xenon enhanced ct for analysis of cerebral integrity, perfusion, and blood flow

BP Drayer, SK Wolfson, OM Reinmuth, M Dujovny, M Boehnke and EE CookXenon enhanced CT for analysis of cerebral integrity, perfusion, and blood flow

1524-4628 Copyright © 1978 American Heart Association. All rights reserved. Print ISSN: 0039-2499. Online ISSN:

Stroke is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX 725141978, 9:123-130Stroke 

http://stroke.ahajournals.org/content/9/2/123.citationlocated on the World Wide Web at:

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

http://www.lww.com/reprintsReprints: Information about reprints can be found online at   journalpermissions@lww.com410-528-8550. E-mail: Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Fax: Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters  http://stroke.ahajournals.org//subscriptions/Subscriptions: Information about subscribing to Stroke is online at

at VA MED CTR BOISE on July 9, 2011http://stroke.ahajournals.org/Downloaded from

Xenon Enhanced CT for Analysis of CerebralIntegrity, Perfusion, and Blood Flow

BURTON P. DRAYER, M.D., SIDNEY K. WOLFSON, Jr., M.D.,

OSCAR M. REINMUTH, M.D., MANUEL DUJOVNY, M.D.,

MANFRED BOEHNKE, M.D., AND EUGENE E. COOK, B.S.

123

SUMMARY Enhancement of the brain substance for CT evalua-tion using inhaled Xenon is confirmed. This technique was applied tothe study of the normal and the embolized adolescent baboon.Healthy cerebral tissue enhances symmetrically, while abnormalareas show significantly diminished enhancement. At maximal

enhancement, an indication of gross comparative cerebral perfusion isobtained. By obtaining serial CT scans over a 10 minute time interval,the clearance rate of Xenon (cerebral blood flow) may be evaluated.Xenon-enhanced CT enables a visual and numerical analysis of bothbrain morphology and physiology.

SINCE THE INITIAL application by Kety and Schmidt,1

various techniques to measure cerebral blood flow have beenutilized with varying success. Although intra-arterialmethods appear reasonably reproducible,2'3 they may bedifficult, time-consuming and morbid. Inhalation tech-niques,4 though far less invasive, present difficulties in dataanalysis due to recirculation and extracranial activity. Allradionuclide methods, whether done by intra-arterial injec-tion or inhalation, are limited by the non-morphologicnature of isotope studies. This makes it difficult to dis-tinguish precisely which brain locality has produced the ex-ternally detected activity.

The introduction of cranial computed tomography (CT)has enabled accurate morphologic definition of the brain andits surroundings. As the iodinated contrast agents routinelyused for intravenous CT enhancement do not diffuse into thebrain substance, they are not ideal for the measurement ofcerebral blood flow.5 CT enhancement may also be obtainedusing diffusible Xenon administered by inhalation.6 Thus,problems generated by the third component of circulationexternal to the brain and by the lack of anatomic specificityare readily overcome by CT imaging. This preliminary studywill describe a CT technique to define cerebral perfusion andcerebral blood flow in the nonhuman primate as monitoredby Xenon-enhanced CT.

Methods

Experimental Subjects

Cerebral blood flow measurements were obtained on 2adolescent baboons (Papio anubis/cynocephalus) weighing7.8 and 8.6 kg. One animal was a healthy control with noevidence of neurologic dysfunction. The second baboon wasstudied 103 days following successful embolization of theleft middle cerebral artery via injection of a silastic-tantalumembolus (205PE) through the left internal carotid artery.7

The position of the embolus was monitored by CT (fig. 1) as

From the Departments of Radiology (Dr. Drayer), Surgery (Dr. Wolfson),Neurology (Drs. Drayer and Reinmuth) and Neurosurgery (Drs. Wolfson andDujovny), of the University of Pittsburgh Health Center, Presbyterian-University Hospital and the Laboratory of Surgical Research (Dr. Wolfsonand Mr. Cook) and Department of Radiology (Dr. Boehnke) of theMontefiore Hospital of Pittsburgh, PA.

This work was supported in part by a grant from the Western PennsylvaniaHeart Association.

Reprints: Dr. Drayer, Radiology Dept., U. of Pittsburgh School ofMedicine, Presbyterian-University Hospital, 230 Lothrop St., Pittsburgh, PA15261

well as by cerebral angiography. The baboon was clinicallyevaluated daily after the embolization.

Anesthesia

General anesthesia was used for all CT (including cerebralblood flow) studies. Following sedation with 8 mg of phen-cyclidine HCL (Sernalyn*), endotracheal intubation with acuffed tube was carried out. Anesthesia was induced with ahalothane (4%)-Oxygen (100%) mixture by face mask.Halothane (1%) was given throughout the scanningprocedures to keep the animal immobile. Continuous car-diac and respiratory monitoring was routine.

Serial CT

Serial CT scans were performed 4, 7, 10, and 45 daysfollowing embolization of the left middle cerebral arteryboth with and without intravenous contrast enhancement.Another CT scan (with no intravenous contrast) was donejust prior to the cerebral blood flow study (day 103). CT wasperformed just before the cerebral blood flow study in thehealthy control baboon.

An EMI 5005 head-body scanner with a 320 X 320matrix and 13 mm collimation was used. Scanning time foreach slice was 40 seconds and the scans were done at 32 maand 120 Kvp. A scan angle of 0 to 5% above the or-bitomeatal baseline was routine. Head movement was not aproblem with adequate anesthesia. Careful packing of thehead was done with bolus bags.

Xenon CT Cerebral Blood Flow

Anesthesia with halothane was carried out as describedabove using a Harvard respirator on a closed rebreathingventilation circuit. A baseline CT scan was performedchoosing a section where the frontal horns of the lateral ven-tricle were seen to advantage. A multi-channel analyzer(Corning) was used to measure Pao2, Paco2, and pH drawnfrom an arterial line and the O2 content of the expired air atbaseline and during serial scanning.

A 100% Oxygen-1% halothane mixture (5 liter flow rate)was inhaled by the baboon until the nitrogen was almost en-tirely washed out of the respiratory system as determined byan oxygen concentration in the expired air of greater than

•Corning Medical Diagnostics, Medfield, MA.

at VA MED CTR BOISE on July 9, 2011http://stroke.ahajournals.org/Downloaded from

124 STROKE VOL 9, No 2, MARCH-APRIL 1978

FIGURE 1. Embolus, Left Middle Cerebral Artery. A silastic-tanlulum embolus (arrow) is seen by CT in the location of thehorizontal portion of the left middle cerebral artery. Angiographyconfirmed the presence of the embolus completely occluding thehorizontal middle cerebral artery.

95%. Commercial (non-radioactive) Xenon was then addedto the system. A concentration of approximately 80% Xenonwas attained as estimated from the decrease in the measuredoxygen concentration.

When the Xenon concentration equilibrated at 80%, animmediate CT scan (maximum Xenon) was performed atthe same section level as the baseline. The inhalation ofXenon was then abruptly discontinued and replaced by theinhalation of 100% oxygen (5 liter flow rate). As the Xenonwas washing out, serial CT scans (at the same cut level) wereperformed at 0.3, 2.0, 3.7, 5.4, 7.1, 8.8, and 10.5 minutesfollowing discontinuing Xenon inhalation, i.e., 2.0 minutesmeans a scan was begun at 1 minute and 40 seconds andcompleted at 2 minutes and 20 seconds.

A matching image region in the anterior left and righthemispheres, including gray and white matter, was thencompared both by visual (various window settings) andnumerical (print-out averaging) methods to evaluate brainenhancement with Xenon. The mean and histogram wereobtained from comparable 95 to 100 voxel regions in eachhemisphere. These same regions were studied on each of thesuccessive serial CT scans to produce a Xenon clearanceprofile. Cerebral blood flow (ml/100 gm per minute) wasthen analyzed by two methods:8

(100) (X)

2. CBF =

(AA)(0.693) (X) (100)

The partition coefficient of Xenon (X) was based on thehemoglobin level.8 No correction was made for recircula-tion, Paco2, phencyclidine HC1 sedation, or halothane (1%)anesthesia.

Results

Baseline CT and Neurologic Condition

The baseline CT scan and neurologic status of the healthy,

control baboon were normal. The successfully embolized ba-boon (embolus in horizontal portion of left middle cerebralartery) never showed any abnormality in his neurologic ex-amination. The serial CT scans at 4, 7, 10, 45, and 103 dayswere repeatedly normal (fig. 2).

Xenon Brain Enhancement

In the control animal, the maximum CT enhancement atapproximately 80% Xenon concentration was symmetric inthe two cerebral hemispheres (table 1). Both deep andsuperficial brain substance seemed to enhance to the samedegree at this level of Xenon inhalation. In the embolizedbaboon, there was definite asymmetrical enhancement of thetwo cerebral hemispheres at 80% Xenon concentration. Thenon-embolized right hemisphere had a maximal enhance-ment of 11.0 EMI units above baseline. The left hemispherein the distribution of the left middle cerebral artery, bothsuperficial and deep, showed diminished maximal enhance-ment (9.4 EMI units) as compared to the right. A small areain the left opercula region had an enhancement of only 7.8EMI units. This poorly enhanced region was accentuated onthe delayed (3.7, 5.4, 7.1, and 8.8 minute) serial washoutscans (figs. 3, 4).

Cerebral Xenon Clearance

The cerebral clearance curves for Xenon were similar inboth animals as calculated by both compartmental (TV4) andstochastic (H/A) methods (table 1). Due to scanner speedand processing limitations, only 3 early scans (Maximum,0.3 and 2.0 minute) were available for the evaluation of thefast component of Xenon clearance (washout). Nevertheless,a distinct fast and slow component of cerebral blood flowwas readily demonstrated (table 2). In addition, stochasticanalysis defined the combined gray and white flow as similarin both hemispheres of both baboons. Using measure mode(WW 0, WL 27) reproductions, a visual assessment of theXenon washout was also readily obtained (fig. 4).

Gas Analysis

The Pao2 and expired air O2 were never lower than 78 torrwhen breathing pure Xenon thus reflecting some leakage inthe inhalation system. This was not of consequence as an ap-proximately 80% Xenon concentration was more than ade-quate for CT enhancement and washout observations. Inboth baboons, the serial blood gas measurement made at thetime of each washout CT scan showed a progressive increasein Pao2 and expired O2 concentrations (table 1). This in-crease in O2 reflects the decrease in Xenon concentrationand coincided with the diminishing attenuation coefficientnoted by CT scan. The Paco2 remained fairly constant in anessentially physiologic range (table 1).

Discussion

One of the early limitations of CT scanning was itsprimarily anatomic orientation. However, with the use of in-travenous contrast enhancement, physiologic informationcan be derived.9 It has also been suggested that cerebral

at VA MED CTR BOISE on July 9, 2011http://stroke.ahajournals.org/Downloaded from

XENON ENHANCED CJ/Drayer et al. 12S

FIGURE 2. Normal CT scan of a baboon 103 days following middle cerebral artery embolization.Progressively higher (with overlap) CT sections of the infratentorial and supratentorial brain compartmentsshowing no evidence of cerebral infarction. (1) Lens, (2) Vitreous, (3) Medial Rectus Muscle, (4) OpticNerve, (5) Brainslem. (6) Cerebellum, (7) Temporal Lobe, (8) Frontal Lobe, (9) Parietal Lobe, (10) Oc-cipital Lobe, (0) Lateral Ventricles.

at VA MED CTR BOISE on July 9, 2011http://stroke.ahajournals.org/Downloaded from

126 STROKE VOL 9, No 2, MARCH-APRIL 1978

TABLE 1A WNL Objective Findings with Xenon CT Studies

Baboon # 1 WNL Serial CT, WNL Clinical, No embolus

Left1 Right1 PaCOi PaO: pHExpired

Oi HB Hot

BaselineMaximum (75-80% Xenon)0.3 Minutes2.03.75.47.18.8

10.5

rCBF2 Brain (H/A)rCBF' Slow (TM)rCBF* Fast (TV2)

17.828.127.522.521.620.620.219.619.4

30.921.349.1

18.028.327.323.621.420.719.619.118.8

33.527.753.6

37.236.037.332.933.835.643.129.741.8

451108326395416432474418494

7.387.457.447.437.417.407.397.387.38

632125521570585594600595611

13.2 37.7

Expressed in EMI units (mean of specified voxel region).ml/lOOg/min.

TABLE IB Objective Findings with Xenon CT Studies

Baboon #2 WNL Serial CT, WNL Clinical, Embolus left MCA

Left1 Right1 PaCOi PaOi p HExpiredxpire

OJ HB Hct

BaselineMaximum (80-85% Xenon)0.3 Minutes2.03.75.47.18.8

10.5

rCBF* Brain (H/A)rCBF* Slow CVY2)rCBF2 Fast

16.325.723.120.719.619.118.417.817.4

33.227.762.0

16.427.425.522.922.320.620.319.218.4

31.126.451.3

32.629.3———35.1——29.9

481 7.40 653 12.778 7.45 96 —

37.9

492 7.38 605 — —

506 7.47 656 — —

1. Expressed as EMI units (mean of specified voxel region).2. ml/lOOg/min.

blood volumes may be measured using intravenousenhancement.10 However, intravenous water-soluble con-trast agents are not freely diffusible into the brain substance

and are thereby inadequate for the evaluation of cerebralblood flow.5 Intrathecal contrast enhancement withmetrizamide (Amipaque, Sterling-Winthrop Research

TABLE 2. Xenon Enhanced CT Analysis of Cerebral Blood Flow (Compartmental Analysis)

5 6TIME (MINUTES)

5 6 7TIME (MINUTES)

A comparison of the blood flow curves in the right (no embolus) versus the left (embolus horizontal MCA) hemisphere using semilogarithmicanalysis of the information in Table 1B and Figure 4. Although maximal perfusion of the right side is greater than the left, calculated cerebralblood flow (fast and slow component) was very similar.

at VA MED CTR BOISE on July 9, 2011http://stroke.ahajournals.org/Downloaded from

XENON ENHANCED CY/Drayer et al. m

D

FIGURE 3. Baseline CT (Before Xenon Enhancement) A. A CT section of the level of the lateral ventriclesphotographed at WW75 and WL28. B. A specimen radiograph of a baboon at a similar level to CT sectionplane showing the normal anatomic structures. C. The same CT section as in A now photographed usingmeasure mode and WL27. D. A reversal photograph (measure mode, WL27) of Figure 3C (Baseline forFigure 4).

at VA MED CTR BOISE on July 9, 2011http://stroke.ahajournals.org/Downloaded from

[28 STROKE VOL 9, No 2, MARCH-APRIL 1978

FIGURE 4. Cerebral Perfusion and Flow by Xenon CT Serial CT scans (measure mode. WL27 , reversalphotograph) of the anterior half of each cerebral hemisphere at precisely the same section level as thebaseline (fig. 3) were performed at measured time intervals during Xenon washout with oxygen. Thedescribed window settings will show volume elements containing material of 27 EMI units or greater asblack.

The diminished Xenon enhancement in the left middle cerebral artery distribution (embolus horizontalleft MCA, fig. I) as compared to the right is evident on all scans indicating diminished perfusion. Theprogressive diminution of Xenon enhancement in both hemispheres over a 10 minute clearance (washout)interval is readily demonstrated even by visual analysis of the scans. The complimentary numerical data(table 1) permits the calculation of cerebral blood flow. 0.0 Minute: Maximal Xenon Enhancement(estimated 80% Xenon concentration). 0.3 Minute following the inhalation of 100% oxygen to wash out theXenon. 2.0, 3.7, 5.4,7.1,8.8 and 10.5 Minute CT scans show the progressive clearance of Xenon with 100%oxygen inhalation. By 10.5 minutes the CT has almost returned to the baseline appearance (see figure 3.)

1 nstitute) may enable the dynamic evaluation of the cerebro-spinal fluid (CSF)circulation and theCSF-brain barrier.1113

Xenon, an inert gas with an atomic number of 54, is freelydiffusible into the gray and white matter of the brain. Xenonis readily exchanged between blood and tissue and highlysoluble in fat. Commercially available, non-radioactiveXenon is faintly visualized by conventional radiographictechniques and prominently seen using CT scanning.6 Thisgas also has a definite anesthetic effect if given in sufficientconcentrations."1 '*

Xenon133 has proven to be an important radionuclide forthe evaluation of cerebral blood flow. The most consistentdata are derived from the intracarotid injection of the radio-isotope with interpretation of serial washout informationusing a modification of the Kety-Schmidt formula.1 Inhala-tion techniques,4' " although requiring far more complexmanipulation of information, appear to supply similarresults with lower morbidity. Com part mental analysisenables the separation of the brain washout curve into 3basic compartments using inhalation methodology; fast

(predominant gray matter washout), slow (predominantlywhite matter washout), and slower (predominantly extra-cerebral sources).

Actual brain imaging using radionuclides requirescyclotron-produced radionuclides introduced by intra-carotid injection.17' 18 Even with adequate collimation, themorphologic resolution of radionuclide techniques seems ex-tremely limited when compared to CT scanning. Since non-labelled Xenon readily enhances the brain substance byfreely diffusing across the blood-brain barrier, CT scanningprovides a simple and reliable method for monitoring thisenhancement. The presence and rate of clearance of Xenonfrom small regions of gray and white matter may bemonitored by knowing the anatomy of the brain as definedby the CT scan. Even prior to the utilization of thenumerical CT data, a visual analysis can separate the fastergray matter washout from the slower white (fig. 4) as well ascompare the Xenon enhancement of the right versus the lefthemisphere.

Xenon-enhanced CT scanning appears to have 3 basic

at VA MED CTR BOISE on July 9, 2011http://stroke.ahajournals.org/Downloaded from

XENON ENHANCED CT/Drayer et al. 129

applications: (1) evaluation of capillary-brain tissue in-tegrity, (2) evaluation of gross comparative cerebral perfu-sion, and (3) evaluation of cerebral blood flow.

1. A baboon with a cerebral embolus had repeatedly nor-mal CT scans and clinical examinations. However, a Xenonenhanced CT scan revealed a small area of minimally per-fused brain in the opercular region. The location of the ab-normality was consistent with the characteristic location ofinfarction in the baboon following occlusion of the middlecerebral artery as documented by previous CT scan andpathologic studies.7-'" Therefore, Xenon-enhanced CT scan-ning appears to be an important adjunct to routine andenhanced CT scanning, especially when subtle tissue abnor-malities of either gray or white matter are sought.

2. A symmetrical, bilateral, total regional brain perfusionshould occur in a normal primate. The deep and superficialdiminished enhancement in the distribution of the embolizedleft middle cerebral artery is expected even though the lepto-meningeal anastomosis aids in the perfusion of a non-humanprimate hemisphere with vascular compromise. Thus, a sim-ple technique is now available to evaluate the gross perfusionof comparable areas of the same brain. Potential applica-tions include the evaluation of transient ischemic attacks,surgical vascular procedures, cerebral autoregulation, andcerebral edema.

3. The cerebral blood flow estimations as calculated usingserial Xenon-enhanced CT scans correlated closely withthose reported for baboons using the hydrogen clearancemethod.20 The findings of normal cerebral blood flow withchronic infarction in the baboon has been previouslyreported.19'20 The analysis obtained more accurately (i.e.consistent with previous nonhuman primate studies of cere-bral blood flow,19-20) reflects the slow component of flow.This, however, is merely a limitation of scanner speed andthe smallest size of a region of brain from which an accuratemean of attenuation coefficients may be obtained. Scans willbe performed in from 2 to 10 seconds with newer CTscanners. This should enable the acquisition of multipleattenuation values during the initial 2 to 3 minutes permit-ting a more significant fast component analysis. The extra-cranial Xenon flow is obviously not a problem as CTpresents a cross sectional image of the brain. In this study,recirculation was not an important element as it has littleeffect on the slow compartment washout curve.4 Our bloodflow values may need minor correction due to the use ofhalothane anesthesia and the mildly lowered arterial CO2

values. Nonetheless, these are experimental variables and donot detract from the concept of using serial Xenon-enhancedCT scans to evaluate accurately regional cerebral bloodflow.

Xenon enhanced CT scanning has definite shortcomingsas a cerebral blood flow technique. However, these all ap-pear readily surmountable with advancing CT technology.Rapid scanning times are needed for more significant grayand white matter washout curves; newer scanners need only2 to 10 seconds per section. With repeated scans at a similarlevel, the radiation dosage is quite high; future generationscanners may possibly lower the dose. Artifacts may alterthe enhanced scan appearance and especially affect the max-imal enhancement values; faster scan times, careful subjectpositioning, and prevention of movement make artifact aminor problem. The cerebral blood flow for only a single

section of brain has been demonstrated. Newer scan designsmay permit the scanning of multiple brain levelssimultaneously, providing information for reconstructingboth horizontal and coronal cerebral blood flow analysis.Since absolute CT values are dependent on the effectiveenergy of the X-ray beam, which varies with the kilovoltageand type and amount of attenuating material,20 smallvariations may occur in these CT values on presently usedscanners; these variations are quite small when studying thecranial contents and newer scanners should almost com-pletely correct for even these minor shifts. Finally, a com-mon scale and unit for CT attenuation coefficient (u) has notyet been established and varies with each manufacturer'sscanner; however, values may be established for eachspecific scanner and reasonable reproducjj?ility assured forany individual scanner by using phantom calibrations.22

Xenon-enhanced CT scanning now permits the evalua-tion of cerebral blood flow and comparative brain perfusionin both anatomic and physiologic terms. We are presentlyestablishing cerebral washout curves in various physiologicand pathologic states in the baboon using numericallyaveraged CT attenuation coefficients and comparing these toflows obtained using Xenon133 inhalation. With further ex-perience and newer developments in scanner technology,consistent and accurate cerebral blood flow and perfusionmeasurements should readily be obtained using Xenon-enhanced CT.

References

1. Kety SS, Schmidt CF: The nitrous oxide method for the quantitativedetermination of cerebral blood flow in man: Theory, procedure, and nor-mal values. J Clin Invest 27: 476-483, (July) 1948

2. Lassen NA, Hjiedt-Rasmussen K, Sjirensen SC, Skinhjij E, Cronqvist S,Bodforss B, Ingvar DH: Regional cerebral blood flow in man determinedby krypton-85. Neurology 13: 719-727, (Sept) 1963

3. Posner JB: Newer techniques of cerebral blood flow measurement.Stroke 3 : 227-237, (May-June) 1972

4. Mallett BL, Veall N: The measurement of regional cerebral clearancerates in man using Xenon-133 inhalation andextracranial recording. ClinSci 29: 179-191, (Aug) 1965

5. Phelps ME, Kuhl DE: Pitfalls in the measurement of cerebral bloodvolume with computed tomography. Radiology 121: 375-377, (Nov)1976

6. Winkler SS, Sackett JF, Holden JE, Fleming DC, Alexander SC,Madsen M, Kimmel RI: Xenon inhalation as an adjunct to computerizedtomography of the brain: Preliminary study. Invest Radiol 12: 15-18,(Jan-Feb) 1977

7. Drayer BP, Dujovny M, Wolfson SK, Boehnke M, Cook EE,Barrionuevo P, Rosenbaum AE: The capacity for computed tomographydiagnosis of cerebral infarction: An experimental study in the nonhumanprimate. Radiology 125: 393-402 (Nov) 1977

8. HjJedt-Rasmussen K, Sveinsdotter E, Lassen NA: Regional cerebralblood flow in man determined by intra-arterial injection of radioactive in-ert gas. Circ Res 18: 237-247, 1966

9. Kramer RA, Janetas GP, Perlstein G: An approach to contrast enhance-ment in computed tomography of the brain. Radiology 116: 641-647,(Sept) 1975

10. Penn RD, Walser R, Ackerman L: Cerebral blood volume in man. Com-puter analysis of a computerized brain scan. JAMA 234: 1154-1155,(Dec) 1975

11. Drayer BP, Rosenbaum AE, Higman HB: Cerebrospinal fluid imagingusing serial metrizamide CT cisternography. Neuroradiology IS: 7-17,(Mar) 1977

12. Drayer BP, Rosenbaum AE: Amipaque brain penetrance, its correlationwith adverse reactions and value radiodiagnostically. Acta Radiol(Stockh) Suppl, in press

13. Drayer BP, Rosenbaum AE, Reigel DH, Bank WO, Deeb ZL:Metrizamide CT cisternography: Pediatric applications. Radiology 124:349-357, (Aug) 1977

14. Cullen SC, Gross EG: Anesthetic properties of Xenon and Krypton.Science 113: 580-582, (May) 1951

at VA MED CTR BOISE on July 9, 2011http://stroke.ahajournals.org/Downloaded from

130 STROKE VOL 9, No 2, MARCH-APRIL 1978

15. Collins VJ: Pharmacology of inorganic gas anesthetics, pp. 1541-1542.In Principles of Anesthesia, 2nd edition. Philadelphia, Lea and Febiger,1976

16. Obrist WD, Thompson HK, Wang HS, Wilkinson WE: Regionalcerebral blood flow estimated by '"Xenon inhalation. Stroke 6: 245-256,(May-June) 1975

17. Kuhl DE, Reivich M, Alavi A, Nyary I, Staum MM: Local cerebralblood volume determined by three-dimensional reconstruction of radio-nuclide scan data. Circ Res 36: 610-619, (May) 1975

18. Phelps ME, Hoffman EJ, Coleman RE, Welch MJ, Raichle ME, WeissES, Sobel BE, Ter-Pergossian MM: Tomographic images of blood pool

and perfusion in brain and heart. J Nucl Med 17: 603-612, (July) 197619. Symon L, Dorsch NWC, Crockard HA: The production and clinical

features of a chronic stroke model in experimental primates. Stroke 6:476^t81, (Sept-Oct) 1975

20. Pasztor E, Symon L, Dorsch NWC, Branston NM: The hydrogenclearance method in assessment of blood flow in cortex, white matter, anddeep nuclei of baboons. Stroke 4: 556-567, (July-Aug) 1973

21. Zatz LM, Alvarey RE: An inaccuracy in computed tomography: Thedependence of CT values. Radiology 124: 91-97, (July) 1977

22. McCullough EC: Factors affecting the use of quantitative informationfrom a CT scanner. Radiology 124: 99-107, (July) 1977

Is There a Real Treatment for Stroke? Clinicaland Statistical Comparison of Different

Treatments in 300 Patients

SERGIO SANTAMBROGIO, M.D., RENATO MARTINOTTI, M.D., FRANCESCO SARDELLA, M.D.,

FERNANDO PORRO, M.D., AND ANTONIO RANDAZZO, M.D.

SUMMARY In the absence of universally accepted criteria for themedical treatment of stroke, we made a rigorously randomized com-parative study of different treatments in 300 patients. One group ofpatients received only a general supportive treatment designed to en-sure adequate supplies of water, electrolytes and calories, pluswhatever was needed to prevent infection and correct extantassociated pathology. Three other groups of patients were treated inthe same way but were also given, respectively, one of the following

medications: Hydergine (Sandoz) (a mixture of three ergotalkaloids), dexamethasone, and mannitol.

No statistically significant difference emerged among any of thetreatment groups and the reference group in terms of objectivetherapeutic results. The authors concluded that, at least with thedosage used in this study, none of the treatments proved more usefulthan conventional supportive therapy in the first 10 days after astroke.

TREATING PATIENTS with stroke is often unrewarding.As in other fields of medicine, the lack of a basic referencetreatment for stroke has generated a multitude of drugtherapies, each being thought of as a cure, only to be soonabandoned. It is likely that some deaths in the first severaldays after a stroke are due to cerebral edema around the in-farcted area rather than to the infarct itself.13 Edemareaches its peak about 3 to 4 days after the stroke, and canbe responsible for transtentorial herniation of the brain,rostrocaudal deterioration, and impaired cerebral bloodflow. Drugs thought to reduce edema reduce intracranialpressure which can indirectly improve brain circulation.Those used in recent years include mannitol,3"7 cor-ticosteroids,2- *•5-"~21 and Hydergine,3'22~32 each with con-flicting reports of clinical effectiveness. We have studiedthese 3 therapies using 1) a large number of patients; 2) ran-dom selection for treatment; 3) each drug alone; 4) resultsfrom reliable clinical parameters; 5) data evaluated by a cor-rectly designed statistical program.

Materials and Methods

We studied 300 patients with a diagnosis of stroke,hospitalized in the Emergency Medicine Division of thePolyclinic Hospital in Milan, during 1974 and 1975. The

From the Emergency Medical Division, Ospedale Maggiore Policlinico,Via Francesco Sforza 35, Milan, Italy 20122.

Reprint requests to Dr. Santambrogio, O.M. Policlinico, Milan, Italy.

mean age for the whole group was 71 years. There were 110men (mean age 69) and 190 women (mean age 73). Of these241 were diagnosed as having an ischemic stroke, 189 in thecarotid artery territory, 40 in the vertebrobasilar territory.In 12 the classification was uncertain. At hospitalization,each patient was assigned randomly to one of the following 4treatment groups:

Group A. These patients received adequate water andelectrolyte replenishment by intravenous infusion for thefirst 48-72 hours, and then a suitable caloric intake, in theform of a special standard diet if necessary, administered bystomach tube. These patients also received suitable an-tibiotics and additional treatments, such as digitalis,diuretics, tracheal aspiration and instillation of mucolyticagents, and bedsore prevention.

Group B. These patients received the same treatment asthose in group A, plus a proprietary mixture ofdihydroergocornine, dihydroergocristine and dihydroergo-criptine, 0.3 mg each (Hydergine) in a daily dosage of 6 am-poules representing 1.8 mg of each alkaloid in the mixture.

Group C. Patients in group C received the same treatmentas group A plus dexamethasone 24 mg daily, given individed doses.

Group D. Patients received the same treatment as groupA plus a solution of 20% hypertonic mannitol given i.v.0.8-0.9 g/kg daily.

In all groups treatment was instituted between 3 and 24hours after onset of symptoms. The period of observationwas 10 days for all 4 groups.

at VA MED CTR BOISE on July 9, 2011http://stroke.ahajournals.org/Downloaded from