe burst suppression activites with suppressions of < I, 1-3 and ?-~ 3 sec respectively: 5. low voltage burst suppression with bursts < 50/zV and suppressions > 3 sec; 6, isoelectric tracing. Computer analyses of the records were used to validate the visual ratings and to provide objective quantified descriptions of the EEG changes due to anaesthesia, hypoxia and recovery (Prior, Maynard and Brierley 1978). Scores 1-3 represented the light to moderate anaesthesia employed.
Exposure to gas mixtures Each altitude was simulated by an oxygen-nitrogen mixture, the fraction of oxygen being derived from the formula (Luft 1964) Fx :
(Bn - - 47) ( B o - 47)
× 0.2094
where Fx =: fraction of oxygen in final mixture, Bn ----mm Hg at altitude, Bo : m m Hg at sea level, 47 = water vapour pressure at 37 °C and 0.2094 ----fraction of oxygen in air. Thus, 10,000 ft = 14 ~ oxygen ("cabin altitude") at ambient pressure, and 37,500 ft -- 3.2 % oxygen at ambient pressure. These oxygen-nitrogen mixtures flowed at 3-5 1/min from cylinders (British Oxygen Co.) through a loop of wide-bore plastic tubing vented through a window. The side-arm of a T-piece at the bottom of each loop was closed while the tube was purged of air. Hypoxia began when the endotracheal tube was connected to the T-piece.
Neuropathological examination Under deep pentobarbitone anaesthesia in the heparinized animal, perfusionfixation with FAM (formaldehyde, glacial acetic acid, absolute methanol, 1 : I : 8, 30003500 ml in 20-25 min) at 120-140 mm Hg was carried out after clamping the descending thoracic aorta and a brief wash-out with saline. After detaching the hind brain, the cerebral hemispheres were cut in 8 mm slices in the stereotactic coronal plane. Occipital, temporal and cerebellar blocks were embedded in Paraplast (Shandon Scientific Co.) and the remainder in low viscosity nitrocellulose. Paraffin sections were stained with cresyl violet, cresyl violet and Luxol fast blue, Mallory's phosphotungstic acid haematoxylin and with haematoxylin and eosin. Celloidin sections were stained additionally with Woelke's modification of Heidenhain's method for myelin sheaths. RESULTS
Pre-hypoxic values (Table 1) There was a steady physiological state for at least 2 h before the first hypoxic exposure. Pre-hypoxic values were derived from the 10 PA and 14 MM employed. There was no relationship between the type of anaesthetic (Saffan in all PA and 8 MM, pentobarbitone in 6 MM) and any of the pre-hypoxic values. Under light anaesthesia (EEG scores 1-3), mean arterial blood pressure (MABP), cerebral perfusion pressure (MABP venous sinus pressure : CPP), paCO2 and pvCOz were significantly higher -
TABLE 1 PRE-HYPOXIC (ABOUT 15 MIN BEFORE FIRST EXPOSURE) VALUES FOR BLOOD GASES, pH AND RELATED PHYSIOLOGICAL DATA IN PA AND MM. In this and subsequent tables, mean 4- 1 SD are given with n and range in brackets. The significance of differences between means (Student's t-test) is * = P < 0.05; ** = P < 0.01 ; n.s. = not significant. 10 P. anubis Respiration/rain Heart rate/min MABP (ram Hg) CPP (mm Hg) paO~(mmHg) pvO~(mm Hg) paCO~(mmHg) pvCO2(mmHg) pall pvH A-VmlOz/100ml
29.6
-4- 6.2 (10) (21-39) 163 -4- 17.8 (10) (144-196) 123 i 17.0 (10) (97-145) 118 4- 19.2 (10) (93-148) 80.2 ± 7.3 (10) (67.5-88.8) 49.9 4- 4.6 (10) (39.9-56.2) 36.9 ± 3.1 (10) (31.5-40.8) 41.2 ± 3.6 (10) (34.2-48.4) 7.405 ± 0.037 (10) (7.368-7.471) 7.369 ± 0.032 (10) (7.332-7.427) 3.3824- 1.179(9) ( I. 525-5.527)
14 M. mulatta 31.3
-4- 5.8 (14) (21-39) 174 -+- 20.0 (14) (132-204) 105 ± 14.8 (13) (86-143) 101 -4- 14.6 (12) (81-137) 88.0 ± 8.7 (13) (69.4-103.0) 45.8 ± 4.3 (13) (39.6-63.5) 33.1 4- 4.0 (13) (27.5-47.1) 39.1 ± 3.7 (13) (34.2-48.6) 7.444 ± 0.039 (13) (7.356-7.504) 7.403 4- 0.032 (13) (7.328-7.471) 3.667± 1.132(13) (2.405-5.723)
P n.s. n.s. * * * * * * * * *
in P A t h a n in M M , p aO~, p a l l a n d p v H were significantly higher in M M . Thus, the m o n key e x h i b i t e d a relative h y p e r v e n t i l a t i o n . The m e a n c o n t r o l h a e m o g l o b i n c o n c e n t r a t i o n was 11.8 (10.1-13.6) g/100 ml, n ---- 10 in P A a n d 10.0 (7.4-13.5) g/100 ml, n = 13 in M M . S o m e fall ( m a x i m u m 3.2 g/100 ml) o c c u r r e d d u r i n g each e x p e r i m e n t due to r e m o v a l o f b l o o d samples.
I. Physiological and Neuropathological Effects of Exposure to Profound Hypoxia (3.2. % Oxygen) from Air A l l these e x p o s u r e s e n d e d in a decline in E E G b a c k g r o u n d activity l e a d i n g r a p i d l y to an isoelectric trace. R e s p i r a t i o n slowed a n d b e c a m e increasingly gasping u p to the " l a s t b r e a t h " . R a p i d m e c h a n i c a l v e n t i l a t i o n with air was b e g u n within 70 see after the " l a s t b r e a t h " to restore c a r d i a c f u n c t i o n a n d respiration. M u l t i p l e e x p o s u r e s were s e p a r a t e d b y intervals o f air breathing. I n M M 1,4 a n d 5 (Fig. 2), these intervals were t o o b r i e f to p e r m i t full physiological, a n d in p a r t i c u l a r E E G recovery. Physiological analysis was only c a r r i e d o u t if the interval between exp o s u r e s was m o r e t h a n 15 min, i.e. 14 e x p o s u r e s in 6 M M (Fig. 2). Typical g r a p h i c a l analyses in the two species are shown in Figs. 3 a n d 4. In multiple exposures p h y s i o l o g -
ical alterations were related to values immediately before the start of each exposure (Figs. 5 and 8). Nine PA and 6 M M survived the profound hypoxia and the EEG was recorded (1-8 days later) under light anaesthesia: Saffan (0.75-1.0 ml/kg), but phencyclidinc hydrochloride (5.0-7.5 mg/kg) in the first 2 MM.
(l) Initial physiological alterations (a) Respiration, heart rate and MABP (Fig. 5). Hyperventilation indicated by a fall in ET pCO~ began at 5-10 sec in both species and was unrelated to an~.esthetic depth (within EEG scores t-3). Respiratory rate increased by a mean of 32 (9-67) °,ii of the immediate pre-hypoxic rates in PA and by 57 (9-177)~o in MM. Peak rates occurred at a mean of 61 (37-90) sec from the start of hypoxia in the baboon and at 57 (30-120) sec in the monkey. Heart rate increased within 5-10 sec by a mean of 20 (8-33) ~o of the pre-hypoxic rate in PA and 20 (5-48)~o in MM. Peak heart rates occurred at a mean of 66 (18-90) sec in PA and 49 (20-95) sec in MM. In PA cardiac arrhythmias occurred in 6 exposures (5 animals) and began at 45-154 sec. They were usually ventricular ectopic beats lasting a few seconds and not affecting blood pressure. In MM cardiac arrhythmias occurring in 13 of the 14 exposures began at 14-93 sec, lasted 3-245 sec and were present at the "last breath" in 11. Heart block, usually 2: 1, occurred in 7 exposures. In 5 of these there were also ectopic beats which occurred alone in the other 6 exposures. Most episodes of arrhythmia resulted in a transient fall in MABP but no alteration in the EEG. In PA there was no rise in MABP in 5 exposures and a mean increase of 18 (4-25) ~ in the other 13. In MM, MABP increased by a mean of 37 (18-67)~ of the pre-hypoxic level. Initial increases occurred after 6-t3 sec with peak values after a mean of 46 (15-90) sec in PA and 38 (15-75) sec in MM. (b) Bloodgases andpH. The rate of fall ofpaOz up to 2 rain was similar in the two species (Fig. 6). In PA there was a further slight fall up to the "last breath" while in MM with a wider range of times to "last breath" paO~ fell to the range 10-19 mm Hg. In PA, paCO2 at 60 and 90 sec was higher and pall lower than in MM. (c) Early changes in the EEG. Transient alterations in the EEG occurring within the first minute in 78 ~o of exposures in each species suggested a lighter level of consciousness or "activation". They comprised high voltage, generalised, continuous, mainly monorhythmic activity at 3-7 c/s with a 50 ~ incidence of spikes. The onset of unequivocal "activation" was earlier in M M (mean 34 ~ 10 sec) than in PA (mean 42 ~ 11 sec) and its mean duration was 57 ± 17 sec in MM and 46 :~ 11 sec in PA.
(2) The final physiological state Each exposure ended in a cardio-respiratory crisis comprising a rapid failure of respiration and circulation and a decline in the EEG. Apnoea usually coincided with an isoelectric EEG but the heart always continued to beat after the "last breath". In 17 of the 18 exposures in PA the sequence was a fall in respiratory rate, bradycardia and a fall in blood pressure. The minimum ET pCO~ during exposure was 28 ~ 4 mm Hg in PA and 18 & 2 mm Hg in MM. In all 18 exposures in PA the "last breath" was preceded
by periodic breathing which was seen in only 2 o f the 14 analysed exposures in M M . In PA, T-wave elevation began at 35-105 sec in 14 o f the 17 exposures. It lasted f r o m 1/4-12 min and, therefore, well into the period o f mechanical ventilation. In M M , T-wave elevation in the E C G usually began at a b o u t 1 min and persisted until just after the start o f mechanical ventilation. The E E G (Table 2) showed a clear and progressive decline in a m o u n t and voltage o f activity with increasing periods o f suppression (scores 2-5) unrelated to changes in anaesthesia. A wide range o f frequencies re-appeared with polymorphic delta waves increasing but not dominating the bursts o f activity. This decline began at about the same time in the two species and immediately after "activation" when M A B P , C P P and heart rate were significantly higher and respiratory rate and the incidence o f cardiac arrhythmias lower in PA. Subsequent deterioration to an isoelectric tracing (Table 2) was rapid, occurring significantly earlier in P A than in M M . A l t h o u g h the mean time to isoelectric state preceded the mean time to "last breath" by 9 sec in P A and 7 sec in M M , E E G activity persisted into apnoea for a few seconds in 7 exposures in P A and 11 in M M . At the onset o f the isoelectric state, with similar arterial oxygen tensions, mean TABLE 2 TIMES TO "LAST BREATH", EEG DECLINE AND ISOELECTR1C EEG WITH RELATED PHYSIOLOGICAL DATA (3.2 ~ OXYGEN FROM AIR)
Time to "last breath" (sec) Time to EEG decline (sec)
10 P. anubis
6 M. mulatta
(18 exposures)
(14 exposures)
P
142 4- 33.2 (18) (89-205) 83 4- 16.2 (18) (44-106)
199 ± 135.7 (14) ¢93-570) 91 4- 13.5 (14) (68-125)
n.s.
116 4- 24.9 (18) (72-153) 106 d_ 27.3 (17) (65-153) 187 4- 18.3 (18) (155-220) 6 32 4- 9.0 (17) (15-52)
88 4- 22.3 (12) (50-115) 81 4- 23.7 (12) (42-110) 132 4- 70.2 (14) (57-200) 86% 41 4- 12.4 (14) (20-57)
**
192 4- 94.3 (14) (98~414)
*
64 4- 37.1 (12) (32-138) 55 4- 34.2 (12) (26-125) 97 4- 21.0 (14) (60-144) 64~ 26 ± 5.8 (8) (18-30)
*
n.s.
Concomitants o f decline
MABP (mm Hg) CPP (ram Hg) Heart rate/rain Arrhythmias Respiration/min
Time to isoelectric EEG (sec) 133 ± 31.3 (17) (92-195)
* ** *
Concomitants o f isoelectric EEG
MABP (ram Hg) CPP (mm Hg) Heart rate/rain Arrhythmias Respiration/rain
99 d_ 28.7 (17) (65-148) 96 4- 27.4 (16) (53-134) 153 ± 27.6 (16) (53-134) 24 16 4- 3.5 (2) (13-18)
* ** **
heart rates, MABP and CPP were significantly higher and respiratory rate and incidence of cardiac arrhythmias lower in PA. (a) The "last breath" (Fig. 7). The range of times to "last breath" was greater in M M than in PA and in the l0 exposures excluded from the group analysis, the "last breath" occurred at 95-382 sec, 3 being outside the baboon range. At the time of the "last breath", heart rate, MABP and CPP (Table 3) were significantly higher in PA than in MM. ( b ) Physiological changes during apnoea. Mechanical ventilation was begun from 19 sec after the"last breath" (mean 31 ± 12, n -- 18 in PA and 35 ± 15, n = l I in MM). Rapid mechanical ventilation was often needed within 30 sec after the "last breath" to avert cardiac failure. During apnoea, heart rate, MABP and CPP were greater in PA than in MM (Fig. 5). The decline in each was continuous in M M while in PA they were well maintained up to a rapid terminal decline. A rise in venous sinus pressure during apnoea, greater in PA than in MM, contributed to the observed reduction in CPP. The data from the blood samples during apnoea (8-70 sec after the "last breath") (Table 3 and Fig. 8) must be interpreted in the light of the above changes. There were no inter-species differences in respect of arterial and venous oxygen tensions but p aCO2 and pvCO2 were significantly higher in PA. TABLE 3 AS TABLE 2. TIMES TO "LAST BREATH" WITH RELATED PHYISOLOGICAL DATA. ALSO BLOOD GASES, pH AND RELATED DATA IN BLOOD SAMPLES DURING APNOEA 10 P. anubis (18 exposures)
6 M. mulatta (14 exposures)
P
Time to "last breath" (sec)
142
±33.2 (18) (89-205)
199
i 136.0 (14) (93-570)
n.s.
At "last breath" Heart rate/min
141
± 25.8 (18) (10(0192) ± 27.4 (18) (48-150) i 26.0 (15) (51-135)
101
± 21.5 (13) (78-144) :~ 33.9 (12) (37-142) ~ 30.9 (12) (33-131)
MABP (mm Hg)
96
CPP (mm Hg)
85
Blood samples during apnoea paO2 (mm Hg) pvO~(mm Hg) p.COa (mm Hg) pvCO2 (mm Hg) paH pvH A-V ml O~/100 ml
17.2
dz 2.4 (16) (13.4-21.3) 10.6 ~ 3.4 (14) (5.4-17.0) 33.2 i 5.8 (16) (23.0-45.9) 42.0 4- 4.7 (14) (34.7-50.7) 7.451 ± 0.053 (16) (7.363-7.555) 7.382 ± 0.046 (14) (7.313-7.435) 0.833 ± 0.467 (10) (0.220-1.520)
69 60
16.1
i 4.2 (9) (10.4-24.4) 11.5 4- 2.9 (9) (7.5-15.5) 23.3 ± 3.9 (9) (15.5-29.3) 29.4 i 5.3 (9) (21.3-37.7) 7.480 -4- 0.066(9) (7.370-7.573) 7.426 ! 0.057(9) (7.336-7.506) 0.722 4- 0.256(6) (0.431-0.989)
n.s. n.s.
n.s.
n.s.
(3} Mechanical ventilation and recovery In both species the initial rapid mechanical ventilation with air (60-110/min) reduced ET pCO3 briefly to low levels (minimum 11 m m Hg) and had to be maintained until heart rate and blood pressure began to improve. Ventilatory rate was reduced only when circulatory status was regarded as satisfactory. Mechanical ventilation was usually terminated at the "break-through" of a few spontaneous breaths, i.e. after a mean of 258 4- 147 sec in PA and 128 4- 118 sec in MM. One PA died of heart failure less than 1 h after the "last breath" in spite of rapid mechanical ventilation. Recovery was characterized by a considerable rise in MABP to a mean of 167 419 m m Hg in PA and 109 4- 51 m m Hg in M M. In PA the highest recorded blood pressure was 290/150 m m Hg 30 sec after the start of mechanical ventilation and lasted 40 sec. In M M the corresponding values were 240/125 m m Hg at 80 sec for 30 sec. In the majority of exposures in both species, heart rate fell to a mean of 82 (range 63-110 in PA, 60-140 in MM)/min at 0-50 sec. In both species, heart rates then increased rapidly. Cardiac arrhythmias during and after mechanical ventilation occurred in 12 exposures in PA and consisted of ectopic or coupled beats lasting up to 21/3 min. In contrast, there were a few missed beats in 4 exposures in MM. Venous sinus pressure, rising during apnoea in all baboons and most monkeys, continued to rise during the first 2 min of mechanical ventilation so that the rise in CPP was often less than that of MABP. The E E G was isoelectric for about 1 min (mean 67 4- 32 see, n -- 16 in PA and 52 4- 49 sec, n = 13 in MM), but for 448 sec in the third exposure in M M 2. Return of activity usually coincided with the rise in MABP and the duration of the isoelectric state was unrelated to the type of depth of anaesthesia before hypoxia. An E E G score of 4 was regained at about 11/3 min from the start of mechanical ventilation. In both species the distribution of E E G recovery times was bimodal, distinct morphological patterns being associated with early and late recoveries. Early recovery, characterized by a return of moderate voltage burst suppression activity was often associated with a lighter level of pre-hypoxic anaesthesia and occurred at a mean of 18 4- 10 sec, n : 9 in PA and 53 4- 51 sec, n : 11 in MM. Late iecovery, characterized by long periods with asynchronous arcoid phase-locked 7 and 14 c/s spindles at 10-20 ktV, usually followed deeper anaesthesia and occurred at a mean of 179 4- 120 sec, n : 9 in PA and at 242 4- 218 sec, n : 3 in MM. The EEGs returned to their pre-hypoxic form except in M M 1 in which infrequent focal parietal spikes (R > L) appeared after the third and fourth exposures.
Plasma lactate and pyruvate Pre-hypoxic values. These are derived from 10 PA and 5 MM. Mean pre-hypoxic arterial plasma lactate was 0.78 4- 0.11 mmol/L, n ---- 10 in PA and 0.66 4- 0.24 mmol/l, n -- 5 in MM. Corresponding pyruvate values were 0.038 4- 0.018 mmol/1, n : 10 in PA and 0.029 4- 0.027 mmol/1, n : 4 in MM. Mean prehypoxic venous plasma lactate was 0.90 4- 0.22 mmol/l, n = 10 in PA and 0.67 4- 0.17 mmol/1, n : 5 in MM. Pyruvate values were 0.046 4- 0.018 mmol/1, n : 10 in PA and 0.035 ~ 0.022 mmol/l, n : 4 in MM.
10
Profoundhypoxia. Because of cumulative effects, the time courses of blood lactate and pyruvate were only defined during and after the first hypoxic exposure unless further exposures were late enough to allow return to normal values. Plasma lactate and pyruvate levels increased slowly up to 1 min and then more rapidly. In final blood samples in PA (86-180 sec) mean arterial lactate was 3.02 i 0.72 mmol/l, n = 14. The mean arterial plasma pyruvate had risen to 0.097 ± 0.041 mmol/1, n =-- 14. Cerebral arterio-venous differences for lactate and pyruvate did not differ significantly from pre-hypoxic values. In PA arterial lactate returned to the pre-hypoxic control range at a mean of 37 mJn (n z 16 exposures) after the "last breath". In MM 3, arterial lactate values at the end of 3 exposures were 2.59, 3.30 and 3.92 mmol/l at 31/2, 4 and 7 min respectively, with recovery to the pre-hypoxic range between each. Corresponding pyruvate values were 0.057, 0.07 and 0.037 mmol/1. Clinical recovery. Of the 10 baboons, one (PA 8) died of cardiac failure after exposure to 3.2 ~o oxygen from air and the other two (PA 7 and 10) died after subsequent exposures to 3.2 ~ oxygen from 14~o oxygen (Fig. !). The remaining 7 animals were returned to their cages in a state of light anaesthesia and the following morning were alert, active and had taken food and water. There were no neurological signs. Follow-up EEGs under light anaesthesia were all normal. Perfusion-fixation was carried out after survivals of 21-48 h. Of the 6 MM exposed to 3.2 ~ oxygen from air, one (MM l) was killed after 8 days, the others within 22-48 h. The latter were regaining consciousness when returned to their cages. The following morning 4 were alert, active, had taken food and water and sbowed no neurological signs. The animal (MM 1) exposed 4 times to 3.2 ~ oxygen from air was sitting up in its cage on the evening of the experiment and appeared normal for tbe next 5 days. On the 6th day it was relatively inactive, its movements were slowed, but there were no abnormal neurological signs. It was killed after a follow-up EEG which showed occasional right parietal spikes. In the remaining 5 MM follow-up EEGs under light anaesthesia were normal. Neuropathologicalexamination. The brains of the 7 surviving baboons and 5 of the 6 monkeys were normal. The brain o f M M 1 (86 g) showed no swelling. In coronal slices there was some patchy cortical discolouration around each intraparietal sulcus. In sections, typical ischaemic neuronal damage with a glio-mesodetmal reaction was restricted to the cerebral cortex. Neurones were reduced to shrunken, darkly-staining nuclei surrounded by cytoplasm staining pale pink with eosin and mauve with Luxol fast blue. There were abundant lipid phagocytes with many mitotic figures and also fibrous astrocytes. Histological alterations were symmetrical and began posteriorly in the deeper halves of each intraparietal sulcus and were centered on the 3rd layer. There was similar damage in the floor of the left inferior occipital sulcus. The upper ischaemic lesions (lying along the boundary zones between the territories of the anterior and middle cerebral arteries) decreased anteriorly so that involvement of the sensory and motor cortices was minimal.
min
Na
i
2,0
4p
i
3
I
°
I
7
60
I
I
I
I
I
I
i
100 i
120 i i i
l:!:!:!:!:!:!:!:!:!:i:i:!:!l
Wt 220 (k!
S (h)
5.')
44
8.C
23
7,0 ~!:~:~:~:~:~:i:::::::~:~:::::::;:;:::;:!:!iiii:!!~)!iiii:..'~" __.....
46
7.0
21
9.3
39
8,3
44
14% Oxygen
7,8
HF
I ~'~" °xy~e°
7..2
HF
52
48
5.3
HF
140 i i i
~
~~
160 18(3 l l l l
~
~
I
| g :::::::::::::::::::::::::::::::::::::::
I
I ::::::::::::::::::::::::::::::::::::::::::::::: IIIIIIIIII~IIIIIIIIIII
10
30
200
~
I
1C ~
80
50
70
90
10
130
150
170
190
~0 min
Fig. 1. Ten P. anubis showing exposures to moderate and profound hypoxia. * - excluded from analysis because of period o f mixed gas inhalation. Also body weights and survival (S). H F = death from heart failure. NO.
20
40
1
60
80
2
1(30
120
140
rain 160
~ I
~_
~
~
I
Wt (kg)
S (h)
3.8
192
5.2
41
~.~ ~
4
63
5~11
I|
I
II
48
5.844
° ~ _ _ l _ _ _ m m
g.,, 48
7 :::::::::::::::::::::::::::::
4..4 42
9 ~ ! 1
,6 0
~
~
14 */* Oxygen
i
32 ./. Oxygen
48
,624
11 ~
6.2
22
12 ~
9.8
24
13 ~
11.5
HF
4.4
22
14
~ l
10
• ,
i
30
I
I
50
I
I
70
I
I
90
I
a
110
i
,
130
i
,
150
i
,
170 rain Fig. 2. Fourteen M . rnulatta showing exposures to moderate and p r o f o u n d hypoxia. * = excluded from detailed analysis because o f proximity to previous exposure. Also body weights and survival (S). H F = death from heart failure.
12 m mol/l Pyr,. 3 70.1
Lact.
2 tO.08 1 t 0.06"'" 0 -LO.04 -7.5 7.4 7.3
art, plasma pyruvaLe
~
a
t
e
] --Arterial -- -- venous
80 mm
60 pC02
Hg
40 2O 0 mm
40
mm 1OO Hg
end-tidal p
f --
50
~
J
end-tidal pO 2 =
20 10 O mm 2O0 :300 f H20 IO0 O 16o 180 I $41D
120 100 80
]
85
45
2O
~), ~ , 50,2,o/mi~ ....
_-"_ ....
Z-
]
ven~Js SinUSpr'essure heart r~ /mir /rain
arrhythmias
mm 180
Hg
140 100 6O
supr~ess~n isoelectric
rterial blOOd pressure
I" 6
~
I activ
O
1
2
3 4 5 6 7 8 9 1011 12 min
Fig. 3. PA 4. Graphical analysis of physiological data. MV = mechanical ventilation at rates shown. S = blood sample during hypoxia.
11. Physiological and Neuropathological Effects of Exposure to Profound Hypoxia (3.2 % oxygen)following Moderate Hypoxia (14 % Oxygen) In 10 PA there were 18 exposures (1-2 per animal). In 12 M M there were t5 exposures (1-2 per animal) (Figs. 1 and 2). Prehypoxic values are given in Table 1. Nine PA and 10 M M survived the profound hypoxia and the EEG was recorded up to 48 h later under light anaesthesia (Ketamine hydrochtoride 20 mg/kg i n M M I0 and 11 and Saffan 0.75-1.0 ml/kg in all other animals except M M 1 4 ~ which no anaesthesia was required). (A) Exposure to 14 % Oxygen (Table 4 and Figs. 5 and 8) Heart rate increased significantly only in M ~ cardiac ~ h y t h m i a s did not occur in either species and blood gases and pH attained a s t a ~ state at~a~r 5Ll0 ~ n . While
13 mmol/I
Loct. Pyr. 3 ~0.07 2 tO.05 1 -1:0.03-0
0,07-
7.5 7.4 7.3
----
Ng
30 20 10
mm
70 60
mm
3O 2O 10 40 f
mm
--
I \ '.. ~
-ts
150 f mm 100 Hg 58
_
I
P02
I
ts --'''-~' ts is ts
end- tid°l
H20
mm
Hg
continuous burst suppression isoelectric
venous
r
..... /
]
~ J
I heart arrhyth.
1I [!
background \ activity 3.2 % 02
EEG I
~
gasps ,, .......... sinus pressu~..re_~u_~
.-
F I- - F L F I-
6
-] ~)~554pT/rniL
60
220 180 140 100 60 200 160 120
J
t
DO2
• mm
__v.ooo /.
Air a i
L-x .
Air i
0
1
2
3
,
i
,
i
i
i
i
4 5 6 7 8 91011 rain
Fig. 4. M M 3.
Graphical
analysis.
Symbols
as
in Fig. 3.
the values for paO2 were similar in the two species (means of 53 and 52 m m Hg), paCO2 and pvCO2 were significantly greater, and pall and PvH were significantly less in PA than in M M indicating a relative hyperventilation in the latter. The A - V oxygen content differences increased from 3.38 to 4.65 ml oxygen/100 rnl (P < 0.05) in PA and from 3.67 to 5.53 ml oxygen/100 ml (P < 0.02) in MM. At the end of pre-exposure to 14 % oxygen only plasma lactate in PA was significantly increased from control values, mean arterial lactate being 1.08 4- 0.30 mmol/1, n = 8 (P < 0.02) and mean cerebral venous lactate 1.17 4- 0.33 mmol/l, n ----- 7 (P < 0.05) in samples taken at 20 min. M M showed no significant changes in plasma lactate or pyruvate. The EEG was unchanged. B. Exposure to 3.2 % Oxygen from 14 % Oxygen In 18 exposures in PA the mean time to "last breath" was 136 sec. It was 880 sec in 15 exposures in M M (Fig. 7).
P. anubis
M. mulatta
;2
-
n= 18 18 18
19 19 18 18
14 14 14
15 15 15 15
resp. r a t e / r a i n 22O
160
":>: :.~.:
140
:
1
PROFOUND HYPOXIA IN Papio anubis AND M a c a c a mulatta - - PHYSIOLOGICAL AND NEUROPATHOLOGICAL EFFECTS I. Abrupt Exposure following Normoxia II. Abrupt Exposure following Moderate Hypoxia
J. B.
BRIERLEY,PAMELA F. PRIOR, JANE CALVERLEYand A. W. BROWN
Medical Research Coucil Laboratories (JBB, JC and A WB) , Carshalton, Surrey, and The EEG Department (PFP), The London Hospital, Whitechapel, London E1 1BB (Great Britain)
SUMMARY Lightly anaesthetized and spontaneously breathing P. anubis (PA) and M. mulatta (MM) inhaled at ambient pressure 3. 2~ oxygen (= 37,500 ft or 11,430 m)
from air and also after pre-exposure to 14 ~ oxygen ( - 10,000 ft or 3,048 m). The EEG, ECG, respiratory rate, arterial and cerebral venous sinus pressures, end-tidal pO~ and pCOz and body temperature were recorded. Arterial and cerebral venous sinus blood gases, pH and pyruvate and lactate contents were estimated. Before hypoxia, MM showed a relative hyperventilation. Profound hypoxia, from air, ended with the "last breath" at 89-205 sec in PA and at 93-570 sec in MM. Brain damage was restricted to one MM (4 exposures). Profound hypoxia after exposure to 14 ~ oxygen ended with the "last breath" at 87-210 sec in PA and at 120 sec-94 min (including 9 exposures over 5 min) in MM. Brain damage was restricted to one MM ("last breath" at 94 min). In the two MM with brain damage there was evidence of reduction in cerebral perfusion near the end of profound hypoxia. Brain damage in one animal contrasts with the frequent and often severe brain damage in MM after equivalent sub-atmospheric decompressions preceded by exposure to moderate altitude (10,000 ft).
INTRODUCTION At present there is no convincing evidence to show that pure hypoxic hypoxia (i.e. a reduced arterial oxygen tension with normal cerebral blood flow) can cause damage in the brains of intact spontaneously breathing primates including man. Schnei-
2 der (1963) stated that "In hypoxaemia produced by lowering of the pO2 of the inspired air there is a rapid development of a vicious circle when the critical threshold is e~ceeded: a hypoxidosis is then quickly enhanced by a decrease in respiration" (p. 151. Brierley and Meldrum (unpublished data) showed that circulatory and respiratory failure were closely associated. Lightly anaesthetized M. mutatta inhaled pure nitrogen via an endotracheal tube. Respiratory and heart rates slowed, blood pressure fell and the "last breath" occurred within 60 sec. Rapid mechanical ventilation with air restored circulation, respiration and the EEG. The brains of all 7 animals were normal. In view of the evidence for considerable cardio-respiratory impairment at the end of anoxia or profound hypoxia, it was notable that profound hypoxia represented by sub-atmospheric decompression to high altitude (using a decompression chamber) resulted in brain damage (often severe) in Rhesus monkeys (Nicholson, Freeland and Brierley 1970; Blagbrough, Brierley and Nicholson 1973). There was no recorded evidence of circulatory failure and manual ventilation was required in only 2 animals. Brain damage, of ischaemic type, lay along arterial boundary zones of cerebrum and cerebellum, but was variable in the basal ganglia. The decompressions were of squarewave profile. In the 18 monkeys of the first study exposed to an altitude of 37,500 ft (11,430 m) for 10-16 rain there was brain damage in 7. In the 21 of the second study exposed to the same altitude, 10 showed brain damage and 2 required manual ventilation. Decompressions to high altitude always followed an initial exposure to "cabin altitude" representing cabin pressurization for aircraft flying at or above 37,500 ft. This was 10,000 ft (3,048 m) for 10-15 min in all animals. Similar brain damage occurs in man as a consequence of abrupt systemic hypotension (Adams, Brierley, Connor and Treip 1966) and was produced in the normoxaemic Rhesus monkey by a rapid reduction in cerebral perfusion which had to be maintained up to apnoea and during mechanical ventilation (Brierley, Brown, Excell and Meldrum 1969). As the limit of tolerance to profound hypoxia is indicated by respiratory failure it was essential in the present investigation to employ spontaneously breathing and lightly anaesthetized animals in order to record physiological data before and after the "last breath". Pre-exposure to an altitude of 10,000 ft might have deferred the time to "last breath" in Rhesus monkeys exposed to an altitude of 37,500 ft in the decompression chamber. The consequences of exposure to high altitude (profound hypoxia) with and without immediate pre-exposure to the lesser altitude (moderate hypoxia) were, therefore, compared. The altitudes of I0,000 ft and 37,500 ft were converted to equivalent fractions of inspired oxygen (Luft 1964), i.e. 14 and 3.2 ~ oxygen respectively, the remainder being nitrogen. The study was divided into two parts. I. Physiological and neuropathological effects of exposure to profound hypoxia (3.2 ~o oxygen)from air, and II. Physiological and neuropathological effects of exposure to profound hypoxia (3.2~ oxygen) immediately after pre-exposure to moderate hypoxia (14~ oxygen). Papio anubis (PA) as well as Macaca mulatta (MM) were submitted to the two types of exposure in order to assess possible species differences.
MATERIALSAND METHODS Ten adult Papio anubis (weight range 5.7-9.3 kg) and 14 young adult Macaca mulatta (weight range 3.8-11.5 kg) were used. Each experiment was carried out in two stages with a 5-day interval (BrJerley and Excell 1966; Brierley et al. 1969; Brierley, Prior, Calverley and Brown 1977). The animals were fasted overnight and aseptic conditions with antibiotic cover were used for the first stage but not for the second as survival did not exceed 48 h. In MM 1 the second stage involved no surgery and survival was 8 days.
Stage I Under pentobarbitone anaesthesia (45 mg/kg intraperitoneally) an in-dwelling cannula was inserted via a femoral artery. The skull was trephined anterior to the torcula for blood sampling from the superior longitudinal sinus and silver-silver chloride epidural ball electrodes (2 parasagittal rows of 7, each 13 mm from the mid-line with inter-electrode distances of 9 or 10 mm) were connected to a microconnector mounted on the skull to record the EEG. Stage H Anaesthesia was induced with pentobarbitone in 12 MM (but with phencyclidine hydrochloride, Sernylan, Parke Davis: 5.75 mg/kg intramuscularly in MM 2 and ketamine hydrochloride, Vetalar, Parke Davis: 19.6 mg/kg intramuscularly in MM 10). Atropine 0.6 mg was injected intramuscularly. In MM 1, 2 and 10 anaesthesia was maintained by intermittent intravenous injections of pentobarbitone. In all other animals a continuous intravenous infusion of the steroid anaesthetic alphaxolone-alpbadolone (Althesin; veterinary preparation, Saffan, Glaxo) was regulated to maintain the EEG features of light to moderate anaesthesia, i.e. continuous EEG activity or definite but not pronounced burst suppression. At this level small alterations of infusion rate did not affect blood pressure or paCO2. In MM the standard solution of Saffan (12 mg/ml total steroids) was diluted with 6 % dextran in saline and given at a mean rate of 0.02 (range 0.006-0.093) ml/min of the undiluted solution. The baboons required infusion of the undiluted solution at a mean rate of 0.07 (range 0.0058-0.186) ml/min. Occasional supplementary bolus injections (0.03-0.5 ml) led to a rapid tall in mean arterial blood pressure of about 5 mm Hg lasting up to 30 sec. After the start of profound hypoxia no further pentobarbitone was given but the infusion of Saffan was continued at the pre-hypoxic rate. Physiological monitoring (details in Brierley et al. 1977) before, during and after the hypoxic exposures included the epidural EEG, ECG, heart rate (audible signal and digital display), respiratory rate and excursion (thoracic cuff or side-arm in the endotracheal tube), arterial blood pressure, venous sinus pressure and haemoglobin content, oxygen saturation, pOz, pCO2, pH, lactate and pyruvate in simultaneous samples of arterial and cerebral venous sinus blood. End-tidal (ET) pCO2 was recorded in all animals and ET pO2 in 8 PA and 2 MM (Beckman gas analysers). Bipolar recordings of EEG background activity were rated visually on a 6-point scale. A score was given for
4 each minute of recording: 1, continuous activity; 2, 3 and 4, mild, moderate and severe burst suppression activites with suppressions of < I, 1-3 and ?-~ 3 sec respectively: 5. low voltage burst suppression with bursts < 50/zV and suppressions > 3 sec; 6, isoelectric tracing. Computer analyses of the records were used to validate the visual ratings and to provide objective quantified descriptions of the EEG changes due to anaesthesia, hypoxia and recovery (Prior, Maynard and Brierley 1978). Scores 1-3 represented the light to moderate anaesthesia employed.
Exposure to gas mixtures Each altitude was simulated by an oxygen-nitrogen mixture, the fraction of oxygen being derived from the formula (Luft 1964) Fx :
(Bn - - 47) ( B o - 47)
× 0.2094
where Fx =: fraction of oxygen in final mixture, Bn ----mm Hg at altitude, Bo : m m Hg at sea level, 47 = water vapour pressure at 37 °C and 0.2094 ----fraction of oxygen in air. Thus, 10,000 ft = 14 ~ oxygen ("cabin altitude") at ambient pressure, and 37,500 ft -- 3.2 % oxygen at ambient pressure. These oxygen-nitrogen mixtures flowed at 3-5 1/min from cylinders (British Oxygen Co.) through a loop of wide-bore plastic tubing vented through a window. The side-arm of a T-piece at the bottom of each loop was closed while the tube was purged of air. Hypoxia began when the endotracheal tube was connected to the T-piece.
Neuropathological examination Under deep pentobarbitone anaesthesia in the heparinized animal, perfusionfixation with FAM (formaldehyde, glacial acetic acid, absolute methanol, 1 : I : 8, 30003500 ml in 20-25 min) at 120-140 mm Hg was carried out after clamping the descending thoracic aorta and a brief wash-out with saline. After detaching the hind brain, the cerebral hemispheres were cut in 8 mm slices in the stereotactic coronal plane. Occipital, temporal and cerebellar blocks were embedded in Paraplast (Shandon Scientific Co.) and the remainder in low viscosity nitrocellulose. Paraffin sections were stained with cresyl violet, cresyl violet and Luxol fast blue, Mallory's phosphotungstic acid haematoxylin and with haematoxylin and eosin. Celloidin sections were stained additionally with Woelke's modification of Heidenhain's method for myelin sheaths. RESULTS
Pre-hypoxic values (Table 1) There was a steady physiological state for at least 2 h before the first hypoxic exposure. Pre-hypoxic values were derived from the 10 PA and 14 MM employed. There was no relationship between the type of anaesthetic (Saffan in all PA and 8 MM, pentobarbitone in 6 MM) and any of the pre-hypoxic values. Under light anaesthesia (EEG scores 1-3), mean arterial blood pressure (MABP), cerebral perfusion pressure (MABP venous sinus pressure : CPP), paCO2 and pvCOz were significantly higher -
TABLE 1 PRE-HYPOXIC (ABOUT 15 MIN BEFORE FIRST EXPOSURE) VALUES FOR BLOOD GASES, pH AND RELATED PHYSIOLOGICAL DATA IN PA AND MM. In this and subsequent tables, mean 4- 1 SD are given with n and range in brackets. The significance of differences between means (Student's t-test) is * = P < 0.05; ** = P < 0.01 ; n.s. = not significant. 10 P. anubis Respiration/rain Heart rate/min MABP (ram Hg) CPP (mm Hg) paO~(mmHg) pvO~(mm Hg) paCO~(mmHg) pvCO2(mmHg) pall pvH A-VmlOz/100ml
29.6
-4- 6.2 (10) (21-39) 163 -4- 17.8 (10) (144-196) 123 i 17.0 (10) (97-145) 118 4- 19.2 (10) (93-148) 80.2 ± 7.3 (10) (67.5-88.8) 49.9 4- 4.6 (10) (39.9-56.2) 36.9 ± 3.1 (10) (31.5-40.8) 41.2 ± 3.6 (10) (34.2-48.4) 7.405 ± 0.037 (10) (7.368-7.471) 7.369 ± 0.032 (10) (7.332-7.427) 3.3824- 1.179(9) ( I. 525-5.527)
14 M. mulatta 31.3
-4- 5.8 (14) (21-39) 174 -+- 20.0 (14) (132-204) 105 ± 14.8 (13) (86-143) 101 -4- 14.6 (12) (81-137) 88.0 ± 8.7 (13) (69.4-103.0) 45.8 ± 4.3 (13) (39.6-63.5) 33.1 4- 4.0 (13) (27.5-47.1) 39.1 ± 3.7 (13) (34.2-48.6) 7.444 ± 0.039 (13) (7.356-7.504) 7.403 4- 0.032 (13) (7.328-7.471) 3.667± 1.132(13) (2.405-5.723)
P n.s. n.s. * * * * * * * * *
in P A t h a n in M M , p aO~, p a l l a n d p v H were significantly higher in M M . Thus, the m o n key e x h i b i t e d a relative h y p e r v e n t i l a t i o n . The m e a n c o n t r o l h a e m o g l o b i n c o n c e n t r a t i o n was 11.8 (10.1-13.6) g/100 ml, n ---- 10 in P A a n d 10.0 (7.4-13.5) g/100 ml, n = 13 in M M . S o m e fall ( m a x i m u m 3.2 g/100 ml) o c c u r r e d d u r i n g each e x p e r i m e n t due to r e m o v a l o f b l o o d samples.
I. Physiological and Neuropathological Effects of Exposure to Profound Hypoxia (3.2. % Oxygen) from Air A l l these e x p o s u r e s e n d e d in a decline in E E G b a c k g r o u n d activity l e a d i n g r a p i d l y to an isoelectric trace. R e s p i r a t i o n slowed a n d b e c a m e increasingly gasping u p to the " l a s t b r e a t h " . R a p i d m e c h a n i c a l v e n t i l a t i o n with air was b e g u n within 70 see after the " l a s t b r e a t h " to restore c a r d i a c f u n c t i o n a n d respiration. M u l t i p l e e x p o s u r e s were s e p a r a t e d b y intervals o f air breathing. I n M M 1,4 a n d 5 (Fig. 2), these intervals were t o o b r i e f to p e r m i t full physiological, a n d in p a r t i c u l a r E E G recovery. Physiological analysis was only c a r r i e d o u t if the interval between exp o s u r e s was m o r e t h a n 15 min, i.e. 14 e x p o s u r e s in 6 M M (Fig. 2). Typical g r a p h i c a l analyses in the two species are shown in Figs. 3 a n d 4. In multiple exposures p h y s i o l o g -
ical alterations were related to values immediately before the start of each exposure (Figs. 5 and 8). Nine PA and 6 M M survived the profound hypoxia and the EEG was recorded (1-8 days later) under light anaesthesia: Saffan (0.75-1.0 ml/kg), but phencyclidinc hydrochloride (5.0-7.5 mg/kg) in the first 2 MM.
(l) Initial physiological alterations (a) Respiration, heart rate and MABP (Fig. 5). Hyperventilation indicated by a fall in ET pCO~ began at 5-10 sec in both species and was unrelated to an~.esthetic depth (within EEG scores t-3). Respiratory rate increased by a mean of 32 (9-67) °,ii of the immediate pre-hypoxic rates in PA and by 57 (9-177)~o in MM. Peak rates occurred at a mean of 61 (37-90) sec from the start of hypoxia in the baboon and at 57 (30-120) sec in the monkey. Heart rate increased within 5-10 sec by a mean of 20 (8-33) ~o of the pre-hypoxic rate in PA and 20 (5-48)~o in MM. Peak heart rates occurred at a mean of 66 (18-90) sec in PA and 49 (20-95) sec in MM. In PA cardiac arrhythmias occurred in 6 exposures (5 animals) and began at 45-154 sec. They were usually ventricular ectopic beats lasting a few seconds and not affecting blood pressure. In MM cardiac arrhythmias occurring in 13 of the 14 exposures began at 14-93 sec, lasted 3-245 sec and were present at the "last breath" in 11. Heart block, usually 2: 1, occurred in 7 exposures. In 5 of these there were also ectopic beats which occurred alone in the other 6 exposures. Most episodes of arrhythmia resulted in a transient fall in MABP but no alteration in the EEG. In PA there was no rise in MABP in 5 exposures and a mean increase of 18 (4-25) ~ in the other 13. In MM, MABP increased by a mean of 37 (18-67)~ of the pre-hypoxic level. Initial increases occurred after 6-t3 sec with peak values after a mean of 46 (15-90) sec in PA and 38 (15-75) sec in MM. (b) Bloodgases andpH. The rate of fall ofpaOz up to 2 rain was similar in the two species (Fig. 6). In PA there was a further slight fall up to the "last breath" while in MM with a wider range of times to "last breath" paO~ fell to the range 10-19 mm Hg. In PA, paCO2 at 60 and 90 sec was higher and pall lower than in MM. (c) Early changes in the EEG. Transient alterations in the EEG occurring within the first minute in 78 ~o of exposures in each species suggested a lighter level of consciousness or "activation". They comprised high voltage, generalised, continuous, mainly monorhythmic activity at 3-7 c/s with a 50 ~ incidence of spikes. The onset of unequivocal "activation" was earlier in M M (mean 34 ~ 10 sec) than in PA (mean 42 ~ 11 sec) and its mean duration was 57 ± 17 sec in MM and 46 :~ 11 sec in PA.
(2) The final physiological state Each exposure ended in a cardio-respiratory crisis comprising a rapid failure of respiration and circulation and a decline in the EEG. Apnoea usually coincided with an isoelectric EEG but the heart always continued to beat after the "last breath". In 17 of the 18 exposures in PA the sequence was a fall in respiratory rate, bradycardia and a fall in blood pressure. The minimum ET pCO~ during exposure was 28 ~ 4 mm Hg in PA and 18 & 2 mm Hg in MM. In all 18 exposures in PA the "last breath" was preceded
by periodic breathing which was seen in only 2 o f the 14 analysed exposures in M M . In PA, T-wave elevation began at 35-105 sec in 14 o f the 17 exposures. It lasted f r o m 1/4-12 min and, therefore, well into the period o f mechanical ventilation. In M M , T-wave elevation in the E C G usually began at a b o u t 1 min and persisted until just after the start o f mechanical ventilation. The E E G (Table 2) showed a clear and progressive decline in a m o u n t and voltage o f activity with increasing periods o f suppression (scores 2-5) unrelated to changes in anaesthesia. A wide range o f frequencies re-appeared with polymorphic delta waves increasing but not dominating the bursts o f activity. This decline began at about the same time in the two species and immediately after "activation" when M A B P , C P P and heart rate were significantly higher and respiratory rate and the incidence o f cardiac arrhythmias lower in PA. Subsequent deterioration to an isoelectric tracing (Table 2) was rapid, occurring significantly earlier in P A than in M M . A l t h o u g h the mean time to isoelectric state preceded the mean time to "last breath" by 9 sec in P A and 7 sec in M M , E E G activity persisted into apnoea for a few seconds in 7 exposures in P A and 11 in M M . At the onset o f the isoelectric state, with similar arterial oxygen tensions, mean TABLE 2 TIMES TO "LAST BREATH", EEG DECLINE AND ISOELECTR1C EEG WITH RELATED PHYSIOLOGICAL DATA (3.2 ~ OXYGEN FROM AIR)
Time to "last breath" (sec) Time to EEG decline (sec)
10 P. anubis
6 M. mulatta
(18 exposures)
(14 exposures)
P
142 4- 33.2 (18) (89-205) 83 4- 16.2 (18) (44-106)
199 ± 135.7 (14) ¢93-570) 91 4- 13.5 (14) (68-125)
n.s.
116 4- 24.9 (18) (72-153) 106 d_ 27.3 (17) (65-153) 187 4- 18.3 (18) (155-220) 6 32 4- 9.0 (17) (15-52)
88 4- 22.3 (12) (50-115) 81 4- 23.7 (12) (42-110) 132 4- 70.2 (14) (57-200) 86% 41 4- 12.4 (14) (20-57)
**
192 4- 94.3 (14) (98~414)
*
64 4- 37.1 (12) (32-138) 55 4- 34.2 (12) (26-125) 97 4- 21.0 (14) (60-144) 64~ 26 ± 5.8 (8) (18-30)
*
n.s.
Concomitants o f decline
MABP (mm Hg) CPP (ram Hg) Heart rate/rain Arrhythmias Respiration/min
Time to isoelectric EEG (sec) 133 ± 31.3 (17) (92-195)
* ** *
Concomitants o f isoelectric EEG
MABP (ram Hg) CPP (mm Hg) Heart rate/rain Arrhythmias Respiration/rain
99 d_ 28.7 (17) (65-148) 96 4- 27.4 (16) (53-134) 153 ± 27.6 (16) (53-134) 24 16 4- 3.5 (2) (13-18)
* ** **
heart rates, MABP and CPP were significantly higher and respiratory rate and incidence of cardiac arrhythmias lower in PA. (a) The "last breath" (Fig. 7). The range of times to "last breath" was greater in M M than in PA and in the l0 exposures excluded from the group analysis, the "last breath" occurred at 95-382 sec, 3 being outside the baboon range. At the time of the "last breath", heart rate, MABP and CPP (Table 3) were significantly higher in PA than in MM. ( b ) Physiological changes during apnoea. Mechanical ventilation was begun from 19 sec after the"last breath" (mean 31 ± 12, n -- 18 in PA and 35 ± 15, n = l I in MM). Rapid mechanical ventilation was often needed within 30 sec after the "last breath" to avert cardiac failure. During apnoea, heart rate, MABP and CPP were greater in PA than in MM (Fig. 5). The decline in each was continuous in M M while in PA they were well maintained up to a rapid terminal decline. A rise in venous sinus pressure during apnoea, greater in PA than in MM, contributed to the observed reduction in CPP. The data from the blood samples during apnoea (8-70 sec after the "last breath") (Table 3 and Fig. 8) must be interpreted in the light of the above changes. There were no inter-species differences in respect of arterial and venous oxygen tensions but p aCO2 and pvCO2 were significantly higher in PA. TABLE 3 AS TABLE 2. TIMES TO "LAST BREATH" WITH RELATED PHYISOLOGICAL DATA. ALSO BLOOD GASES, pH AND RELATED DATA IN BLOOD SAMPLES DURING APNOEA 10 P. anubis (18 exposures)
6 M. mulatta (14 exposures)
P
Time to "last breath" (sec)
142
±33.2 (18) (89-205)
199
i 136.0 (14) (93-570)
n.s.
At "last breath" Heart rate/min
141
± 25.8 (18) (10(0192) ± 27.4 (18) (48-150) i 26.0 (15) (51-135)
101
± 21.5 (13) (78-144) :~ 33.9 (12) (37-142) ~ 30.9 (12) (33-131)
MABP (mm Hg)
96
CPP (mm Hg)
85
Blood samples during apnoea paO2 (mm Hg) pvO~(mm Hg) p.COa (mm Hg) pvCO2 (mm Hg) paH pvH A-V ml O~/100 ml
17.2
dz 2.4 (16) (13.4-21.3) 10.6 ~ 3.4 (14) (5.4-17.0) 33.2 i 5.8 (16) (23.0-45.9) 42.0 4- 4.7 (14) (34.7-50.7) 7.451 ± 0.053 (16) (7.363-7.555) 7.382 ± 0.046 (14) (7.313-7.435) 0.833 ± 0.467 (10) (0.220-1.520)
69 60
16.1
i 4.2 (9) (10.4-24.4) 11.5 4- 2.9 (9) (7.5-15.5) 23.3 ± 3.9 (9) (15.5-29.3) 29.4 i 5.3 (9) (21.3-37.7) 7.480 -4- 0.066(9) (7.370-7.573) 7.426 ! 0.057(9) (7.336-7.506) 0.722 4- 0.256(6) (0.431-0.989)
n.s. n.s.
n.s.
n.s.
(3} Mechanical ventilation and recovery In both species the initial rapid mechanical ventilation with air (60-110/min) reduced ET pCO3 briefly to low levels (minimum 11 m m Hg) and had to be maintained until heart rate and blood pressure began to improve. Ventilatory rate was reduced only when circulatory status was regarded as satisfactory. Mechanical ventilation was usually terminated at the "break-through" of a few spontaneous breaths, i.e. after a mean of 258 4- 147 sec in PA and 128 4- 118 sec in MM. One PA died of heart failure less than 1 h after the "last breath" in spite of rapid mechanical ventilation. Recovery was characterized by a considerable rise in MABP to a mean of 167 419 m m Hg in PA and 109 4- 51 m m Hg in M M. In PA the highest recorded blood pressure was 290/150 m m Hg 30 sec after the start of mechanical ventilation and lasted 40 sec. In M M the corresponding values were 240/125 m m Hg at 80 sec for 30 sec. In the majority of exposures in both species, heart rate fell to a mean of 82 (range 63-110 in PA, 60-140 in MM)/min at 0-50 sec. In both species, heart rates then increased rapidly. Cardiac arrhythmias during and after mechanical ventilation occurred in 12 exposures in PA and consisted of ectopic or coupled beats lasting up to 21/3 min. In contrast, there were a few missed beats in 4 exposures in MM. Venous sinus pressure, rising during apnoea in all baboons and most monkeys, continued to rise during the first 2 min of mechanical ventilation so that the rise in CPP was often less than that of MABP. The E E G was isoelectric for about 1 min (mean 67 4- 32 see, n -- 16 in PA and 52 4- 49 sec, n = 13 in MM), but for 448 sec in the third exposure in M M 2. Return of activity usually coincided with the rise in MABP and the duration of the isoelectric state was unrelated to the type of depth of anaesthesia before hypoxia. An E E G score of 4 was regained at about 11/3 min from the start of mechanical ventilation. In both species the distribution of E E G recovery times was bimodal, distinct morphological patterns being associated with early and late recoveries. Early recovery, characterized by a return of moderate voltage burst suppression activity was often associated with a lighter level of pre-hypoxic anaesthesia and occurred at a mean of 18 4- 10 sec, n : 9 in PA and 53 4- 51 sec, n : 11 in MM. Late iecovery, characterized by long periods with asynchronous arcoid phase-locked 7 and 14 c/s spindles at 10-20 ktV, usually followed deeper anaesthesia and occurred at a mean of 179 4- 120 sec, n : 9 in PA and at 242 4- 218 sec, n : 3 in MM. The EEGs returned to their pre-hypoxic form except in M M 1 in which infrequent focal parietal spikes (R > L) appeared after the third and fourth exposures.
Plasma lactate and pyruvate Pre-hypoxic values. These are derived from 10 PA and 5 MM. Mean pre-hypoxic arterial plasma lactate was 0.78 4- 0.11 mmol/L, n ---- 10 in PA and 0.66 4- 0.24 mmol/l, n -- 5 in MM. Corresponding pyruvate values were 0.038 4- 0.018 mmol/1, n : 10 in PA and 0.029 4- 0.027 mmol/1, n : 4 in MM. Mean prehypoxic venous plasma lactate was 0.90 4- 0.22 mmol/l, n = 10 in PA and 0.67 4- 0.17 mmol/1, n : 5 in MM. Pyruvate values were 0.046 4- 0.018 mmol/1, n : 10 in PA and 0.035 ~ 0.022 mmol/l, n : 4 in MM.
10
Profoundhypoxia. Because of cumulative effects, the time courses of blood lactate and pyruvate were only defined during and after the first hypoxic exposure unless further exposures were late enough to allow return to normal values. Plasma lactate and pyruvate levels increased slowly up to 1 min and then more rapidly. In final blood samples in PA (86-180 sec) mean arterial lactate was 3.02 i 0.72 mmol/l, n = 14. The mean arterial plasma pyruvate had risen to 0.097 ± 0.041 mmol/1, n =-- 14. Cerebral arterio-venous differences for lactate and pyruvate did not differ significantly from pre-hypoxic values. In PA arterial lactate returned to the pre-hypoxic control range at a mean of 37 mJn (n z 16 exposures) after the "last breath". In MM 3, arterial lactate values at the end of 3 exposures were 2.59, 3.30 and 3.92 mmol/l at 31/2, 4 and 7 min respectively, with recovery to the pre-hypoxic range between each. Corresponding pyruvate values were 0.057, 0.07 and 0.037 mmol/1. Clinical recovery. Of the 10 baboons, one (PA 8) died of cardiac failure after exposure to 3.2 ~o oxygen from air and the other two (PA 7 and 10) died after subsequent exposures to 3.2 ~ oxygen from 14~o oxygen (Fig. !). The remaining 7 animals were returned to their cages in a state of light anaesthesia and the following morning were alert, active and had taken food and water. There were no neurological signs. Follow-up EEGs under light anaesthesia were all normal. Perfusion-fixation was carried out after survivals of 21-48 h. Of the 6 MM exposed to 3.2 ~ oxygen from air, one (MM l) was killed after 8 days, the others within 22-48 h. The latter were regaining consciousness when returned to their cages. The following morning 4 were alert, active, had taken food and water and sbowed no neurological signs. The animal (MM 1) exposed 4 times to 3.2 ~ oxygen from air was sitting up in its cage on the evening of the experiment and appeared normal for tbe next 5 days. On the 6th day it was relatively inactive, its movements were slowed, but there were no abnormal neurological signs. It was killed after a follow-up EEG which showed occasional right parietal spikes. In the remaining 5 MM follow-up EEGs under light anaesthesia were normal. Neuropathologicalexamination. The brains of the 7 surviving baboons and 5 of the 6 monkeys were normal. The brain o f M M 1 (86 g) showed no swelling. In coronal slices there was some patchy cortical discolouration around each intraparietal sulcus. In sections, typical ischaemic neuronal damage with a glio-mesodetmal reaction was restricted to the cerebral cortex. Neurones were reduced to shrunken, darkly-staining nuclei surrounded by cytoplasm staining pale pink with eosin and mauve with Luxol fast blue. There were abundant lipid phagocytes with many mitotic figures and also fibrous astrocytes. Histological alterations were symmetrical and began posteriorly in the deeper halves of each intraparietal sulcus and were centered on the 3rd layer. There was similar damage in the floor of the left inferior occipital sulcus. The upper ischaemic lesions (lying along the boundary zones between the territories of the anterior and middle cerebral arteries) decreased anteriorly so that involvement of the sensory and motor cortices was minimal.
min
Na
i
2,0
4p
i
3
I
°
I
7
60
I
I
I
I
I
I
i
100 i
120 i i i
l:!:!:!:!:!:!:!:!:!:i:i:!:!l
Wt 220 (k!
S (h)
5.')
44
8.C
23
7,0 ~!:~:~:~:~:~:i:::::::~:~:::::::;:;:::;:!:!iiii:!!~)!iiii:..'~" __.....
46
7.0
21
9.3
39
8,3
44
14% Oxygen
7,8
HF
I ~'~" °xy~e°
7..2
HF
52
48
5.3
HF
140 i i i
~
~~
160 18(3 l l l l
~
~
I
| g :::::::::::::::::::::::::::::::::::::::
I
I ::::::::::::::::::::::::::::::::::::::::::::::: IIIIIIIIII~IIIIIIIIIII
10
30
200
~
I
1C ~
80
50
70
90
10
130
150
170
190
~0 min
Fig. 1. Ten P. anubis showing exposures to moderate and profound hypoxia. * - excluded from analysis because of period o f mixed gas inhalation. Also body weights and survival (S). H F = death from heart failure. NO.
20
40
1
60
80
2
1(30
120
140
rain 160
~ I
~_
~
~
I
Wt (kg)
S (h)
3.8
192
5.2
41
~.~ ~
4
63
5~11
I|
I
II
48
5.844
° ~ _ _ l _ _ _ m m
g.,, 48
7 :::::::::::::::::::::::::::::
4..4 42
9 ~ ! 1
,6 0
~
~
14 */* Oxygen
i
32 ./. Oxygen
48
,624
11 ~
6.2
22
12 ~
9.8
24
13 ~
11.5
HF
4.4
22
14
~ l
10
• ,
i
30
I
I
50
I
I
70
I
I
90
I
a
110
i
,
130
i
,
150
i
,
170 rain Fig. 2. Fourteen M . rnulatta showing exposures to moderate and p r o f o u n d hypoxia. * = excluded from detailed analysis because o f proximity to previous exposure. Also body weights and survival (S). H F = death from heart failure.
12 m mol/l Pyr,. 3 70.1
Lact.
2 tO.08 1 t 0.06"'" 0 -LO.04 -7.5 7.4 7.3
art, plasma pyruvaLe
~
a
t
e
] --Arterial -- -- venous
80 mm
60 pC02
Hg
40 2O 0 mm
40
mm 1OO Hg
end-tidal p
f --
50
~
J
end-tidal pO 2 =
20 10 O mm 2O0 :300 f H20 IO0 O 16o 180 I $41D
120 100 80
]
85
45
2O
~), ~ , 50,2,o/mi~ ....
_-"_ ....
Z-
]
ven~Js SinUSpr'essure heart r~ /mir /rain
arrhythmias
mm 180
Hg
140 100 6O
supr~ess~n isoelectric
rterial blOOd pressure
I" 6
~
I activ
O
1
2
3 4 5 6 7 8 9 1011 12 min
Fig. 3. PA 4. Graphical analysis of physiological data. MV = mechanical ventilation at rates shown. S = blood sample during hypoxia.
11. Physiological and Neuropathological Effects of Exposure to Profound Hypoxia (3.2 % oxygen)following Moderate Hypoxia (14 % Oxygen) In 10 PA there were 18 exposures (1-2 per animal). In 12 M M there were t5 exposures (1-2 per animal) (Figs. 1 and 2). Prehypoxic values are given in Table 1. Nine PA and 10 M M survived the profound hypoxia and the EEG was recorded up to 48 h later under light anaesthesia (Ketamine hydrochtoride 20 mg/kg i n M M I0 and 11 and Saffan 0.75-1.0 ml/kg in all other animals except M M 1 4 ~ which no anaesthesia was required). (A) Exposure to 14 % Oxygen (Table 4 and Figs. 5 and 8) Heart rate increased significantly only in M ~ cardiac ~ h y t h m i a s did not occur in either species and blood gases and pH attained a s t a ~ state at~a~r 5Ll0 ~ n . While
13 mmol/I
Loct. Pyr. 3 ~0.07 2 tO.05 1 -1:0.03-0
0,07-
7.5 7.4 7.3
----
Ng
30 20 10
mm
70 60
mm
3O 2O 10 40 f
mm
--
I \ '.. ~
-ts
150 f mm 100 Hg 58
_
I
P02
I
ts --'''-~' ts is ts
end- tid°l
H20
mm
Hg
continuous burst suppression isoelectric
venous
r
..... /
]
~ J
I heart arrhyth.
1I [!
background \ activity 3.2 % 02
EEG I
~
gasps ,, .......... sinus pressu~..re_~u_~
.-
F I- - F L F I-
6
-] ~)~554pT/rniL
60
220 180 140 100 60 200 160 120
J
t
DO2
• mm
__v.ooo /.
Air a i
L-x .
Air i
0
1
2
3
,
i
,
i
i
i
i
4 5 6 7 8 91011 rain
Fig. 4. M M 3.
Graphical
analysis.
Symbols
as
in Fig. 3.
the values for paO2 were similar in the two species (means of 53 and 52 m m Hg), paCO2 and pvCO2 were significantly greater, and pall and PvH were significantly less in PA than in M M indicating a relative hyperventilation in the latter. The A - V oxygen content differences increased from 3.38 to 4.65 ml oxygen/100 rnl (P < 0.05) in PA and from 3.67 to 5.53 ml oxygen/100 ml (P < 0.02) in MM. At the end of pre-exposure to 14 % oxygen only plasma lactate in PA was significantly increased from control values, mean arterial lactate being 1.08 4- 0.30 mmol/1, n = 8 (P < 0.02) and mean cerebral venous lactate 1.17 4- 0.33 mmol/l, n ----- 7 (P < 0.05) in samples taken at 20 min. M M showed no significant changes in plasma lactate or pyruvate. The EEG was unchanged. B. Exposure to 3.2 % Oxygen from 14 % Oxygen In 18 exposures in PA the mean time to "last breath" was 136 sec. It was 880 sec in 15 exposures in M M (Fig. 7).
P. anubis
M. mulatta
;2
-
n= 18 18 18
19 19 18 18
14 14 14
15 15 15 15
resp. r a t e / r a i n 22O
160
":>: :.~.:
140
:
