INTRODUCTION

There are several crucial questions to be answered about asthma. What is the basic pathological process? Why is asthma prevalence increasing? Why does it aggregate in families, and are these familial factors central to aetiology and pathogenesis? Are there asthma genes, and what would the gene products be? Is asthma heterogenous in nature? That is, should we refer to asthmas, rather than asthma?

Before even attempting to answer these questions, we must first summarize the longstanding arguments about the definition of asthma.

Definitions

It has been traditionally stated that to perform science one must be able to quantify an entity, and in order to do this one must be able to define it. Asthma is one condition where some difficulty has been met in producing a consistent common definition. This is a problem with any diagnostic category, defined in terms of symptoms experienced by a patient, where the underlying pathological processes are not fully understood. Within the syndrome called asthma, a set of symptoms commonly accepted as a unique disease in some individuals, for example exercise induced wheeze in children, might be due to other known pathological processes in individuals such as elderly smokers. Furthermore the episodic nature of symptoms experienced by patients quite reasonably labelled as asthmatic by a physician, might well mean that even quite discriminating physiological measures might, at an inappropriate time, fail to demonstrate pathological change in lung function [Sears 1986; Salome et al 1987]. A further point is that the dichotomous trait asthma is usually defined in terms of continuous measures such as bronchial responsiveness and response to therapy - diagnostic cutpoints based on such measures will invariably lead to false positives and false negatives.

In terms of population based epidemiology, the search has been for a set of questions that might reliably and validly replicate the clinical diagnosis of asthma, and extend that to milder states that might precede asthma. In terms of clinical diagnosis, especially in the paediatric field, the definition of asthma has been extended to include syndromes such as persistent nocturnal cough in the absence of wheeze (cough variant asthma), or cough and wheeze exclusively associated with lower respiratory tract infection - largely on the basis of epidemiological work [William & McNichol 1969], and on therapeutic response. In cough variant asthma, Koh et al [1993] have recently shown that these patients differ from "classical" asthma merely in the fact that they do not audibly wheeze until a higher degree of bronchoconstriction is present - a cardinal symptom does not reflect identical airway narrowing in different individuals.

Definitions of asthma

Older definitions:

The word "asthma" or "asma" derives from the "azo" or "azein" meaning "panting" or "breathing hard". Hippocrates seems to use the term to denote all types of dyspnoea, saying it became more common with increasing age. Paulus Aeginata [Adams 1844] summarises the classical (Galenical) understanding of the disease as "those who breathe thick without fever, like those who have run fast, are said to be asthmatic, that is to say, to pant for breath; and from their being obliged to keep the chest erect for fear of being suffocated, they are called orthopnoic. The affection arises from thick and viscid humours becoming infarcted in the lungs. Dyspnoea is a common symptom which accompanies these and many other complaints". Arataeus of Cappadocia describes "a rale even when awake, and still more so when asleep...constant and difficult coughing, with little expectoration", which can be brought on by "...running, gymnastic exercise, or any other work", and may be intermittent in nature [Adams 1844; Sakula 1988]. Treatment included squills, oxymel, arsenicals, and belladonna. It seems likely from the above that cardiac asthma was not fully differentiated from bronchial asthma.

Floyer in 1698 wrote "A treatise of the asthma" [Sakula 1988], differentiating the "periodic asthma" that he suffered, due to "Constriction of the Bronchia", from "continued asthma". He described this condition as being characterised by "fits of asthma", with expiratory obstruction and "Wheezing noise", and commented "I have observed the fit always to happen after sleep in the night". He also states that he was thankful that his asthma was not inherited from his ancestors, and so was not passed onto his sons.

Heberden's [1802] description of a century later emphasises many of the features considered characteristic in modern descriptions of the syndrome. "The first fit of the asthma has been experienced at all times, from the earliest infancy to extreme old age...Pleurisies, peripneumonies, and frequent catarrhs, often end in this distemper. A mal-conformation of the breast [? increased AP diameter], and a cough returning every winter, and becoming worse and worse, bring on an incurable asthma. In some it comes on suddenly, without any manifest previous illness. Very violent fits of it will allow long intervals of apparently perfect freedom from any difficulty of breathing...In some it never fails to be brought on in certain situations, or houses, and is not felt in others; though at so small a distance, that it is difficult to satisfy ourselves in fixing upon the circumstance which could make the difference. There have been those who have lived with an asthma for fifty years; and others have died of it in a few months. Some few constitutions have of themselves either outgrown, or, assisted by some judicious methods of cure, have entirely conquered the asthma. It is usually, but not universally, attended with a cough...Whenever there is any degree of asthma, it rarely fails of showing itself just upon waking out of the first sleep... Several asthmas cannot bear the country air...but the far greater number are impatient of cities, and are always easier in the country...So summer is to not a few the time of their breathing with most difficulty; though winter be most generally the dangerous season". He summarises that asthma differs from pulmonary tuberculosis for example in the "violent fits of suffocation", the intermittent nature of the disease, and the heterogeneity of precipitants such as sleep or extreme emotions.

Peshkin, Rackemann, Taub and others:

Swineford [1965] has reviewed definitions of asthma put forward this century. He cites Peshkin (1926) as the start of "this important controversy". This definition reads "a clinical syndrome, quite readily diagnosed on clinical examination...a recurring dyspnoea, more marked on expiration and associated with wheezing...[the] exciting cause of which may or may not be protein sensitisation [ie allergy]". Rackemann, who formulated the intrinsic/extrinsic asthma dichotomy, felt that "it is best to regard the word as denoting a symptom...in the same category as headache or nausea, [though] it can almost be regarded as a disease entity...produced by a wide variety of causes...operating to cause bronchial spasm". Taub (1945) suggested "a symptom complex characterised by attacks of dyspnoea followed by wheezing and coughing, whose aetiology is hypersensitivity [allergy]". Herxheimer (1952) felt that "bronchial asthma means a transient and usually repeated attack of breathlessness caused by bronchial obstruction...due to spasm of the bronchial muscle, or to secretion into the bronchial lumen, or to oedema of the mucous membrane, or to any combination of these factors".

Ciba symposium:

The 1959 CIBA Guest Symposium was an important milestone in respiratory epidemiology. The modern definitions for a number of chronic lung diseases derive from this meeting. Asthma was defined as [Swineford 1965]:

"...a widespread narrowing of the bronchial airways, which changes in severity over short periods of time either spontaneously or under treatment, and is not due to cardiovascular disease".

Some authors have derived an acronym based on this definition viz. Variable Obstructive Intrabronchial Disease, in imitation of the more accepted Chronic Obstructive Pulmonary Disease.

American Thoracic Society:

The American Thoracic Society produced the most widely quoted modern definition of asthma in 1963, revised in 1987 [ATS 1987]. This reads:

"Asthma is a clinical syndrome characterised by increased responsiveness of the tracheobronchial tree to a variety of stimuli. The major symptoms of asthma are paroxysms of dyspnoea, wheezing, and cough, which may vary from mild and almost undetectable to severe and unremitting (status asthmaticus). The primary physiological manifestation of this hyperresponsiveness is variable airways obstruction. This can take the form of spontaneous fluctuations in the severity of obstruction, substantial improvements in the severity of obstruction following bronchodilators or corticosteroids, or increased obstruction caused by drugs or other stimuli."

Points to note in this definition are that it is operational, no mention is made of possible pathogenesis, and allergic phenomena are not mentioned except as "other stimuli". This differs from older definitions where allergy is sometimes taken as a sine qua non. The 1962 definition has an extensive coda, and mentions "the basic defect in asthma appears to be an alteration in the host...[which] in some instances seems clearly related to an altered immunological state such as atopy...[though] in others the underlying cause for the defect cannot be determined". This earlier definition also included some exclusion criteria, some of which might be questioned eg acute bronchitis.

Other definitions:

Epidemiologists have used several operational definitions of asthma. A widely used one is "doctor-diagnosed" asthma - where the individual has been told by a doctor that they have asthma. More seriously, a self-reported history of intermittent wheezing is used, or history of wheezing plus bronchial hyperresponsiveness on testing [Peat et al 1992, 1993]. Two other definitions are worth examining.

"Eosinophilic bronchitis":

Some workers have suggested that we have sufficient understanding of asthma to define it pathologically, and have coined the term "eosinophilic bronchitis". Unfortunately this term has been used to denote other disease ie steroid responsive chronic cough with sputum eosinophilia and normal bronchial responsiveness. This still leaves the problem of identifying a non-invasive diagnostic method to validate using this definition. Induced (eg by hypertonic saline inhalation) sputum cytology looking for the presence of eosinophils is one method currently being explored.

"CARA":

The definitions so far examined all stem essentially from the 1959 CIBA symposium's division of respiratory disease. Many (mainland) European respiratory physicians have, by contrast, not clearly distinguished between chronic bronchitis and asthma and a composite entity "Chronische Aspecificke Respiratoire Aandoeningen" (CARA) has been formulated. This reflects an underlying difference between the natural histories of COPD posited by the "Dutch" hypothesis [Sparrow et al 1988; Sluiter et al 1991], where allergic and/or childhood processes underlie susceptibility to irreversible airflow limitation as well as to asthma, and the "British" hypothesis where pollutants lead to COPD while asthma is (more) related to allergic disease. Asthmatics do have an accelerated age-related decline in lung function, and so can develop chronic obstruction, but smoking related airways disease (chronic bronchitis and small airway disease) differs pathologically in several ways from asthma [Jeffery 1991; Burrows 1991].

Definition of bronchial hyperresponsiveness

Operational:

As the ATS definition of asthma suggests, asthma is closely associated with dynamic changes in the calibre of airways. These changes can be of a rapid nature, mediated by bronchial smooth muscle constriction and relaxation, or slowly reversible, caused by increased mucus secretion, bronchial wall oedema or peribronchial vasodilatation. Bronchial responses can be nonspecific, that is provoked by a variety of pharmacologically or chemically active molecules (eg histamine, sulphur dioxide), or by physical or physicochemical agents (exercise, hyper/hypo- osmolar mists); or they can be specific to a given allergen.

Bronchial responsiveness is quantified as the alteration in lung function, mainly expiratory flow, provoked by the given agent. It is thus a continuous variable, and is continuously distributed in the population. Normal values for bronchial responsiveness must be determined. One method is to use statistical criteria based on randomly selected population samples, such as the 95th percentile of the frequency distribution [Hopp et al 1985]. It is found that asthmatics tend to differ significantly from the rest of the population [Cockcroft 1977, 1992; Salome et al 1987], responding to a greater extent to the same stimulus than nonasthmatics.

Via association with clinically diagnosed asthma:

A normative approach such as this will detect some individuals who have high levels of bronchial responsiveness but who would not be diagnosed as asthmatic by the usual clinical criteria. The alternative approach is to perform some type of discriminant analysis using measured bronchial responsiveness to predict whether a subject would be clinically diagnosed as asthmatic or not such as Popa and Singleton [1988], or Gilbert and Auchincloss [1990]. Josephs et al [1990] reviewed studies of bronchial responsiveness in different classes of severity of asthma, and concluded that a linear relationship existed. In a longitudinal study however, Josephs et al [1989] found that the level of bronchial responsiveness was directly but only partially correlated to severity of asthmatic symptoms experienced in the immediate period preceding testing.

Association with other respiratory disease:

Nonspecific bronchial responsiveness is increased in a number of other disease states. Specific bronchial hyperresponsiveness is only seen in asthma and allergic disease, and is in fact quite specific to asthma.

COPD:

Individuals suffering from COPD exhibit higher than normal levels of nonspecific bronchial responsiveness [Sparrow et al 1988]. This leads to a higher prevalence of hyperresponsiveness in the elderly compared to those in midlife, and in smokers compared to nonsmokers - probably mediated by small airway inflammation.

Allergy:

A history of allergic rhinitis or eczema is associated with an elevated level of bronchial responsiveness [Townley et al 1975]. The pattern of response to increasing dose of methacholine is often quite specific, exhibiting a "plateau" phenomenon, or level of response that cannot be increased by further increase in drug dose. Other authors [Woolcock et al 1984; Macklem 1990] state all nonasthmatics show a plateau; therefore the plateau occurs at a lower dose in allergic nonasthmatics than those completely "normal". Dav‚ et al [1990] have shown that allergic nonasthmatic subjects will demonstrate persistent bronchial responsiveness for up to 11 years in 60- 70% of cases.

Heart Disease:

Bronchial hyperresponsiveness has been demonstrated to be present in congestive cardiac failure, whether due to ischaemic myocardial disease or valvular disease. In the case of mitral stenosis, bronchial responsiveness is correlated with pulmonary artery pressure, but mitral valve replacement leads only to partial improvement.

Definition of allergy

Von Pirquet was the first to use the term "Allergie" in 1906, derived from "allos", and implying altered reactivity [Bahna 1989]. He had previously published a monograph on serum sickness, and used the term allergy to denote the fact that exposure of animals to antigens rather than inducing tolerance could lead to hypersensitivity on certain occasions. Other workers subsequently drew a parallel between asthma in humans and anaphylaxis in guinea pigs to suggest that the human disease was an anaphylactic phenomenon. In modern usage, it refers to any IgE mediated (Coombs Type I) hypersensitivity.

Atopic allergy (Coca & Cooke):

In 1922, Arthur Coca and Robert Cooke presented [reached print 1923] a classification of immune "hypersensitiveness" that they felt was more useful than earlier nosologies. This was divided into two broad categories: Normal, such as "ordinary" serum sickness (where rechallenge often leads to a more severe reaction); and Abnormal, that is anaphylaxis, hypersensitiveness of infection, and idiosyncratic hypersensitiveness or atopy. The latter term was coined (atopia, "a strange disease") to designate the asthma and hayfever group of diseases that were

  1. exhibited against non-precipitinogenic as well as precipitinogenic substances;
  2. greatly lessened but not completely removed by the suitable injection of the active substance (hyposensitisation);
  3. inherited, via an autosomal dominant gene [Cooke & Vander Veer, 1916];
  4. expressed in pathological changes the more important of which are different from those of the anaphylactic reactions in the guinea-pig, the rabbit or the dog;
  5. not passively transferable to normal individuals with the blood of sensitive individuals - "unfortunately" undermined by the discovery of the Prausnitz-Kustner test in 1922.

The critical factors distinguishing atopic disease from other immune mediated diseases were the absence of an antigen-antibody mechanism as was understood at that time, and the familial clustering of these diseases. Cooke and Vander Veer [1916] had suggested from their family studies that asthma, allergic rhinitis, urticaria, angioedema and acute gastroenteritis due to food were definitely atopic, and "deferred a decision regarding eczema, migraine and epilepsy which had been suggested for inclusion by others" [Nelson 1985]. The classification was heavily criticised, and Coca was soon forced by his studies on ragweed allergy to revise this definition to include the presence of reagin. Reagin was finally identified as an immunoglobulin in 1967 (because it was not quantifiable by earlier methods - one ten thousandth the amount of IgG), but the genetics and clinical picture of these diseases still differentiate them from other hypersensitivities.

Asthma as an allergic disease

The modern definitions of asthma cited above make little or no mention of allergic processes. Burrows [1989] and others [eg Chapman et al 1988] have argued the more traditional viewpoint that regards asthma mainly as an allergic disease. Evidence for this standpoint is pathological, epidemiological and genetic.

The eosinophil:

This is an important effector cell in the pathological processes of asthma and in allergy (to be reviewed in Chapter 3). This is the case even in asthma where the individual does not express any signs of allergy or other hypersensitivity - intrinsic asthma.

Markers of allergy:

Skin prick test reactivity and total serum IgE levels are strongly associated with asthma. This is the case in children [Sears et al 1991], and in adults [Burrows et al 1989; Burrows et al, 1991; Ohman et al, 1993]. Burrows and coworkers [1991] have argued that the magnitude of these allergic markers in known atopic individuals falls off with age, and that this attenuation has led to disease in older individuals being labelled as nonallergic. They found that in elderly individuals in the Tucson Obstructive Lung Disease study developing asthma for the first time, appropriate age adjustment of sIgE showed a correlation between asthma and sIgE. Ohman et al [1993] have similarly found that new onset wheezing in men (mean age 64 years) in the Normative Aging study was associated with the presence in peripheral blood of house dust mite specific IgE in 15% of cases versus 7% of controls, although total sIgE was not elevated.

Bronchial challenge:

Aerosols of allergens that the subject exhibits a skin response to will lead to bronchoconstriction in up to 80% of asthmatic individuals. This offers the mechanism for the association between allergic sensitisation and asthma.

Genetics:

Family and twin studies show that asthma and allergic diseases arise at least partly from the same genes. This seems to be true even for asthmatics where no good clinical evidence exists for an allergic pathogenesis [Schwartz 1952].

Bronchial hyperresponsiveness:

Allergic diseases eg allergic rhinitis, are associated with increased bronchial responsiveness, even when overt asthma is not present. Sears [1991] found that markers of allergy were increased in children exhibiting bronchial hyperresponsiveness who had never expressed any allergic symptoms.

Allergic and perennial rhinitis

While recognisable descriptions of asthma date back to the Greeks, hayfever or allergic rhinitis seems to be a relatively modern condition, save in the case of that precipitated by roses - "the rose cold", where an aptly titled paper "on the reason why the heads of people swell at the time of roses and produce catarrh" appeared in the ninth century AD [Emanuel 1988]. Allergic rhinitis is characterised by the (protective) symptoms of rhinorrhea and coryza, as well as pruritis of the nose, pharynx and conjunctivae by aeroallergen exposure - classically pollens (hayfever, summer catarrh). The effect of allergen is most usually diagnosed anamnestically, by a history of temporal association with particular seasons, or exposures (eg cats).

Perennial rhinitis refers to non-seasonal or chronic symptoms of the same nature. This will be due in most cases to continual exposure to allergens such as house dust - perhaps three-quarters of patients [Pastorello et al 1985], but in many cases no allergy is demonstrable - "vasomotor rhinitis". This condition is more commonly associated with nasal polyps and involvement of the sinuses.

Atopic dermatitis

Atopic dermatitis or atopic eczema is the third condition making up the allergic triad. It differs from the other eczemas [Sampson 1992], as extrinsic asthma differs from intrinsic asthma, in having an early age-at-onset - 95% of all cases before the age of five years; an association with the other atopic conditions - only 20% of atopic dermatitis sufferers do not suffer from hayfever or asthma; and exacerbation by environmental allergens, including aeroallergens.

Hanifin and Rajka [1980; Larsen & Hanifin 1992] have defined diagnostic criteria for atopic dermatitis, based on traditional clinical major and minor features. These require (3 of 4 major features): a history of a chronic recurring rash; classically affecting (in order of specificity) the flexural creases of the elbows and knees, extensor aspects wrists and ankles, hands and feet, face and neck, and less so the trunk (with flexural lichenification or linearity in adults); pruritis, especially when sweating; and a personal and family history of (other) atopic disease. Svensson et al [1985] add seasonal variation in symptoms, and exacerbation by stress - though these are seen in other skin conditions. Clinical signs (minor features) of importance were xerosis, facial pallor or erythema, knuckle dermatitis, keratosis pilaris, nipple eczema (said to be quite specific by other authors), nummular eczematous patches, pompholyx, ichthyosis, and the Dennie-Morgan (suborbital) fold. Diepgen et al [1989] constructed scoring systems for these criteria using 110 atopic dermatitis patients and 527 controls. Their top five criteria turned out to be "itch when sweating" (Odds Ratio=33), "intolerance to wool" (OR=27), xerosis (OR=72), white dermatoglyphism (OR=42, a sign Svensson et al [1985] discarded as worthless), and Hertoghe's sign (absence or thinning of the lateral eyebrows, OR=63). They note that a family history of atopy was common in controls (though not as high as in cases OR=3.7), and so less specific. More recently, these criteria have been used to design a disease definition and a questionnaire by the UK Atopic Dermatitis Diagnostic Criteria Working Party [1994a, 1994b, 1994c, 1996]. Their final criteria are relatively simple, requiring responses by the patient or parent to five questions, and a clinical examination. These were:

  1. An itchy skin condition (or parental report of scratching or rubbing in a child)
  2. plus 3 or more of the following
    1. History of involvement of the skin creases such as folds of elbows, behind the knees, fronts of ankles or around the neck (including cheeks of children under 10
    2. A personal history of asthma or hayfever (or history of atopic disease in a first-degree relative in children under 4)
    3. A history of a general dry skin in the last year
    4. Visible flexural eczema (or eczema involving the cheeks/forehead and outer limbs in children under 4).
    5. Onset under the age of 2 (not used if child is under 4).

In a community-based validation study of 695 schoolchildren [Williams et al 1996], these criteria had a sensitivity of 80% and specificity of 97% (Youden's index=0.77), compared to an experienced dermatologist's history and examination supplemented by clinical records.

Chronic Obstructive Pulmonary Disease (COPD)

The Global Initiative for Chronic Obstructive Lung Disease has promulgated a definition of COPD [2001,2003] that chimes nicely with that of asthma:

"Based on current knowledge, a working definition of COPD is a disease state characterized by airflow limita- tion that is not fully reversible. The airflow limitation is usually both progressive and associated with an abnor- mal inflammatory response of the lungs to noxious parti- cles or gases. Symptoms, functional abnormalities, and complications of COPD can all be explained on the basis on this underlying inflammation and the resulting pathology."

Diagnosis is based on traditional criteria for chronic bronchitis (see next chapter) and simple spirometry. They then add,

"COPD can coexist with asthma, the other major chronic obstructive airway disease characterized by an underlying airway inflammation. Asthma and COPD have their major symptoms in common, but these are generally more variable in asthma than in COPD. The underlying chronic airway inflammation is also very different: that in asthma is mainly eosinophilic and driven by CD4+ T lymphocytes, while that in COPD is neutrophilic and characterized by the presence of increased numbers of macrophages and CD8+ T lymphocytes."