22
Monosialoganglioside in Subarachnoid Hemorrhage I. Papo, MD; A. Benedetti, MDf; A. Carteri, MD; G.A. Merli, MD; S. Mingrino, MD; and R. Bruno, MD We studied 119 patients with disturbance of consciousness following subarachnoid hemorrhage, due mostly to verified aneurysm rupture, admitted to five Italian neurosurgical departments over 18 months. Level of consciousness as assessed by score on the Glasgow Coma Scale ranged from 8 to 14 before the beginning of treatment; level of consciousness was assessed again 7, 14, and 21 days later. Patients were randomly allocated to treatment with monosialoganglioside or placebo according to a double-blind experimental design. The two treatment groups were homogeneous at entry with regard to the main clinical parameters. Both groups improved, but the rate and degree of improvement were greater in the monosialogangliosidetreated group. The difference was significant on days 14 (/»=0.04) and 21 (p=0.02). Our results seem to confirm the hypothesis that monosialoganglioside reduces brain edema and provides nonspecific neuronal membrane protection. (Stroke 1991^22:22-26)
S
ubarachnoid or intracerebral hemorrhage due to aneurysm rupture may lead to disturbance of consciousness of different degrees and with multiple mechanisms and is often associated with increased intracranial pressure and reduced cerebral perfusion pressure.1-4 Parenchymal hypoperfusion produces hypoxia with tissue acidosis, alteration of the lactate/pyruvate ratio, and, consequently, further membrane alterations. In this way, a vicious cycle is initiated, producing further cellular membrane damage. The severity and duration of the disturbance of consciousness is closely related to the level of hypoxia and, hence, to the impairment of cellular function. Monosialoganglioside (GM1) is a natural component of neuronal membranes and plays an important role in neuronal transmission. Animal experiments have investigated the therapeutic effect of GM1 on the ionic and enzymatic equilibrium of neuronal cells and its effect on plasma membranes and the damaged blood-brain barrier.5-9 In addition, the role of GM1 in protecting nerve cell membranes against toxic metabolites such as glutamate has been demonstrated.7-8-10-12 Some clinical
From the Neurosurgical Department (I.P.), Ospedale Le Torrette, Ancona; the Neurosurgical Department (A.B.), Ospedale Civile, Vicenza; the Neurosurgical Department (A.C.), Ospedale Civile, Treviso; the NeurosurgicaJ Department (G.A.M.), Ospedale Civile, Modena; the Neurosurgical Department (S.M.), Ospedale Civile, Padova; and the Clinical Research Department (R.B.), Fidia, Abano Terme, Italy. tDeceased. Address for correspondence: Dr. Roberto Bruno, Clinical Research Department, FIDIA S.pA., Via Ponte della Fabbrica, 3/a, 35031 Abano Terrae PD, Italy. Received March 20, 1990; accepted September 26, 1990.
trials have confirmed this hypothesis, demonstrating the effectiveness of GM1 in patients suffering from acute stroke13 and its sequelae.14-15 The aim of our study was to assess the therapeutic effect of GM1 on the time course of disturbances of consciousness, mostly involving a diffuse neuronal impairment of the type mentioned above, after subarachnoid hemorrhage (SAH). Subjects and Methods We recruited 137 consecutive patients with welldocumented SAH admitted to five Italian neurosurgical departments from December 1986 to May 1988. Detailed inclusion criteria were established, and a study protocol was developed and followed. The study was carried out according to a doubleblind, randomized, placebo-controlled design. A randomization list was prepared by an independent statistical group for each department, and no person involved in the trial (including physicians and the monitoring group) had access to the code until the data set was frozen. The treatment solutions were identical in appearance and pH. Only adults with clear evidence of SAH were included, and only patients with a Glasgow Coma Scale (GCS) score of 8-14 were randomized and treated. No account was taken of the necessity for surgical intervention or the time elapsing between SAH and the initiation of treatment. Half of the patients were given GM1 by slow intravenous infusion daily for 7 days as follows: day 1, 500 mg; day 2, 300 mg; day 3, 200 mg; and days 4-7, 100 mg. The large initial dose was given to achieve a steady state as soon as possible. The remaining
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
Papo et al
GMl in SAH
23
TABLE 1. Evaluable Patients in Study of GMl in Subarachnoid Hemorrhage: Contribution of Each Neurosurglcal Department Department Treatment
Treviso
Vicenza
Modena
20 19 39
10 14 24
8 5 13
GMl Placebo Total
Ancona 8 7 15
Padova
Combined
14 14 28
60 59 119
GMl, monosialoganglioside.
patients received placebo (saline) infusions. All patients received routine treatment, both medical and surgical, as needed. The SAH was detected by computed tomography or lumbar puncture. Angiography was performed in most patients. Sequential blood samples were taken for routine hematology and biochemistry studies to assess the safety of the drug. The level of consciousness (as score on the GCS) was assessed, and a complete neurologic examination was performed daily during the first week and then on days 14 and 21. Treatment efficacy was detected as changes from baseline in the GCS score and analyzed by a nonparametric method. Assessments 7, 14, and 21 days after the start of treatment were analyzed taking into account the diminishing group sizes caused by deaths and dropouts. The extended Mantel-Haenszel test16 was used to determine whether differences between the groups were significant. This method does not require any a priori assumption other than randomization. To guarantee comparability of the treatment groups, the GCS scores at baseline were tested for homogeneity. To determine whether the results were consistent among departments, we performed a test for homogeneity.17 The term to test homogeneity O^omog) w a s obtained by subtracting the Mantel-Haenszel statistic, which assesses treatment association, from ;^ota], which assesses total association.18 We compared mortality TABLE 2. Characteristics of Evaluated Patients at Entry Into Study of GMl in Subarachnoid Hemorrhage
GMl («=60)
Age (mean±SD years) Score on Glasgow Coma Scale (mean±SD) Days after SAH
Range
Placebo ("=59)
Treatment GMl
Placebo
Aneurysm Internal carotid artery Anterior communicating artery
43
49
10 14
9 23
Middle cerebral artery Anterior cerebral artery Posterior communicating artery Basilar artery
12 5 2 0 4
13 2 1 1
Finding
Sex (%) Women Men
Mean SD Median
Results The five neurosurgical departments recruited 137 patients. Eighteen were excluded from the analysis because of major protocol violations: two had a predominantly parenchymal hemorrhagic component, 14 had GCS scores outside the required range, and two did not attend the follow-up. Table 1 shows the contribution of each department after exclusion. The clinical parameters and prognostic factors were evenly distributed between the groups (Table 2), and no significant heterogeneity was detected. Angiography demonstrated aneurysms in 92 patients, 90 in the anterior circulation and two in the posterior. Four-vessel examination failed to reveal a source of bleeding in seven patients. In the remaining 20, angiography was incomplete or omitted because of clinical deterioration (Table 3). Of the 119 patients, 78 (39 in the GMl-treated group and 39 in the placebo-treated group) underwent surgery during or immediately before the follow-up period (21 days). Twenty-five patients (11 randomized to the GMl-treated group and 14 randomized to the placebo-treated group) died during the 21 days of observation, and three were lost to follow-up because of transfer to another institution (Table 4). Consequently, analysis was performed on 113 patients at 7 days, on 101 at 14 days, and on 91 at 21 days. TABLE 3. Angiographic Findings in Patients Evaluated in Study of GMl in Subarachnoid Hemorrhage
Treatment Characteristic
rates for the treatment groups using the Mantel-Haenszel test for dichotomous data.19
73.3 26.7 53±11.2
67.8 32.2 55 ±12.6
10.98±1.93
11.03±2.08
6.8 7.9
6.4
43 1-49
6.0 3.0 1-30
GMl, monosialoganglioside; SAH, subarachnoid hemorrhage.
Negative examination Incomplete/omitted examination Total GMl, monosialoganglioside.
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
13 60
3 7 59
24
Stroke Vol 22, No 1, January 1991
TABLE 4.
Patient Disposition During Study of GMl in Subaracbnoid Hemorrhage
Evaluated Died Lost to follow-up
Days 7-14
Days 1-7
Baseline
Days 14-21
GMl
Placebo
Total
GMl
Placebo
Total
GMl
Placebo
Total
GMl
Placebo
Total
60
59
119
57 3 0
56 3 0
113 6 0
52 5 0
49 6 1
101 11 1
47 3 2
44 5 0
91 8 2
GMl, monosialoganglioside.
The difference in GCS score between the groups was not significant on day 7 (p=0.39) but it was on both days 14 (p=0.04) and 21 (p=0.02) (Table 5). As an estimate of the overall treatment effect, a weighted average of the mean treatment differences for all five departments was calculated using weightings consistent with those of the Mantel-Haenszel test. These differences were 1.32 and 1.29, respectively, on days 14 and 21 (Table 5, Figure 1). The 95% confidence intervals of the treatment differences for each department individually and for all five departments combined are shown for days 14 and 21 in Figure 2. Results of the treatment x department heterogeneity test were not significant at any time (Table 5). There was no significant difference in mortality rates between the groups (11 of 60 [18.3%] in the GMl-treated group and 14 of 59 [23.7%] in the placebo-treated group). One patient receiving an antiepiJeptic drug (in the GMl-treated group) developed a diffuse erythema. No differences were observed between the groups in terms of laboratory data, complications, and new or adverse events (Table 6). Discussion The results of this double-blind trial show a significant improvement in the level of consciousness as measured by score on the GCS for patients treated with GMl compared with placebo. A comparable amount of improvement has been considered clinically relevant in similar studies of naloxone20 and nimodipine21 in patients with SAH. Improvement in the level of consciousness in our patients given GMl was most evident 1 and 2 weeks after the end of treatment, indicating that the pharmacologic effect of the drug induces a more rapid and, probably, more complete functional recovery in subjects with diffuse reversible neuronal impairment. TABLE 5.
Based on the decline in mean GCS score in the placebo-treated group from day 7 to day 14 and the continuous increase in mean GCS score in the GMltreated group, it is possible that GMl prevents or moderates vasospasm in some patients. By means of cytotoxic substances such as arachidonic acid and prostaglandins, SAH can cause many different complications, including vasospasm and both cytotoxic and vasogenic cerebral edema. These complications are often associated with a reduction in cerebral perfusion pressure, and the resulting hypoxia appears to be responsible for disturbances of consciousness, particularly their duration and severity. Substantial experimental data show that GMl protects cell membranes from injury.5-8-22 It is known that exogenous GMl is incorporated into cell membranes and becomes pharmacologically active in protecting the membrane from endogenous and exogenous toxic insults. It also has been demonstrated that GMl has a particular affinity for "incorporation" into neuronal tissue at the locus of injury.12^23-25 GMl can stimulate the activity of NaJC-ATPase and adenylcyclase.9-24"26 Karpiak and Mahadik25 also suggest that GMl protects the membrane from hydrolysis, phospholipase activation, and the toxic effect of free radicals. In addition, GMl reduces membrane permeability to water, in agreement with data obtained with negatively charged phospholipids.27 GMl also neutralizes the toxic effect of glutamate on the membrane and diminishes the efflux of intracellular K+ and the influx of Caz+.s." According to Karpiak and Mahadik,25 GMl limits cerebral edema by modulating free radical concentration, lipid hydrolysis, phospholipase activation, or membrane injury caused by arachidonic acid. From the pathophysiologic and morphologic standpoints, the most relevant effect of GMl is the reduction of edema, both cytotoxic and vasogenic. This phenom-
Comparisons of Treatment Groups on Days 7, 14, and 21 in Study of GMl in Subaracbnoid Hemorrhage Interaction
Treatment GMl
Placebo
Difference
Day
Mean
SEM
Mean
SEM
Mean
SEM
7 14 21
12.53 13.59 14.32
036 0.34 0.20
12.03 12.28 13.02
0.46 0.54 0.49
0.50 1.32 1.29
0.58 0.64 0.53
Treatment x Department P* 0J9 0.04 0.02
Test
df
9.09 5.60 6.69
10 10 9
P 0.52 0.84 0.67
GMl, monosialoganglioside. Descriptive statistics for all five departments combined are based on weighted averages reflecting adjustment for departments through relative contributions to Mantel-Haenszel procedure. 'Based on extended Mantel-Haenszel procedures.
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
Papo et al
GMl in SAH
25
1514-
3
O 4
11
COMBINED
lO-'-r
14
21
-3
-2
-1
15 14S
§
0
13
'
1
U 12 O
2 3 4
lO-'-r
14
21
COMBINED
15 -
14-
U
3
-
2
-
1
0
1
2
3
4
5
6
7
FIGURE 2. Graphs of 95% confidence intervals for treatment differences for each department individually (0, Treviso; 1, Vicenza; 2, Modena; 3, Ancona; and 4, Padova) and for all five departments combined at (top) 14 days and (bottom) 21 days after start of study of monosialoganglioside in subarachnoid hemorrhage.
12
U
11
10
Appendix 1. List of Participants 1
days
14
21
FIGURE 1. Graphs of improvement in score on Glasgow Coma Scale (GCS) in patients with subarachnoid hemorrhage treated with monosialoganglioside (•) or placebo (o). Top: 113 patients at 7 days; center: 101 patients at 14 days; bottom: 91 patients at 21 days.
enon has been experimentally demonstrated in both traumatic and ischemic edema,24-26'27 but the mechanisms whereby GMl plays a role in reducing the duration of functional deficit, as well as the effect of GMl on vasospasm or other pathologic mechanisms, must still be clarified. We believe that SAH causes diffuse hypoxic damage leading to impairment of consciousness due to multiple functional disturbances. The beneficial effect of GMl observed in our patients was probably due to its membrane-protecting effect, resulting in a reduction in primary and secondary cell damage and brain edema. The precise mechanism of this effect is not known, but it accounts for the quicker recovery of neuronal function in GMl-treated patients. A larger trial with a longer follow-up is planned to confirm and validate the benefit of GMl in this type of patient.
Clinical Coordinating Center: I. Papo, MD (chief), and M.A. Recchioni, MD —Ospedale Le Torrette di Ancona. Participating centers in alphabetical order of place: G.A. Merli, MD (chief), and L. Corradini, TABLE 6. Number of Complications and New or Adverse Events in Study of GMl In Subarachnoid Hemorrhage Complication/event Rebleeding Hydrocephalus Headache Nausea/vertigo Seizures Pneumonia Fever Melena Arrhythmia Abnormal liver function Abnormalities in laboratory tests Diffuse erythema GMl, monosialoganglioside.
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
Treatment Placebo GMl 4 2 4 0 1
3 3 5 2
6
2 4
5 1
4 1
2
4
24
27
16 1
22 0
26
Stroke
Vol 22, No 1, January 1991
MD —Ospedale Civile di Modena, Neurosurgical Department; S. Mingrino, MD (chief), and S. Ferla, MD — Ospedale Civile di Padova, Neurosurgical Department; A. Carted, MD (chief), and D. Billed, MD — Ospedale Civile di Treviso, Divisione di Neurochirurgia; A. Benedetti, MD (chief), L. Volpin, MD, and U. Fornezza, MD —Ospedale Civile di Vicenza, Divisione di Neurochirurgia; R. Bruno, MD, and A. Bacchieri, PhD —Clinical Research Department, FIDIA, Abano Terme. Acknowledgments
We gratefully acknowledge the cooperation of Prof. A. Migliore and his group and Dr. G. Curatola for biochemical consulting help in preparing the manuscript. References 1. Kassel NF, Tomer JC, Jane JA: Studio Cooperative) Internazionale sul Timing Chirurgico degli Aneurismi Cerebrali, in Da Pian R, Pasqualin A, Scienza R (eds): Aneurismi e Angiomi Cerebrali: Principi di Trattamento. Verona, Libreria Cortina, 1986, pp 141-146 2. Doczi T: The pathogenetic and prognostic significance of blood-brain barrier damage at the acute stage of aneurysmal subarachnoid haemorrhage: Clinical and experimental studies. Ada Neurochir (Wien) 1985;77:110-132 3. Grote E, Hassler W: The critical first minutes after subarachnoid hemorrage. Neurvsurgery 1988;22:654-661 4. Rothman SM, Olney JW: Glutamate and the pathophysiology of hypoxic-ischemic brain damage. Ann Neurol 1986;19: 105-111 5. Skaper SD, Katoh-Semba R, Varon S: GM1 ganglioside accelerates neurite outgrowth from primary peripheral and central neurons under selected culture conditions. Dev Brain Res 1985;23:19-26 6. Leon A, Dal Toso R,- Presti D, Benvegnu D, Facci L, Kirschner G, Tettamanti G, Toffano G: Development and survival of neurons in dissociated fetal mesencephalic serumfree cell cultures: II. Modulatory effects of gangliosides. / NeurosciRes 1988;8:746-753 7. Vaccarino F, Guidotti A, Costa E: Ganglioside inhibition of glutamate mediated protein kinase C translocation in primary cultures of cerebellar neurons. Proc Natl Acad Sci USA 1987;84:8707-8711 8. Favaron M, Manev H, Alho H, Bertolino M, Ferret B, Guidotti A, Costa E: Gangliosides prevent glutamate and kainate neurotoxicity in primary neuronal cultures of neonatal rat cerebellum and cortex. Proc Natl Acad Sci USA 1988;85: 7351-7355 9. Li YS, Mahadik SP, Rapport MM, Karpiak SE: Acute effects of GM1 ganglioside: Reduction in both behavioral asymmetry and loss of (Na + ,K + ) ATPase after nigrostriatal transection. Brain Res 1986;377:292-297 10. Toffano G, Benvegnu D, Bonetti AC, Facci L, Leon A, Orlando P, Ghidoni R, Tettamanti G: Interactions of GM1 ganglioside with crude rat brain neuronal membranes. J Neurochem 1980^5:861-866 11. Leon A, Tettamanti G, Toffano G: Changes in functional properties of neuron membranes by insertion of exogenous
12. 13.
14. 15.
16. 17. 18. 19. 20. 21.
22.
23. 24. 25. 26.
27.
ganglioside, in Rapport MM, Gorio A (eds): Gangliosides in Neurological and Neuromuscular Function, Development and Repair. New York, Raven Press, Publishers, 1981, pp 45-54 Karpiak SE, Li YS, Mahadik SP: Gangliosides (GM1 and AGF2) reduce mortality due to ischemia: Protection of membrane function. Stroke 1987;18:184-187 Argentino C, Sacchetti ML, Toni D, Savoini G, D'Arcangelo E, Erminio F, Federico F, Ferro Milone F, Gallai V, Gambi D, Mamoli A, Ottonello GA, Ponari O, Rebucci G, Senin U, Fieschi C: GM1 ganglioside therapy in acute ischemic stroke. Stroke 1989;20:l 143-1149 Bassi S, Albizzati MG, Sbacchi M, Frattola L, Massarotti M: Double-blind evaluation of monosialoganglioside (GM1) therapy in stroke. / Neurosci Res 1984;12:493-498 Battistin L, Cesari A, Galligioni F, Marin G, Massarotti M, Paccagnella D, Pellegrini A, Testa G, Tonin P: Effects of GMl-ganglioside in cerebrovascular diseases: A double blind trial in 40 cases. Eur Neurol 1985;24:343-351 Mantel N: Chi-square tests with one degree of freedom: Extensions of the Mantel-Haenszel procedure. J Am StatAssoc 1963^8:690-700 Fleiss JL: Analysis of data from multiclinic trials. Controlled Clin Trials 1986;7:267-275 Koch GG, Impey PB: Randomization model methods, in Analysis of Categorical Data. SMS. Montreal, Les Presses de l'Universite de Montreal, 1985, pp 224-261 Mantel N, Haenszel W: Statistical aspects of the analysis of data from retrospective studies of disease. / Natl Cancer Inst 1959;22:719-748 Bell BA, Miller JD, Neto NGF, O'Neill P, Laughton LM: Effect of naloxone on deficits after aneurysmal subarachnoid hemorrhage. Neurosurgery 1985;16:498-500 Pickard JD, Murray GD, Illingworth R, Shaw MDM, Teasdale GM, Foy PM, Humphrey PRD, Lang DA, Nelson R, Richards P, Sinar J, Bailey S, Skene A: Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British Aneurysm Nimodipine Trial. BrMedJ 1989;298: 636-642 Dal Toso R, Skaper SD, Ferrari G, Vantini G, Toffano G, Leon A: Ganglioside involvement in membrane-mediated transfer of trophic information: Relationship to GM1 effects following CNS injury, in Stein DG, Sabel BA (eds): Pharmacological Approaches to the Treatment of Brain and Spinal Cord Injury. New York, Plenum Press, 1988, pp 143-165 Gorio A, Carmignoto G, Facci L, Finesso M: Motor nerve sprouting induced by ganglioside treatment. Brain Res 1980; 197:236-241 Karpiak SE, Li YS, Mahadik SP: Ganglioside treatment: Reduction of CNS injury and facilitation of functional recovery. Brain Inj 1987;l:161-170 Karpiak SE, Mahadik SP: Reduction of cerebral oedema with GM1 ganglioside. J Neurosci Res 1984;12:485-492 Komatsumoto S, Greenberg JH, Hickey WF, Reivich M: Effect of the ganglioside GM1 on neurologic function, electroencephalogram amplitude, and histology in chronic middle cerebral artery occlusion in cats. Stroke 1988; 19:1027-1035 Cahn J, Borzeix MG, Toffano G: Effect of GM1 ganglioside and of its inner ester derivate in a model of transient cerebral ischemia in the rat, in Tettamanti G, Ledeen RW, Sandhoff K, Nagai Y, Toffano G (eds): Gangliosides and Neuronal Plasticity. Fidia Res Ser 1986;6:345-443
KEY WORDS • gangliosides • subarachnoid hemorrhage
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
Monosialoganglioside in subarachnoid hemorrhage. I Papo, A Benedetti, A Carteri, G A Merli, S Mingrino and R Bruno Stroke. 1991;22:22-26 doi: 10.1161/01.STR.22.1.22 Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1991 American Heart Association, Inc. All rights reserved. Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://stroke.ahajournals.org/content/22/1/22
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Stroke is online at: http://stroke.ahajournals.org//subscriptions/
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
Monosialoganglioside in Subarachnoid Hemorrhage I. Papo, MD; A. Benedetti, MDf; A. Carteri, MD; G.A. Merli, MD; S. Mingrino, MD; and R. Bruno, MD We studied 119 patients with disturbance of consciousness following subarachnoid hemorrhage, due mostly to verified aneurysm rupture, admitted to five Italian neurosurgical departments over 18 months. Level of consciousness as assessed by score on the Glasgow Coma Scale ranged from 8 to 14 before the beginning of treatment; level of consciousness was assessed again 7, 14, and 21 days later. Patients were randomly allocated to treatment with monosialoganglioside or placebo according to a double-blind experimental design. The two treatment groups were homogeneous at entry with regard to the main clinical parameters. Both groups improved, but the rate and degree of improvement were greater in the monosialogangliosidetreated group. The difference was significant on days 14 (/»=0.04) and 21 (p=0.02). Our results seem to confirm the hypothesis that monosialoganglioside reduces brain edema and provides nonspecific neuronal membrane protection. (Stroke 1991^22:22-26)
S
ubarachnoid or intracerebral hemorrhage due to aneurysm rupture may lead to disturbance of consciousness of different degrees and with multiple mechanisms and is often associated with increased intracranial pressure and reduced cerebral perfusion pressure.1-4 Parenchymal hypoperfusion produces hypoxia with tissue acidosis, alteration of the lactate/pyruvate ratio, and, consequently, further membrane alterations. In this way, a vicious cycle is initiated, producing further cellular membrane damage. The severity and duration of the disturbance of consciousness is closely related to the level of hypoxia and, hence, to the impairment of cellular function. Monosialoganglioside (GM1) is a natural component of neuronal membranes and plays an important role in neuronal transmission. Animal experiments have investigated the therapeutic effect of GM1 on the ionic and enzymatic equilibrium of neuronal cells and its effect on plasma membranes and the damaged blood-brain barrier.5-9 In addition, the role of GM1 in protecting nerve cell membranes against toxic metabolites such as glutamate has been demonstrated.7-8-10-12 Some clinical
From the Neurosurgical Department (I.P.), Ospedale Le Torrette, Ancona; the Neurosurgical Department (A.B.), Ospedale Civile, Vicenza; the Neurosurgical Department (A.C.), Ospedale Civile, Treviso; the NeurosurgicaJ Department (G.A.M.), Ospedale Civile, Modena; the Neurosurgical Department (S.M.), Ospedale Civile, Padova; and the Clinical Research Department (R.B.), Fidia, Abano Terme, Italy. tDeceased. Address for correspondence: Dr. Roberto Bruno, Clinical Research Department, FIDIA S.pA., Via Ponte della Fabbrica, 3/a, 35031 Abano Terrae PD, Italy. Received March 20, 1990; accepted September 26, 1990.
trials have confirmed this hypothesis, demonstrating the effectiveness of GM1 in patients suffering from acute stroke13 and its sequelae.14-15 The aim of our study was to assess the therapeutic effect of GM1 on the time course of disturbances of consciousness, mostly involving a diffuse neuronal impairment of the type mentioned above, after subarachnoid hemorrhage (SAH). Subjects and Methods We recruited 137 consecutive patients with welldocumented SAH admitted to five Italian neurosurgical departments from December 1986 to May 1988. Detailed inclusion criteria were established, and a study protocol was developed and followed. The study was carried out according to a doubleblind, randomized, placebo-controlled design. A randomization list was prepared by an independent statistical group for each department, and no person involved in the trial (including physicians and the monitoring group) had access to the code until the data set was frozen. The treatment solutions were identical in appearance and pH. Only adults with clear evidence of SAH were included, and only patients with a Glasgow Coma Scale (GCS) score of 8-14 were randomized and treated. No account was taken of the necessity for surgical intervention or the time elapsing between SAH and the initiation of treatment. Half of the patients were given GM1 by slow intravenous infusion daily for 7 days as follows: day 1, 500 mg; day 2, 300 mg; day 3, 200 mg; and days 4-7, 100 mg. The large initial dose was given to achieve a steady state as soon as possible. The remaining
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
Papo et al
GMl in SAH
23
TABLE 1. Evaluable Patients in Study of GMl in Subarachnoid Hemorrhage: Contribution of Each Neurosurglcal Department Department Treatment
Treviso
Vicenza
Modena
20 19 39
10 14 24
8 5 13
GMl Placebo Total
Ancona 8 7 15
Padova
Combined
14 14 28
60 59 119
GMl, monosialoganglioside.
patients received placebo (saline) infusions. All patients received routine treatment, both medical and surgical, as needed. The SAH was detected by computed tomography or lumbar puncture. Angiography was performed in most patients. Sequential blood samples were taken for routine hematology and biochemistry studies to assess the safety of the drug. The level of consciousness (as score on the GCS) was assessed, and a complete neurologic examination was performed daily during the first week and then on days 14 and 21. Treatment efficacy was detected as changes from baseline in the GCS score and analyzed by a nonparametric method. Assessments 7, 14, and 21 days after the start of treatment were analyzed taking into account the diminishing group sizes caused by deaths and dropouts. The extended Mantel-Haenszel test16 was used to determine whether differences between the groups were significant. This method does not require any a priori assumption other than randomization. To guarantee comparability of the treatment groups, the GCS scores at baseline were tested for homogeneity. To determine whether the results were consistent among departments, we performed a test for homogeneity.17 The term to test homogeneity O^omog) w a s obtained by subtracting the Mantel-Haenszel statistic, which assesses treatment association, from ;^ota], which assesses total association.18 We compared mortality TABLE 2. Characteristics of Evaluated Patients at Entry Into Study of GMl in Subarachnoid Hemorrhage
GMl («=60)
Age (mean±SD years) Score on Glasgow Coma Scale (mean±SD) Days after SAH
Range
Placebo ("=59)
Treatment GMl
Placebo
Aneurysm Internal carotid artery Anterior communicating artery
43
49
10 14
9 23
Middle cerebral artery Anterior cerebral artery Posterior communicating artery Basilar artery
12 5 2 0 4
13 2 1 1
Finding
Sex (%) Women Men
Mean SD Median
Results The five neurosurgical departments recruited 137 patients. Eighteen were excluded from the analysis because of major protocol violations: two had a predominantly parenchymal hemorrhagic component, 14 had GCS scores outside the required range, and two did not attend the follow-up. Table 1 shows the contribution of each department after exclusion. The clinical parameters and prognostic factors were evenly distributed between the groups (Table 2), and no significant heterogeneity was detected. Angiography demonstrated aneurysms in 92 patients, 90 in the anterior circulation and two in the posterior. Four-vessel examination failed to reveal a source of bleeding in seven patients. In the remaining 20, angiography was incomplete or omitted because of clinical deterioration (Table 3). Of the 119 patients, 78 (39 in the GMl-treated group and 39 in the placebo-treated group) underwent surgery during or immediately before the follow-up period (21 days). Twenty-five patients (11 randomized to the GMl-treated group and 14 randomized to the placebo-treated group) died during the 21 days of observation, and three were lost to follow-up because of transfer to another institution (Table 4). Consequently, analysis was performed on 113 patients at 7 days, on 101 at 14 days, and on 91 at 21 days. TABLE 3. Angiographic Findings in Patients Evaluated in Study of GMl in Subarachnoid Hemorrhage
Treatment Characteristic
rates for the treatment groups using the Mantel-Haenszel test for dichotomous data.19
73.3 26.7 53±11.2
67.8 32.2 55 ±12.6
10.98±1.93
11.03±2.08
6.8 7.9
6.4
43 1-49
6.0 3.0 1-30
GMl, monosialoganglioside; SAH, subarachnoid hemorrhage.
Negative examination Incomplete/omitted examination Total GMl, monosialoganglioside.
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
13 60
3 7 59
24
Stroke Vol 22, No 1, January 1991
TABLE 4.
Patient Disposition During Study of GMl in Subaracbnoid Hemorrhage
Evaluated Died Lost to follow-up
Days 7-14
Days 1-7
Baseline
Days 14-21
GMl
Placebo
Total
GMl
Placebo
Total
GMl
Placebo
Total
GMl
Placebo
Total
60
59
119
57 3 0
56 3 0
113 6 0
52 5 0
49 6 1
101 11 1
47 3 2
44 5 0
91 8 2
GMl, monosialoganglioside.
The difference in GCS score between the groups was not significant on day 7 (p=0.39) but it was on both days 14 (p=0.04) and 21 (p=0.02) (Table 5). As an estimate of the overall treatment effect, a weighted average of the mean treatment differences for all five departments was calculated using weightings consistent with those of the Mantel-Haenszel test. These differences were 1.32 and 1.29, respectively, on days 14 and 21 (Table 5, Figure 1). The 95% confidence intervals of the treatment differences for each department individually and for all five departments combined are shown for days 14 and 21 in Figure 2. Results of the treatment x department heterogeneity test were not significant at any time (Table 5). There was no significant difference in mortality rates between the groups (11 of 60 [18.3%] in the GMl-treated group and 14 of 59 [23.7%] in the placebo-treated group). One patient receiving an antiepiJeptic drug (in the GMl-treated group) developed a diffuse erythema. No differences were observed between the groups in terms of laboratory data, complications, and new or adverse events (Table 6). Discussion The results of this double-blind trial show a significant improvement in the level of consciousness as measured by score on the GCS for patients treated with GMl compared with placebo. A comparable amount of improvement has been considered clinically relevant in similar studies of naloxone20 and nimodipine21 in patients with SAH. Improvement in the level of consciousness in our patients given GMl was most evident 1 and 2 weeks after the end of treatment, indicating that the pharmacologic effect of the drug induces a more rapid and, probably, more complete functional recovery in subjects with diffuse reversible neuronal impairment. TABLE 5.
Based on the decline in mean GCS score in the placebo-treated group from day 7 to day 14 and the continuous increase in mean GCS score in the GMltreated group, it is possible that GMl prevents or moderates vasospasm in some patients. By means of cytotoxic substances such as arachidonic acid and prostaglandins, SAH can cause many different complications, including vasospasm and both cytotoxic and vasogenic cerebral edema. These complications are often associated with a reduction in cerebral perfusion pressure, and the resulting hypoxia appears to be responsible for disturbances of consciousness, particularly their duration and severity. Substantial experimental data show that GMl protects cell membranes from injury.5-8-22 It is known that exogenous GMl is incorporated into cell membranes and becomes pharmacologically active in protecting the membrane from endogenous and exogenous toxic insults. It also has been demonstrated that GMl has a particular affinity for "incorporation" into neuronal tissue at the locus of injury.12^23-25 GMl can stimulate the activity of NaJC-ATPase and adenylcyclase.9-24"26 Karpiak and Mahadik25 also suggest that GMl protects the membrane from hydrolysis, phospholipase activation, and the toxic effect of free radicals. In addition, GMl reduces membrane permeability to water, in agreement with data obtained with negatively charged phospholipids.27 GMl also neutralizes the toxic effect of glutamate on the membrane and diminishes the efflux of intracellular K+ and the influx of Caz+.s." According to Karpiak and Mahadik,25 GMl limits cerebral edema by modulating free radical concentration, lipid hydrolysis, phospholipase activation, or membrane injury caused by arachidonic acid. From the pathophysiologic and morphologic standpoints, the most relevant effect of GMl is the reduction of edema, both cytotoxic and vasogenic. This phenom-
Comparisons of Treatment Groups on Days 7, 14, and 21 in Study of GMl in Subaracbnoid Hemorrhage Interaction
Treatment GMl
Placebo
Difference
Day
Mean
SEM
Mean
SEM
Mean
SEM
7 14 21
12.53 13.59 14.32
036 0.34 0.20
12.03 12.28 13.02
0.46 0.54 0.49
0.50 1.32 1.29
0.58 0.64 0.53
Treatment x Department P* 0J9 0.04 0.02
Test
df
9.09 5.60 6.69
10 10 9
P 0.52 0.84 0.67
GMl, monosialoganglioside. Descriptive statistics for all five departments combined are based on weighted averages reflecting adjustment for departments through relative contributions to Mantel-Haenszel procedure. 'Based on extended Mantel-Haenszel procedures.
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
Papo et al
GMl in SAH
25
1514-
3
O 4
11
COMBINED
lO-'-r
14
21
-3
-2
-1
15 14S
§
0
13
'
1
U 12 O
2 3 4
lO-'-r
14
21
COMBINED
15 -
14-
U
3
-
2
-
1
0
1
2
3
4
5
6
7
FIGURE 2. Graphs of 95% confidence intervals for treatment differences for each department individually (0, Treviso; 1, Vicenza; 2, Modena; 3, Ancona; and 4, Padova) and for all five departments combined at (top) 14 days and (bottom) 21 days after start of study of monosialoganglioside in subarachnoid hemorrhage.
12
U
11
10
Appendix 1. List of Participants 1
days
14
21
FIGURE 1. Graphs of improvement in score on Glasgow Coma Scale (GCS) in patients with subarachnoid hemorrhage treated with monosialoganglioside (•) or placebo (o). Top: 113 patients at 7 days; center: 101 patients at 14 days; bottom: 91 patients at 21 days.
enon has been experimentally demonstrated in both traumatic and ischemic edema,24-26'27 but the mechanisms whereby GMl plays a role in reducing the duration of functional deficit, as well as the effect of GMl on vasospasm or other pathologic mechanisms, must still be clarified. We believe that SAH causes diffuse hypoxic damage leading to impairment of consciousness due to multiple functional disturbances. The beneficial effect of GMl observed in our patients was probably due to its membrane-protecting effect, resulting in a reduction in primary and secondary cell damage and brain edema. The precise mechanism of this effect is not known, but it accounts for the quicker recovery of neuronal function in GMl-treated patients. A larger trial with a longer follow-up is planned to confirm and validate the benefit of GMl in this type of patient.
Clinical Coordinating Center: I. Papo, MD (chief), and M.A. Recchioni, MD —Ospedale Le Torrette di Ancona. Participating centers in alphabetical order of place: G.A. Merli, MD (chief), and L. Corradini, TABLE 6. Number of Complications and New or Adverse Events in Study of GMl In Subarachnoid Hemorrhage Complication/event Rebleeding Hydrocephalus Headache Nausea/vertigo Seizures Pneumonia Fever Melena Arrhythmia Abnormal liver function Abnormalities in laboratory tests Diffuse erythema GMl, monosialoganglioside.
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
Treatment Placebo GMl 4 2 4 0 1
3 3 5 2
6
2 4
5 1
4 1
2
4
24
27
16 1
22 0
26
Stroke
Vol 22, No 1, January 1991
MD —Ospedale Civile di Modena, Neurosurgical Department; S. Mingrino, MD (chief), and S. Ferla, MD — Ospedale Civile di Padova, Neurosurgical Department; A. Carted, MD (chief), and D. Billed, MD — Ospedale Civile di Treviso, Divisione di Neurochirurgia; A. Benedetti, MD (chief), L. Volpin, MD, and U. Fornezza, MD —Ospedale Civile di Vicenza, Divisione di Neurochirurgia; R. Bruno, MD, and A. Bacchieri, PhD —Clinical Research Department, FIDIA, Abano Terme. Acknowledgments
We gratefully acknowledge the cooperation of Prof. A. Migliore and his group and Dr. G. Curatola for biochemical consulting help in preparing the manuscript. References 1. Kassel NF, Tomer JC, Jane JA: Studio Cooperative) Internazionale sul Timing Chirurgico degli Aneurismi Cerebrali, in Da Pian R, Pasqualin A, Scienza R (eds): Aneurismi e Angiomi Cerebrali: Principi di Trattamento. Verona, Libreria Cortina, 1986, pp 141-146 2. Doczi T: The pathogenetic and prognostic significance of blood-brain barrier damage at the acute stage of aneurysmal subarachnoid haemorrhage: Clinical and experimental studies. Ada Neurochir (Wien) 1985;77:110-132 3. Grote E, Hassler W: The critical first minutes after subarachnoid hemorrage. Neurvsurgery 1988;22:654-661 4. Rothman SM, Olney JW: Glutamate and the pathophysiology of hypoxic-ischemic brain damage. Ann Neurol 1986;19: 105-111 5. Skaper SD, Katoh-Semba R, Varon S: GM1 ganglioside accelerates neurite outgrowth from primary peripheral and central neurons under selected culture conditions. Dev Brain Res 1985;23:19-26 6. Leon A, Dal Toso R,- Presti D, Benvegnu D, Facci L, Kirschner G, Tettamanti G, Toffano G: Development and survival of neurons in dissociated fetal mesencephalic serumfree cell cultures: II. Modulatory effects of gangliosides. / NeurosciRes 1988;8:746-753 7. Vaccarino F, Guidotti A, Costa E: Ganglioside inhibition of glutamate mediated protein kinase C translocation in primary cultures of cerebellar neurons. Proc Natl Acad Sci USA 1987;84:8707-8711 8. Favaron M, Manev H, Alho H, Bertolino M, Ferret B, Guidotti A, Costa E: Gangliosides prevent glutamate and kainate neurotoxicity in primary neuronal cultures of neonatal rat cerebellum and cortex. Proc Natl Acad Sci USA 1988;85: 7351-7355 9. Li YS, Mahadik SP, Rapport MM, Karpiak SE: Acute effects of GM1 ganglioside: Reduction in both behavioral asymmetry and loss of (Na + ,K + ) ATPase after nigrostriatal transection. Brain Res 1986;377:292-297 10. Toffano G, Benvegnu D, Bonetti AC, Facci L, Leon A, Orlando P, Ghidoni R, Tettamanti G: Interactions of GM1 ganglioside with crude rat brain neuronal membranes. J Neurochem 1980^5:861-866 11. Leon A, Tettamanti G, Toffano G: Changes in functional properties of neuron membranes by insertion of exogenous
12. 13.
14. 15.
16. 17. 18. 19. 20. 21.
22.
23. 24. 25. 26.
27.
ganglioside, in Rapport MM, Gorio A (eds): Gangliosides in Neurological and Neuromuscular Function, Development and Repair. New York, Raven Press, Publishers, 1981, pp 45-54 Karpiak SE, Li YS, Mahadik SP: Gangliosides (GM1 and AGF2) reduce mortality due to ischemia: Protection of membrane function. Stroke 1987;18:184-187 Argentino C, Sacchetti ML, Toni D, Savoini G, D'Arcangelo E, Erminio F, Federico F, Ferro Milone F, Gallai V, Gambi D, Mamoli A, Ottonello GA, Ponari O, Rebucci G, Senin U, Fieschi C: GM1 ganglioside therapy in acute ischemic stroke. Stroke 1989;20:l 143-1149 Bassi S, Albizzati MG, Sbacchi M, Frattola L, Massarotti M: Double-blind evaluation of monosialoganglioside (GM1) therapy in stroke. / Neurosci Res 1984;12:493-498 Battistin L, Cesari A, Galligioni F, Marin G, Massarotti M, Paccagnella D, Pellegrini A, Testa G, Tonin P: Effects of GMl-ganglioside in cerebrovascular diseases: A double blind trial in 40 cases. Eur Neurol 1985;24:343-351 Mantel N: Chi-square tests with one degree of freedom: Extensions of the Mantel-Haenszel procedure. J Am StatAssoc 1963^8:690-700 Fleiss JL: Analysis of data from multiclinic trials. Controlled Clin Trials 1986;7:267-275 Koch GG, Impey PB: Randomization model methods, in Analysis of Categorical Data. SMS. Montreal, Les Presses de l'Universite de Montreal, 1985, pp 224-261 Mantel N, Haenszel W: Statistical aspects of the analysis of data from retrospective studies of disease. / Natl Cancer Inst 1959;22:719-748 Bell BA, Miller JD, Neto NGF, O'Neill P, Laughton LM: Effect of naloxone on deficits after aneurysmal subarachnoid hemorrhage. Neurosurgery 1985;16:498-500 Pickard JD, Murray GD, Illingworth R, Shaw MDM, Teasdale GM, Foy PM, Humphrey PRD, Lang DA, Nelson R, Richards P, Sinar J, Bailey S, Skene A: Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British Aneurysm Nimodipine Trial. BrMedJ 1989;298: 636-642 Dal Toso R, Skaper SD, Ferrari G, Vantini G, Toffano G, Leon A: Ganglioside involvement in membrane-mediated transfer of trophic information: Relationship to GM1 effects following CNS injury, in Stein DG, Sabel BA (eds): Pharmacological Approaches to the Treatment of Brain and Spinal Cord Injury. New York, Plenum Press, 1988, pp 143-165 Gorio A, Carmignoto G, Facci L, Finesso M: Motor nerve sprouting induced by ganglioside treatment. Brain Res 1980; 197:236-241 Karpiak SE, Li YS, Mahadik SP: Ganglioside treatment: Reduction of CNS injury and facilitation of functional recovery. Brain Inj 1987;l:161-170 Karpiak SE, Mahadik SP: Reduction of cerebral oedema with GM1 ganglioside. J Neurosci Res 1984;12:485-492 Komatsumoto S, Greenberg JH, Hickey WF, Reivich M: Effect of the ganglioside GM1 on neurologic function, electroencephalogram amplitude, and histology in chronic middle cerebral artery occlusion in cats. Stroke 1988; 19:1027-1035 Cahn J, Borzeix MG, Toffano G: Effect of GM1 ganglioside and of its inner ester derivate in a model of transient cerebral ischemia in the rat, in Tettamanti G, Ledeen RW, Sandhoff K, Nagai Y, Toffano G (eds): Gangliosides and Neuronal Plasticity. Fidia Res Ser 1986;6:345-443
KEY WORDS • gangliosides • subarachnoid hemorrhage
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
Monosialoganglioside in subarachnoid hemorrhage. I Papo, A Benedetti, A Carteri, G A Merli, S Mingrino and R Bruno Stroke. 1991;22:22-26 doi: 10.1161/01.STR.22.1.22 Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1991 American Heart Association, Inc. All rights reserved. Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://stroke.ahajournals.org/content/22/1/22
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Stroke is online at: http://stroke.ahajournals.org//subscriptions/
Downloaded from http://stroke.ahajournals.org/ by guest on June 24, 2016
