4 Allergy and Immunology
Disorders of the immune system are diverse and range from mild to severe in their manifestations and impact on normal function. In this chapter, we review the physical findings and characteristic symptoms of children with hypersensitivity reactions and immune deficiencies, as well as diagnostic techniques and radiographic findings. Topics have been chosen on the basis of their prevalence and importance in the pediatric population and their association with characteristic physical findings.
Hypersensitivity disorders of the human immune system have been classified by Gell and Coombs into four groups (Table 4-1). Type I reactions occur promptly after the sensitized individual is exposed to an antigen and are mediated by specific IgE antibody. Cross-linking of IgE on the surface of mast cells and basophils leads to release of histamine and other inflammatory mediators. This mechanism is responsible for the common disorders of immediate hypersensitivity, such as allergic rhinitis and urticaria. So-called “anaphylactoid” reactions are clinically similar, but are caused by degranulation of mast cells and basophils in the absence of specific IgE. Type II reactions involve antibodies directed against antigenic components of peripheral blood or tissue cells or foreign antigens, resulting in cell destruction. Examples of this type include autoimmune hemolytic anemia and Rh and ABO hemolytic disease of the newborn. In type III reactions, antigen–antibody complexes form and are deposited in the lining of blood vessels, stimulating tissue inflammation mediated by complement or activated white blood cells. Examples of this type of reaction are serum sickness and the immune complex–mediated renal diseases. Type IV reactions involve T cell–mediated tissue inflammation and typically occur 24 to 48 hours after exposure. Examples of this type are tuberculin (purified protein derivative, PPD) reactions and contact dermatitis (see Chapter 8).
The development of type I hypersensitivity depends on hereditary predisposition, sensitization by exposure to an antigen, and subsequent reexposure to the antigen leading to an allergic reaction. Antigens that stimulate allergic reactions are known as allergens, and the mechanism of allergen-induced mediator release in type I hypersensitivity reactions is shown in Figure 4-1. IgE antibodies directed toward specific allergens are bound to the high-affinity IgE receptor on mast cells and basophils. When allergen causes cross-linking of IgE antibodies on the cell surface, the cell becomes activated, leading to the release of preformed mediators and the generation of the early and late mediators of anaphylaxis. The preformed mediators include histamine, tryptase, chymase, heparin, and other proteases that drive the earliest symptoms of anaphylaxis. A serum tryptase level is currently the best biologic marker of anaphylaxis, but it is still a relatively insensitive test. Serum tryptase levels should be obtained close to the onset of anaphylaxis because levels peak in 1 hour and remain elevated for only 4 to 24 hours. The early and late mediators generated by mast cell activation include prostaglandins, leukotrienes, and cytokines. These mediators, which are generated over minutes to hours, continue to drive the clinical symptoms of the allergic reaction and initiate an inflammatory cascade that leads to the recruitment of eosinophils, basophils, and lymphocytes.
Figure 4-1 IgE is bound to the mast cell surface, and when cross-linked by antigen, the mast cell becomes activated. The activated mast cell releases preformed mediators and generates additional mediators over minutes to hours. FcεRI, high-affinity IgE receptor; IL-4 and IL-13, interleukin-4 and interleukin-13; TNF-α, tumor necrosis factor-α.
(From Broide DH: Molecular and cellular mechanisms of allergic disease, J Allergy Clin Immunol 108:S65-S71, 2001.)
Type I reactions may occur in one or more target organs including the upper and lower respiratory tracts, cardiovascular system, skin, conjunctivae, and gastrointestinal (GI) tract. Manifestations depend on the systems involved, as shown in Figure 4-2. The most common manifestation of type I reaction is seasonal allergic rhinitis with a prevalence of at least 25%. The most serious manifestation of type I hypersensitivity is anaphylaxis, which can simultaneously involve all of the organ systems mentioned above.
Type I hypersensitivity has been diagnosed by skin testing for more than 100 years. The percutaneous skin test, also known as either the scratch or prick test, is an in vivo method to detect the presence of IgE antibody to specific allergens. The skin prick test is the safest and most specific test and correlates best with symptoms. The skin prick test is typically performed with a plastic lancet on either the forearm or upper back and involves a superficial disruption of the epidermis that is nearly painless (Fig. 4-3). The test leaves a barely visible mark, and when performed properly, the prick site should not bleed. The test is interpreted after 15 to 20 minutes by measuring the maximal diameter of both the wheal and the flare. Skin test results are compared with a negative control, which is usually saline, and a positive control, which is typically histamine. Historically, intradermal tests have been considered to be more sensitive than prick tests, but the specificity is poor and these tests should be used only when ruling out allergic disease is essential.
Figure 4-3 Allergy prick tests. A, The skin prick test is typically performed with a plastic lancet on either the forearm or upper back. B, The test leaves a barely visible mark and should be nearly painless. C, The test is interpreted after 15 to 20 minutes by measuring the maximal diameter of both the wheal and the flare.
Allergy skin testing is contraindicated in four clinical situations: (1) when antihistamines have been used in the recent past—this will typically manifest as a negative histamine control; (2) when skin disease limits the area available for testing; (3) during either an asthma exacerbation or episode of anaphylaxis; and (4) when a patient is taking a β-blocking medicine, because these can interfere with epinephrine treatment in rare cases of test-induced anaphylaxis. When skin testing is not possible, in vitro testing is a good alternative. The in vitro tests can be accomplished with just a few milliliters of serum and also may be advantageous for some patients who have a difficult time sitting through allergy tests. In vitro allergy tests are more expensive and less sensitive than allergy prick tests (Table 4-2). In vitro tests are especially helpful in the evaluation of patients with possible food allergies; this is discussed further below.
|Variable||Skin Test||In Vitro|
|Risk of allergic reaction||Rare||No|
|Affected by antihistamines||Yes||No|
|Affected by corticosteroids||Not usually||No|
|Affected by extensive dermatitis or dermatographism||Yes||No|
|Broad selection of antigens||Yes||Yes|
Anaphylaxis results from widespread degranulation of mast cells after cross-linking of IgE on the mast cell surface. A clinically similar reaction in which mast cells degranulate without cross-linking of antigen-specific IgE is termed anaphylactoid. Anaphylaxis is a clinical diagnosis, and a report issued jointly by the National Institute of Allergy and Infectious Diseases and the Food Allergy and Anaphylaxis Network established criteria to define likely anaphylactic episodes (Table 4-3). Onset of anaphylaxis is typically rapid and often explosive after bee stings, drug administration, or food ingestion (Fig. 4-4). The pattern of organ system involvement can vary, based on the antigen, dose, and route of exposure, and can range from isolated urticaria to cardiovascular collapse (Fig. 4-5). Airway obstruction and hypotension are the most severe manifestations of anaphylaxis. The upper or lower airway, or both, can be affected. Upper airway obstruction is due to laryngeal edema, whereas lower airway involvement is due to edema and bronchospasm. Hypotension is caused by vasodilation, which may be complicated by loss of intravascular volume. Vascular collapse may be aggravated by decreases in myocardial function. In addition to the airway and cardiovascular system, other organ systems are also involved in anaphylaxis. The skin is the most commonly involved organ, with urticaria being nearly universal and angioedema often present. GI involvement can manifest as vomiting, diarrhea, and abdominal pain due to gut edema.
BP, blood pressure; PEF, peak expiratory flow.
* Low systolic blood pressure for children is defined as less than 70 mm Hg from 1 month to 1 year, less than (70 mm Hg + [2 × age in yr]) from 1 to 10 years, and less than 90 mm Hg from 11 to 17 years.
(From Novembre E, Cianteroni A, Bernardini R, et al: Anaphylaxis in children: clinical and allergological features, Pediatrics 101:E8, 1998.)
(Modified from Lieberman P: Anaphylaxis: how to quickly narrow the differential diagnosis, J Respir Dis 20:221-232, 1999.)
For all reactions but isolated skin symptoms, intramuscular epinephrine is the therapy of choice. Studies have demonstrated the superiority of intramuscular injections compared with subcutaneous injections of epinephrine (Fig. 4-6). Patients at risk should carry self-injectable epinephrine and be trained in its use. There are currently four different epinephrine autoinjector devices: EpiPen, Adrenaclick, Twinject, and a generic autoinjector (Fig. 4-7). All come in two doses: either 0.15 mg or 0.3 mg. Operating technique varies somewhat among the devices, so it is important for families to become familiar with their specific device. Epinephrine is most effective when it is used within 30 to 60 minutes of the onset of anaphylaxis, and other medications should not delay prompt delivery of epinephrine, which is often lifesaving. In cases of hypotension, large-volume fluid resuscitation and intravenous epinephrine may be required. Any administration of epinephrine should be followed by a call to 911 and observation in an emergency department. Albuterol inhalation may be useful for lower airway symptoms, and steroids may prevent late-phase reactions. Antihistamines can be used for reactions confined to the skin and as an adjunct to epinephrine in more severe reactions.
Figure 4-6 During anaphylaxis, intramuscular delivery of epinephrine in the lateral aspect of the thigh produces the highest serum levels of epinephrine. A, anterior deltoid; IM, intramuscular; SC, subcutaneous; T, thigh.
(From Simons FE: Epinephrine absorption in adults: intramuscular versus subcutaneous injection, J Allergy Clin Immunol 108:871-873, 2001.)
Figure 4-7 Anaphylaxis. The EpiPen requires three steps: (1) Remove the gray cap to activate the device; (2) press the black tip firmly against the lateral aspect of the thigh (do not touch the end of the EpiPen); and (3) hold the EpiPen in place for 10 seconds.
Contrary to common opinion, life-threatening reactions to Hymenoptera are rare in childhood. The most common reactions are large local reactions or generalized urticaria. Large local reactions represent a late-phase IgE-mediated response to the sting. The swelling associated with large local reactions is contiguous with the sting, begins 12 to 24 hours after the sting, peaks in 2 to 3 days, lasts approximately 1 week, and should be treated with antihistamines and nonsteroidal antiinflammatory drugs (NSAIDs). Patients with large local reactions have an excellent long-term prognosis, do not need allergy testing, and do not need to carry epinephrine.
Children 16 years of age or younger who have had generalized urticaria and no respiratory, cardiovascular, or GI symptoms after an insect sting also have a good long-term prognosis. The risk for systemic anaphylaxis from stinging insects in children with symptoms confined to the skin is equal to the risk in the general population. These children do not need allergy testing or desensitization with immunotherapy. At present, consensus among allergists suggests that children with a history of generalized urticaria from stinging insects carry epinephrine.
Children with a history of systemic anaphylaxis after a Hymenoptera sting should undergo allergy testing and, if the test is positive, should receive immunotherapy. Testing should take place 4 weeks or more after the reaction. When testing is done immediately after the reaction, there is an increased rate of false-negative test results. Allergen immunotherapy for Hymenoptera sensitivity is the most effective form of allergen immunotherapy available. Venom immunotherapy for wasps, yellow jackets, and hornets provides complete protection from anaphylaxis, while on immunotherapy, in 95% to 100% of patients. Venom immunotherapy for honeybees is slightly less effective, providing complete protection to 80% of patients while on therapy. Most venom immunotherapy protocols require weekly injections for about 16 weeks and then every 1 to 3 months while on maintenance therapy. The duration of immunotherapy should be 3 to 5 years in most cases.
The Hymenoptera that have been associated with anaphylaxis come from three subfamilies: Apidae (honeybees); Vespidae (yellow jackets, wasps, and white- and yellow-faced hornets); and Formicidae (fire ants). In the United States, yellow jackets are the most common cause of Hymenoptera-induced anaphylaxis. Yellow jackets typically nest in the ground, are scavengers for food, and, consequently, are frequently encountered at picnics and around garbage cans. The yellow jacket is small with tight yellow and black bands. The honeybee is the least aggressive of the Hymenoptera family. Stings from honeybees occur most commonly in beekeepers and after accidental contact. The honeybee’s stinger is barbed and is retained in the skin after stings (Fig. 4-8). If the stinger is visible, it should be quickly flicked away from the skin with a fingernail. The stinger and venom sac should not be removed by pinching between the thumb and forefinger as this process can express additional venom. Wasps and hornets are very territorial and will sting to protect their nests. The fire ant is an increasingly important cause of Hymenoptera-induced anaphylaxis. Fire ants are found in the southeastern United States, but their natural habitat appears to be expanding.
Food allergy can be divided into two broad groups on the basis of the mechanism of disease: (1) type I (IgE-mediated) hypersensitivity and (2) other immunologically mediated reactions (Fig. 4-9). Other food reactions (e.g., lactose intolerance) do not have an immune basis and are referred to as intolerant reactions. Food allergic reactions are highly reproducible (i.e., occur with any exposure to a food), and often are triggered by ingestions of very small quantities. Anaphylaxis is the best described and understood type I hypersensitivity to food. Although contact reactions (erythema, pruritus, hives) are common if an individual touches her or his food allergen, the most severe allergic reactions occur with ingestion. Milk, egg, wheat, soy, and peanut account for more than 90% of food allergic reactions in children. The onset of type I hypersensitivity to milk and egg is almost always in the first year of life. Fortunately, these two foods rarely cause more than generalized urticaria and the sensitivity is eventually outgrown in the vast majority of patients, frequently by school age.
Figure 4-9 Reactions to foods can be broadly divided into two mechanistic groups: (1) type I (IgE-mediated) hypersensitivity; and (2) immune (non–IgE-mediated) reactions. Atopic dermatitis and eosinophilic gastroenteritis can be caused by either or both mechanisms.
Most allergic reactions to foods occur with the first few ingestions of a food. In contrast to type I hypersensitivity to milk and egg, less than 25% of patients will outgrow their peanut or tree nut sensitivity. Peanuts and tree nuts cause the majority of life-threatening reactions to foods, and all patients with this hypersensitivity should carry epinephrine. In addition to peanut or tree nut allergy, other risk factors for life-threatening reactions from food-induced anaphylaxis include asthma, adolescence, and the delayed administration of epinephrine. A past history of mild reactions does not rule out the possibility of a future life-threatening episode of food-induced anaphylaxis. A substantial minority of patients with type I hypersensitivity to peanut will develop type I hypersensitivity to one or more tree nuts; and conversely, patients with hypersensitivity to a tree nut have an increased risk of developing peanut hypersensitivity. Nuts are also frequently processed together, increasing the risk of cross-contamination exposures with nuts. Many allergists therefore recommend avoiding all nuts if a patient is allergic to either peanuts or tree nuts, although recommendations are individualized and many tree nut–allergic children continue to ingest peanut butter as long as they are cautious with this and avoid peanut-containing foods that may contain tree nuts.
An increasingly common form of type I–mediated hypersensitivity to food is the oral allergy syndrome (also known as pollen food syndrome). Patients with oral allergy syndrome have underlying seasonal allergic rhinitis and develop pruritus and angioedema of the oropharynx when ingesting fresh fruits and vegetables. The reaction is due to cross-reactivity between heat-labile proteins in some fruits and vegetables and outdoor seasonal pollens. The reaction is often eliminated by heating the vegetable or fruit. The reaction does not typically extend beyond the oropharynx, and patients typically do not need to carry epinephrine, although in a minority of cases reactions can be systemic and may be severe, so taking a good history is essential before recommendations are made for avoidance and the need for epinephrine.
Type I hypersensitivity to foods can be an unrecognized trigger in up to one third of children with severe atopic dermatitis and is an uncommon trigger in children with mild atopic dermatitis. It has rarely been described in adults with atopic dermatitis. The evaluation of type I hypersensitivity to foods in children with atopic dermatitis should be reserved for patients whose disease cannot be managed with good skin care and intermittent use of low- to moderate-potency topical antiinflammatory medications. When food hypersensitivity plays a role in poorly controlled atopic dermatitis, six foods account for the vast majority of reactions: milk, egg, wheat, soy, peanut, and fish.
One of the challenges with diagnosing type I hypersensitivity to foods is the poor specificity of both the skin prick and in vitro tests. The “gold standard” for diagnosis of food allergy is a double-blind, placebo-controlled food challenge (DBPCFC). A distinction is drawn between patients who are “sensitized,” based on skin or serum testing, and those who are clinically allergic, based on a history of reactions to ingestions. IgE-mediated reactions occur quickly and reproducibly with ingestion of the triggering food. Research using in vitro allergy tests has identified levels of specific IgE against certain foods that predict with 95% certainty a reaction during a DBPCFC. In vitro food-specific IgE levels are most useful clinically when used in association with a patient’s ingestion history and (if indicated) skin test results; in a nationally representative survey study, approximately 17% of the overall population in the United States displayed serum sensitization to either cow’s milk, egg, peanut, or shrimp, and this prevalence rate was even higher (28%) among children ages 1 to 5 years. Yet, on the basis of these serum levels in the same study, only approximately 2.5% of the U.S. population was predicted to be clinically allergic to one of these foods. The data concerning positive predictive levels depend on the age of the patient and are not available for all foods (Table 4-4). The in vitro allergy test may also be repeated over time to try to help with the identification of patients who may have outgrown their sensitivity. The sensitivity of both the skin prick and in vitro tests is excellent, but not 100%. If there is a strong clinical history of a type I reaction to a food, it may be necessary to perform an oral challenge in a medically supervised setting before food hypersensitivity can be fully ruled out.
Eosinophilic gastroenteritis is an especially difficult problem to evaluate because eosinophilic infiltration of the gut may be caused by either type I (IgE-mediated) hypersensitivity or other immune pathways. Patients with type I hypersensitivity causing eosinophilic gastroenteritis can be sensitive to either foods or inhaled allergens (pollens/molds) that are inadvertently swallowed. Because eosinophilic gastroenteritis can also be caused by non–IgE-mediated pathways, a negative skin prick or in vitro test does not eliminate the possibility that food allergy may be causing the disease; similarly, a positive skin test to a food does not definitely establish causation. Patch testing has been studied in patients with eosinophilic gastroenteritis to try to identify foods that are causing eosinophilic inflammation through a non–IgE-mediated pathway; at the present time, this method of testing with foods is nonstandardized and still felt to be experimental. Patients with eosinophilic inflammation of the upper GI tract tend to have problems with dysphagia, food impaction, abdominal pain, and vomiting. Patients with eosinophilic inflammation of the lower GI tract tend to have diarrhea, failure to thrive, and abdominal pain. Eosinophilic gastroenteritis appears to have an increasing incidence, and the long-term prognosis is not completely understood, although the disease often seems to follow a relapsing and remitting course.
The term food allergy also encompasses the food reactions elicited through non–IgE-mediated immune pathways. The classic example of a non–IgE-mediated hypersensitivity to food is seen in the milk protein enterocolitis of infancy (also known as food-induced eosinophilic proctocolitis or dietary protein proctitis). These children usually present in the first few months of life with bloody diarrhea that improves within days of removing milk proteins from the diet. Many of these children are breast-fed, and react to either cow’s milk or another protein in the maternal diet. Milk protein enterocolitis is not IgE mediated, so neither allergy skin prick nor in vitro testing is helpful in the evaluation of this condition. Approximately half of children with milk protein enterocolitis will also experience symptoms with soy protein–based formulas. Almost all infants will outgrow milk protein enterocolitis by 1 to 2 years of age.
Food protein–induced enterocolitis syndrome (FPIES) is another non–IgE-mediated reaction with a more severe presentation. Infants typically present two or more hours after an ingestion of the triggering food, with profuse vomiting often to the point of dehydration, hypotension, and lethargy. Fecal leukocytes, a rise in the peripheral absolute neutrophil count, acidosis, and methemoglobinemia may also be seen, whereas skin findings are typically not present. Cow’s milk and soy are the most common triggers of FPIES, although it has been described with a variety of other foods as well. Because the presentation is so different from IgE-mediated allergic reactions, it is often not recognized and infants frequently undergo multiple rule-out sepsis workups before a diagnosis is made.
Drug allergy reactions are immunologically mediated responses that result in the production of drug-specific antibodies, T cells, or both. Type I hypersensitivity to drugs can range from mild reactions involving only the skin (generalized urticaria and/or angioedema) to laryngeal edema, bronchospasm, hypotension, and cardiovascular collapse. The reactions are mediated by drug-specific IgE bound to mast cells and basophils. IgE is cross-linked by the drug, leading to mast cell and basophil activation and release of mediators that drive the clinical reaction. These reactions occur within minutes of exposure and require prior sensitization. IgE-mediated reactions occur early in the treatment course, usually after the first or second dose of a drug. A rash that develops after more than a few days of therapy is seldom due to IgE-mediated mechanisms.
Penicillin is the only drug that has been widely studied as a cause of type I hypersensitivity, and consequently, it is the only drug for which well-standardized allergy testing is available. Once penicillin enters the body, it undergoes spontaneous conversion to a reactive intermediate that then binds to nearby proteins, forming penicilloyl amides. Penicilloyl amides account for 95% of tissue-bound penicillin and are known as the major antigenic determinant. A small percentage of penicillin can be found in the body as the native drug, penicillin, or other metabolites called minor antigenic determinants (penicilloate and penilloate). Skin testing done with the major antigenic determinant, penicillin, and the minor determinants has a negative predictive value approaching 100%. At present, in the United States only the major determinant and the drug itself, penicillin G, are available for skin testing. Although some studies have shown that the minor determinants alone account for up to 10% to 20% of positive skin tests, a negative test to the penicilloyl amides and penicillin G has excellent negative predictive value.
The use of cephalosporins in patients with a history of penicillin allergy is controversial. Before 1980, between 10% and 20% of penicillin-allergic patients reacted to cephalosporins; but since 1980, that number has dropped to 2%. Much of the concern about cross-reactivity between these two classes of drugs is based on the presence of their common molecular structure, known as a β-lactam ring. When cephalosporins were first introduced there was occasional contamination with trace amounts of penicillin. Although clinical cross-reactivity between these two drugs has lessened over the years, there are clear reports of anaphylactic reactions after the use of cephalosporins in penicillin-allergic patients. The combination of all these factors has contributed to the persistent concern about the safety of cephalosporins in patients with penicillin allergy. Unfortunately, skin testing for cephalosporins has not been standardized and the negative predictive value of these tests is not known. There is no consensus regarding the use of cephalosporins in patients with penicillin allergy. In patients with a remote clinical history and mild reactions, it may be reasonable to treat with second- or third-generation cephalosporins without either testing or challenge dosing. Patients who have had either severe or recent reactions should be treated the most cautiously. One option is to test for penicillin, and if the test is negative then the history is not confirmed and precautions with cephalosporins are not necessary. If testing cannot be performed then the options are as follows: (1) find an alternative drug; (2) challenge dose with cephalosporin; or (3) treat by using an induction-of-tolerance protocol.
Anaphylaxis from drugs other than penicillin must be diagnosed on the basis of clinical presentation. Patients with type I hypersensitivity to drugs can still receive the drug with the induction of drug tolerance, also called drug desensitization. Inducing drug tolerance involves slow delivery of increasing drug doses over a period of 6 to 12 hours. The mechanism by which tolerance is induced is not well understood. Patients remain tolerant of the drug as long as they are receiving the medication, but once the drug is discontinued future treatment requires repeating the protocol to induce tolerance.
A variety of drugs can cause anaphylactoid (also called pseudoallergic) reactions. Anaphylactoid reactions are due to generalized mast cell mediator release not caused by cross-linking of specific IgE on mast cell surfaces. The exact mechanism of anaphylactoid reactions is not known, and no diagnostic testing is available. The two most common drug-induced causes of anaphylactoid reactions are radiocontrast media and nonsteroidal antiinflammatory drugs.
Drugs can also induce type II, III, and IV Gell and Coombs reactions. The classic Gell and Coombs type II reaction is due to IgG antibodies directed against drug bound to cell surfaces, causing either hemolytic anemia or thrombocytopenia. Serum sickness represents a Gell and Coombs type III reaction. Serum sickness usually begins 1 to 3 weeks after drug exposure and involves various symptoms including fever, malaise, arthralgias, arthritis, urticaria, and lymphadenopathy. Many patients with serum sickness will develop a characteristic serpiginous, erythematous, or purpuric eruption at the junction of the palmar or plantar and dorsolateral aspects of the hands and feet, respectively (Fig. 4-10). Patients with serum sickness may have reduced levels of complement components C3 and C4 along with the presence of circulating immune complexes. Serum sickness can be treated with antihistamines, and if the patient fails to improve systemic steroids can be used. Symptoms usually resolve within a few weeks. Future avoidance of the drug inciting the reaction is recommended.
Figure 4-10 Cutaneous eruptions on the sides of the hands and feet of patients with serum sickness. A, A scalloped band of erythema can be seen on the side of the finger at the margin of the palmar skin. B, A band of purpura is seen at the margin of the plantar skin. The purpura was preceded by a band of erythema.
(From Lawley TJ, Bielory L, Gascon P, et al: Prospective clinical and immunological analysis of patients with serum sickness, N Engl J Med 311:1407-1413, 1984. Copyright © 1984 Massachusetts Medical Society. All rights reserved.)
Erythema multiforme minor (EM) and Stevens-Johnson syndrome (SJS) are drug reactions that do not conform to the Gell and Coombs classification scheme. EM can be caused by drugs or infection or be idiopathic. EM lesions typically begin as dusky, red macules or erythematous papules that evolve into target lesions after 24 to 48 hours (see Chapter 8). The diagnosis is made on the basis of the clinical appearance and the absence of significant mucosal involvement. The symptoms can persist for a few weeks, and if discomfort is significant, either antihistamines or steroids can be used. Patients with EM do not experience long-term sequelae from their disease. Patients with Stevens-Johnson syndrome (SJS) have skin symptoms similar to those of EM, but also manifest significant mucosal involvement as well as elevated temperature, and constitutional symptoms. Severe disease with extensive skin and mucous membrane involvement may produce long-term complications. Skin findings with SJS evolve from targetoid lesions to vesicles and bullae followed by sloughing of the skin. SJS and toxic epidermal necrolysis (TEN) are most likely manifestations of a single disease that exists along a clinical spectrum. Patients with SJS have less than 10% of the body affected by epidermal detachment and those with TEN have more than 30% of the body affected. TEN is almost always drug induced whereas SJS may be induced by either drugs or infection with Mycoplasma or herpes simplex virus. When drug exposure is the inciting cause, the exposure precedes the onset of skin findings by 1 to 3 weeks. There is no universally accepted therapy, with some studies showing a benefit from high-dose intravenous immunoglobulin and other studies showing no benefit. Glucocorticoids are relatively contraindicated in this condition in children because some studies have shown a deleterious effect.
Another drug reaction that does not easily conform to a Gell and Coombs category is the syndrome known as DRESS (drug rash with eosinophilia and systemic symptoms). The mechanism of this clinical syndrome is unknown. DRESS was initially described with anticonvulsant therapies but is now known to be caused by a number of different drugs. DRESS reactions usually begin 2 to 6 weeks into drug therapy. Clinical characteristics involve a generalized rash, fever, eosinophilia (found in only 50% of cases), and multiorgan dysfunction. The liver is the most commonly involved organ, but any internal organ can be involved. Treatment involves discontinuation of the inciting drug and supportive therapy. The use of systemic glucocorticoids has not been well studied but is believed to be effective for this condition.
Allergic rhinitis, characterized by inflammation, edema, and weeping of the nasal mucosa, is the most common allergic disorder and occurs in up to 25% of the population. Diagnosis is based on characteristic history, physical findings, and testing for antigen-specific IgE. Common presenting symptoms include nasal congestion and pruritus, clear rhinorrhea, and paroxysms of sneezing. Whereas older children may blow their noses frequently, younger children do not. Instead, they sniff, snort, and repetitively clear their throats. Nasal pruritus stimulates grimacing and twitching (Fig. 4-11) and picking or rubbing of the nose (“allergic salute”). Picking, repetitive sneezing, and blowing along with underlying inflammation may produce enough irritation to cause epistaxis. In the case of allergy to seasonal pollens, the symptoms may be acute, have a sudden onset, and be confined to the period during which the particular airborne pollen is detectable. Seasonal allergic rhinitis tends to be due to trees in the spring, grasses in the summer, and ragweed or other pollens in the fall.
Interestingly, seasonal allergic rhinitis is often called “hay fever,” although hay is not involved and fever is not a symptom. However, the manifestations can be severe enough to induce flulike symptoms of fatigue and malaise. In contrast, symptoms may be chronic and more indolent in the case of allergy to perennial allergens including molds, house dust mites, and animal dander.
Many patients have prominent itching and watering of the eyes with nasal symptoms, and some experience pruritus of the throat or ears. Associated symptoms include (1) disturbed sleep and snoring; (2) morning dryness and irritation of the throat as a result of mouth breathing; (3) lassitude, fatigue, and irritability from sleep interruption; (4) early nighttime cough; and (5) if the maxillary, frontal, and ethmoidal sinuses are affected, a sensation of pressure over the cheeks, forehead, and bridge of the nose. The extent to which allergic rhinitis affects a patient’s quality of life is often underappreciated.
Many children with long-standing allergic rhinitis can be recognized by their facial characteristics. Ocular manifestations of the allergic disposition include cobblestoning of the conjunctivae (see Fig. 4-33), the allergic shiner, and Dennie’s sign. Allergic shiners, that is, bluish discolorations or dark circles beneath the eyes, are commonly observed in patients with allergic rhinitis (Fig. 4-12). This finding represents chronic venous congestion secondary to inflammation. Dennie’s sign refers to prominent folds or creases on the lower eyelid (Fig. 4-13) running parallel to the lower lid margin. Although these lines were originally thought to indicate a predisposition to allergy, data suggest that they may be present in any condition associated with periocular pruritus and scratching or chronic nasal congestion. Frequent upward rubbing of the nose with the palm of the hand (the allergic salute; Fig. 4-14) promotes development of a transverse nasal crease across the lower third of the nose (Fig. 4-15). Chronic obstruction produced by nasal mucosal edema may result in mouth breathing and a typical open-mouthed, adenoid-type facies (Fig. 4-16).
(Courtesy Meyer B. Marks, MD, Miami, Fla.)
On nasal examination, attention should be focused on the position of the nasal septum; nasal patency; mucosal appearance; and presence and character of secretions, polyps, or foreign bodies (see Chapter 23). The typical physical examination findings in allergic rhinitis include a marked decrease in nasal patency resulting from swollen inferior turbinates, which appear pale, edematous, and bluish gray (Fig. 4-17). The mucosa appears edematous, and secretions are clear and watery to mucoid in character.
Figure 4-17 Pale, edematous inferior nasal turbinate of a patient with allergic rhinitis, as seen through a fiberoptic rhinoscope. Even though this tool is not routinely used in evaluations, the physical findings are well illustrated, including watery nasal secretions.
Depending on the specific allergens, allergic rhinitis may be acute, recurrent, or chronic and must be distinguished from a number of nonallergic conditions. This necessitates a thorough medical and family history and careful examination. In some instances, response to a trial of medication and/or observations over time may be necessary to confirm the diagnosis. When symptoms are seasonal or regularly associated with exposure to specific allergens, the distinction is generally clear. In evaluating patients with perennial or recurrent but nonseasonal symptoms, allergy, recurrent infection, the nonallergic rhinitis with eosinophilia syndrome (NARES), and vasomotor rhinitis must be considered.
Children with frequent upper respiratory infections and/or persistent nasal congestion can present a diagnostic challenge. In some cases the phenomenon is due to recurrent viral infections, particularly in children in their first year of day care or nursery school. In other patients, tonsillar and adenoidal hypertrophy provides favorable conditions for recurrent infections (see Chapter 23). Atopic (allergic) children may have increased risk of infection because of impaired flow of secretions due to mucosal edema. Although viral infections tend to produce clear or white discharge in contrast to bacterial infections, which typically produce a purulent yellow or green discharge, there is considerable overlap, limiting the value of this distinction.
Other forms of rhinitis that must be distinguished from allergic rhinitis are enumerated in Table 4-5. Although characterized by eosinophilia, NARES does not produce nasal pruritus, and patients lack specific IgE antibodies as measured by skin testing or in vitro serum testing. Patients with vasomotor rhinitis do not complain of pruritus, have a clear discharge without eosinophils, and also lack specific IgE antibodies. Vasomotor rhinitis is thus considered a form of noninflammatory rhinitis, the etiology of which is unknown, although it is often triggered by nonspecific stimuli such as cold air exposure or smoke. The condition is diagnosed most frequently in adults but may affect children. Congestion or rhinorrhea may predominate in this disorder. Rhinitis medicamentosa is a condition seen in patients who have been using α-adrenergic vasoconstrictor nose drops (phenylephrine or oxymetazoline) as decongestants for prolonged treatment periods. The disorder is characterized by rebound vasodilation that produces an erythematous, edematous mucosa in association with a profuse clear nasal discharge.
Some children with perennial allergic rhinitis have congestion so constant and severe that it produces signs of chronic nasal obstruction. This must be distinguished from other acquired and congenital causes (see Chapter 23). The history, physical findings, and results of allergy tests for specific IgE and nasal smears, along with therapeutic trials of antihistamines and intranasal corticosteroids, will all help lead to a diagnosis.
Many patients with allergic rhinitis have mild symptoms that are adequately controlled by intermittent antihistamine administration and/or environmental controls. In many of these patients the pattern of symptoms suggests the responsible allergens, and testing for specific IgE is not indicated. Those with severe symptoms only partially alleviated by antihistamines, topical antiinflammatory agents, and environmental controls and those with perennial symptoms who require daily therapy should be referred for specific IgE testing. Desensitization (allergy shots) is an effective treatment for patients who are not well controlled with medications or who prefer not to use them regularly.
Respiratory distress in children (tachypnea with or without grunting, flaring, retractions, and cyanosis) should be promptly evaluated and treated. The first step in approaching respiratory distress is to differentiate upper from lower airway disorders. At times, various degrees of upper and lower airway obstruction may coexist, as in laryngotracheobronchitis.
Upper airway obstruction causes difficulty moving air into the chest, whereas lower airway obstruction causes difficulty moving air out of the chest. This difference results in characteristic physical findings. In general, lower airway obstruction produces prolongation of the expiratory phase of respiration and typical expiratory wheezing, whereas upper airway obstruction prolongs the inspiratory phase. Wheezing is defined as musical or whistling auscultatory sounds heard more often on expiration than on inspiration, although in severe obstruction both inspiratory and expiratory wheezing are often present. Inspiratory stridor, seen with upper airway obstruction, can mimic wheezing. Both can be detected concomitantly. Stridor is defined as a crowing sound usually heard during the inspiratory phase of respiration. It tends to be loud when the obstruction is subglottic and quiet when obstruction is supraglottic. Mild to moderate increases in respiratory and heart rates are common in upper airway obstruction, whereas lower airway disorders such as pneumonitis and asthma often lead to markedly increased respiratory and heart rates. Retractions are often generalized (suprasternal, intracostal, and subcostal) in severe airway obstruction of any etiology.
Asthma is the most common chronic respiratory condition affecting children. Defining characteristics of asthma, as elucidated by the National Heart, Lung, and Blood Institute (NHLBI) Guidelines for the Diagnosis and Management of Asthma (see Bibliography), include the following: (1) lower airway obstruction that is partially or fully reversible either spontaneously or with bronchodilator or antiinflammatory treatments, (2) the presence of lower airway inflammation, and (3) increased lower airway responsiveness (bronchial hyperreactivity). The last is characterized by inherent hyperreactivity of the airways to stimuli including allergens, infection, exercise, chemical agents such as methacholine, cold or dry air, emotions, and weather changes. Many cases of asthma, particularly in children, have an atopic basis. Specific allergens implicated in atopic patients are pollen, mold spores, house dust mites, and animal dander, whereas drugs, food, and insect venoms typically cause similar symptoms of wheezing and respiratory distress as part of anaphylaxis (see earlier discussion). On exposure, these allergens, via cross-linking specific IgE, produce the characteristic features of asthma: mucosal edema, increased mucus production, and smooth muscle contraction that result in airway inflammation, airway hyperreactivity, and bronchoconstriction. These responses combine to produce a state of reversible obstruction of the large and small airways that is the hallmark of asthma.
Patients with asthma should be evaluated to determine the important triggers for their disease. Affected individuals are often aware of the specific stimuli that trigger exacerbations of their asthma. Viruses are the most common precipitants of acute asthma in children, especially rhinoviruses, respiratory syncytial virus, and parainfluenza viruses. These infections usually affect the upper and lower airways, producing rhinorrhea, nasal congestion, and often fever in addition to wheezing, which tends to develop insidiously. In contrast, allergen-triggered episodes typically lack fever and have a more abrupt onset of wheezing.
Asthma is one of the leading causes of pediatric morbidity. Indeed, approximately 10% to 12% of children in the United States show signs and symptoms compatible with asthma at some time during childhood. Peak incidence of onset is before the age of 5 years (Fig. 4-18). In childhood, boys are affected more often than girls and tend to have more severe disease. Beyond puberty, the gender distribution is equal because onset in the teenage years is more common in girls, perhaps due to hormonal factors involved in menarche. Asthmatic children with respiratory allergy and eczema usually have more severe courses than those who wheeze only with upper respiratory infections.
(Modified from Yunginger JW, Reed CE, O’Connell EJ, et al: A community-based study of the epidemiology of asthma: Incidence rates, 1964-1983, Am Rev Respir Dis 146:888-894, 1992.)