48
Journal of the Royal Society of Medicine Volume 72 January 1979
Discussion In 1958 Goodall reported on a 20-years' retrospective study of children in a Yorkshire practice; 22.5% wheezed at some time, 47% of whom wheezed only once, and 5% of the study eventually developed asthma. In Fry's 10-years' prospective follow up of children in a South London practice (Fry 1961), 16% wheezed at some time, of whom 63% wheezed only once, and 8% of the study wheezed three to four times. However, the incidence of asthma in this practice was 8% in-children in the 5-15 years' age range and 6% in the 0-20 years' age range, studied over a period from 1970-73 (Blair 1974). Horn & Gregg (1973) found after a five-years' follow up that at least a third of young children with wheezing or asthmatic bronchitis with no constitutional factor (no history or presence of eczema or allergic rhinitis, no positive skin tests and no family history of atopic disease) were undoubtedly potential asthmatics triggered by infection and, they presumed, with an immunological disturbance. These findings confirmed Boesen's (1953) results. The natural history of childhood asthma may be best determined by a prospective long-term follow up. Short-term follow ups of such patients will fail to include those patients whose asthma remits and subsequently relapses - not necessarily triggered by the same factors as initially. Retrospective follow-up studies, especially if incomplete, are of limited value as patients' memories cannot be relied upon. Not all children outgrow their asthma; in this series 48% of childhood asthmatics seen between the years 1948 and 1952 were still asthmatic twenty years later. The factors, as shown in the various tables, which adversely affected the long-term prognosis, were the degree of inheritance as indicated by the family history, the absence or limitation of breast-feeding, and the history or presence of other atopic disease in the patient, especially infantile eczema. The patient's age at onset of the asthma did not affect the long-term prognosis. Positive skin tests, whether single or multiple, were associated with the severity of the asthma at the onset, but not with the long-term prognosis. Complete relief from symptoms after twenty years is not synonymous with complete cure as there may be a reduction in the peak flow levels and modification of the lability index in such cases. References Blair H (1969) Proceedings of the Royal Society of Medicine 62, 1008 Blair H (1974) Clinical Allergy 4, 389 Blair H (1977) Archives of Disease in Childhood 52, 613-619 Boesen I (1953) Acta paediatrica 42, 87 Fry J (1961) British Medical Journal i, 711 Goodall J F (1958) Journal of the Royal College of General Practitioners 1, 51 Horn M E C & Gregg I (1973) Chest 4, Suppl; p 445 Jacoby N M (1966) Lancet i, 1354
Allergy and wheezy children Michael F D'Souza MB MRCGP Cardiothoracic Institute, Brompton Hospital, London SW3
Allergy has developed into such an expanding discipline in the last few years that very few practising general clinicians could have had any hope of keeping up to date, so what follows is more an exercise in synthesizing some of the current theories and findings in this field, rather than a serious attempt at an authoritative review or indeed any display of original work. It is proposed, therefore, to attempt a delineation of both wheezing and allergy, to speculate on their interrelationship and, finally, to discuss how these concepts might influence current practice for the general practitioner looking after wheezing children. 0 1 41-0768/79/010048-07/$O 1.00/0
4.
1979 The
Royal Society of Medicine
Journal of the Royal Society of Medicine Volume 72 January 1979
49
Wheezing Many will now accept Forgacs' interpretation of wheezing (Forgacs 1967), namely that the musical rhonchus heard through the stethescope (whether polyphonic or monophonic) is produced in the same way as snoring. Both have their origins in the vibration of air as it passes through an intermittently closing airway. Mere narrowing of the airways does not engender wheeze nor indeed does the pitch of the wheeze bear any measurable relationship to the size of the airways involved. It goes without saying that wheezing can often be claimed to be present (particularly by patients filling out questionnaires) when there is no clinical evidence for its existence. Equally it is well known that during very severe asthmatic attacks there might be little or no wheeze, i.e. when the airways are completely blocked rather than intermittently closing. It is widely accepted that the airway closure that results in wheezing is mainly caused by three pathophysiological changes: smooth muscle bronchospasm; bronchial wall thickening and oedema; and intraluminal sticky secretions. It is still uncertain how much blame can be apportioned to each of these factors. Although most drug company advertisements seem concerned with the smooth muscle bronchospasm, the oedema and intraluminal secretions seem much the most important factors, particularly in prolonged or fatal attacks of wheezing. Allergy It was von Pirquet in 1906 who first used the word 'allergy' to describe 'a general concept of changed reactivity'. He observed that although on first contact with a substance that could cause an allergic reaction (allergen) initially there might be no reaction, on subsequent exposures with the same material the organism reacted, often violently; i.e. with respect to this allergen the organism had altered its reactivity or, in other words, had become sensitized. The parallel that this allergic capacity bore to the normal process of developing immunological defences was obvious, and with the term 'allergy' von Pirquet had coined a unifying term for both immunology and hypersensitivity. In recent years, Gell & Coombs (1968) have provided a very useful classification of all allergic reactions according to the nature of the immunological changes provoked in the organism by the allergen. Gell & Coombs Type I allergy, is the major group that most people think of when they use the term allergy. This is synonymous with 'acute anaphylaxis' or immediate hypersensitivity, so named because the reaction occurs within minutes of exposure to the causative allergen. A great deal is now known about the detailed mechanism of this reaction. Ishizaka et al. (1966) have demonstrated that the reaginic antibody involved in the Type I reaction is usually IgE. More recently, Parish (1970) has shown that a short-latency IgG might also be involved. Following work of Porter (1959), the overall basic structure of antibodies is now known to resemble a Y-shape. This is composed of two light and two heavy chains of amino acids. The spreading arms of this Y-shaped molecule are called the Fab portions while the upright is termed the Fc portion. As illustrated in Figure 1, IgE attaches itself by the footpiece of the Fc portion of its molecule to mast cells (up to 40 000 molecules can attach to each cell). The Fab arm-pieces have specific configurations that enable the antibody to bind tightly to its specific allergen. The result of this allergen-antibody union is that the mast cell is degranulated, thus secreting a large number of biological potent materials which have the effect of causing local oedema and muscle-constriction, and attracting basophils and eosinophils to the site of the
reaction. In Gell & Coombs Type II reaction it is the allergen that is part of a cell wall while the antibody is free in the serum. The classical situation where this type of reaction occurs is in rhesus-incompatibility. In Type III reactions both allergen and antibody are free of any cellular attachments. The complex they form when they bind together attracts the chain reaction of complement-fixation. This reaction is tissue-destructive, mediated by the lysosomes released from phagocytes (polymorphs). In some instances it may also trigger off the secretion of mast-cell granules, as in the Type I reaction. This Type III reaction has been equated with Arthus' phenomenon or 'late reaction', so termed because it takes some 5-6 hours after exposure to the allergen before it clinically manifests itself.
50
Journal of the Royal Society of Medicine Volume 72 January 1979
Other
\
*ine
Ig E
AST CELL *~~~~~~ ~ \ granulation * \* *-, J @ Beta-rec
Histamine
B
a
vagus
h
SRSA
ProstaglaQdins) (Serotonin) (Kinins) Eosinophilotaxins
00
05
Figure 1. The Gell & Coombs Type I allergic response (after Gell & Coombs 1968)
The last type of reaction is the Gell & Coombs Type IV reaction, w4hich in evolutionary terms is probably the oldest. What distinguishes this reaction is that instead of involving antibodies it is mediated by specifically-sensitized small lymphocytes. This reaction characteristically takes 18-48 hours to develop fully and consequently has been called the delayed reaction. It is responsible for such well known clinical reactions as tuberculin sensitivity and contact dermatitis. It is tissue-destructive and possibly is also able to cause mast-cell degranulation, although there is little evidence for this. All these reactions (and possibly others not yet delineated) may be triggered off singly or in combination in the sensitized organism, and may occasion disease when they are called 'hypersensitivity' reactions, or may prevent disease by, for example, neutralizing bacterial invasion, in which case they are called 'immune responses'. In the popular imagination and, indeed, in some medical writing as well, the term allergy has been used synonymously with hypersensitivity and in general the role of allergy in the wheezing patient has been regarded as predominantly a pathological one, even though its protective or disease-preventing functions may be of equal or greater significance.
Relationship of allergy and wheezing The two main types of reaction that have been implicated in the pathogenesis of wheeze are Type I and Type III. The evidence that Type III reactions cause the wheezing in diseases such as allergic bronchopulmonary aspergillosis would appear to be somewhat contentious. In a review of some 50 patients with this condition (Safirstein et al. 1973), it was found that although many patients had severe bronchial damage due to type III activity as manifested by extensive proximal bronchiectasis, their asthmatic symptoms were very slight. Furthermore, in these patients there was always evidence of Type I sensitivity. One of the main reasons for postulating the existence of Type III asthma was the observation that on bronchial-challenge testing a delayed wheeze may occur 5-6 hours after challenge (Pepys et al. 1968). This reaction is often less responsive to simple bronchodilators than the immediate wheeze, but it does respond to corticosteroid therapy. Although these features reflect what would be expected from a Type III reaction, there are alternative theoretical explanations for these phenomena such as the local persistence of allergen after challenge and possibly also the consequences of basophils migrating to the site of the reaction (Figure 1) (Kimura et al. 1974). However the main reason for doubting the Type III asthma theory is that, apart from in allergic bronchopulmonary aspergillosis, there is a singular lack of bronchial tissue damage in most cases of even severe asthma. So although conceding that there might be
Journal of the Royal Society of Medicine Volume 72 January 1979
51
some effect due to Type III and even Type IV reactions, this discussion will be confined to the role of allergy in asthma due to Type I reactions.
Unstable mast cell and other factors The consequences of mast-cell degeneration alone seem ample cause for all the observed pathology of wheezing. However, the clinical effects of atropine and observations on vagotomized allergic dogs do provide some evidence of a modest effect of smooth muscle bronchoconstriction mediated by the vagus, and inevitably there are other factors that can be causative, such as the oedema produced directly by infection. The unstable mast cell, however, can be advanced as being the cornerstone to the pathophysiology of most asthma, and the role that allergy plays in rendering the mast cell unstable is of considerable importance, perhaps even more so in childhood than in adult life. Three theoretically interacting factors which may originate wheeze are listed in Table 1. The Table 1. Factors that may determine wheeze Allergy (Type I and ? others) (a) Produce a 'trigger effect' (b) A 'predisposing effect' - mast-cell instability (c) 'Protective effect' - mast-cell stabilization by non-inhalant IgE; blocking IgG (d) Short latency IgG Mast cell and smooth muscle cell beta-receptor blocking etvent (a) ? Viral infection leading to beta-blockade and antibody adjuvant effect (b) ? Toxins and other environmental agents Genetic predisposition (a) ? Impairment of factors that limit allergic reactions (b) Sex effect (c) Idiosyncrasies
first and best understood remains Type I Allergy. It is clearly possible for this allergy to have two quite different actions. The first action is making the mast cell unstable in such a way that other normally innocuous factors like exercise and mild irritants might be sufficient to produce degranulation. The other action is that of triggering immediate or delayed degranulation itself. There is a third action in which Type I allergy is possibly protective against the development of asthma. In tropical countries the prevalence of childhood asthma has frequently been reported to be very low (Warrell et al. 1975). The characteristic of these populations is endemic parasitic infestation with very high levels of IgE. It has been convincingly reasoned that the IgE specific to the worm allergens has occupied the majority of the available bronchial mast cells and thereby deprived that IgE specific to inhaled allergens of a footing. Thus there is less chance of IgE/allergen combination on the bronchial mast cells which would render them unstable or induce degranulation directly, although this has not been supported by all available evidence. It has been speculated (teleologically), that the whole of the Type I allergic system might well have evolved for the purpose of coping with problems such as superficial parasitic infestation the itching in the skin promoting scratching and possible removal of skin parasites and, likewise, the itching in the respiratory tract, with its consequent sneezing and coughing, helping to expel local respiratory invaders. It is, however, a little difficult to view the full-blown wheezing reaction as being particularly useful for survival. It seems likely that although the oedema fluid is undoubtedly useful for improving local immunity, the wheeze may be due to the absence of some control mechanisms which normally prevent this reaction going too far. This somewhat simplistic view of the pathophysiology of wheeze is far from complete. The
52
Journal of the Royal Society of Medicine Volume 72 January 1979
relationship of the various drugs to the physiology of the cells involved in these reactions is only partially understood. The role of adenyl cyclase and cyclic adenosine phosphate might well be only a part of the picture, since guanyl cyclase and cyclic guanosine monophosphate appear to have the opposite action on cells and so appear to be equally important (Goldberg et al. 1975). Why some people show large allergic reactions in their skin or their noses and yet do not wheeze, or vice versa, is still a mystery. It may well be that some important control mechanisms for limiting the allergic reactions are in fact organ specific. The second factor in Table 1 derives largely from the work of Szentivanyi (1971) who observed that the injection of smooth cultures of Bordetella pertussis into certain strains of rats induces transitory hypersensitivity. On further investigation of this phenomenon it appeared that this was due to two separate effects. Firstly it appeared that a beta-adrenergic blocking event had taken place (similar to that induced by propranolol). Secondly, there was an 'adjuvant effect' resulting in the increased production of antibody, particularly mast-cell sensitizing antibody. Interest obviously lies in establishing whether infections in man, particularly rhinovirus infections, may be associated with similar consequences. The final factor listed is genetic predisposition. This may well be involved in the other two, but the clinical phenomenon of wheezing is too varied to permit explanation solely by the first two factors (for example, the fact that boys have more asthma than girls cannot be explained on a purely allergic basis) and this final factor must be included for completeness. In Table 2 some of the more important other clinical factors known to precipitate wheeze are Table 2. Otherfactors implicated in wheezing Emotions (a) ? Act by direct vagal action (b) ? 'Release' effect on unstable mast cells
Infections (a) Possibly total cause of wheeze in small babies (tiny airways easily blocked) (b) Trigger effect on unstable mast cells Irritants (temperature changes, smoke, etc.) (a) Trigger effect on unstable mast cells Exercise
(a) Complex effect (why does swimming cause less wheeze than running?) (b) ? Combined emotional, 'shaking' and other factors causing trigger effect on unstable mast cells
listed with brief speculative notes on how they might generate wheeze. Particularly contentious is the role of emotions. In speculating that emotions exhibit at least some of their observed effects as a 'release phenomenon', it is implied that during an acute emotional experience there would be a generalized increase in corticosteroids and adrenalin etc., all of which would make wheezing less likely, but as the level of these hormones fell the unstable mast cell would be 'released' from their control suddenly and hence degranulate. Although this is purely speculative, such a theory would predict that the height of any wheezing attack associated with emotion would occur shortly after the worst of the emotional experience - a finding that, in general, accords with my clinical experience. Clinical implications Although it is enjoyable to struggle with biochemical jigsaw puzzles, the really important issue for the practising clinician is what difference these ideas are going to make to the management of the patient. In practical-terms the value of this knowledge would appear at present to be
Journal of the Royal Society of Medicine Volume 72 January 1979
53
limited. Although the pharmacological management of asthma has improved dramatically in the past ten years, there have been few or no specifically allergic therapeutic breakthroughs, though sodium cromoglycate (Intal) has been largely developed in the context of allergic research. Allergic diagnosis In my own management of the wheezy child, the first and probably most important use I make of such allergic concepts is in diagnosis. All wheezy children are investigated for possible allergic causes by careful history-taking. Enquiry directed towards identifying possible ingested or inhaled allergens occasionally produces immediate dividends with the discovery of a pet or old eiderdown as one of the chief precipitating causes of the problem. More often than not, however, even after a thorough history, one is left in considerable doubt as to the role played by allergens. Here I employ the skin prick test using the modified skin prick method (Pepys 1972). (In general, skin tests in children under five are not reliable; i.e. they may be skintest negative due only to their age and may become positive if retested at a later age). I personally use only five skin tests in my routine set: control, Aspergillus fumigatus, grass pollens, Dermatophagoides pteronyssinus and house dust. These were chosen as a consequence of a survey carried out at the Brompton Hospital on 656 asthmatics (Hendricks et al. 1975) which showed that just these five tests would identify 99% of those asthmatics positive to one or more of the 22 allergens routinely used. These five may be augmented by specific tests for cat, dog etc. as suggested by the history. Why should the clinician go to the bother of doing such skin prick tests? I believe this is not an easy question to answer. Certainly, as with other tests in medicine, the skin test is open to. abuse and gross misinterpretation, not least because we are still a long way from having 'pure' allergens to use for testing (or treatment) purposes. My reasons for using the tests are twofold. Firstly, I wish to identify a subject as atopic, by which I mean capable of producing a positive ( > 1 mm) skin prick test reaction to any of the four allergens. This is merely of use in making me more suspicious that any allergy may be of importance in causing the patient's disease. Davis (1976) in his excellent studies on children favours multiple positivity to more than one group of allergens to define atopy. I would agree that such a finding is very significant, but even one positive prick test has been shown to be associated with significantly higher levels of total serum IgE (D'Souza & Davies, 1977). The second reason I do these tests is to look for direct evidence that a specific allergen may be of relevance. It has been suggested in a number of studies that when a specific allergen is established by bronchial-challenge testing as being 'truly causative' in generating asthma, either a positive history or a positive prick test alone would have predicted only about 40% of the cases, but when both are shown to be positive at the same time, this predicts 80-90% of those positive on bronchial challenge. I do not favour the use of large numbers of skin tests in a blind search for allergic causes. The end result of such an exercise often discredits the true worth of prick testing and even today results in futile attempts at hyposensitization with allergen cocktails (a practice that resulted in allergic management being looked upon as a form of quackery and set back its scientific appraisal for many years). One of the main values of allergic investigation lies in helping the patient or his parents to understand more about the nature of his disability. Simply showing someone that his nerves are not all to blame for his wheezing can be of impressive benefit psychologically, as well as leading to a more rational treatment of the problems.
Allergic therapy The choice of allergic therapy quite naturally depends on the diagnosis. The use of sodium cromoglycate and specific allergen avoidance remains the mainstay of treatment. Food allergies which appear to be commonest in very young children are both nebulous to diagnose and difficult to treat. Milk-free diets etc. can be purgatory for mothers and are only rarely worth the effort compared with a direct pharmacological approach. The quantity of house dust and other inhalant allergens can be reduced in the environment of the wheezy child but never
54
Journal of the Royal Society of Medicine Volume 72 January 1979
totally eliminated. It is most important that common-sense prevails in matching the efforts at allergen elimination with the clinical benefits achieved. Hyposensitization has, in my clinical practice, very little place. I only regularly prescribe such injections for insect sting anaphylaxis where the results are excellent. Occasionally, pollen hyposensitization can benefit summer seasonal wheezing, but it can scarcely be looked upon as an ideal therapy. After a detailed study of mite hyposensitization (D'Souza et al. 1973), I am convinced that such treatment is clinically effective. However, the degree to which improvement occurs only rarely justifies the time, discomfort and risks of the treatment. More recently J 0 Warner (1977, personal communication) has confirmed that such injections are clinically effective, but he has emphasized that selection of patients should be by bronchial-challenge testing, which at present rules this therapy out of most routine clinical practice.
Summary Allergy must be seen as playing an important role in both predisposing children to wheeze and actually precipitating many attacks of asthma. The current specifically allergic treatments are, however, somewhat disappointing.
References Davis J B (1976) Clinical Allergy 6, 329-338 D'Souza M F & Davies R J (1977) American Review of Respiratory Diseases 115, 211 D'Souza M F, Pepys J, Wells I D, Tai E, Palmer F, Overell B G, McGrath I T & Megson M (1973) Clinical Allergy 3, 177-193 Forgacs P (1967) Lancet i, 203-205 Gell P G H & Coombs R R A (1968) Clinical Aspects of Immunology. 2nd edn. Blackwell, Oxford Goldberg N D, Haddox M K, Nicol S E, Sanford C H & Glass D B (1975) In: New Directions in Asthma. Ed. M Stein. American College of Chest Physicians, Illinois; p 103-124 Hendricks D J, Davies R J, D'Souza M F & Pepys J (1975) Thorax 30, 2-8 Ishizaka K, Ishizaka T & Hornbrook M M (1966) Journal ofImmunology 97, 840 Kimura I, Tanizaki Y, Takahashi K, Saito K, Ueda N & Sato S (1974) Clinical Allergy 4, 281-290
Parish W E (1970) Lancet ii, 591 Pepys J (1972) Proceedings of the Royal Society of Medicine 65, 271 Pepys J, Hargreaves F E, Chan M & McCarthy D S (1968) Lancet 2, 134 Porter R R (1959) Biochemical Journal 73, 119 Safirstein B H, D'Souza M F, Simon G, Tai E H & Pepys J (I1973) American Review of Respiratory Diseases 108, 450459
Szentivanyi A (1971) In: Immunological Diseaes. 2nd edn. Ed. M Samter et al. Little Brown, Boston; p 356-374 von Pirquet C (1906) Munchen Medizinische Wochenschrift 54, 1457 Warrell D A, Fawcett I W, Harrison B D W, Agamah A J, Ibu J O, Pope H M & Maberly D J (1975) Quarterly Journal of Medicine 174, 325-347
Infections in the wheezy child D G Sims MB MRCP1
Booth Hall Children's Hospital, Manchester M9 2AA
Viruses are the pathogens most commonly associated with wheezing in childhood. Table 1 shows viruses demonstrated by immunofluorescence or culture techniques in the respiratory tracts of 2225 infants and 2132 older children over a 5-year period to April 1974 in Newcastle upon Tyne. Respiratory syncytial (RS) virus was found in 83.5% of virus-positive infants who wheezed but in only 42% of older children. Of the infants with RS virus 75.8% wheezed, but only 46.8% of older children did so. Parainfluenza viruses were the next most common viruses found in wheezy infants and children. Influenza, like RS virus a winter pathogen, accounted for only 1.500 of those wheezy infants in whom a virus was demonstrated, and only 1 7.6% infected with influenza wheezed. The cause of the greatly increased relationship of wheezing with RS virus in infancy is unknown but McIntosh (1976) has recently reviewed current theories. 1
Present address: Oldham & District General Hospital, Oldham OLI 2JH
0 141-0768/79/010054-03/$01.00/0
,lr-"I 1979 The Royal Society of Medicine
Journal of the Royal Society of Medicine Volume 72 January 1979
Discussion In 1958 Goodall reported on a 20-years' retrospective study of children in a Yorkshire practice; 22.5% wheezed at some time, 47% of whom wheezed only once, and 5% of the study eventually developed asthma. In Fry's 10-years' prospective follow up of children in a South London practice (Fry 1961), 16% wheezed at some time, of whom 63% wheezed only once, and 8% of the study wheezed three to four times. However, the incidence of asthma in this practice was 8% in-children in the 5-15 years' age range and 6% in the 0-20 years' age range, studied over a period from 1970-73 (Blair 1974). Horn & Gregg (1973) found after a five-years' follow up that at least a third of young children with wheezing or asthmatic bronchitis with no constitutional factor (no history or presence of eczema or allergic rhinitis, no positive skin tests and no family history of atopic disease) were undoubtedly potential asthmatics triggered by infection and, they presumed, with an immunological disturbance. These findings confirmed Boesen's (1953) results. The natural history of childhood asthma may be best determined by a prospective long-term follow up. Short-term follow ups of such patients will fail to include those patients whose asthma remits and subsequently relapses - not necessarily triggered by the same factors as initially. Retrospective follow-up studies, especially if incomplete, are of limited value as patients' memories cannot be relied upon. Not all children outgrow their asthma; in this series 48% of childhood asthmatics seen between the years 1948 and 1952 were still asthmatic twenty years later. The factors, as shown in the various tables, which adversely affected the long-term prognosis, were the degree of inheritance as indicated by the family history, the absence or limitation of breast-feeding, and the history or presence of other atopic disease in the patient, especially infantile eczema. The patient's age at onset of the asthma did not affect the long-term prognosis. Positive skin tests, whether single or multiple, were associated with the severity of the asthma at the onset, but not with the long-term prognosis. Complete relief from symptoms after twenty years is not synonymous with complete cure as there may be a reduction in the peak flow levels and modification of the lability index in such cases. References Blair H (1969) Proceedings of the Royal Society of Medicine 62, 1008 Blair H (1974) Clinical Allergy 4, 389 Blair H (1977) Archives of Disease in Childhood 52, 613-619 Boesen I (1953) Acta paediatrica 42, 87 Fry J (1961) British Medical Journal i, 711 Goodall J F (1958) Journal of the Royal College of General Practitioners 1, 51 Horn M E C & Gregg I (1973) Chest 4, Suppl; p 445 Jacoby N M (1966) Lancet i, 1354
Allergy and wheezy children Michael F D'Souza MB MRCGP Cardiothoracic Institute, Brompton Hospital, London SW3
Allergy has developed into such an expanding discipline in the last few years that very few practising general clinicians could have had any hope of keeping up to date, so what follows is more an exercise in synthesizing some of the current theories and findings in this field, rather than a serious attempt at an authoritative review or indeed any display of original work. It is proposed, therefore, to attempt a delineation of both wheezing and allergy, to speculate on their interrelationship and, finally, to discuss how these concepts might influence current practice for the general practitioner looking after wheezing children. 0 1 41-0768/79/010048-07/$O 1.00/0
4.
1979 The
Royal Society of Medicine
Journal of the Royal Society of Medicine Volume 72 January 1979
49
Wheezing Many will now accept Forgacs' interpretation of wheezing (Forgacs 1967), namely that the musical rhonchus heard through the stethescope (whether polyphonic or monophonic) is produced in the same way as snoring. Both have their origins in the vibration of air as it passes through an intermittently closing airway. Mere narrowing of the airways does not engender wheeze nor indeed does the pitch of the wheeze bear any measurable relationship to the size of the airways involved. It goes without saying that wheezing can often be claimed to be present (particularly by patients filling out questionnaires) when there is no clinical evidence for its existence. Equally it is well known that during very severe asthmatic attacks there might be little or no wheeze, i.e. when the airways are completely blocked rather than intermittently closing. It is widely accepted that the airway closure that results in wheezing is mainly caused by three pathophysiological changes: smooth muscle bronchospasm; bronchial wall thickening and oedema; and intraluminal sticky secretions. It is still uncertain how much blame can be apportioned to each of these factors. Although most drug company advertisements seem concerned with the smooth muscle bronchospasm, the oedema and intraluminal secretions seem much the most important factors, particularly in prolonged or fatal attacks of wheezing. Allergy It was von Pirquet in 1906 who first used the word 'allergy' to describe 'a general concept of changed reactivity'. He observed that although on first contact with a substance that could cause an allergic reaction (allergen) initially there might be no reaction, on subsequent exposures with the same material the organism reacted, often violently; i.e. with respect to this allergen the organism had altered its reactivity or, in other words, had become sensitized. The parallel that this allergic capacity bore to the normal process of developing immunological defences was obvious, and with the term 'allergy' von Pirquet had coined a unifying term for both immunology and hypersensitivity. In recent years, Gell & Coombs (1968) have provided a very useful classification of all allergic reactions according to the nature of the immunological changes provoked in the organism by the allergen. Gell & Coombs Type I allergy, is the major group that most people think of when they use the term allergy. This is synonymous with 'acute anaphylaxis' or immediate hypersensitivity, so named because the reaction occurs within minutes of exposure to the causative allergen. A great deal is now known about the detailed mechanism of this reaction. Ishizaka et al. (1966) have demonstrated that the reaginic antibody involved in the Type I reaction is usually IgE. More recently, Parish (1970) has shown that a short-latency IgG might also be involved. Following work of Porter (1959), the overall basic structure of antibodies is now known to resemble a Y-shape. This is composed of two light and two heavy chains of amino acids. The spreading arms of this Y-shaped molecule are called the Fab portions while the upright is termed the Fc portion. As illustrated in Figure 1, IgE attaches itself by the footpiece of the Fc portion of its molecule to mast cells (up to 40 000 molecules can attach to each cell). The Fab arm-pieces have specific configurations that enable the antibody to bind tightly to its specific allergen. The result of this allergen-antibody union is that the mast cell is degranulated, thus secreting a large number of biological potent materials which have the effect of causing local oedema and muscle-constriction, and attracting basophils and eosinophils to the site of the
reaction. In Gell & Coombs Type II reaction it is the allergen that is part of a cell wall while the antibody is free in the serum. The classical situation where this type of reaction occurs is in rhesus-incompatibility. In Type III reactions both allergen and antibody are free of any cellular attachments. The complex they form when they bind together attracts the chain reaction of complement-fixation. This reaction is tissue-destructive, mediated by the lysosomes released from phagocytes (polymorphs). In some instances it may also trigger off the secretion of mast-cell granules, as in the Type I reaction. This Type III reaction has been equated with Arthus' phenomenon or 'late reaction', so termed because it takes some 5-6 hours after exposure to the allergen before it clinically manifests itself.
50
Journal of the Royal Society of Medicine Volume 72 January 1979
Other
\
*ine
Ig E
AST CELL *~~~~~~ ~ \ granulation * \* *-, J @ Beta-rec
Histamine
B
a
vagus
h
SRSA
ProstaglaQdins) (Serotonin) (Kinins) Eosinophilotaxins
00
05
Figure 1. The Gell & Coombs Type I allergic response (after Gell & Coombs 1968)
The last type of reaction is the Gell & Coombs Type IV reaction, w4hich in evolutionary terms is probably the oldest. What distinguishes this reaction is that instead of involving antibodies it is mediated by specifically-sensitized small lymphocytes. This reaction characteristically takes 18-48 hours to develop fully and consequently has been called the delayed reaction. It is responsible for such well known clinical reactions as tuberculin sensitivity and contact dermatitis. It is tissue-destructive and possibly is also able to cause mast-cell degranulation, although there is little evidence for this. All these reactions (and possibly others not yet delineated) may be triggered off singly or in combination in the sensitized organism, and may occasion disease when they are called 'hypersensitivity' reactions, or may prevent disease by, for example, neutralizing bacterial invasion, in which case they are called 'immune responses'. In the popular imagination and, indeed, in some medical writing as well, the term allergy has been used synonymously with hypersensitivity and in general the role of allergy in the wheezing patient has been regarded as predominantly a pathological one, even though its protective or disease-preventing functions may be of equal or greater significance.
Relationship of allergy and wheezing The two main types of reaction that have been implicated in the pathogenesis of wheeze are Type I and Type III. The evidence that Type III reactions cause the wheezing in diseases such as allergic bronchopulmonary aspergillosis would appear to be somewhat contentious. In a review of some 50 patients with this condition (Safirstein et al. 1973), it was found that although many patients had severe bronchial damage due to type III activity as manifested by extensive proximal bronchiectasis, their asthmatic symptoms were very slight. Furthermore, in these patients there was always evidence of Type I sensitivity. One of the main reasons for postulating the existence of Type III asthma was the observation that on bronchial-challenge testing a delayed wheeze may occur 5-6 hours after challenge (Pepys et al. 1968). This reaction is often less responsive to simple bronchodilators than the immediate wheeze, but it does respond to corticosteroid therapy. Although these features reflect what would be expected from a Type III reaction, there are alternative theoretical explanations for these phenomena such as the local persistence of allergen after challenge and possibly also the consequences of basophils migrating to the site of the reaction (Figure 1) (Kimura et al. 1974). However the main reason for doubting the Type III asthma theory is that, apart from in allergic bronchopulmonary aspergillosis, there is a singular lack of bronchial tissue damage in most cases of even severe asthma. So although conceding that there might be
Journal of the Royal Society of Medicine Volume 72 January 1979
51
some effect due to Type III and even Type IV reactions, this discussion will be confined to the role of allergy in asthma due to Type I reactions.
Unstable mast cell and other factors The consequences of mast-cell degeneration alone seem ample cause for all the observed pathology of wheezing. However, the clinical effects of atropine and observations on vagotomized allergic dogs do provide some evidence of a modest effect of smooth muscle bronchoconstriction mediated by the vagus, and inevitably there are other factors that can be causative, such as the oedema produced directly by infection. The unstable mast cell, however, can be advanced as being the cornerstone to the pathophysiology of most asthma, and the role that allergy plays in rendering the mast cell unstable is of considerable importance, perhaps even more so in childhood than in adult life. Three theoretically interacting factors which may originate wheeze are listed in Table 1. The Table 1. Factors that may determine wheeze Allergy (Type I and ? others) (a) Produce a 'trigger effect' (b) A 'predisposing effect' - mast-cell instability (c) 'Protective effect' - mast-cell stabilization by non-inhalant IgE; blocking IgG (d) Short latency IgG Mast cell and smooth muscle cell beta-receptor blocking etvent (a) ? Viral infection leading to beta-blockade and antibody adjuvant effect (b) ? Toxins and other environmental agents Genetic predisposition (a) ? Impairment of factors that limit allergic reactions (b) Sex effect (c) Idiosyncrasies
first and best understood remains Type I Allergy. It is clearly possible for this allergy to have two quite different actions. The first action is making the mast cell unstable in such a way that other normally innocuous factors like exercise and mild irritants might be sufficient to produce degranulation. The other action is that of triggering immediate or delayed degranulation itself. There is a third action in which Type I allergy is possibly protective against the development of asthma. In tropical countries the prevalence of childhood asthma has frequently been reported to be very low (Warrell et al. 1975). The characteristic of these populations is endemic parasitic infestation with very high levels of IgE. It has been convincingly reasoned that the IgE specific to the worm allergens has occupied the majority of the available bronchial mast cells and thereby deprived that IgE specific to inhaled allergens of a footing. Thus there is less chance of IgE/allergen combination on the bronchial mast cells which would render them unstable or induce degranulation directly, although this has not been supported by all available evidence. It has been speculated (teleologically), that the whole of the Type I allergic system might well have evolved for the purpose of coping with problems such as superficial parasitic infestation the itching in the skin promoting scratching and possible removal of skin parasites and, likewise, the itching in the respiratory tract, with its consequent sneezing and coughing, helping to expel local respiratory invaders. It is, however, a little difficult to view the full-blown wheezing reaction as being particularly useful for survival. It seems likely that although the oedema fluid is undoubtedly useful for improving local immunity, the wheeze may be due to the absence of some control mechanisms which normally prevent this reaction going too far. This somewhat simplistic view of the pathophysiology of wheeze is far from complete. The
52
Journal of the Royal Society of Medicine Volume 72 January 1979
relationship of the various drugs to the physiology of the cells involved in these reactions is only partially understood. The role of adenyl cyclase and cyclic adenosine phosphate might well be only a part of the picture, since guanyl cyclase and cyclic guanosine monophosphate appear to have the opposite action on cells and so appear to be equally important (Goldberg et al. 1975). Why some people show large allergic reactions in their skin or their noses and yet do not wheeze, or vice versa, is still a mystery. It may well be that some important control mechanisms for limiting the allergic reactions are in fact organ specific. The second factor in Table 1 derives largely from the work of Szentivanyi (1971) who observed that the injection of smooth cultures of Bordetella pertussis into certain strains of rats induces transitory hypersensitivity. On further investigation of this phenomenon it appeared that this was due to two separate effects. Firstly it appeared that a beta-adrenergic blocking event had taken place (similar to that induced by propranolol). Secondly, there was an 'adjuvant effect' resulting in the increased production of antibody, particularly mast-cell sensitizing antibody. Interest obviously lies in establishing whether infections in man, particularly rhinovirus infections, may be associated with similar consequences. The final factor listed is genetic predisposition. This may well be involved in the other two, but the clinical phenomenon of wheezing is too varied to permit explanation solely by the first two factors (for example, the fact that boys have more asthma than girls cannot be explained on a purely allergic basis) and this final factor must be included for completeness. In Table 2 some of the more important other clinical factors known to precipitate wheeze are Table 2. Otherfactors implicated in wheezing Emotions (a) ? Act by direct vagal action (b) ? 'Release' effect on unstable mast cells
Infections (a) Possibly total cause of wheeze in small babies (tiny airways easily blocked) (b) Trigger effect on unstable mast cells Irritants (temperature changes, smoke, etc.) (a) Trigger effect on unstable mast cells Exercise
(a) Complex effect (why does swimming cause less wheeze than running?) (b) ? Combined emotional, 'shaking' and other factors causing trigger effect on unstable mast cells
listed with brief speculative notes on how they might generate wheeze. Particularly contentious is the role of emotions. In speculating that emotions exhibit at least some of their observed effects as a 'release phenomenon', it is implied that during an acute emotional experience there would be a generalized increase in corticosteroids and adrenalin etc., all of which would make wheezing less likely, but as the level of these hormones fell the unstable mast cell would be 'released' from their control suddenly and hence degranulate. Although this is purely speculative, such a theory would predict that the height of any wheezing attack associated with emotion would occur shortly after the worst of the emotional experience - a finding that, in general, accords with my clinical experience. Clinical implications Although it is enjoyable to struggle with biochemical jigsaw puzzles, the really important issue for the practising clinician is what difference these ideas are going to make to the management of the patient. In practical-terms the value of this knowledge would appear at present to be
Journal of the Royal Society of Medicine Volume 72 January 1979
53
limited. Although the pharmacological management of asthma has improved dramatically in the past ten years, there have been few or no specifically allergic therapeutic breakthroughs, though sodium cromoglycate (Intal) has been largely developed in the context of allergic research. Allergic diagnosis In my own management of the wheezy child, the first and probably most important use I make of such allergic concepts is in diagnosis. All wheezy children are investigated for possible allergic causes by careful history-taking. Enquiry directed towards identifying possible ingested or inhaled allergens occasionally produces immediate dividends with the discovery of a pet or old eiderdown as one of the chief precipitating causes of the problem. More often than not, however, even after a thorough history, one is left in considerable doubt as to the role played by allergens. Here I employ the skin prick test using the modified skin prick method (Pepys 1972). (In general, skin tests in children under five are not reliable; i.e. they may be skintest negative due only to their age and may become positive if retested at a later age). I personally use only five skin tests in my routine set: control, Aspergillus fumigatus, grass pollens, Dermatophagoides pteronyssinus and house dust. These were chosen as a consequence of a survey carried out at the Brompton Hospital on 656 asthmatics (Hendricks et al. 1975) which showed that just these five tests would identify 99% of those asthmatics positive to one or more of the 22 allergens routinely used. These five may be augmented by specific tests for cat, dog etc. as suggested by the history. Why should the clinician go to the bother of doing such skin prick tests? I believe this is not an easy question to answer. Certainly, as with other tests in medicine, the skin test is open to. abuse and gross misinterpretation, not least because we are still a long way from having 'pure' allergens to use for testing (or treatment) purposes. My reasons for using the tests are twofold. Firstly, I wish to identify a subject as atopic, by which I mean capable of producing a positive ( > 1 mm) skin prick test reaction to any of the four allergens. This is merely of use in making me more suspicious that any allergy may be of importance in causing the patient's disease. Davis (1976) in his excellent studies on children favours multiple positivity to more than one group of allergens to define atopy. I would agree that such a finding is very significant, but even one positive prick test has been shown to be associated with significantly higher levels of total serum IgE (D'Souza & Davies, 1977). The second reason I do these tests is to look for direct evidence that a specific allergen may be of relevance. It has been suggested in a number of studies that when a specific allergen is established by bronchial-challenge testing as being 'truly causative' in generating asthma, either a positive history or a positive prick test alone would have predicted only about 40% of the cases, but when both are shown to be positive at the same time, this predicts 80-90% of those positive on bronchial challenge. I do not favour the use of large numbers of skin tests in a blind search for allergic causes. The end result of such an exercise often discredits the true worth of prick testing and even today results in futile attempts at hyposensitization with allergen cocktails (a practice that resulted in allergic management being looked upon as a form of quackery and set back its scientific appraisal for many years). One of the main values of allergic investigation lies in helping the patient or his parents to understand more about the nature of his disability. Simply showing someone that his nerves are not all to blame for his wheezing can be of impressive benefit psychologically, as well as leading to a more rational treatment of the problems.
Allergic therapy The choice of allergic therapy quite naturally depends on the diagnosis. The use of sodium cromoglycate and specific allergen avoidance remains the mainstay of treatment. Food allergies which appear to be commonest in very young children are both nebulous to diagnose and difficult to treat. Milk-free diets etc. can be purgatory for mothers and are only rarely worth the effort compared with a direct pharmacological approach. The quantity of house dust and other inhalant allergens can be reduced in the environment of the wheezy child but never
54
Journal of the Royal Society of Medicine Volume 72 January 1979
totally eliminated. It is most important that common-sense prevails in matching the efforts at allergen elimination with the clinical benefits achieved. Hyposensitization has, in my clinical practice, very little place. I only regularly prescribe such injections for insect sting anaphylaxis where the results are excellent. Occasionally, pollen hyposensitization can benefit summer seasonal wheezing, but it can scarcely be looked upon as an ideal therapy. After a detailed study of mite hyposensitization (D'Souza et al. 1973), I am convinced that such treatment is clinically effective. However, the degree to which improvement occurs only rarely justifies the time, discomfort and risks of the treatment. More recently J 0 Warner (1977, personal communication) has confirmed that such injections are clinically effective, but he has emphasized that selection of patients should be by bronchial-challenge testing, which at present rules this therapy out of most routine clinical practice.
Summary Allergy must be seen as playing an important role in both predisposing children to wheeze and actually precipitating many attacks of asthma. The current specifically allergic treatments are, however, somewhat disappointing.
References Davis J B (1976) Clinical Allergy 6, 329-338 D'Souza M F & Davies R J (1977) American Review of Respiratory Diseases 115, 211 D'Souza M F, Pepys J, Wells I D, Tai E, Palmer F, Overell B G, McGrath I T & Megson M (1973) Clinical Allergy 3, 177-193 Forgacs P (1967) Lancet i, 203-205 Gell P G H & Coombs R R A (1968) Clinical Aspects of Immunology. 2nd edn. Blackwell, Oxford Goldberg N D, Haddox M K, Nicol S E, Sanford C H & Glass D B (1975) In: New Directions in Asthma. Ed. M Stein. American College of Chest Physicians, Illinois; p 103-124 Hendricks D J, Davies R J, D'Souza M F & Pepys J (1975) Thorax 30, 2-8 Ishizaka K, Ishizaka T & Hornbrook M M (1966) Journal ofImmunology 97, 840 Kimura I, Tanizaki Y, Takahashi K, Saito K, Ueda N & Sato S (1974) Clinical Allergy 4, 281-290
Parish W E (1970) Lancet ii, 591 Pepys J (1972) Proceedings of the Royal Society of Medicine 65, 271 Pepys J, Hargreaves F E, Chan M & McCarthy D S (1968) Lancet 2, 134 Porter R R (1959) Biochemical Journal 73, 119 Safirstein B H, D'Souza M F, Simon G, Tai E H & Pepys J (I1973) American Review of Respiratory Diseases 108, 450459
Szentivanyi A (1971) In: Immunological Diseaes. 2nd edn. Ed. M Samter et al. Little Brown, Boston; p 356-374 von Pirquet C (1906) Munchen Medizinische Wochenschrift 54, 1457 Warrell D A, Fawcett I W, Harrison B D W, Agamah A J, Ibu J O, Pope H M & Maberly D J (1975) Quarterly Journal of Medicine 174, 325-347
Infections in the wheezy child D G Sims MB MRCP1
Booth Hall Children's Hospital, Manchester M9 2AA
Viruses are the pathogens most commonly associated with wheezing in childhood. Table 1 shows viruses demonstrated by immunofluorescence or culture techniques in the respiratory tracts of 2225 infants and 2132 older children over a 5-year period to April 1974 in Newcastle upon Tyne. Respiratory syncytial (RS) virus was found in 83.5% of virus-positive infants who wheezed but in only 42% of older children. Of the infants with RS virus 75.8% wheezed, but only 46.8% of older children did so. Parainfluenza viruses were the next most common viruses found in wheezy infants and children. Influenza, like RS virus a winter pathogen, accounted for only 1.500 of those wheezy infants in whom a virus was demonstrated, and only 1 7.6% infected with influenza wheezed. The cause of the greatly increased relationship of wheezing with RS virus in infancy is unknown but McIntosh (1976) has recently reviewed current theories. 1
Present address: Oldham & District General Hospital, Oldham OLI 2JH
0 141-0768/79/010054-03/$01.00/0
,lr-"I 1979 The Royal Society of Medicine
