BRITISH MEDICAL JOURNAL
1 MAY 1973
produce a leucopenia and there may be a common single factor incorporated into many viruses which may produce this effect. As many viruses also reach the CNS there is still a credible possibility that several of them may be important in creating the clinical picture of MS. This possibility would help to rationalise the results of surveys which might otherwise seem to conflict. 1
Perier, 0, and Gregoire, A, Brain, 1965, 88, 937. Rinne, U K, Annales Medicinae Internae Fenniae, 1968, 58, 179. 3Perier, 0, International Archives of Allergy, 1969, 36, Suppl. 452. 4Suzuki, K, et al, Laboratory Investigation, 1969, 20, 444. 5 Andrews, J M, M'ltiple Sclerosis, 1972. 6 Field, E J, et al, Lancet, 1972, 2, 280. 7Prineas, J, Science, 1972, 178, 760. 8 Adams, J M, and Iwagawa, D T, Proceedings of the Society for Experimental Biology and Medicine, 1962, 111, 562. 9 Brody, J A, Lancet, 1972, 2, 173. 10 ter Meulen, V, et al, Lancet, 1972, 2, 1. 11 Rogers, N G, et al, Nature, 1967, 216, 446. 12 Carp, R I, et al, Journal of Experimental Medicine, 1972, 136, 618. 13 Carp, R I, Merz, G S, and Licursi, P C, Infection and Immunity, 1974, 9, 1011. 14 Koldovsky, U, et al, Infection and Immunity, 1975, 12, 1355. 15 Henle, G, et al, Infection and Immunity, 1975, 12, 1367. 16 Licursi, P C, et al, Infection and Immunity, 1972, 6, 370. 2
Depression in old age Depression, in all its forms, is undoubtedly the most common mental disorder affecting persons aged between 65 and 75. Only after 75 does the prevalence of senile psychosis start to become greater; and eminently treatable depressive illnesses are not unknown in octogenarians. Because the symptoms of depression in old age so often mimic dementia the most careful diagnostic scrutiny is called for. Even in younger patients depression affects intellectual function, leading to poor concentration, difficulty in solving everyday problems, indecisiveness, and impaired memory. When superimposed upon the ageing brain, where on account of organic factors there may already be some loss of mental agility leading to difficulty in grasping and coming to terms with new ideas, these symptoms may suggest a greater degree of dementia than is actually present and, accordingly, a gloomier than justified prognosis. A positive family history of depression or a history of a previous attack may be helpful. Incontinence may be a useful pointer, for, apart from accidents, this is unlikely in uncomplicated depression. In all cases of doubt antidepressive treatment may well be worth a trial. In younger patients much diagnostic energy is often expended in deciding whether a depressive illness is endogenous or reactive (often a useless pursuit), and in the elderly this may be an even more futile exercise. While the depressions of later life contain many elements-endogenous and reactive, physical and psychological-it is the interplay among them which is of the greatest importance and which carries with it considerable implications for treatment. Charatan1 has pointed out that loss is the central theme of depression in old ageloss which can take many forms. As such losses are likely to be multiple and successive, the ageing individual tends to become increasingly sensitised and possibly, on this account, progressively depressed until finally "sans everything" may paradoxically bring relief. While the most important of these losses is usually bereavement, leading to loss of companionship of the spouse, other such losses may be due to the deaths of old friends or to
1031
retirement, when contacts with workmates may cease. Loss of income may prevent relatives from visiting owing to the rising cost of travel. Social isolation may be further enhanced by loss of initiative or be due to disorders which interfere with locomotion-chronic cardiorespiratory disease, arthritis, or neurological disorders. To these failing eyesight and hearing may add some degree of sensory deprivation. Clearly, then, the treatment of depression in old age is by no means simple but calls for the relief of many different disablements. Some of these, such as constipation, insomnia, lack of appetite, and loss in weight, together with a variety of rheumatic aches and pains, will require medical treatment; others may require the mobilisation of the resources of the various social agencies. Voluntary organisations, too, may have an important part to play, with such measures as meals-onwheels, visiting, arranging old people's outings, and so onall of which are designed to combat increasing social isolation and loneliness. But, because the starting point is so often a visit to the surgery or a domiciliary consultation, it often falls to the family doctor and sometimes to the psychiatrist to initiate proceedings. More specific forms of treatment are not necessarily precluded by age, though by themselves they are likely to be insufficient. Even in the very old electroconvulsive therapy can work wonders, though it may impair memory more than in younger people. But this, perhaps, may matter less. Antidepressant drugs are also of value, though it is essential to be cautious about dosage. Special risks in the elderly include hypotension; hypothermia; the precipitation of glaucoma where this is incipient; urinary retention; and cardiac arrhythmias, which may be a hazard to those with pre-existing cardiac damage. All such drugs, particularly in overdosage, may give rise to or exacerbate the transient confusional states to which the elderly are prone. To add pharmacogenic disease to their other burdens is something to be avoided; for those who are older are less resilient, and attention to these and many other details may well count most in the long run. 1
Charatan, F B, New York State journal of Medicine, 1975, 75, 2505, 2307.
Caffeine, coffee, and cancer What twist is it in man's devious make-up that makes him round on the seemingly more wholesome and pleasurable aspects of his environment and suspect them of being causes of his misfortunes? Whatever it is, stimulants of all kinds (and especially coffee and caffeine) maintain a position high on the list for suspicion despite a continuing lack of real evidence of any hazard to health. The range of possible serious toxic effects for any chemical found in our environment is wide, but fears are usually allayed once it has been shown not to be teratogenic, mutagenic, or carcinogenic. Unfortunately coffee cannot claim such a clean bill of health, for (at intake levels manifestly above those of even the most ardent coffee drinker) caffeine has been shown in the laboratory to be both teratogenic and mutagenic.1 Nishimura and Nakai2 saw embryonic deaths and various congenital abnormalities in the progeny of mice given a single intraperitoneal dose of caffeine equivalent to about 100 cups of coffee. Later, other workers found similar results in rats, rabbits, and chickens. Nevertheless, no embryotoxic effects were seen in mice given daily doses of caffeine
1032
equivalent of up to 40 cups of coffee,' and none were seen in a four-generation mouse study of animals exposed to caffeine in the drinking water at rates equivalent of up to 30 cups of coffee per day.3 Caffeine also got off to a bad start in tests for mutagenicity, with positive results in both Escherichia coli4 and Drosophila.5 Nevertheless, subsequent studies in mice with the dominant lethal assay6 7 were negative, and so were cytogenetic studies in mice.8 Like almost everything else, in high concentrations caffeine is cytotoxic to human lymphocytes maintained in tissue culture9-at concentrations 100 times higher than that achievable in the blood by drinking eight cups of coffee in quick succession. At lower concentrations nothing more than slight stimulation or inhibition of mitosis occurs. Laboratory tests of caffeine for carcinogenicity in animals, carried out a long time ago, gave essentially negative results.'0 These tests, however, would be regarded as inadequate by modern standards, and there is, moreover, a theoretical possibility that caffeine could act as a cocarcinogen by catalysing the formation of N-nitrosamines from secondary amines and nitrites in the stomach." To counterbalance the fear this possibility engenders there is evidence of anticarcinogenic effects of caffeine in animals exposed to known carcinogens'2-'5 and the fact that caffeine is rapidly and efficiently metabolised in man.16 Suspicion of a cancer risk for man has centred on the urinary tract since Cole17 18 reported a higher incidence of coffeedrinking in men (1 24: 1) and women (2-58: 1) with bladder cancer than among matched controls, and Shennan'9 drew attention to a strong correlation between coffee consumption and national mortality rates for renal cancer. The latest development in this area is a report from Sir Richard Doll and his colleagues in Oxford20 of their failure to find any association between either adenocarcinoma of the renal parenchyma (106 patients) or carcinoma of the renal pelvis (33 patients) and coffee consumption. They explain that because of the small numbers of subjects available for study a two- to threefold enhancement of cancer incidence by coffee cannot be excluded; but confidence in the reliability of their negative finding is increased by two facts. Firstly, Wynder and his colleagues21 also saw no evidence of an association between coffee consumption and renal cancer; and, secondly, the findings of the Oxford workers are consistent with there being a 1 8-fold excess incidence of cancer of the renal pelvis in men who smoked compared with non-smokers and a definite trend towards increased risk of this form of cancer in association with heavier cigarette consumption. These latter findings were to be expected, because agents which cause cancer in the urinary bladder commonly affect the renal pelvis similarly, and there is now a well-established association between bladder cancer risk and smoking of about the same order-a twofold excess. For the diehard puritan who cannot bring himself to accept that coffee might be harmless so far as cancer is concerned there is still the let-out that Schmauz and Cole'8 saw a particularly high risk of cancer of the renal pelvis among drinkers of seven or more cups of coffee per day: none of the patients or control subjects studied by the Oxford workers consumed as much as this. But for those who, not being gluttons for the brew, find the very smell of it pleasurable and mouth-watering the news from Oxford will provide at least some grounds for
comfort.
'Mulvihill, J J, Teratology, 2
1973, 8, 69. Nishimura, H, and Nakai, K, Proceedings of the Society for Experimental Biology and Medicine, 1960, 104, 140.
BRITISH MEDICAL JOURNAL
1 mAY 1976
3Thayer, P S, and Kensler, C J, Toxicology and Applied Pharmacology, 1973, 25, 169. 4Demerec, M, Bertani, G, and Flint, J, American Naturalist, 1951, 85, 119. 5 Bateman, A J, Mutation Research, 1969, 7, 475. 6 Epstein, S S, et al, Food and Cosmetics Toxicology, 1970, 8, 381. 7Adler, I-D, Humangenetik, 1969, 7, 137. 8 Adler, I-D, and Rohrborn, G, Humangenetik, 1969, 8, 81. 9 Timson, J, British Journal of Pharmacology, 1970, 38, 731. 10 Boughton, L L, and Stoland, 0 OJ,ournal of the American Pharmaceutical Association, 1943, 32, 187. 11 Challis, B C, and Bartlett, C D, Nature, 1975, 254, 522. 12 Weil-Malherbe, H, 1946, 40, 351. 13 Booth, J, and Boyland, E, Biochimica et Biophysica Acta, 1953, 12, 75. 14 Leiter, J, and Shear, M J,Jrournal of the National Cancer Institute, 1943, 3, 455. 15 Rothwell, K, Nature, 1974, 252, 69. 16 Axelrod, J, and Reichenthal, J, Journal of Pharmacology and Experimental Therapeutics, 1953, 107, 519. 17 Cole, P, Lancet, 1971, 1, 1335. 18 Schmauz, R, and Cole, P, Journal of the National Cancer Institute, 1974, 52, 1431. 19 Shennan, D H, British Journal of Cancer, 1973, 28, 473. 20 Armstrong, B, Garrod, A, and Doll, R, British Journal of Cancer, 1976, 33, 127. 21 Wynder, E L, Mabuchi, K, and Whitmore, W F, Jrournal of the National Cancer Institute, 1974, 53, 1619.
Biochemical,Journal,
Anaesthetists and infection A recent issue of the British Journal of Anaesthesial contains a symposium on infection as it concerns the anaesthetist. The journal quotes an author in the BMJr 103 years ago2 who condemned the use of inhalers as "a disgusting evil," and claimed that his own was the only one cleaned between one patient and the next: "sweet seventeen is made to follow a bearded devotee to Bacchus, saturated with the smoke of cigars and the exhalations of cognac." Within living memory little attention was paid to rubber face masks (and none to tubing or bags) unless perhaps they had been used for patients with known open tuberculosis; furthermore, perfunctory immersion in 5% phenol without adequate rinsing was known to damage the facial skin of the next patient. All this has changed, and the first paper3 in the symposium is about the "decontamination" (not necessarily sterilisation) of anaesthetic equipment, now far more complex than it used to be. The order of preference of methods of disinfection is heat, then ethylene oxide, and lastly liquid disinfectants. Apparently the problem of damage from residual disinfectant still exists, and with more reason, from ethylene oxide and glutaraldehyde, particularly with rubber materials. The responsibilities of anaesthetists in intensive care units are discussed by Gaya,4 who argues that the extreme susceptibility to infection of patients in these units is sometimes compounded by overcrowding, inadequate staffing, and the indiscriminate use of antibiotics favouring the spread of resistant pathogens. Design is important: ideally these units should include single rooms with space for the bulky equipment required and with positive pressure ventilation for protective isolation and exhaust ventilation for containment; these two should be distinct, since a reversible ventilation system may be misused and many in any case have secondary effects elsewhere in the ward. Bulky equipment, often impossible to sterilise, tends to accumulate in ICUs, and ease of sterilisation should be a paramount requirement in its choice. Gaya discourages the use of bacteriological monitoring except for endotracheal aspirates and peritoneal dialysis fluids, two examples where it may give a timely warning and point to
1 MAY 1973
produce a leucopenia and there may be a common single factor incorporated into many viruses which may produce this effect. As many viruses also reach the CNS there is still a credible possibility that several of them may be important in creating the clinical picture of MS. This possibility would help to rationalise the results of surveys which might otherwise seem to conflict. 1
Perier, 0, and Gregoire, A, Brain, 1965, 88, 937. Rinne, U K, Annales Medicinae Internae Fenniae, 1968, 58, 179. 3Perier, 0, International Archives of Allergy, 1969, 36, Suppl. 452. 4Suzuki, K, et al, Laboratory Investigation, 1969, 20, 444. 5 Andrews, J M, M'ltiple Sclerosis, 1972. 6 Field, E J, et al, Lancet, 1972, 2, 280. 7Prineas, J, Science, 1972, 178, 760. 8 Adams, J M, and Iwagawa, D T, Proceedings of the Society for Experimental Biology and Medicine, 1962, 111, 562. 9 Brody, J A, Lancet, 1972, 2, 173. 10 ter Meulen, V, et al, Lancet, 1972, 2, 1. 11 Rogers, N G, et al, Nature, 1967, 216, 446. 12 Carp, R I, et al, Journal of Experimental Medicine, 1972, 136, 618. 13 Carp, R I, Merz, G S, and Licursi, P C, Infection and Immunity, 1974, 9, 1011. 14 Koldovsky, U, et al, Infection and Immunity, 1975, 12, 1355. 15 Henle, G, et al, Infection and Immunity, 1975, 12, 1367. 16 Licursi, P C, et al, Infection and Immunity, 1972, 6, 370. 2
Depression in old age Depression, in all its forms, is undoubtedly the most common mental disorder affecting persons aged between 65 and 75. Only after 75 does the prevalence of senile psychosis start to become greater; and eminently treatable depressive illnesses are not unknown in octogenarians. Because the symptoms of depression in old age so often mimic dementia the most careful diagnostic scrutiny is called for. Even in younger patients depression affects intellectual function, leading to poor concentration, difficulty in solving everyday problems, indecisiveness, and impaired memory. When superimposed upon the ageing brain, where on account of organic factors there may already be some loss of mental agility leading to difficulty in grasping and coming to terms with new ideas, these symptoms may suggest a greater degree of dementia than is actually present and, accordingly, a gloomier than justified prognosis. A positive family history of depression or a history of a previous attack may be helpful. Incontinence may be a useful pointer, for, apart from accidents, this is unlikely in uncomplicated depression. In all cases of doubt antidepressive treatment may well be worth a trial. In younger patients much diagnostic energy is often expended in deciding whether a depressive illness is endogenous or reactive (often a useless pursuit), and in the elderly this may be an even more futile exercise. While the depressions of later life contain many elements-endogenous and reactive, physical and psychological-it is the interplay among them which is of the greatest importance and which carries with it considerable implications for treatment. Charatan1 has pointed out that loss is the central theme of depression in old ageloss which can take many forms. As such losses are likely to be multiple and successive, the ageing individual tends to become increasingly sensitised and possibly, on this account, progressively depressed until finally "sans everything" may paradoxically bring relief. While the most important of these losses is usually bereavement, leading to loss of companionship of the spouse, other such losses may be due to the deaths of old friends or to
1031
retirement, when contacts with workmates may cease. Loss of income may prevent relatives from visiting owing to the rising cost of travel. Social isolation may be further enhanced by loss of initiative or be due to disorders which interfere with locomotion-chronic cardiorespiratory disease, arthritis, or neurological disorders. To these failing eyesight and hearing may add some degree of sensory deprivation. Clearly, then, the treatment of depression in old age is by no means simple but calls for the relief of many different disablements. Some of these, such as constipation, insomnia, lack of appetite, and loss in weight, together with a variety of rheumatic aches and pains, will require medical treatment; others may require the mobilisation of the resources of the various social agencies. Voluntary organisations, too, may have an important part to play, with such measures as meals-onwheels, visiting, arranging old people's outings, and so onall of which are designed to combat increasing social isolation and loneliness. But, because the starting point is so often a visit to the surgery or a domiciliary consultation, it often falls to the family doctor and sometimes to the psychiatrist to initiate proceedings. More specific forms of treatment are not necessarily precluded by age, though by themselves they are likely to be insufficient. Even in the very old electroconvulsive therapy can work wonders, though it may impair memory more than in younger people. But this, perhaps, may matter less. Antidepressant drugs are also of value, though it is essential to be cautious about dosage. Special risks in the elderly include hypotension; hypothermia; the precipitation of glaucoma where this is incipient; urinary retention; and cardiac arrhythmias, which may be a hazard to those with pre-existing cardiac damage. All such drugs, particularly in overdosage, may give rise to or exacerbate the transient confusional states to which the elderly are prone. To add pharmacogenic disease to their other burdens is something to be avoided; for those who are older are less resilient, and attention to these and many other details may well count most in the long run. 1
Charatan, F B, New York State journal of Medicine, 1975, 75, 2505, 2307.
Caffeine, coffee, and cancer What twist is it in man's devious make-up that makes him round on the seemingly more wholesome and pleasurable aspects of his environment and suspect them of being causes of his misfortunes? Whatever it is, stimulants of all kinds (and especially coffee and caffeine) maintain a position high on the list for suspicion despite a continuing lack of real evidence of any hazard to health. The range of possible serious toxic effects for any chemical found in our environment is wide, but fears are usually allayed once it has been shown not to be teratogenic, mutagenic, or carcinogenic. Unfortunately coffee cannot claim such a clean bill of health, for (at intake levels manifestly above those of even the most ardent coffee drinker) caffeine has been shown in the laboratory to be both teratogenic and mutagenic.1 Nishimura and Nakai2 saw embryonic deaths and various congenital abnormalities in the progeny of mice given a single intraperitoneal dose of caffeine equivalent to about 100 cups of coffee. Later, other workers found similar results in rats, rabbits, and chickens. Nevertheless, no embryotoxic effects were seen in mice given daily doses of caffeine
1032
equivalent of up to 40 cups of coffee,' and none were seen in a four-generation mouse study of animals exposed to caffeine in the drinking water at rates equivalent of up to 30 cups of coffee per day.3 Caffeine also got off to a bad start in tests for mutagenicity, with positive results in both Escherichia coli4 and Drosophila.5 Nevertheless, subsequent studies in mice with the dominant lethal assay6 7 were negative, and so were cytogenetic studies in mice.8 Like almost everything else, in high concentrations caffeine is cytotoxic to human lymphocytes maintained in tissue culture9-at concentrations 100 times higher than that achievable in the blood by drinking eight cups of coffee in quick succession. At lower concentrations nothing more than slight stimulation or inhibition of mitosis occurs. Laboratory tests of caffeine for carcinogenicity in animals, carried out a long time ago, gave essentially negative results.'0 These tests, however, would be regarded as inadequate by modern standards, and there is, moreover, a theoretical possibility that caffeine could act as a cocarcinogen by catalysing the formation of N-nitrosamines from secondary amines and nitrites in the stomach." To counterbalance the fear this possibility engenders there is evidence of anticarcinogenic effects of caffeine in animals exposed to known carcinogens'2-'5 and the fact that caffeine is rapidly and efficiently metabolised in man.16 Suspicion of a cancer risk for man has centred on the urinary tract since Cole17 18 reported a higher incidence of coffeedrinking in men (1 24: 1) and women (2-58: 1) with bladder cancer than among matched controls, and Shennan'9 drew attention to a strong correlation between coffee consumption and national mortality rates for renal cancer. The latest development in this area is a report from Sir Richard Doll and his colleagues in Oxford20 of their failure to find any association between either adenocarcinoma of the renal parenchyma (106 patients) or carcinoma of the renal pelvis (33 patients) and coffee consumption. They explain that because of the small numbers of subjects available for study a two- to threefold enhancement of cancer incidence by coffee cannot be excluded; but confidence in the reliability of their negative finding is increased by two facts. Firstly, Wynder and his colleagues21 also saw no evidence of an association between coffee consumption and renal cancer; and, secondly, the findings of the Oxford workers are consistent with there being a 1 8-fold excess incidence of cancer of the renal pelvis in men who smoked compared with non-smokers and a definite trend towards increased risk of this form of cancer in association with heavier cigarette consumption. These latter findings were to be expected, because agents which cause cancer in the urinary bladder commonly affect the renal pelvis similarly, and there is now a well-established association between bladder cancer risk and smoking of about the same order-a twofold excess. For the diehard puritan who cannot bring himself to accept that coffee might be harmless so far as cancer is concerned there is still the let-out that Schmauz and Cole'8 saw a particularly high risk of cancer of the renal pelvis among drinkers of seven or more cups of coffee per day: none of the patients or control subjects studied by the Oxford workers consumed as much as this. But for those who, not being gluttons for the brew, find the very smell of it pleasurable and mouth-watering the news from Oxford will provide at least some grounds for
comfort.
'Mulvihill, J J, Teratology, 2
1973, 8, 69. Nishimura, H, and Nakai, K, Proceedings of the Society for Experimental Biology and Medicine, 1960, 104, 140.
BRITISH MEDICAL JOURNAL
1 mAY 1976
3Thayer, P S, and Kensler, C J, Toxicology and Applied Pharmacology, 1973, 25, 169. 4Demerec, M, Bertani, G, and Flint, J, American Naturalist, 1951, 85, 119. 5 Bateman, A J, Mutation Research, 1969, 7, 475. 6 Epstein, S S, et al, Food and Cosmetics Toxicology, 1970, 8, 381. 7Adler, I-D, Humangenetik, 1969, 7, 137. 8 Adler, I-D, and Rohrborn, G, Humangenetik, 1969, 8, 81. 9 Timson, J, British Journal of Pharmacology, 1970, 38, 731. 10 Boughton, L L, and Stoland, 0 OJ,ournal of the American Pharmaceutical Association, 1943, 32, 187. 11 Challis, B C, and Bartlett, C D, Nature, 1975, 254, 522. 12 Weil-Malherbe, H, 1946, 40, 351. 13 Booth, J, and Boyland, E, Biochimica et Biophysica Acta, 1953, 12, 75. 14 Leiter, J, and Shear, M J,Jrournal of the National Cancer Institute, 1943, 3, 455. 15 Rothwell, K, Nature, 1974, 252, 69. 16 Axelrod, J, and Reichenthal, J, Journal of Pharmacology and Experimental Therapeutics, 1953, 107, 519. 17 Cole, P, Lancet, 1971, 1, 1335. 18 Schmauz, R, and Cole, P, Journal of the National Cancer Institute, 1974, 52, 1431. 19 Shennan, D H, British Journal of Cancer, 1973, 28, 473. 20 Armstrong, B, Garrod, A, and Doll, R, British Journal of Cancer, 1976, 33, 127. 21 Wynder, E L, Mabuchi, K, and Whitmore, W F, Jrournal of the National Cancer Institute, 1974, 53, 1619.
Biochemical,Journal,
Anaesthetists and infection A recent issue of the British Journal of Anaesthesial contains a symposium on infection as it concerns the anaesthetist. The journal quotes an author in the BMJr 103 years ago2 who condemned the use of inhalers as "a disgusting evil," and claimed that his own was the only one cleaned between one patient and the next: "sweet seventeen is made to follow a bearded devotee to Bacchus, saturated with the smoke of cigars and the exhalations of cognac." Within living memory little attention was paid to rubber face masks (and none to tubing or bags) unless perhaps they had been used for patients with known open tuberculosis; furthermore, perfunctory immersion in 5% phenol without adequate rinsing was known to damage the facial skin of the next patient. All this has changed, and the first paper3 in the symposium is about the "decontamination" (not necessarily sterilisation) of anaesthetic equipment, now far more complex than it used to be. The order of preference of methods of disinfection is heat, then ethylene oxide, and lastly liquid disinfectants. Apparently the problem of damage from residual disinfectant still exists, and with more reason, from ethylene oxide and glutaraldehyde, particularly with rubber materials. The responsibilities of anaesthetists in intensive care units are discussed by Gaya,4 who argues that the extreme susceptibility to infection of patients in these units is sometimes compounded by overcrowding, inadequate staffing, and the indiscriminate use of antibiotics favouring the spread of resistant pathogens. Design is important: ideally these units should include single rooms with space for the bulky equipment required and with positive pressure ventilation for protective isolation and exhaust ventilation for containment; these two should be distinct, since a reversible ventilation system may be misused and many in any case have secondary effects elsewhere in the ward. Bulky equipment, often impossible to sterilise, tends to accumulate in ICUs, and ease of sterilisation should be a paramount requirement in its choice. Gaya discourages the use of bacteriological monitoring except for endotracheal aspirates and peritoneal dialysis fluids, two examples where it may give a timely warning and point to
