- Email: [email protected]
Medical treatment of acute ischaemic stroke
537
Medical treatment of acute ischaemic stroke
The aim of early treatment for acute ischaemic stroke is to reverse or limit the degree of brain dysfunction. We will review medical treatments that are given as soon as possible (within hours, rather than days or weeks) after the onset of symptoms; we will not discuss treatments initiated after the acute phase (eg, oral anticoagulants) or those given for long-term secondary prevention. Our comments are based on a systematic search for completed randomised clinical trials of medical treatments for acute stroke, over 100 of which have been identified so far. TableI lists the main therapeutic strategies together with the number of completed randomised controlled clinical trials of each strategy.
The ideal medical treatment There
main types of medical treatment for ischaemic stroke: (a) simple, cheap, patients widely practicable treatments that can be given with little or no monitoring to virtually all patients; and (b) complex treatments that are more costly and difficult to administer (and monitor) and can be given only. to a few carefully selected patients. Examples of the former category are aspirin, fixed-dose subcutaneous heparin, oral calcium antagonists, and corticosteroids. Examples of the latter are fibrinolytic drugs, isovolaemic haemodilution, intravenous naftidrofuryl, hyperbaric oxygen, and intravenous inhibitors of excitatory aminoacids. Research, both in laboratory animals and in randomised clinical trials, has focused more on the complex high technology approach. The potential public health impact of both types of treatments is shown in table n. are two
with
acute
The example given in table II highlights the dilemma in research and practice: highly effective but very specific treatments may provide striking benefits for a few selected individuals whereas simpler treatments with more modest effects given to virtually all stroke patients will probably have a greater impact on public health.
Importance of large clinical trials If a medical therapy is to be used widely in patients with stroke, randomised trials should have provided clear evidence that such treatment reduces mortality. The risk of early death in unselected groups of patients with ischaemic stroke is low (ie, 10% after 30 days in the Oxfordshire Community Stroke Project1 and 12% after 3 weeks in untreated controls in the TRUST study2). If a simple, widely practicable treatment such as an oral calcium antagonist or oral aspirin reduced early mortality by only one sixth from 12% to 10%, this would represent 20 deaths avoided per 1000 patients treated. This reduction would be worthwhile, especially if it were accompanied by a similar proportional improvement in survival free of disability. Reliable detection of such moderate benefits requires trials with at least a thousand deaths in each treatment group.3 acute
TABLE II-PUBLIC HEALTH IMPACT OF TWO POSSIBLE TREATMENTS FOR THE 1 000 000 PATIENTS WITH ACUTE ISCHAEMIC STROKE EACH YEAR IN EUROPE
Example Target patient
Simple treatment:
Complex treatment:
oral/rectal once daily aspirin
intravenous thrombolytic agent
Any ischaemic stroke within 48 h of onset
TABLE!—MED!CAL TREATMENTS FOR EARLY ACUTE ISCHAEMIC : STROKE COMPLETED CLINICAL TRIALS Number available for treatment/yr* Number dead within 3 weeks without treatment* Number dead within 3 weeks with treatmentt Deaths avoided or
delayed
Ischaemic stroke due to occlusion of a major cerebral vessel, confirmed by computed tomography and available for treatment within 6 h of onset
800 000
50 000
96000
15 000
80 000
10 000
16000
5000
*Approximate estimates derived from several sources 1,2,8,12 tAssuming aspirin reduced 3-week mortality by about one-sixth from 12to 10% and fibrinolytic therapy reduced 3-week mortality by about one-third from 30 to 20%
However, in the two largest acute stroke trials undertaken so far-the TRUST trial of an oral calcium antagonist2 and the Italian haemodilution trial4-there were only 323 and 296 deaths, respectively, and no clear evidence of a reduction in mortality could be shown. The confidence intervals around the estimate of effect on mortality in both trials were wide and could not exclude the possibility that treatrhent reduced mortality or, equally, that it increased the risk of death by 10-20%. Of course, the mortality result was not statistically ADDRESS: Department of Clinical Neurosciences, Western General Hospital, Edinburgh EH4 2XU, UK (P. Sandercock, FRCP, H. Willems). Correspondence to Dr Sandercock. *Actlve drug vs placebo
or
open control.
538
significant, but if there were a small but real adverse effect on mortality and such treatments were widely adopted (perhaps because of claims of reduced neurological disability in survivors5) widespread use of either treatment might
effect, and also identify the specific categories of patients for whom treatment is especially beneficial or especially hazardous. treatment
Experimental treatments just entering
few tens of thousands of extra deaths from stroke each year in Europe alone. cause a
Survival free of severe
disability
Although it is important to ensure that treatment reduces mortality--or at least does not increase it-it is equally important to ensure that treatment reduces disability in survivors. For example, a trial with dextran 40 in 1976 by Matthews et al6 showed that treatment reduced mortality but increased disability among survivors. For these reasons, the effect of treatment should be assessed in terms of the proportion of patients who survive free of severe disability; this measure is likewise the most important from the patients’ point of view, but has been used in few trials .2,4.6
Need for
large simple trials and overviews
Lack of evidence of benefit is not the same as evidence of lack of benefit. For some treatments (eg, antiplatelet therapy for acute ischaemic stroke) the data are insufficient whereas for others (eg, haemodilution) enough trials have been undertaken to formulate reliable conclusions after application of standard overview methods.7-9
Promising treatments: very large randomised trials needed
Antiplatelet drugs Although antiplatelet drugs such as aspirin have been tested widely (207 randomised controlled trials with over 100 000 patients), only 2 very small trials with a total of 59 patients included subjects with acute stroke (ref 10 and data on ticlopidine held by Sanofi). Aspirin, started as soon as possible after the onset of acute myocardial infarction, reduces one-month mortality by one fifth." Since many elements of the vascular pathophysiology of acute cerebral infarction are common to the pathophysiology of acute myocardial infarction, it is logical to test the value of aspirin in the cerebral condition. Thus two very large studies, each aiming for 20 000 patients, are being established-the International Stroke Trial and the National Study of Stroke in China-to test the value of early aspirin therapy in acute ischaemic stroke.
Heparin There have been 10 randomised controlled trials of heparin in over 1000 patients with acute ischaemic stroke, of whom 173 died. A formal overview of the data from these trials (Sandercock et al, unpublished) suggests promising effects on prevention of deep venous thrombosis and pulmonary embolism but uncertain effects on mortality and haemorrhagic transformation of the cerebral infarct. Two large trials of heparin are under way: the International Stroke Trial includes a heparin vs control comparison in a factorial design with aspirin vs no aspirin, and the Treatment of Acute Stroke Trial tests a low molecular weight heparinoid (Org 10172) against control (Adams HP,
personal communication). Thrombolytic agents Several randomised trials of different thrombolytic regimens are underway in Europe, North America, and Australia. Formal overviews of the results should help to provide the best estimate of the size and direction of
clinical trials There are several neuroprotective agents that act at different sites in the pathophysiological chain between cell ischaemia and cell death (see Pulsinelli, p 533). Modulators of excitatory aminoacid neurotransmitters, ion fluxes, or prostanoid metabolism, free radical scavengers, and other agents are being tested in small clinical trials or soon will be.12 These agents (table I) show great promise of benefit, but many may not be widely applicable.
Treatments of uncertain value: more randomised trials and overviews needed Calcium antagonists There have been 15 randomised controlled trials with 4000 patients,with over 500 deaths. No single trial has been large enough on its own to provide clear evidence of a reduction in mortality. A non-standard overview of the older trials suggested there might be modest benefits,’3 but an up-to-date overview with standard methods, incorporating the much larger recent trials, is clearly needed to determine whether treatment is beneficial or whether further trials are needed. over
Corticosteroids A crude analysis of the 10 trials conducted so far suggests that treatment is unlikely to reduce mortality-103/284 (36%) deaths among treated patients vs 104/287 (36%) among controls-but a formal overview and probably further trials are justified.
Other treatments
Glycerol, GM-1 ganglioside, naftidrofuryl, naloxone, pentoxifylline, prostacyclin, and beta-blockers have all been tested in randomised trials, but no clear evidence of benefit (or lack of benefit) has emerged from any single trial. Some additional trials are in progress. Again, formal overviews are required to decide whether additional trials are justified. Trials of blood pressure reduction, of manipulation of blood 0, CO2, and glucose concentrations, and of intravenous rehydration therapy9 are all needed.
Unpromising treatments unlikely to be valuable Haemodilution There have been 15 randomised trials in just under 3000 subjects, of whom 471 died. 12 trials were formally reviewed by use of a standard overview technique’ by Asplund.9 The observed effect on mortality was very close to zero with narrow confidence intervals and, furthermore, there was no evidence that disability was reduced in survivors.
Hyperbaric oxygen A pilot study by Anderson et al 14 was too small to provide evidence of efficacy, but clearly showed the treatment (in the form tested) was unpleasant, expensive, and time consuming and therefore unsuitable for large scale use. Conclusions There is currently no medical therapy which recommended for routine
use
in
most
patients
can
with
be
acute
539
ischaemic stroke. Haemodilution should be avoided. Data on all other agents are inconclusive. However, the trials (and overviews) in progress show every promise of identifying effective treatments in the near future.
REFERENCES 1. Bamford JM, Sandercock PAG, Dennis MS, Burn JPS, Warlow CP. A
prospective study of acute cerebrovascular disease in the community: the Oxfordshire Community Stroke Project 1981-1986. 2. Incidence, case fatality rates and overall outcome at one year of cerebral infarction, primary intracerebral and subarachnoid haemorrhage. J Neural Neurosurg Psychiatry 1990; 53: 16-22.
WB, Oxbury JM, Grainger KM, Greenhall RC. A blind controlled trial of dextran 40 in the treatment of ischaemic stroke. Brain
6. Matthews
1976; 99: 193-206. 7. Peto R. Why do we need systematic overviews of randomized trials? Stat Med 1987; 6: 233-40. 8. Antiplatelet Trialists’ Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. Br Med J 1988; 296: 320-31. 9. Asplund K. Hemodilution in acute stroke. Cerebrovasc Dis 1991; 1 (suppl 1): 129-38. 10. Ciuffetti G, Aisa G, Mercuri M, et al. Effects of ticlopidine on the neurologic outcome and the hemorheologic pattern in the postacute phase of ischemic stroke: a pilot study. Angiology 1990; 41: 505-11. 11. ISIS-2
Collaborative
Group.
Randomised
trial
of intravenous
2. TRUST
Study Group. Randomised, double-blind, placebo-controlled trial of nimodipine in acute stroke. Lancet 1990; 336: 1205-09. 3. Yusuf S, Collins RC, Peto R. Why do we need some large, simple randomised trials? Stat Med 1984; 3: 409-20. 4. Italian Acute Stroke Study Group. Haemodilution in acute stroke: results
streptokinase, oral aspirin, both, or neither among 17 187 cases of suspected acute myocardial infarction: ISIS-2. Lancet 1988; ii: 349-60. 12. Editorial. Treatment for stroke? Lancet 1991; 337: 1129-30. 13. Gelmers HJ, Hennerici M. Effect of nimodipine on acute ischemic stroke: pooled results from five randomised trials. Stroke 1990; 21 (suppl IV):
of the Italian Haemodilution Trial. Lancet 1988; i: 318-21. 5. Gelmers HJ, Goner K, de Weerdt CJ, Wiezer HJ. A controlled trial of nimodipine in acute ischemic stroke. N Engl J Med 1988; 318: 203-07.
IV-81-84. 14. Anderson DC, Bottini AG, Jagiella WM, et al. A pilot study of hyperbaric oxygen in the treatment of human stroke. Stroke 1991; 22: 1137-42.
HYPOTHESIS
Immune surveillance, organophosphorus exposure, and lymphomagenesis DAVID S. NEWCOMBE
Prevalence of lymphoproliferative disorders is increased in populations with various chemical exposures,
including
organophosphorus
compounds. Lymphomas are also more common in individuals with a substantially decreased monocyte esterase activity. Organophosphorus compounds inhibit esterases associated with monocytes, natural killer (NK) cells, lymphokine-activated killer (LAK) cells, and cytotoxic T lymphocytes, and these inhibitory effects impair immune surveillance and cytotoxic functions mediated by such cells. Lymphoma development is also associated with Epstein-Barr virus (EBV) and human herpesvirus-6 (HHV-6) infections, which are regulated by cytotoxic immune responses mediated by monocytes, T cells, and NK cells. My hypothesis is that lymphomagenesis is a multistep process, and the absence or inhibition of monocyte esterase and perhaps other immune cell esterases alters esterasedependent detoxification of a factor critical for the early steps of oncogenesis. Also, such an enzyme deficit might impair the processes that regulate the dissemination and limit the total burden of pathogens such as the lymphoma-associated herpesviruses. An added risk to any viralmediated lymphoproliferation might be an
organophosphorus-induced change.
oncogenic
genetic
Lympnoma epidemiology and organophosphorus exposure Certain population groups that have or are likely to have been exposed to organophosphorus agents are at increased
risk of lymphoma.l-4 According to some studies, mortality from lymphoma in farm workers is high;1,3 such workers are exposed to various fertilisers and pest-control substances, including organophosphorus compounds. Grain mill workers also have an excess of lymphatic and haemopoietic tumours, including lymphomas.s These workers are exposed to malathion and presumably its
contaminant, O,O,S-trimethyl immunosuppressive a Furthermore, phosphorothioate.6,7 statistically significant increase in non-Hodgkin lymphoma has been recorded in blue-collar white men employed in motor vehicle and resin product manufacturing processes;2 these men may have been exposed to organophosphorus flame retardants and plasticisers from resins and resin use in moulding automobile parts.
Esterases, immune functions, and organophosphorus agents Cytotoxic T lymphocytes and natural killer (NK) cells are essential for cell-mediated immunity in man. These cell types and monocyte cytotoxic activities are responsible for the elimination of virus-infected cells and provide immune surveillance for premalignant cells. Both functions have an important role in the regulation oflymphoproliferation and lymphoma development. Serine esterases, a ubiquitous family of enzymes found in cytotoxic T cells, monocytes, and NK cells, seem to be an integral component of the cytolytic mechanisms of these cells. Since some of these serine esterases are inhibited by organophosphorus compounds, organophosphorus agents that enter a person’s bloodstream might impair critical cell-mediated ADDRESS: Department of Environmental Health Sciences, 615 North Wolfe Street, Johns Hopkins University School of Hygiene and Public Health, Baltimore, Maryland 21205, USA (Prof D. S. Newcombe, MD).
Medical treatment of acute ischaemic stroke
The aim of early treatment for acute ischaemic stroke is to reverse or limit the degree of brain dysfunction. We will review medical treatments that are given as soon as possible (within hours, rather than days or weeks) after the onset of symptoms; we will not discuss treatments initiated after the acute phase (eg, oral anticoagulants) or those given for long-term secondary prevention. Our comments are based on a systematic search for completed randomised clinical trials of medical treatments for acute stroke, over 100 of which have been identified so far. TableI lists the main therapeutic strategies together with the number of completed randomised controlled clinical trials of each strategy.
The ideal medical treatment There
main types of medical treatment for ischaemic stroke: (a) simple, cheap, patients widely practicable treatments that can be given with little or no monitoring to virtually all patients; and (b) complex treatments that are more costly and difficult to administer (and monitor) and can be given only. to a few carefully selected patients. Examples of the former category are aspirin, fixed-dose subcutaneous heparin, oral calcium antagonists, and corticosteroids. Examples of the latter are fibrinolytic drugs, isovolaemic haemodilution, intravenous naftidrofuryl, hyperbaric oxygen, and intravenous inhibitors of excitatory aminoacids. Research, both in laboratory animals and in randomised clinical trials, has focused more on the complex high technology approach. The potential public health impact of both types of treatments is shown in table n. are two
with
acute
The example given in table II highlights the dilemma in research and practice: highly effective but very specific treatments may provide striking benefits for a few selected individuals whereas simpler treatments with more modest effects given to virtually all stroke patients will probably have a greater impact on public health.
Importance of large clinical trials If a medical therapy is to be used widely in patients with stroke, randomised trials should have provided clear evidence that such treatment reduces mortality. The risk of early death in unselected groups of patients with ischaemic stroke is low (ie, 10% after 30 days in the Oxfordshire Community Stroke Project1 and 12% after 3 weeks in untreated controls in the TRUST study2). If a simple, widely practicable treatment such as an oral calcium antagonist or oral aspirin reduced early mortality by only one sixth from 12% to 10%, this would represent 20 deaths avoided per 1000 patients treated. This reduction would be worthwhile, especially if it were accompanied by a similar proportional improvement in survival free of disability. Reliable detection of such moderate benefits requires trials with at least a thousand deaths in each treatment group.3 acute
TABLE II-PUBLIC HEALTH IMPACT OF TWO POSSIBLE TREATMENTS FOR THE 1 000 000 PATIENTS WITH ACUTE ISCHAEMIC STROKE EACH YEAR IN EUROPE
Example Target patient
Simple treatment:
Complex treatment:
oral/rectal once daily aspirin
intravenous thrombolytic agent
Any ischaemic stroke within 48 h of onset
TABLE!—MED!CAL TREATMENTS FOR EARLY ACUTE ISCHAEMIC : STROKE COMPLETED CLINICAL TRIALS Number available for treatment/yr* Number dead within 3 weeks without treatment* Number dead within 3 weeks with treatmentt Deaths avoided or
delayed
Ischaemic stroke due to occlusion of a major cerebral vessel, confirmed by computed tomography and available for treatment within 6 h of onset
800 000
50 000
96000
15 000
80 000
10 000
16000
5000
*Approximate estimates derived from several sources 1,2,8,12 tAssuming aspirin reduced 3-week mortality by about one-sixth from 12to 10% and fibrinolytic therapy reduced 3-week mortality by about one-third from 30 to 20%
However, in the two largest acute stroke trials undertaken so far-the TRUST trial of an oral calcium antagonist2 and the Italian haemodilution trial4-there were only 323 and 296 deaths, respectively, and no clear evidence of a reduction in mortality could be shown. The confidence intervals around the estimate of effect on mortality in both trials were wide and could not exclude the possibility that treatrhent reduced mortality or, equally, that it increased the risk of death by 10-20%. Of course, the mortality result was not statistically ADDRESS: Department of Clinical Neurosciences, Western General Hospital, Edinburgh EH4 2XU, UK (P. Sandercock, FRCP, H. Willems). Correspondence to Dr Sandercock. *Actlve drug vs placebo
or
open control.
538
significant, but if there were a small but real adverse effect on mortality and such treatments were widely adopted (perhaps because of claims of reduced neurological disability in survivors5) widespread use of either treatment might
effect, and also identify the specific categories of patients for whom treatment is especially beneficial or especially hazardous. treatment
Experimental treatments just entering
few tens of thousands of extra deaths from stroke each year in Europe alone. cause a
Survival free of severe
disability
Although it is important to ensure that treatment reduces mortality--or at least does not increase it-it is equally important to ensure that treatment reduces disability in survivors. For example, a trial with dextran 40 in 1976 by Matthews et al6 showed that treatment reduced mortality but increased disability among survivors. For these reasons, the effect of treatment should be assessed in terms of the proportion of patients who survive free of severe disability; this measure is likewise the most important from the patients’ point of view, but has been used in few trials .2,4.6
Need for
large simple trials and overviews
Lack of evidence of benefit is not the same as evidence of lack of benefit. For some treatments (eg, antiplatelet therapy for acute ischaemic stroke) the data are insufficient whereas for others (eg, haemodilution) enough trials have been undertaken to formulate reliable conclusions after application of standard overview methods.7-9
Promising treatments: very large randomised trials needed
Antiplatelet drugs Although antiplatelet drugs such as aspirin have been tested widely (207 randomised controlled trials with over 100 000 patients), only 2 very small trials with a total of 59 patients included subjects with acute stroke (ref 10 and data on ticlopidine held by Sanofi). Aspirin, started as soon as possible after the onset of acute myocardial infarction, reduces one-month mortality by one fifth." Since many elements of the vascular pathophysiology of acute cerebral infarction are common to the pathophysiology of acute myocardial infarction, it is logical to test the value of aspirin in the cerebral condition. Thus two very large studies, each aiming for 20 000 patients, are being established-the International Stroke Trial and the National Study of Stroke in China-to test the value of early aspirin therapy in acute ischaemic stroke.
Heparin There have been 10 randomised controlled trials of heparin in over 1000 patients with acute ischaemic stroke, of whom 173 died. A formal overview of the data from these trials (Sandercock et al, unpublished) suggests promising effects on prevention of deep venous thrombosis and pulmonary embolism but uncertain effects on mortality and haemorrhagic transformation of the cerebral infarct. Two large trials of heparin are under way: the International Stroke Trial includes a heparin vs control comparison in a factorial design with aspirin vs no aspirin, and the Treatment of Acute Stroke Trial tests a low molecular weight heparinoid (Org 10172) against control (Adams HP,
personal communication). Thrombolytic agents Several randomised trials of different thrombolytic regimens are underway in Europe, North America, and Australia. Formal overviews of the results should help to provide the best estimate of the size and direction of
clinical trials There are several neuroprotective agents that act at different sites in the pathophysiological chain between cell ischaemia and cell death (see Pulsinelli, p 533). Modulators of excitatory aminoacid neurotransmitters, ion fluxes, or prostanoid metabolism, free radical scavengers, and other agents are being tested in small clinical trials or soon will be.12 These agents (table I) show great promise of benefit, but many may not be widely applicable.
Treatments of uncertain value: more randomised trials and overviews needed Calcium antagonists There have been 15 randomised controlled trials with 4000 patients,with over 500 deaths. No single trial has been large enough on its own to provide clear evidence of a reduction in mortality. A non-standard overview of the older trials suggested there might be modest benefits,’3 but an up-to-date overview with standard methods, incorporating the much larger recent trials, is clearly needed to determine whether treatment is beneficial or whether further trials are needed. over
Corticosteroids A crude analysis of the 10 trials conducted so far suggests that treatment is unlikely to reduce mortality-103/284 (36%) deaths among treated patients vs 104/287 (36%) among controls-but a formal overview and probably further trials are justified.
Other treatments
Glycerol, GM-1 ganglioside, naftidrofuryl, naloxone, pentoxifylline, prostacyclin, and beta-blockers have all been tested in randomised trials, but no clear evidence of benefit (or lack of benefit) has emerged from any single trial. Some additional trials are in progress. Again, formal overviews are required to decide whether additional trials are justified. Trials of blood pressure reduction, of manipulation of blood 0, CO2, and glucose concentrations, and of intravenous rehydration therapy9 are all needed.
Unpromising treatments unlikely to be valuable Haemodilution There have been 15 randomised trials in just under 3000 subjects, of whom 471 died. 12 trials were formally reviewed by use of a standard overview technique’ by Asplund.9 The observed effect on mortality was very close to zero with narrow confidence intervals and, furthermore, there was no evidence that disability was reduced in survivors.
Hyperbaric oxygen A pilot study by Anderson et al 14 was too small to provide evidence of efficacy, but clearly showed the treatment (in the form tested) was unpleasant, expensive, and time consuming and therefore unsuitable for large scale use. Conclusions There is currently no medical therapy which recommended for routine
use
in
most
patients
can
with
be
acute
539
ischaemic stroke. Haemodilution should be avoided. Data on all other agents are inconclusive. However, the trials (and overviews) in progress show every promise of identifying effective treatments in the near future.
REFERENCES 1. Bamford JM, Sandercock PAG, Dennis MS, Burn JPS, Warlow CP. A
prospective study of acute cerebrovascular disease in the community: the Oxfordshire Community Stroke Project 1981-1986. 2. Incidence, case fatality rates and overall outcome at one year of cerebral infarction, primary intracerebral and subarachnoid haemorrhage. J Neural Neurosurg Psychiatry 1990; 53: 16-22.
WB, Oxbury JM, Grainger KM, Greenhall RC. A blind controlled trial of dextran 40 in the treatment of ischaemic stroke. Brain
6. Matthews
1976; 99: 193-206. 7. Peto R. Why do we need systematic overviews of randomized trials? Stat Med 1987; 6: 233-40. 8. Antiplatelet Trialists’ Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. Br Med J 1988; 296: 320-31. 9. Asplund K. Hemodilution in acute stroke. Cerebrovasc Dis 1991; 1 (suppl 1): 129-38. 10. Ciuffetti G, Aisa G, Mercuri M, et al. Effects of ticlopidine on the neurologic outcome and the hemorheologic pattern in the postacute phase of ischemic stroke: a pilot study. Angiology 1990; 41: 505-11. 11. ISIS-2
Collaborative
Group.
Randomised
trial
of intravenous
2. TRUST
Study Group. Randomised, double-blind, placebo-controlled trial of nimodipine in acute stroke. Lancet 1990; 336: 1205-09. 3. Yusuf S, Collins RC, Peto R. Why do we need some large, simple randomised trials? Stat Med 1984; 3: 409-20. 4. Italian Acute Stroke Study Group. Haemodilution in acute stroke: results
streptokinase, oral aspirin, both, or neither among 17 187 cases of suspected acute myocardial infarction: ISIS-2. Lancet 1988; ii: 349-60. 12. Editorial. Treatment for stroke? Lancet 1991; 337: 1129-30. 13. Gelmers HJ, Hennerici M. Effect of nimodipine on acute ischemic stroke: pooled results from five randomised trials. Stroke 1990; 21 (suppl IV):
of the Italian Haemodilution Trial. Lancet 1988; i: 318-21. 5. Gelmers HJ, Goner K, de Weerdt CJ, Wiezer HJ. A controlled trial of nimodipine in acute ischemic stroke. N Engl J Med 1988; 318: 203-07.
IV-81-84. 14. Anderson DC, Bottini AG, Jagiella WM, et al. A pilot study of hyperbaric oxygen in the treatment of human stroke. Stroke 1991; 22: 1137-42.
HYPOTHESIS
Immune surveillance, organophosphorus exposure, and lymphomagenesis DAVID S. NEWCOMBE
Prevalence of lymphoproliferative disorders is increased in populations with various chemical exposures,
including
organophosphorus
compounds. Lymphomas are also more common in individuals with a substantially decreased monocyte esterase activity. Organophosphorus compounds inhibit esterases associated with monocytes, natural killer (NK) cells, lymphokine-activated killer (LAK) cells, and cytotoxic T lymphocytes, and these inhibitory effects impair immune surveillance and cytotoxic functions mediated by such cells. Lymphoma development is also associated with Epstein-Barr virus (EBV) and human herpesvirus-6 (HHV-6) infections, which are regulated by cytotoxic immune responses mediated by monocytes, T cells, and NK cells. My hypothesis is that lymphomagenesis is a multistep process, and the absence or inhibition of monocyte esterase and perhaps other immune cell esterases alters esterasedependent detoxification of a factor critical for the early steps of oncogenesis. Also, such an enzyme deficit might impair the processes that regulate the dissemination and limit the total burden of pathogens such as the lymphoma-associated herpesviruses. An added risk to any viralmediated lymphoproliferation might be an
organophosphorus-induced change.
oncogenic
genetic
Lympnoma epidemiology and organophosphorus exposure Certain population groups that have or are likely to have been exposed to organophosphorus agents are at increased
risk of lymphoma.l-4 According to some studies, mortality from lymphoma in farm workers is high;1,3 such workers are exposed to various fertilisers and pest-control substances, including organophosphorus compounds. Grain mill workers also have an excess of lymphatic and haemopoietic tumours, including lymphomas.s These workers are exposed to malathion and presumably its
contaminant, O,O,S-trimethyl immunosuppressive a Furthermore, phosphorothioate.6,7 statistically significant increase in non-Hodgkin lymphoma has been recorded in blue-collar white men employed in motor vehicle and resin product manufacturing processes;2 these men may have been exposed to organophosphorus flame retardants and plasticisers from resins and resin use in moulding automobile parts.
Esterases, immune functions, and organophosphorus agents Cytotoxic T lymphocytes and natural killer (NK) cells are essential for cell-mediated immunity in man. These cell types and monocyte cytotoxic activities are responsible for the elimination of virus-infected cells and provide immune surveillance for premalignant cells. Both functions have an important role in the regulation oflymphoproliferation and lymphoma development. Serine esterases, a ubiquitous family of enzymes found in cytotoxic T cells, monocytes, and NK cells, seem to be an integral component of the cytolytic mechanisms of these cells. Since some of these serine esterases are inhibited by organophosphorus compounds, organophosphorus agents that enter a person’s bloodstream might impair critical cell-mediated ADDRESS: Department of Environmental Health Sciences, 615 North Wolfe Street, Johns Hopkins University School of Hygiene and Public Health, Baltimore, Maryland 21205, USA (Prof D. S. Newcombe, MD).