522 STROKE VOL 10, No 5, SEPTEMBER-OCTOBER 1979

Dissertations No 8: 1-50, Umea, Centraltryckeriet, 1972 14.9. Nibbelink DW, Torner JC, Henderson WG: Intracranial

aneurysms and subarachnoid hemorrhage. A cooperative study.Antifibrinolytic therapy in recent onset subarachnoid 15.hemorrhage. Stroke 6: 622-629, 1975

10. Sengupta RP, So SC, Villarejo-Ortega FJ: Use of epsilonaminocaproic acid (EACA) in the preoperative management of 16.ruptured intracranial aneurysms. J Neurosurg 44: 479-484,1976

11. Boterell EH, Lougheed WM, Scott JW, Vanderwater SL:Hypothermia, and interruption of carotid, or carotid and 17.vertebral, circulation in the surgical management of in-tracranial aneurysms. J Neurosurg 13: 1-42, 1956

12. Fodstad H, Liliequist B, Schannong M, Thulin CA: Tranex-amic acid in the preoperative management of ruptured in- 18.tracranial aneurysms. Surg Neurol 10: 9-15, 1978

13. Girvin JP: Failure of antifibrinolytic agents to improveoperative treatment of ruptured intracranial aneurysms, pp 19.279-281. In Morley TP (ed) Current Controversies inNeurosurgery. Philadelphia, London, Toronto, W.B. Saunders 20.Company 1976

Maurice-Williams RS: Prolonged antifibrinolysis: an effectivenon-surgical treatment for ruptured intracranial aneurysms?Brit Med J 1: 945-947, 1978Kaste M, Troupp H: Subarachnoid haemorrhage: long-termfollow-up results of late surgical versus conservative treatmentBrit Med J 1: 1310-1311, 1978van Rossum J, Wintzen AR, Endtz LJ, Schoen JHR, de JongeH: Effect of tranexamic acid on re-bleeding after subarachnoidhemorrhage: a double-blind controlled clinical trial. AnnNeurol 2: 242-245, 1977Knibestol M, Karadayi A, Tovi D: Echo-encephalographicstudy ofventricular dilatation after subarachnoid haemorrhagewith special reference to the effect of antifibrinolytic treatment.Acta Neurol Scand 54: 57-70, 1976Kagstrom E, Palma L: Influence of antifibrinolytic treatmenton the morbidity in patients with subarachnoid haemorrhage.Acta Neurol Scand 48: 257-258, 1972Rydin E, Lundberg PO: Tranexamic acid and intracranialthrombosis. Lancet II: 49: 1976Davies D, Howell DA: Tranexamic acid and arterial throm-bosis. Lancet I: 49, 1977

Detrimental Effect of Prolonged Hypothermiain Cats and Monkeys With and Without

Regional Cerebral Ischemia

PETTER A. STEEN, M.D., EDWARD H. SOULE, M.D.,

AND JOHN D. MICHENFELDER, M.D.

SUMMARY In a previous study occlusion of a middle cerebral artery (MCA) followed by 48 h of hypother-mia (29°) was lethal in 5 of 5 monkeys as compared to only 3 of 9 normothermic animals. The present studyextended these observations in monkeys and cats with or without MCA occlusion. In monkeys MCA occlusionplus 48 h of hypothermia was consistently lethal. Without MCA occlusion 2 of 3 monkeys survived, but werecomatose the first 12 h post-hypothermia. In normothermic cats, MCA occlusion was lethal in only one of 5animals whereas hypothermia was lethal in 20 of 21 cats with or without MCA occlusion. The detrimentaleffects of hypothermia were not favorably influenced either by hemodilution or by deliberate alterations inPacOj. The effect of 48 h of hypothermia and rewarming on cerebral blood flow (CBF) and cerebralmetabolites was evaluated in 6 normal monkeys. CBF was reduced 60 to 70 percent at 29°C and returned toonly a maximum of 50 percent of control with re-warming. Prior to re-warming distribution of CBF was in-homogeneous. Cerebral metabolites were borderline normal prior to re-warming but energy stores decreasedwhile lactate increased with re-warming.

Stroke Vol 10, No 5,1979

MANY MEASURES have been proposed as poten-tially efficacious in the treatment of acute regionalcerebral ischemia; few of these have been found tofavorably influence the outcome for patients sufferingfrom acute stroke.1 The cerebral protective effect ofhypothermia is well established, particularly as atechnique for prolonging the brain's tolerance toperiods of complete cerebral ischemia.2* RosomofPalso reported a beneficial effect of hypothermia in anacute canine stroke model produced by occlusion of amiddle cerebral artery (MCA). A study from this

Dr. Steen is Fellow in Anesthesiology, Dr. Soule is Professor ofPathology and Dr. Michenfelder is Professor of Anesthesiology,Mayo Medical School, Rochester, MN 55901.

This work was supported in part by Research Grant NS-7507from the National Institutes of Health, Public Health Service.

laboratory8 did not confirm these findings in monkeyswith MCA occlusion subjected to 48 h of hypother-mia; a detrimental effect was observed instead. It wasspeculated that hypothermia might ultimatelydiminish oxygen delivery to the region of ischemia byan effect on blood viscosity with a resulting decrease incollateral flow. Decreased oxygen delivery could alsoresult from a temperature effect on oxygen-hemoglobin dissociation resulting in decreased releaseof oxygen to the tissues. The present study wasdesigned to examine these possibilities as well as theeffect of hypothermia and rewarming on cerebralblood flow and cerebral metabolites.

Materials and Methods

Twelve Macaca Java or Macaca Phillipina (3-5 kg)and 26 cats (2.5-5 kg) of both sexes, unmedicated and

by guest on July 13, 2018http://stroke.ahajournals.org/

Dow

nloaded from

DETRIMENTAL EFFECT OF HYPOTHERMIA/5/een et al. 523

fasting, were studied. In all animals anesthesia was in-duced and maintained during surgery with halothane 1percent in nitrous oxide 70 percent and oxygen. Pan-curonium 0.3 mg/kg was given to produce muscleparalysis and to facilitate intubation of the tracheawith a cuffed endotracheal tube; ventilation was con-trolled with a Harvard pump. A femoral artery andvein were exposed and cannulated for pressuremeasurements, blood sampling, and drug and fluid ad-ministration. A urinary catheter and esophageal ther-mistor were inserted and secured.

Survival Studies. Six monkeys and all cats were usedfor survival studies. Previously reported survivalstudies in 14 monkeys (Macaca Java) were includedfor comparison purposes and comprised groups I andII.8 The 6 new monkeys were divided in 2 groups(Monkeys III and IV) and the cats were divided in 6groups (Cats I-VI). For detailed breakdown of thegroups see table 1. In 3 monkeys (III) and 23 cats(I-V), the right middle cerebral artery (MCA) was ex-posed via a transorbital approach using the operatingmicroscope. A miniaturized Mayfield clip was placedacross the MCA distal to the first anterior branchwhich supplies an anterior-inferior portion of the fron-tal lobe, except for 5 cats (V) where the clip was placedon the dura (sham operation). The body of the clipremained extradurally. After placement of the clip (orat an equivalent time in animals not subjected toMCA occlusion) halothane was discontinued in allgroups and the animals were sedated with diazepam0.1 mg/kg. The dural incision was sealed withSurgicel® and glue (alpha cyanoacrylate) and thewound was closed. Streptomycin (100 mg) andpenicillin (300,000 units) were administered intra-muscularly.

Thereafter all animals were maintained for 48 h onan intensive care protocol with continuous attendanceby 2 technicians under the supervision of a physician.The mean arterial blood pressure (MAP), centralvenous pressure (CVP), heart rate (HR), temperature,and urine output were monitored continuously andrecorded every 30 min. The EEG and ECG wererecorded every 2 h, and blood gases and hematocritwere determined every 4 h. Plasma lactate, pyruvate,

sodium, potassium and osmolarity were measured inall animals at the start and end of the 48 h period. Inhypothermic animals not subjected to operation(monkeys IV and cats VI) plasma corticosteroids werealso measured at the start and the end. All animalsreceived pancuronium 0.05 mg/kg h, diazepam 0.1mg/kg/h and 5 percent dextrose in 0.45 percentsaline, 5.0 ml/kg/h. Additional pancuronium wasgiven as needed to avoid any shivering, and the degreeof muscle paralysis was evaluated with a nervestimulator. Bicillin 200,000 units was given in-tramuscularly every 24 h. If MAP fell below 60 mmHg, the animals were supported with phenylephrine orepinephrine as needed. Hypothermia, 29°C was ac-complished by surface cooling in all monkeys and incats, groups II-VI. In animals subjected to operationthis was initiated 30 min after MCA occlusion (table1). The temperature was controlled with icepacks orheat lamps as needed. Ventilation was initially con-trolled with oxygen 40 percent, and nitrogen 60percent (humidified) and adjusted according to bloodgases. Paco2 was kept at 35 ± 2 mm Hg (temperaturecorrected) in monkey group IV, and cat groups I,IV-VI. Monkey group III and cat groups II and IIIwere ventilated to maintain a Paco2 35 ± 2 mm Hg,uncorrected for temperature. The trachea was suc-tioned as needed. The extremities were wrapped andthe animals were turned every 4 h.

One monkey group (III) and 2 cat groups (III, IV)were hemodiluted during cooling in order tocounteract the known effects of temperature onviscosity. Assuming near Newtonian behavior and anormal hematocrit of 40, dilution to a hematocrit of30-31 should result in near normal viscosity at 29°C*

•In humans and temperate zone animals, viscosity increases 2-3percent/°C,7 thus at 29°C viscosity should be increased ap-proximately 23 percent at all shear rates from 1 to 100 sec 1 . ' 1 ' Therelative viscosity is also constant at shear rates above 1 sec"1, in-dicating that plasma is the only part of the system affected bytemperature, and blood with an hematocrit of 30 or less displaysnear Newtonian behavior at all rates of shear.* 10 A change inhematocrit should, therefore, not affect the effect upon viscosity of achange in temperature. For Newtonian behavior, viscosity can beexpressed as a function of hematocrit (hct) according to Vand " by:i) = plasma IJ (1 + 0.025 hct + 7.3 • 10' hct2).

TABLE 1 Management of Animals in Survival Studies

Group

Monkeys I*

Monkeys II*

Monkeys III

Monkeys IV

Cats I

Cats II

Cats III

Cats IV

Cats V

Cats VI

n

9

5

3

3

5

5

2

6

5

3

MCA Occlusion

Yes

Yes

Yes

No

Yes

Yes

Yes

Yes

Sham

No

Hemodilution

No

No

Yes

No

No

No

Yes

Yes

No

No

Hypothermia

No

Yes

Yes

Yes

No

Yes

Yes

Yes

Yes

Yes

PaCOi (35 mm Hg)t

—

Uncorrected

Uncorrected

Corrected

—

Uncorrected

Uncorrected

Corrected

Corrected

Corrected

•Monkey Groups I and II were reported previously.0tA Pacos of 35 mm Hg was considered to be normal for normothermic animals. In hypothermic animals ventilation was varied

among groups so as to produce either temperature corrected (see table 2) or uncorrected values of Pacoi near 35 mm Hg (see text).

by guest on July 13, 2018http://stroke.ahajournals.org/

Dow

nloaded from

524 STROKE VOL 10, No 5, SEPTEMBER-OCTOBER 1979

Dilution was achieved by infusion of albumin andblood withdrawal as needed.

One sham operated cat was re-warmed after 24 h,one after 36 h. All other hypothermic animals were re-warmed after 48 h with heat lamps and warmingblankets. Drugs and fluids were discontinued after 48h in all animals. If spontaneous ventilation was judgedadequate, the endotracheal tube and all catheters wereremoved, monitoring was discontinued, and theanimals were returned to their cages for observationfor 5 days. If spontaneous ventilation remained in-adequate (despite reversal of muscle relaxants), theanimals were allowed to die.

CBFand Cerebral Metabolite Studies. These were donein 6 monkeys. Cerebral blood flow was measured witha modified Kety-Schmidt technique.12 A catheter wasinserted retrograde into the common carotid arteryvia the sublingual artery for injection of 133Xe; the ex-ternal carotid artery was tied off immediately distal tothe catheter. The soft tissue around the ipsilateral or-bit was incised to avoid extracerebral contaminationwith l33Xe via possible anastomoses between theophthalmic artery and extracerebral vessels. Formeasurement of CBF 0.5 mC; 1S3Xe in 0.2 ml salinewas injected, and CBF calculated from the initial onemin slope of the washout curve using a tissue/bloodpartition coefficient of 0.87.13

After completion of surgery, halothane was discon-tinued and diazepam 0.1 mg/kg was given. Controlmeasurements of CBF were taken 30 min later.Hypothermia, 29°C, was then induced by surfacecooling. Thereafter the monkeys were maintained for48 h using the same protocol as in the survival studies.No blood chemical analyses were performed in orderto minimize blood loss. CBF was measured every 4 h.

In 3 monkeys a craniotomy was done at the end ofthe 48 h period. The dura overlying the parietal cortexwas excised and biopsies of the parietal cortex weretaken using a technique that deposits a sample ofbrain (200-400 mg) into liquid nitrogen within onesec.14 The tissue was stored at -76°C and preparedfor analysis in a refrigerated box (-25°C) asdescribed by Folbergrova et al.16 Tissue extracts wereanalyzed with enzymatic fluorometric methods forphosphocreatine (PCr), ATP, ADP, AMP, andglucose,16 lactate and pyruvate.17 The sum of theadenine nucleotides was calculated as 2 Ad = [ATP]+ [ADP] + [AMP]. The energy state of the tissueswas expressed as the energy charge potential of theadenine nucleotide pool according to Atkinson:18 ECP= [ATP] + 0.5 [ADP]/2 Ad.

In these same 3 monkeys the distribution of CBFwas assessed at the conclusion of the experiment byautoradiography using 14C-antipyrine.19 Seventy-five/iCi/kg of the isotope was infused at a constant rateover the last minute prior to the taking of brain biop-sies; immediately following biopsy the monkeys werekilled with KC1 intravenously. The whole brain wasrapidly removed and frozen in acetone-dry ice. Twen-ty micron sections were cut coronally every 8 mm in acryostate at -15°C. The sections were immediatelydried on a hot plate at 60°C. Immediately adjacent

sections were prepared for histology (H & E stain) orautoradiographs. The latter were placed in contactwith a high resolution, single emulsion x-ray film, andfollowing a 14-day exposure, the films were developed.

In the 3 other monkeys re-warming was institutedas in the survival studies. CBF and the other variableswere recorded with every 2 degrees increase intemperature. At 37°C brain biopsies were taken andanalyzed as described above.

All blood gases are reported corrected fortemperature. All values are presented as mean ± SEM.Student's Mest for paired data was used for statisticalcomparison of data obtained from the same animals.Student's f-test for unpaired data was used for com-parison of different groups of animals, p < 0.05 wasregarded as significant. No statistical comparisonswere done between animal groups with n < 5.

Results

Survival Studies. Six of 9 normothermic monkeyswith MCA occlusion (Group 1) survived 7 days whileall 8 hypothermic monkeys with MCA occlusion(Groups II, III) died (with or without hemodilution)(table 2.) One of the 3 hypothermic monkeys withoutMCA occlusion (Group IV) died, while the remaining2 survived without any apparent deficit, although bothremained comatose for the first 12 hours post-hypothermia. AH deaths in the hypothermic monkeysoccurred during or shortly after re-warming.

Four of 5 normothermic cats with MCA occlusion(Group 1) survived 7 days, while 20 of 21 hypothermiccats (Groups II-VI) died (table 2). MCA occlusion,hemodilution, or level of Paco2 had no effect on sur-vival. Three cats required inotrope support at somepoint during hypothermia. Two cats (Group III) diedat 32 and 33 hours of hypothermia. One cat died 2days post-hypothermia, all other non-survivors diedduring or shortly after re-warming.

There were no significant changes from controls inplasma lactate, pyruvate, L/P ratio, potassium, os-molarity or corticosteroid levels when measured after48 h (not tabulated). There was a decrease in plasmasodium in all groups, from 154 ± 1 to 140 ± 2 meqv/1(mean ± SEM for all groups combined). Mean 48 hvalues for other variables are presented in table 2.Differences between groups are largely accounted forby the expected effects of the various physiologic in-terventions used (hypothermia, hemodilution, and/orcontrolled ventilation). A mild metabolic acidosis wasevident in many of the animals at the end of the 48 h.

CBF and Metabolite Studies. As in the survivalstudies, hypothermia in this group resulted in amodest reduction in blood pressure and a significantdecrease in heart rate (table 3). With re-warming,heart rate returned toward control while bloodpressure remained below control. Pao2 and Paco2remained steady throughout the study, but ametabolic acidosis developed with significantdecreases in pH and buffer base (BB+).

With cooling, mean CBF decreased immediately to40 percent of control (fig. 1), and after 16 hours to ap-

by guest on July 13, 2018http://stroke.ahajournals.org/

Dow

nloaded from

DETRIMENTAL EFFECT OF HYPOTHERMIA/S/ee/j et al. 525

TABLE 2 Seven Day Survival and Physiologic Variables. Mean 48 h Intensive Care Values for Animals in Survival Studies. Mean± SEM

Group7 day MABP

survival (mm Hg)CVP HR

(mm Hg) (beats/min)Temp(°C)

Hct%

PaOs*(mm Hg)

PaCOs*(mm Hg) pH

BB+(meq/L)

Monkeys I

Monkeys II

Monkeys III

Monkeys IV

Cats I

Cats II

Cats III

Cats IV

CatsV

Cats VI

9

5

3

3

5

5

2

6

5

3

6

0

0

2

4

0

0

1

0

0

98±4

91±3

94±4

115±6

114±9111±860

106±493

±7122

±15

2±0

2±0

3±1

4±1

3±0

2±0

4±1

5±1

3±0

196±9

128±4

126±2

113±3

196±13

120±2121±6

114±3

117±13120±2

37.1±0.0

29.4±0.1

28.9±0.1

29.2±0.0

37.0±0.1

29.2±0.1

29.0±0.1

29.3±0.1

29.1±0.1

29.1±0.0

(~40)t

(~40)t

31±0

46±434

±243±3

21±6

32±0

32±0

47±5

160±5

116±2182

±10

218±31

145±3

170±7145

±20

181±16

195±20

201±21

35±1

24±1

24±0

37±1

35±1

24±0

23±1

36±135

±236

±1

7.55±0.01

7.55±0.01

7.52±0.08

7.37±0.01

7.39±0.00

7.48±0.01

7.47±0.01

7.30±0.01

7.24±0.03

7.28±0.02

53±1

49±1

44±4

42±1

42±1

41±0

40±1

38±1

35±236

±1

•Temperature corrected.fEstimated from hemoglobin concentration.

proximately 30 percent of control. Thereafter CBFremained relatively stable. With re-warming, CBF ini-tially increased to about 50 percent of control (at33°C) but with continued re-warming to 37°C, CBFagain decreased to 30 percent of control.

In the monkeys not re-warmed, autoradiographicstudies showed non-homogeneous distribution of CBFin 2 of 3 animals after 48 h of hypothermia (fig. 2). In4 of the combined 10 coronal sections taken fromthese 2 monkeys there were areas of cerebral cortexwith very poor perfusion adjacent to well perfusedareas. In the third monkey, CBF distribution was nor-mal in all 5 sections. Histologic study of these sectionsshowed no apparent abnormalities in any of themonkeys. There were no histologic differencesbetween adjacent regions of well perfused and poorlyperfused cortex (fig. 3).

The metabolic measurements (table 4) were consis-tent with the CBF and histologic findings. Prior to re-warming the cerebral energy state was near normal

(primarily as reflected by the ECP) although a degreeof lactate accumulation is evident. After re-warmingthe cerebral metabolic state was clearly abnormal withloss of energy stores and further accumulation of lac-tate.

Discussion

It is readily apparent from these results that aprolonged period (48 h) of moderate hypothermia(29°C) has severe deleterious effects in both monkeysand cats. In monkeys, these effects were exaggeratedby, but not dependent upon, the prior creation of aregional cerebral ischemic lesion (by MCA occlusion).In cats, prolonged hypothermia was almost uniformlylethal (upon rewarming) whether or not a regionalischemic lesion had been created. This confirms theresults of a previous study6 in which 48 h of hypother-mia was found to aggravate rather than ameliorate theeffects of regional cerebral ischemia in monkeys.Combining the results of that study with the present

TABLE 3 Physiologic Variables Before and After 48 h of Hypothermia for Animals in CBF and MetaboliteStudies

Time

Control 37°C

24 hr 29°C

48 hr 29°C

Rewanned 37°C

n

6

6

6

3

MABP(mm Hg)

101±898

±3

71±5

74±16

HR(beats/min)

199±13

118±498

±4

167±7

Hct(%)

40±1

48±3

42±2

~35*

P&OJ(mm Hg)

185±9

163±15

159±10

150±9

PaCO.(mm Hg)

38±1

38±3

39±1

40±3

pH

7.46±0.02

7.33±0.02

7.30±0.02

7.28±0.00

BB(meq/L)

49±140

±1

40±1

39±1

•Estimated from hemoglobin concentration.

by guest on July 13, 2018http://stroke.ahajournals.org/

Dow

nloaded from

526 STROKE VOL 10, No 5, SEPTEMBER-OCTOBER 1979

CBF Before, During And After 48 Hours Of Hypothermia

100

c

1CDOO

|

CBF

80

60

40

20

TlSE

FIGURE 1. CBF before and during 48 h ofhypothermia, 29°C, in 6 monkeys, and duringsubsequent re-warming in 3 of them. The valuesat 48 h were the same in monkeys re-warmed (3animals) and not re-warmed. The data aretherefore presented as one continuous graph.

37 29 29 29 29 29 29 29 29 29 29 29 29 29 31 33 35 37

Temperature °C

0 4 8 12 16 20 24 28 32 36 40 44 48

Duration of hypothermia (hr)

Control p Hypothermia »4*Rewarmingd

results reveals that MCA occlusion was lethal (at 48 h)in 8 of 8 hypothcrmic monkeys as compared to only 3of 9 normothcrmic monkeys.

These results would seem to conflict with the many

ft

FIGURE 2. Coronal sections of monkey brain showingautoradiographs (top) and adjacent sections stained with H& E (bottom). Arrow indicates a gyrus of cortical graywhich is grossly underperfused compared to remainder ofcortex.

favorable reports,1"8 both experimental and clinical,describing the beneficial effects of hypothermia. Toour knowledge, all of these reports have been con-cerned with short term application of hypothermiaonly (up to a few hours). Results following hypother-mia of longer duration have been equivocal or evennegative.10"" Rosomoff* reported that one hour ofhypothermia (24°C) in dogs reduced the neurologicdeficit resulting from MCA occlusion while othershave consistently found that hypothermia acutely in-creases the tolerance of the brain to periods of com-plete global ischemia.*""* The presumed basis for suchprotection is the known direct effect of temperature onmetabolic rates.1"4 Thus, in complete ischemia, atemperature induced reduction in metabolism shouldprolong brain viability by reducing the rates of energyutilization and lactate production.14' " In incompleteischemia, as produced by MCA occlusion, a reductionin brain O, requirements should increase the brain'stolerance for a reduction in Ot delivery. ThatRosomoff5 was able to demonstrate protection in anacute model of MCA occlusion, whereas we observedan opposite effect in a chronic model, suggests that theduration of hypothermia per se may be a critical fac-tor. In addition to an effect of time, we considered andtested for the possible deleterious effects of reducedtemperature on blood viscosity, acid-base balance,and oxygen-hemoglobin dissociation.

In both monkeys and cats, hemodilution of a degreecalculated to result in normal viscosity at 29°C did notalter the lethality of MCA occlusion followed by 48 hof hypothermia. Possibly any beneficial effect ofhemodilution was countered by the detrimental effectof reduced Os carrying capacity. In normal dogsKoster et al." reported that the increase in CBFresulting from hemodilution during hypothermia wasbalanced by the reduced Oa carrying capacity suchthat O, delivery was unaltered. Whether this wouldalso be true in regional ischemia as produced by MCAocclusion is not known.

by guest on July 13, 2018http://stroke.ahajournals.org/

Dow

nloaded from

DETRIMENTAL EFFECT OF HYPOTHERMIA/Ste«» et al. 527

V « .

* . • •

A %

B.*>• -

FIGURE 3. A)H & E stain of thick (20 m n) frozen sectionshowing neurons from a cortical area with poor perfusion.X ¥00. B) Similarly prepared section from a well perfusedarea of the cortex. Cell preservation in both areas are essen-tially comparable. X 400.

At temperatures other than 37CC, the optimumlevel of ventilation, and hence Paco, and pH, for non-hibernating homeotherms is unknown. It has been es-tablished that poikilotherms do not adjust ventilationwith a decrease in temperature and thus Pacoaremains constant (uncorrected for temperature)." Ifthis is appropriate for homeotherms (with a normalPacOj of 35 mm Hg) then at 29°C the appropriatecorrected Paco2 would be 24 mm Hg. Such a relativerespiratory alkalosis would further shift the oxygen-hemoglobin dissociation curve to the left (hypother-mia per se results in a leftward shift of the curve) and

might aggravate tissue hypoxia by limiting off-loadingof oxygen. On the other hand, hibernating animalstend to maintain a constant pH with reduction in bodytemperature.28 If this is appropriate for homeothermsthen corrected Paco, at 29 C should remain close to35 mm Hg (equivalent to a Pacd of about 50 mm Hguncorrected) and the leftward shift of oxygen hemo-globin dissociation would be reduced. An additionalimportant variable altered by CO, is the cerebralblood flow. In normal hypothermic rats, CBF remainsresponsive to CO3. HSgerdal et al." reported thatCBF was reduced to 15" percent of control at a Paco,of 15 mm Hg (corrected) and a temperature of 22°C.They observed no adverse metabolic effects resultingfrom this low CBF state. Again the possible effects onCBF in areas of regional ischemia is unknown. Wespeculated that maintenance of a constant uncorrectedPaco, (as in poikilotherms) might, by the combinedeffect on oxygen-hemoglobin dissociation and CBF,aggravate regional ischemia, thus accounting for theuntoward effects of hypothermia. However, survival inanimals maintained at a corrected Paco, of 35 mmHg was not improved. In these latter animals, we didnot completely normalize the effect of temperature onoxygen-hemoglobin dissociation. This would have re-quired deliberate production of a degree of metabolicacidosis or a further reduction in ventilation. Still, inmost of the animals maintained at a corrected Paco3of 35 mm Hg, a modest metabolic acidosis did developwith time. Thus, any leftward shift in oxygen-hemo-globin dissociation would be further reduced.

A nearly uniform observation in both the monkeyand cat survival studies was the apparent unmaskingof a deleterious effect of hypothermia only upon re-warming. Prior to re-warming, most animalsappeared stable as judged by vital signs, electrolytebalance, acid-base status, and blood gases. This hasbeen previously reported. Blair et al.*° described acutecirculatory collapse during re-warming in spon-taneously ventilating hypothermic dogs. Popovic81

reported that rats would survive maximally 5.5 hoursof hypothermia (15°C) if re-warmed, but for 9 hours ifnot re-warmed. He suggested that hypothermia wasprotecting against its own potentially injurious effectsand that re-warming unmasks these effects.

In order to determine whether injurious effects ex-isted prior to re-warming as well as during and follow-ing re-warming, we carried out additional studies todetermine the effect of both hypothermia and re-warming on CBF and cerebral metabolites. With theonset of hypothermia, the reduction in CBF was

TABLE 4

Temp

29"C

37°C

Cerebral

n

CO

C

O

MetabolitesPCrt

Ounol/g)

2.76±0.04

2.28±0.20

after 48 h of Hypothermia and after Re-warming

ATPn(jimol/g)

1.87±0.03

1.73±0.11

ADPt(j*mol/g)

0.31±0.01

0.40±0.02

AMPt0imol/g)

0.039±0.004

0.067±0.004

ECPt

0.9140.005

0.8780.007

Gluooeet

9.24±0.69

9.05±1.44

Lactate+t0imol/g)

2.70±0.27

5.46±2.10

L/Pratlo*t

44±12

69±16

*Norm«l primate (Seimiri soiureua) T«1UM, thia laboratory: ATP - 2.00 ± 0.07; Lactate - 2.14 ± 0.16; L/P - 9.0 ± 1.8.tNormal canine values, thi« laboratory; PCr - 3.04 ± 0.17; ATP - 2.14 ± 0.10; ADP - 0.30 ± 0.01; AMP - 0.06 ± 0.01; ECP - 0.92 ± 0.01; Gltt-

oose - 2.21 ± 0.18; L*otate - 1.04 ± 0.14; L/P - 17 ± 1.

by guest on July 13, 2018http://stroke.ahajournals.org/

Dow

nloaded from

528 STROKE VOL. 10, No 5, SEPTEMBER-OCTOBER 1979

nearly appropriate for the expected reduction inmetabolic rate (approximately 50 percent24). By 16 h,CBF stabilized at a moderately reduced level (30 per-cent of control) but, assuming homogeneous distribu-tion, adequate for metabolic needs. With re-warming,CBF initially increased but could not be maintained attemperatures greater than 35°C. From the autoradio-graphic studies done prior to re-warming, it is evidentthat distribution of CBF was not homogeneous in atleast 2 of 3 animals. Inhomogeneity of flow duringhypothermia has been found in other vascular bedssuch as the mesentery of rats32 and dogs.33 The brainmetabolites prior to re-warming were consistent witha marginally abnormal cerebral status. Following re-warming, these abnormalities were clearly magnifiedpresumably as a reflection of an inadequate CBF, bothregionally and globally.

These findings are consistent with and perhapssufficient to explain the deleterious effects of hypother-mia encountered in the survival studies. Although theunderlying mechanism remains undetermined, it isprobable that such effects are a function of both timeand temperature. Whether there is a "critical" timefor a given temperature is unknown. Possibly thetechnique of re-warming may also be critical. Ouranimals were progressively re-warmed over a 2-3 hperiod. Perhaps if re-warming was instead "staged"with deliberate pauses at 2-3°C increments, thedeleterious effects might be minimized. It is apparent,too, that species differences exist. Cats are extremelysensitive such that even in the absence of regionalcerebral ischemia, hypothermia was lethal upon re-warming. Normal monkeys were less sensitive, but thedetrimental effects of hypothermia were easilydemonstrated in monkeys with regional cerebralischemia. In man, with normal cerebral function,prolonged periods of hypothermia (28-30°C) aregenerally assumed to be well tolerated. However, it isinteresting to note that in Fay's" pioneer humanstudies done in 1940 of prolonged hypothermia fortreatment of metastatic carcinoma, the mortality ratein 169 "treatments" (administered to 124 patients)was 11.2 percent. Of these, only 2 patients died whilehypothermic, 4 died during re-warming and 13 diedwithin 24 h of re-warming. We are unaware of anypublished reports concerning the effects of hypother-mia in patients suffering from acute stroke. In one un-reported series,36 12 acute stroke patients were cooledto 28-30°C for 4-7 days; following re-warming 10 ofthese died. The once common use of prolongedhypothermia as a therapeutic tool in head injured andpost-cardiac arrest patients has been largely aban-doned. A review of the literature does not provide asatisfactory explanation for this change in practice. Itis reasonable to conclude that morbidity and mortalitywere not improved by hypothermia. The present studysupports such a conclusion and suggests, in addition, apossible detrimental effect.

References1. Hoff JT: Resuscitation in focal brain ischemia. Crit Care Med

6: 245-253, 1978

2. Boyd RS, Connolly JE: Tolerance of anoxia of the dog's brainat various temperatures. Surgical Forum 12: 408-410, 1961

3. Marshall SB, Owens JC, Swan H: Temporary circulatoryocclusion to the brain of the hypothermic dog. Arch Surg 72:98-106, 1956

4. Wolin LR, Massopust LC Jr, White RJ: Behavioral effects ofautocerebral perfusion, hypothermia and arrest of cerebralblood flow in the rhesus monkey. Exp Neurol 39: 336-341, 1973

5. Rosomoff HL: Hypothermia and cerebral vascular lesions. II.Experimental interruption followed by induction of hypother-mia. Arch Neurol Psychiatry 78: 454-464, 1957

6. Michenfelder JD, Milde JH: Failure of prolonged hypocapnia,hypothermia, or hypertension to favorably alter acute stroke inprimates. Stroke 8: 87-91, 1977

7. Merrill EW, Gilliland ER, Cokelet G, Shin H, Britten A, WellsRE Jr: Rheology of human blood, near and at zero flow. Effectsof temperature and hematocrit level. Biophysical J 3: 199-213,1963

8. Guard CL, Murrish DE: Effects of temperature on the viscousbehavior of blood from Antarctic birds and mammals. CompBiochem Physiol 52A: 287-290, 1975

9. Harkness J, Whittington RB: Blood-plasma viscosity: An ap-proximate temperature — invariant arising from generalizedconcepts. Biorheology 6: 169-187, 1970

10. Merrill EW: Rheology of blood. Physiological Reviews 49:863-888, 1969

11. Vand V: Viscosity of solutions and suspensions. Part II. J PhysColloid Chem 52: 300-313, 1948

12. Kety SS, Schmidt CF: The determination of cerebral bloodflow in man by the use of nitrous oxide in low concentrations.Am J Physiol 143: 53-66, 1945

13. Olesen J, Paulson OB, Lassen NA: Regional cerebral bloodflow in man determined by the initial slope of the clearance ofintra-arterially injected 1MXe. Stroke 2: 519-540, 1971

14. Kramer RS, Sanders AP, Lesage AM et al: The effect ofprofound hypothermia on preservation of cerebral ATP contentduring circulatory arrest. J Thorac Cardiovasc Surg 56:699-709, 1968

15. Folbergrova J, MacMillan V, Siesjo BK: The effect ofmoderate and marked hypercapnia upon the energy state andupon the cytoplasmic NAD/NADH+ ratio of the rat brain. JNeurochem 19: 2497-2505, 1972

16. Lowry OH, Passonneau JV, Hasselberger FX et al: Effect ofischemia on known substrates and cofactors of the glycolyticpathway in brain. J Biol Chem 239: 18-30, 1964

17. Lowry OH, Passonneau JV: A Flexible System of EnzymaticAnalysis. New York, London, Academic Press, 1972

18. Atkinson DE: The energy charge of the adenylate pool as aregulatory parameter. Interaction with feedback modifiers.Biochemistry 7: 4030-4034, 1968

19. Reivich M, Jehle J, Sokoloff L et al: Measurement of regionalcerebral blood flow with antipyrine -I4C in awake cats. J ApplPhysiol 27: 296-300, 1969

20. Strong MJ, Keats AS: Induced hypothermia following cere-bral anoxia. Anesthesiology 28: 920-923, 1967

21. Mullan S, Raimondi AJ, Suwanwela C: Effect of hypothermiaupon cerebral injuries in dogs. Arch Neurol 5: 545-551, 1961

22. Gray TC: Reflections on circulatory control. Lancet 1:383-389, 1957

23. Connaughton PJ, Holt G, Lewis FJ: Prolonged hypothermiabelow 25CC for periods over 20 hours: Influence of therapy andtechnique on survival. Surgery 50: 372-381, 1961

24. Michenfelder JD, Theye RA: Hypothermia: Effect on caninebrain and whole-body metabolism. Anesthesiology 30:1107-1112, 1968

25. Michenfelder JD, Theye RA: The effects of anesthesia andhypothermia on canine cerebral ATP and lactate during anoxiaproduced by decapitation. Anesthesiology 33: 430-439, 1970

26. Koster JK, Van de Vanter SH, Bean J, Collins JJ, Cohn LH:Effect of hemodilution and profound hypothermic circulatoryarrest on blood flow and oxygen consumption of the brain. SurgForum 27: 235-237, 1976

27. Rahn H: Body temperature and acid-base regulation.Pneumonologie 151: 87-94, 1974

28. Reeves RB: Role of body temperature in determining the acid-

by guest on July 13, 2018http://stroke.ahajournals.org/

Dow

nloaded from

DETRIMENTAL EFFECT OF HYPOTHERMIA/Sfeen et al. 529

base state of vertebrates. Fed Proc 28: 1204-1208, 196929. Hagerdal M, Harp JR, Siesjo BK: Influence of changes in

arterial Pco2 on cerebral blood flow and cerebral energy stateduring hypothermia in the rat. Acta Anaesth Scand Suppl 47:25-33, 1975

30. Blair E, Montgomery AV, Swan H: Post-hypothermic cir-culatory failure. I. Physiologic observations on the circulation.Circ 13: 909-915, 1956

31. Popovic V: Lethargic hypothermia in hibernators and non-

hibernators. Ann NY Acad Sci 80: 320-331, 195932. Lynch HF, Adolph EF: Blood flow in small blood vessels during

deep hypothermia. J Appl Physiol 11: 192-196, 195733. Bond TP, Derrick JR, Guest MM: Microcirculation during

hypothermia. Arch Surg 89: 887-890, 196434. Fay T: Early experiences with local and generalized refrigera-

tion of the human brain. J Neurosurg 16: 239-260, 195935. Fields WS: Personal communication, Department of

Neurology, University of Texas, Houston, Texas

Prognosis in Patients With Infarction and TIAin Carotid Territory During and After

Anticoagulant Therapy

H A N S LINK, M.D., GUSTAF LEBRAM, M.D., INGEGERD JOHANSSON, M.D.,

AND CLAES RADBERG, M.D.

SUMMARY One hundred seventeen patients, 31 with TIA and 86 with cerebral infarction, had angio-graphically verified atherosclerosis within the relevant carotid artery territory and normal CSF. They weretreated with anticoagulants for a mean of 11.1 months. No TIA but 1 cerebral infarction, appearing during in-adequate anticoagulant therapy, was registered. Seventy-six of the patients, 20 with TIA and 56 with infarc-tion, were followed for a mean of 4.4 months after cessation of anticoagulants or during inadequate anti-coagulant treatment. Ten patients, 1 with initial TIA and 9 with initial infarction, developed cerebral infarctionnecessitating re-institution of anticoagulant therapy.

Long-term, anticoagulant treatment can be recommended in carefully selected patients with TIA, and alsowith infarction in the carotid territory.

Stroke Vol 10, No 5, 1979

EVIDENCE has been reported that patients withtransient ischemic attacks (TIA) in the carotidterritory benefit from treatment with anticoagulantdrugs.1-2i 3 Patients with cerebral infarction are com-monly not treated in the same way because of the riskof cerebral hemorrhage. Studies have shown that, inunselected cases, the risks of anticoagulant therapycan exceed the benefits.4

A careful selection of patients with cerebral infarc-tion seems warranted in order to study the effect ofanticoagulant treatment in this patient category. Thisstudy describes the outcome in a selected group of pa-tients with cerebral infarction or TIA in the part of thebrain supplied by the carotid artery. All patients wereinvestigated by cerebral angiography and werefollowed during and after anticoagulant therapy.

Materials and Methods

Patients

One hundred sixty-two consecutive patients, 47females and 115 males, 36 patients with TIA, and 126with cerebral infarction were included. The age dis-tribution is given in table 1. In patients with cerebralinfarction, neurological signs and symptoms had per-

From the Departments of Neurology (Drs. Link and Lebram)and Radiology (Drs. Johansson and Radberg), University Hospital,S-581 85 Linkoping, Sweden

sisted for more than 24 h, and the neurological ex-amination usually revealed persistent, though oftenonly slightly disabling, symptoms. TIA was defined asan attack with symptoms and signs of neurologicaldysfunction of less than 24 h duration. The patientswere admitted to the neurology department during a 2year period (Oct. 1, 1973 to Sept. 30, 1975). Onlythose patients were included who were considered can-didates for carotid endarterectomy and/or anti-coagulant treatment, i.e. the biological age was accept-ably low, the patients were considered able to managetreatment with anticoagulants, and they were not liv-ing too far from facilities for control of treatment.Patients with disorders which contraindicated anti-coagulant therapy were excluded. High blood pres-sure was not considered a contraindication, but wasalways lowered to a value below 160/100 mm Hgbefore anticoagulant therapy was started. Electrocar-diograms and determinations of serum transaminaseswere performed in all 162 patients on day 1 and 3 afteradmission. Patients with a probable source of em-bolism other than the carotid artery or its brancheswere excluded. These included patients with recentmyocardial infarction or artrial fibrillation. CSFstudies and cerebral angiography was carried out in allpatients.

CSF Investigations

In all patients lumbar puncture was performed on

by guest on July 13, 2018http://stroke.ahajournals.org/

Dow

nloaded from

P A Steen, E H Soule and J D Michenfelderregional cerebral ischemia.

Deterimental effect of prolonged hypothermia in cats and monkeys with and without

Print ISSN: 0039-2499. Online ISSN: 1524-4628 Copyright © 1979 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Stroke doi: 10.1161/01.STR.10.5.522

1979;10:522-529Stroke. 

http://stroke.ahajournals.org/content/10/5/522World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://stroke.ahajournals.org//subscriptions/

is online at: Stroke Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer available in the

Permissions in the middle column of the Web page under Services. Further information about this process isOnce the online version of the published article for which permission is being requested is located, click Request

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Stroke Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on July 13, 2018http://stroke.ahajournals.org/

Dow

nloaded from