Chapter 91. Abnormal Responses to Ultraviolet Radiation: Idiopathic, Probably Immunologic, and Photoexacerbated

Abnormal responses to ultraviolet radiation (UVR) exposure divide into four categories (Table 91-1): (1) acquired idiopathic, presumed immunologic; (2) DNA repair defect disorders; (3) chemical and drug photosensitivity, including the porphyrias; and (4) photoexacerbated dermatoses not caused directly by UVR. The first and last categories and an approach to assessing photosensitivity are covered in this chapter. The other categories are covered in other chapters in this book. Diagnosis of photodistributed eruptions is discussed in Box 91-1 and Figure 91-1, and the entities are discussed in the following sections.

Polymorphic (Polymorphous) Light Eruption

Epidemiology

Polymorphic light eruption (PMLE) is common worldwide, but thought to occur more frequently in temperate latitudes and rarely in equatorial latitudes. However, a large-scale cross-sectional study has revealed no latitudinal gradient in Europe.1 This study estimated an overall suspected lifetime prevalence of 18% among Europeans. Previous reports indicate a prevalence of 10%–15% among North Americans2 and southern Britons,3 and 5% among southern Australians.3 Women are affected more than twice as frequently as men.1 Fitzpatrick skin type also influences the risk of developing PMLE, with the highest prevalence in people with skin type I and the lowest prevalence in people with skin type IV or higher.1

Etiology and Pathogenesis

A delayed-type hypersensitivity (DTH) response to a sunlight-induced cutaneous photoantigen was first suggested as the cause of PMLE in 1942, based on the delay between UVR exposure and onset of the eruption, as well as on its lesional histologic features.4 In normal skin, UVR is known to induce a transient state of immunosuppression by depleting epidermal Langerhans cells and by promoting the release of immunosuppressive cytokines, including IL-4 and IL-10. In patients with PMLE, the immunosuppression normally associated with UVR is diminished substantially. It is thought that this creates an environment in which hypersensitivity responses to one or more photoinduced neoantigens is permissible. The notion that PMLE represents a DTH response is supported by histopathologic, molecular, and epidemiologic data.

Histopathologic examination of biopsy specimens from lesions of PMLE induced by solar-simulated radiation demonstrates the consistent appearance within hours of a T-cell dominated perivascular infiltrate that peaks by 72 hours. CD4+ T cells are most numerous in early lesions, whereas by 72 hours CD8+ T cells predominate, perhaps abolishing the response.5 Increased numbers of epidermal Langerhans cells and dermal macrophages are also present. Additionally, it has been demonstrated that neutrophil infiltration into the skin following UVB irradiation is deficient, leading to impaired IL-4 and IL-10 release.6 Whereas a Th2 cytokine milieu is favored in normal skin following irradiation, a Th1 cytokine profile is favored in patients with PMLE.7

Molecular studies have revealed increased intercellular adhesion molecule-1 expression on keratinocytes overlying the perivascular infiltrate in PMLE,8 as has been noted in DTH reactions but not in irritant contact dermatitis or after the UVB (290–320 nm) irradiation of normal skin.9

More recently, the induction of allergic contact sensitivity to dinitrochlorobenzene after solar-simulated irradiation of sensitization sites has been shown to occur more easily in PMLE patients than in normal individuals,10,11 which suggests that sensitization to a putative UVR-induced cutaneous antigen may also be easier during disease-inducing exposure. On the other hand, elicitation of allergic contact responses to dinitrochlorobenzene in previously sensitized PMLE patients and normal individuals were equally suppressed by irradiation,12 which explain the frequent development of immunologic tolerance, often called hardening or desensitization, as summer progresses or during prophylactic phototherapy. In fact, normalization in the depletion of epidermal Langerhans cells in response to UVR has been observed in patients with PMLE after undergoing therapeutic hardening.13

Finally, epidemiologic studies also point to hypersensitivity in patients with PMLE. The prevalence of PMLE is much lower in patients with skin cancer, suggesting that UVR-induced immunosuppression that may allow the persistence of malignant cells also inhibits hypersensitivity to photoantigens.14 Also, PMLE is quite uncommon in iatrogenically immunosuppressed transplant recipients, occurring in only 2% of this population.15

The photoinduced neoantigens that initiate PMLE have not been identified, but several molecules may play roles, even in a single patient. The altered molecule(s) itself may hypothetically become antigenic directly by UVR, or secondarily through UVR-induced free radicals that modify nonantigenic peptides in such a way that they become antigenic. Of course, both mechanisms may even take place simultaneously.

UVA radiation (320–400 nm) usually seems more effective than UVB (290–320 nm)12,16 at initiating PMLE. In one study, UVA irradiation elicited the eruption successfully in 56% of patients exposed to UVA or UVB daily for 4–8 days, in 17% of those exposed to UVB, and in 27% of those exposed to both.17 However, other reports have suggested that UVB radiation may be implicated in up to 57% of patients. Therefore, it could broadly be said that approximately 50% of PMLE patients seem sensitive to UVB radiation and 75% to UVA, including in each case approximately 25% who are sensitive to both. Even visible light may be responsible on rare occasions.18 As a result, paradoxically, patients may note that the use of sunscreens, which tend preferentially to remove UVB while transmitting some UVA and all visible light, may even have a PMLE-enhancing effect if exposure times are lengthened. There is most likely genetic predisposition to PMLE,19 but the intensity of initial UVR exposure may also be important in such predisposed individuals.

Clinical Features

History

PMLE usually has onset within the first three decades of life and affects females two to three times more often than males. It occurs in all skin types and racial groups, but is more common in Caucasians. A positive family history is present in about a fifth of patients.19

The typical PMLE eruption (Fig. 91-2) occurs each spring or on sunny vacations after the first substantial UVR exposure following a prolonged period with little exposure. It can also occur after recreational sunbed use or, very rarely, after exposure to visible light,18 and it often abates with continuing exposure. It is rare in winter except after extended outdoor recreational activities. Sufficient exposure may also occur through window glass. The threshold required to trigger PMLE varies from patient to patient and is usually from 30 minutes to several hours, although it may occur several days into a vacation. Lesions appear in hours to days following exposure, but usually in not less than 30 minutes. Patients may note itching as the first sign of an impending PMLE eruption. Once UVR exposure ceases, all lesions gradually resolve fully without scarring over one to several days, occasionally taking a week or two. In any given patient, PMLE outbreaks always tend to affect the same exposed sites. The distribution is generally symmetric. Only some of the exposed skin is usually affected, particularly skin that is habitually covered, and large areas may be spared. An apparent extreme example of PMLE is juvenile spring eruption,20 which tends to affect boys in the spring and is characterized by pruritic papules and vesicles on their ear helices, although typical PMLE sometimes coexists. Systemic symptoms in PMLE are rare, but malaise and nausea may occur.

Cutaneous Lesions

PMLE has many morphologic forms, all probably with similar pathogenesis and prognosis. The term “polymorphous” describes the variability in lesion morphology observed among different patients with the eruption. In fact, the lesions are usually quite monomorphous in an individual patient. Papular, papulovesicular (Fig. 91-3), plaque (Fig. 91-4), vesiculobullous, insect bite-like, and erythema multiforme-like forms have been described, and pruritus alone may even occur, albeit rarely.21 The papular form, characterized by large or small separate or confluent erythematous and edematous papules that generally tend to form clusters, is most common. Papulovesicular and plaque variants occur less frequently, and the other forms are rare. An eczematous form has been claimed but probably refers to mild chronic actinic dermatitis or photoexacerbated seborrheic or atopic dermatitis. Small papular PMLE, generally sparing the face and occurring after several days of continuing exposure, has been designated benign summer light eruption in Europe.22

Laboratory Tests

Histology

The histologic features of PMLE are characteristic but not pathognomonic,23 and they vary with clinical presentation. There is generally a moderate to dense perivascular infiltrate in the upper and mid dermis in all forms. The infiltrate consists predominantly of T cells, with neutrophils and infrequent eosinophils. Other common features are papillary, dermal and perivascular edema with endothelial cell swelling. Epidermal change, not always present, may include variable spongiosis and occasional dyskeratosis, exocytosis, and basal cell vacuolization.

Blood Tests

Assessment for circulating antinuclear antibodies (ANA) and extractable nuclear antibodies (ENA) is advisable to exclude subacute cutaneous or other form of cutaneous lupus erythematosus (LE), and, if there is uncertainty, red blood cell protoporphyrins should be assessed to exclude erythropoietic protoporphyria (EPP). Approximately 11% of patients with PMLE have been found to have a positive ANA, with the vast majority having insignificant titers of less than 1:160.24 An even smaller fraction (less than 1%) of patients with PMLE have anti-Ro antibodies.24 Clinical correlation is necessary in these patients to exclude the possibility of cutaneous LE.

Phototesting

Cutaneous phototesting with a monochromator confirms photosensitivity in up to half of cases,25 but usually does not differentiate PMLE from other photodermatoses. However, provocation testing with a solar simulator or other broadband source, sometimes repeated over 2 or 3 successive days, may induce the eruption, allowing a subsequent skin biopsy. This is most appropriate when the history and clinical features are not diagnostic.

Differential Diagnosis

(See Box 91-1)

Complications

A very few patients with PMLE may develop LE, as there is a higher than normal prevalence of prior PMLE in patients with LE.26 However, the presence of autoantibodies does not portend development of LE. Patients with PMLE may also experience significant disease-related psychosocial morbidity. The rate of both anxiety and depression in patients with PMLE are twice that of the general population, and these rates are similar to those observed in patients with psoriasis and atopic dermatitis.27

Prognosis and Clinical Course

Over 7 years, 57% of 114 patients reported steadily diminishing sun sensitivity, including 11% in whom the condition cleared.28

Prevention

PMLE may often be avoided by moderating sunlight exposure, wearing protective clothing, and applying broad-spectrum high-protection-factor sunscreens regularly. Sunscreens with UVA and UVB protection may prevent PMLE eruptions in photoprovocable patients,29 but sunscreens without UVA protection are generally ineffective. Prophylactic phototherapy each spring or before sunny vacations tends to prevent attacks.

Treatment

The first goal in treating PMLE is to prevent it. As noted above, one should advise restricting midday sunlight exposure and employing frequent applications of broad-spectrum, high-protection sunscreens. If this is not fully effective, patients who have outbreaks only infrequently, such as on vacations, usually respond well to short courses of oral steroids that are prescribed and taken with them to use in the event of an eruption.30 If PMLE does develop, approximately 20–30 mg prednisone taken at the first sign of pruritus and then each morning until the eruption clears usually provides relief within several days, and recurrences are then uncommon during the same vacation. This treatment, if well tolerated, may be repeated safely every few months.

More severely affected individuals who experience repeated attacks of PMLE throughout the summer may require prophylactic courses of low-dose photochemotherapy (psoralen and UVA radiation: PUVA) in the spring. This appears to be more effective than broadband UVB radiation, controlling symptoms in up to 90% compared to approximately 60% of cases.31 Narrowband (311-nm) UVB phototherapy, effective in 70%–80% of cases, is now probably the treatment of choice, because of ease of administration.32 Prophylactic PUVA or UVB irradiation may sometimes trigger the eruption, particularly in severely affected patients, in which case a brief course of oral steroid therapy is indicated.

Various other therapies have also been tried but appear largely ineffective. These include hydroxychloroquine,33 which is perhaps occasionally useful; β-carotene,34 which is rarely effective; nicotinamide,35 which is usually ineffective; and ω-3 polyunsaturated fatty acids, which are perhaps of moderate assistance in some patients.36

A small percentage of patients remain who are unsuitable for, unable to tolerate, or not helped by any of these measures. For these patients, when severely affected, oral immunosuppressive therapy, usually intermittent, with azathioprine or cyclosporine can be helpful.37,38

Actinic Prurigo

Epidemiology

Actinic prurigo (AP) occurs throughout much of the world. Native North and South Americans are particularly affected. The disease is estimated to occur in 2% of the Canadian Inuit population.39 In Mexico, AP is seen most commonly in the indigenous and Mestizo populations living at altitudes greater than 1,000 m.40 Less commonly, inhabitants of Europe, United States, Australia, and Japan are reported to develop AP.

Pathogenesis

AP appears to be UVR-induced in that it is more severe in spring and summer and often demonstrates abnormal skin phototest responses to UVB and/or UVA radiation.41 UVA is implicated more often than UVB.42 The cytokine TNF-α is overexpressed by keratinocytes in lesions of AP, creating a proinflammatory environment.40 Sunlight exposure and solar simulating irradiation may sometimes induce an eruption resembling PMLE in patients with AP, and many patients have close relatives with PMLE.19 A dermal, perivascular mononuclear cell infiltrate similar to that of PMLE may be seen in early lesions.

Therefore, AP may be a slowly evolving, excoriated variant of PMLE, and thus also a DTH reaction. The human leukocyte antigen (HLA) DRB1*0407 (DR4) is found in 60%–70% of patients with AP but in only 4%–8% of normal DR4 positive controls.42 Additionally, HLA DRB1*0401 is found in approximately 20% of affected individuals.42 These immunogenetic features may well be responsible for modulating conventional PMLE into AP. In addition, in some patients, AP appears to transform into PMLE and, in others, PMLE appears to transform into AP,43 all of which suggests a relationship between the two disorders. The cutaneous UVR chromophores responsible for the eruption are not known, but they are likely to be diverse.

Clinical Features

History

AP occurs more commonly in females, with a female to male ratio of about 2:1.42 The eruption has its onset in childhood, usually present by age 10 years.41 A positive family history of either AP or PMLE is present in about a fifth of patients.19 The eruption is often present all year round, but it is commonly worse in summer. Very rarely it is worse in winter or both spring and fall, with immunologic tolerance presumably developing during the summer. Exacerbations tend to begin gradually during sunny weather in general rather than after specific sun exposure, although PMLE-like outbreaks may also occur.

Cutaneous Lesions

The primary lesion of AP is a pruritic papule or nodule that occurs singly or in clusters (Fig. 91-5). Papules and nodules are often excoriated and crusted, and plaques may assume a lichenified or eczematous appearance. Vesicles are not seen unless superinfection is present.42 Sun-exposed areas are most often affected, particularly the forehead, chin, cheeks, ears, and forearms. There is a gradual fading toward habitually covered skin, and the sacral area and buttocks may be mildly affected. Lower lip cheilitis and conjunctivitis are also possible, particularly in Native Americans44 (Figs. 91-6, and 91-7). Healed facial lesions may leave dyspigmentation and sall pitted or linear scars.

Laboratory Tests

Histology

Early papular lesions show changes similar to those of PMLE, namely, mild acanthosis, exocytosis, and spongiosis in the epidermis and a moderate lymphohistiocytic superficial and middermal perivascular infiltrate.23 A dense dermal lymphoid infiltrate and lymphoid follicles may be seen in lesions from the lip.40 In persistent lesions, however, excoriations, more acanthosis, variable lichenification, and a dense mononuclear cell infiltrate produce a nonspecific appearance.

Blood Tests

Assessment of ANA and ENA should be undertaken to exclude subacute cutaneous or other forms of cutaneous LE. The finding of HLA type DRB1*0401 (DR4) or DRB1*0407, especially the latter, supports the diagnosis of AP.

Phototesting

Cutaneous phototesting with a monochromator confirms light sensitivity in up to half of cases,41 but, as in PMLE, does not differentiate other photodermatoses. Most patients with positive monochromator testing have reduced minimal erythema doses (MED) in the UVA spectrum or in the combined UVA/UVB spectra.42 Provocation testing with a solar simulator or other broadband sources induces typical lesions of AP in about two-thirds of patients.42

Differential Diagnosis

(See Box 91-2)

Complications

Mild scarring, especially on the face, and hypopigmentation may result from excoriations associated with AP. Additionally, two cases of primary cutaneous B-cell lymphoma arising on the face in patients with AP have been reported.45

Prognosis and Clinical Course

AP commonly arises in childhood and often improves or resolves in adolescence, although persistence into adult life is possible. Persistent cases may assume features of PMLE in adulthood. More rarely, the disorder arises in adulthood and persists indefinitely.46

Prevention

Prevention begins by restricting midday sunlight exposure and the compulsive use of broad-spectrum sunscreens.46 In addition, topical calcineurin inhibitors tacrolimus and pimecrolimus may be effective in preventing recurrences in patients with previously documented disease. Of course, there is no known way to prevent its initial onset.

Treatment

In less severe cases of AP, sufficient relief may be achieved by restricting sun exposure and by using broad-spectrum, high-protection-factor sunscreens alone.47 Higher potency topical corticosteroids may be used to relieve the inflammation and pruritus associated with the disease. Phototherapy with narrowband UVB or PUVA may occasionally help,48 perhaps more reliably if the skin is cleared first with oral steroids. Topical tacrolimus or pimecrolimus may also help, again if the skin is cleared first. The treatment of choice in more severe or recalcitrant cases is thalidomide, with initial doses of 50–100 mg daily, preferably given intermittently. Responses to thalidomide are evident in most patients within several weeks. The most serious complication associated with thalidomide is teratogenicity, so pregnancy must be rigorously avoided. Other potential adverse effects are typically mild, including drowsiness, headache, constipation, and weight gain. An increased risk of thromboembolism and dose-related peripheral (mostly sensory) neuropathy are other potential adverse effects of thalidomide. In cases where thalidomide is unavailable or otherwise not appropriate, oral immunosuppressive therapy with azathioprine or cyclosporine may also be considered.

Hydroa Vacciniforme

Epidemiology

Hydroa vacciniforme (HV) occurs in North America, Europe, Japan, and very likely elsewhere. However, its rarity and lack of universally acknowledged diagnostic criteria may make the diagnosis difficult to establish.

Etiology and Pathogenesis

The pathogenesis of HV is not known. No chromophores have been identified, and although UVB minimal erythemal dose responses are normal in most patients, some have increased UVA sensitivity.49 Blood, urine, and stool porphyrin concentrations are normal, as are all other laboratory tests. Nevertheless, its clear relationship to sunlight exposure, its distribution, and its early clinical appearances are all similar to that of PMLE, which suggests a relationship. On the other hand, fully developed HV eruptions are more severe than those found in PMLE, are associated with permanent scarring, and are unresponsive to treatments ordinarily effective in PMLE, apart perhaps from sunscreens and occasionally prophylactic phototherapy. Reports from Asia and Mexico have linked HV to chronic Epstein–Barr virus (EBV) infection,50,51 although a relationship between EBV and HV in all populations has not been established. Japanese reports indicate that EBV nucleic acids are found in the cutaneous lesions of HV in 85%–95% of patients but not in lesional skin of control patients.51,52 A recent report from France has now provided substantial evidence that EBV infection persists in adult patients with HV and that it plays an important pathogenic role.53

Clinical Findings

History

HV commonly develops in early childhood and resolves spontaneously by puberty, although, in some patients, it is lifelong. There is male predominance for severe forms, whereas milder disease is more common in females.49,54 Familial incidence is exceptional. One estimate of the prevalence of HV is 0.34 cases per 100,000 with an approximately equal sex ratio; males had a later onset and longer duration of the disorder than females.54

HV eruptions typically occur in summer,41 often with an intense burning or stinging sensation followed by the appearance of individual or confluent papules and then vesicles, all within hours of sunlight exposure (Fig. 91-8). This is followed by umbilication, crusting, and progression to permanent pock scarring within weeks. The eruption affects the cheeks and, to a lesser extent, other areas of the face as well as the backs of the hands and outer aspects of the arms. The distribution tends to be symmetrical.

Cutaneous Lesions

HV is characterized by initial erythema, sometimes with swelling, followed by symmetrically scattered, usually tender papules within 24 hours; vesiculation, occasionally confluent and hemorrhagic (Fig. 91-9); umbilication; then crusting and detachment of the lesions with permanent, depressed, hypopigmented scars within weeks. These scars are invariably present. Oral ulcers and eye signs also occur rarely.55,56

Laboratory Tests

Histology

Early histologic changes include intraepidermal vesicle formation with subsequent focal epidermal keratinocyte necrosis and spongiosis. There is a dermal perivascular neutrophil and lymphocyte infiltrate.23 Older lesions show necrosis, ulceration, and scarring. Vasculitic features have been reported.49

Blood Tests

Blood, urine, and stool porphyrin concentrations should be assessed to exclude cutaneous porphyria, and an ANA and ENA to exclude the small possibility of cutaneous LE.

Phototesting

Phototesting may show increased sensitivity to short-wavelength UVA in some patients, but phototesting usually does not discriminate HV from other photodermatoses. Simulated solar irradiation may also induce erythema at reduced doses or occasionally provoke the typical vesiculation of HV (see Fig. 91-7).

Other Tests

Viral studies for herpes infection or other viral disorders should be undertaken if photoexacerbation or photoinduction of these other disorders seems at all possible.

Differential Diagnosis

(See Box 91-3).

There are reports of severe HV-like eruptions occurring in patients with chronic EBV infection and other associated disorders such as hypersensitivity to mosquito bites and the hemophagocytic syndrome. HV-like eruptions are distinguished from true HV by the development of lesions in both exposed and sun-protected skin and by the presence of systemic symptoms such as fever, hepatosplenomegaly, and wasting.51,57 The distinction between an HV-like eruption and true HV is important because patients with a severe HV-like eruption may rarely go on to develop potentially fatal hematologic malignancy.56–59 Finally, a recent quality-of-life study indicates that HV causes embarrassment and self-consciousness among children with the disease.60 The negative impact of HV on quality of life exceeds previously reported indices for atopic dermatitis and psoriasis.60

Complications

Pock scarring is an inevitable sequela of HV. Cases of severe HV-like eruption may progress to lymphoproliferative disease.

Prognosis and Clinical Course

HV often resolves in adolescence but may occasionally persist into adult life.

Prevention

Sun avoidance and sunscreen use, as well as prophylactic phototherapy annually in spring, tend to prevent HV in some patients.

Treatment

Treatment of HV consists of restricting sunlight exposure and use of high-protection-factor broad-spectrum sunscreens. Occasionally, antimalarials appear to have helped, but their true value has not been established. As with PMLE, prophylactic phototherapy with narrowband UVB or PUVA, particularly the latter, may be helpful but must be administered with care to avoid disease exacerbation.49,53,61 If conservative measures are ineffective, however, as often occurs, topical or intermittent oral steroids, topical calcineurin inhibitors, or even oral immunosuppressive medication may be tried if clinically appropriate, though these agents too are usually ineffective. In patients with chronic EBV infection, antiviral therapy with acyclovir and valacyclovir was reported in a small series of patients to reduce the frequency and severity of eruptions.62

Chronic Actinic Dermatitis

Historical Aspects

Epidemiology

Chronic actinic dermatitis (CAD) has regularly been diagnosed in Europe, North America, Africa, Australia, New Zealand, Japan, and India. The disorder therefore appears to have worldwide distribution, affecting all skin types, although it is perhaps more common in temperate regions.

Etiology and Pathogenesis

Studies of the clinical, histologic, and immunohistochemical features of CAD all show it to resemble the DTH reaction of allergic contact dermatitis,73–75 even in its severe pseudolymphomatous form (formerly called actinic reticuloid), in which the clinical and histologic features duplicate those seen in long-standing allergic contact dermatitis.76 It is therefore highly probable that CAD is an allergic reaction. In addition to hypersensitivity to cutaneous photoantigens, patients with CAD often have concomitant allergic contact dermatitis to airborne or other ubiquitous allergens, including plant compounds, fragrances, and medicaments.77 Commonly implicated allergens include sesquiterpene lactone from plants of the Compositae family and sunscreens. A recent study indicates that sesquiterpene lactone remains the most common allergen in patients with CAD, with positive and clinically relevant photopatch testing to this allergen documented in approximately 20% of patients.78 In addition, cc has emerged as an increasingly common antigen in CAD78 as has balsam of Peru.79

When CAD occurs in the absence of an obvious epicutaneous contact allergen, the relevant novel antigen must be either directly radiation-induced or formed indirectly as a result of secondary oxidative metabolism. Important support for the latter possibility comes from the fact that albumin can become antigenic in vitro through photooxidation of its histidine moieties.80 There is no evidence for a genetic susceptibility to CAD; however, one stimulus for the acquisition of skin reactivity may be concurrent allergic contact dermatitis to recognized exogenous sensitizers or photosensitizers,81,82 often airborne, which may predispose by altering cutaneous immunity, and thus permitting immunological recognition of an endogenous photoantigen. Long-standing endogenous eczema,68,69,83,84 drug-induced photosensitivity,85 human immunodeficiency virus infection,86 and possibly PMLE75 may also play similar roles. On the other hand, in addition or instead, chronic photodamage in frequently sun-exposed elderly outdoor enthusiasts, those who most often develop CAD,75 may impair normal UVR-induced skin immunosuppression sufficiently for endogenous UVR-induced photoantigens to be recognized, as apparently also occurs for genetic reasons in PMLE.19 Clearly, there is much work left to be done to identify the immunologic mechanisms that account for CAD.

Determining the action spectrum for CAD should theoretically help identify the postulated antigens, and the action spectra for CAD have been shown to resemble that of sunburn in many patients.87 However, the eruption in CAD is eczematous, and much lower doses of UVR are required to evoke CAD than to produce erythema. In any event, the UVR chromophore for some patients may be the same as that of sunburn, namely DNA,87 with UVR-damaged DNA serving directly as an antigen in CAD. In other patients with CAD, however, the photoallergen must be different, because these patients react only to UVA radiation,88 and some patients react only to visible light.89

In summary, CAD appears to be an allergic contact dermatitis-like reaction against UVR-altered DNA or similar or associated molecules, perhaps as a result of enhanced immune reactivity due to concomitant airborne contact dermatitis or a reduced immunosuppressive capacity in photodamaged skin. The eruption occurs most often in patients with long-standing exposure to sunlight and airborne contact allergies.

Clinical Features

History

CAD may arise de novo in apparently normal skin or in the skin of patients with previous endogenous eczema, photoallergic or allergic contact dermatitis, or rarely PMLE.75,90 Concurrent allergic contact sensitivity to plant allergens, fragrances, or sunscreens is common.82 The condition usually affects middle-aged or elderly men,75 as CAD is rarely seen in patients under 50 years of age, except for those with prior atopic eczema.83,84 The disorder is worse in summer, developing within minutes to hours after sunlight exposure and producing a pruritic confluent erythematous eruption that occasionally remits over several days with scaling, if exposure ceases and if the reaction is mild. However, severely affected patients frequently do not even recognize that exacerbations are related to sunlight exposure, especially when affected all year round.

Cutaneous Lesions

The lesions of CAD are eczematous, patchy or confluent, and acute, subacute, or chronic (Figs. 91-10 and 91-11). In severe cases, lichenification is common. Less commonly, scattered or widespread erythematous, shiny, infiltrated pseudolymphomatous papules or plaques are present on a background of erythematous, eczematous, or normal skin.75 Habitually exposed areas are most often affected, commonly with sharp cutoff at lines of clothing. There is sparing of deep skin creases, upper eyelids, finger webs, and skin behind the earlobes. In severe disease, eczema of the palms and soles may also be found. Eyebrows, eyelashes, and scalp hair may be stubbly or altogether lost from constant rubbing and scratching. Erythroderma, usually but not always accentuated on exposed sites, supervenes rarely. Variable, sometimes geographic, sparing of exposed areas of the face or elsewhere, as well as irregular hyperpigmentation and hypopigmentation, sometimes vitiligo-like,91 may also occasionally be found.

Laboratory Tests

Histology

Histologic features include epidermal spongiosis and acanthosis, sometimes with hyperplasia. There is usually a predominantly perivascular lymphocytic cellular infiltrate confined to the upper dermis that in milder cases may resemble chronic eczema.23 Severe CAD, however, may mimic cutaneous T-cell lymphoma (CTCL), on occasion being virtually indistinguishable. Features mimicking CTCL include epidermal Pautrier-like microabscesses and deep, dense epidermotropic mononuclear cell infiltration, sometimes with atypia. Typically, there is no marked increase in mitoses. T-cell receptor gene rearrangement studies should be undertaken if there is suspicion of CTCL. However, T-cell receptor clonality may also be observed in benign dermatoses.

Blood Tests

Assessment of the ANA and ENA is advisable in all patients to exclude the unlikely possibility of cutaneous LE. In severe or erythrodermic CAD, there may be large numbers of circulating CD8+ Sézary cells without other suggestions of malignancy.92,93 Human immunodeficiency virus status should be assessed if there is suspicion that this may be a predisposing factor.86 Serum IgE may be elevated, with higher levels of IgE correlating with more severe disease.79

Phototesting

Phototesting is essential, if available, to confirm the diagnosis of CAD. Almost invariably one finds low erythemal thresholds and eczematous or pseudolymphomatous responses after irradiation with UVB, usually with UVA, and rarely with visible wavelengths.75 A small number of patients react only to UVA,88 and fewer still only to visible light,89 in which case, drug photosensitivity must be excluded. Testing should be done on uninvolved skin of the back with no topical or systemic steroid therapy for at least the preceding few days to avoid false-negative results.94 Monochromatic and broad-spectrum sources both induce abnormal responses, with the former determining the action spectrum for disease induction and the latter tending to demonstrate acute eczema.

Patch and Photopatch Testing

Patch and photopatch testing is also essential in suspected CAD, because contact sensitivity to airborne allergens such as Compositae oleoresins, phosphorus sesquisulfide, and colophony alone may resemble CAD or even coexist. In addition, occasional secondary contact or photocontact sensitivity to sunscreens or other topical therapies may complicate the clinical picture further. Positive results with photopatch testing are found in approximately 80% of patients.78,79

Differential Diagnosis

(See Box 91-4)

Complications

A relationship to CTCL seems likely to be coincidental, especially because results of T-cell receptor, immunoglobulin gene rearrangement, and other studies are negative in CAD.92,95,96 In addition, CAD gradually resolves in many patients, there is no higher incidence of malignancies, and life expectancies are thought to be normal.97 However, CTCL itself may present very rarely with severe CAD-like photosensitivity, and careful investigation to exclude CTCL is necessary when the disease suspected.98

Prognosis and Clinical Course

Once established, CAD usually persists for years before resolving gradually.98

Treatment

Treatment of CAD is often difficult and not fully effective. Rigorous avoidance of UVR and exacerbating contact allergens is essential, along with regular application of high protection-factor broad-spectrum topical sunscreens of low irritancy and allergenic potential. Strong topical steroids such as clobetasol propionate are also often needed and frequently produce marked symptomatic relief without adverse effects, even after long-term use, if confined to affected skin. Occasional oral steroid use is often helpful for disease flares. In more resistant disease, the topical calcineurin inhibitors—tacrolimus and pimecrolimus—sometimes produce good results if tolerated.99,100 For refractory CAD, however, oral immunosuppressive therapy is almost always necessary and generally helpful if tolerated. Azathioprine 1.5–2.5 mg/kg/day often produces remission in months,101 after which it may be reduced in dosage, or perhaps discontinued in the winter. Cyclosporine 3.5 to 5.0 mg/kg/day is usually rapidly effective,102 but is more likely to produce adverse effects, which sometimes necessitate withdrawal. Mycophenolate mofetil is less often useful. Finally, long-term low-dose phototherapy with PUVA, usually several times weekly initially followed by maintenance exposures about every 3 weeks may be helpful,103 generally accompanied initially by oral and topical steroid therapy to avoid disease flares.

Prevention

The risk of CAD can probably be reduced by moderating outdoor pursuits, especially those associated with plant allergen exposure such as gardening, even more so for individuals who already have a tendency to develop eczematous eruptions in exposed areas. Avoidance of UVR is critical, and patients should be aware that indoor lighting may also be a source of UVA exposure. Compact fluorescent lamps even emit UVR at wavelengths as low as 254 nm, and some patients with CAD appear to be susceptible to disease flares after exposure to such lamps.104

Solar Urticaria

Epidemiology

Solar urticaria (SU) has been reported in Asia, Europe, and the United States, and, thus, probably occurs worldwide. Perhaps 3 per 100,000 are affected.105 Ultraviolet and visible radiation are causes of this form of urticaria, but SU accounts for less than 0.1% of all cases of chronic urticaria.106

Etiology and Pathogenesis

Primary SU is an immediate type I hypersensitivity response against a cutaneous or circulating photoallergen(s) generated from a precursor at the time of irradiation. Both circulating photoallergens and relevant IgE antibodies have been demonstrated. This has been termed primary SU, for which no genetic basis has been identified. Very rarely, SU may occur in association with drug photosensitivity, such as chlorpromazine and coal tar,106 cutaneous porphyria, or LE. This has been termed secondary SU.

Two types of primary SU have been proposed, both involving immunoglobulin E-mediated hypersensitivity against a neoantigen that is photoinduced. In type 1, the photoallergen is the chromophore, which is generated only in patients with SU. Type 2 is mediated by circulating antibodies found only in patients and directed against a common chromophore-generated antigen.106 The wide range of inducing wavelengths is attributed to the unique action spectra of different photoallergens (chromophores). Patients with type 1 SU appear to have photoallergens of molecular mass 25–34 kilodaltons (kDa) and action spectra within the visible region, whereas type 2 SU has photoallergens of 25–1,000 kDa and variable action spectra.107 The range of eliciting wavelengths can narrow or broaden over months or even years, suggesting that the relevant choromophores may vary over time.

Exposure to visible or UVA irradiation before, during, or after the urticaria-inducing irradiation inhibits whealing in some patients, possibly by inactivation of the initial photoproduct or the inhibition of subsequent reactions.108,109

Clinical Features

History

Primary SU is slightly more common in females and may arise at any age, although most patients develop the disease in childhood or young adulthood.106 The first episode typically occurs after prolonged sunlight exposure or occasionally following tanning bed use. It also seems to be associated with other photosensitive skin disorders more often than expected.110 Typically, 5–10 minutes or, rarely, 20–30 minutes of exposure leads to itching and erythema, followed by patchy or confluent urticarial whealing, with gradual resolution within 2 hours. Rarely, patients report itching alone, and the onset of symptoms may be delayed for up to several hours.111 A rare variant termed “fixed solar urticaria” has been reported and is characterized by urticarial lesions that are induced repeatedly in the same location.112 In patients with mild disease, or in those who quickly recognize their SU onset and avoid further exposure, whealing may not be reported, even if it may be elicited during phototesting. Patients having extensive whealing may also describe headache, nausea, bronchospasm, and syncope, which rarely may be life threatening. Angioedema and anaphylaxis are exceedingly rare but have been reported. Secondary SU should be excluded by ruling out drug photosensitivity, cutaneous porphyria, and LE.

Cutaneous Lesions

SU usually affects all exposed skin with an initial macular erythema, followed by whealing and a surrounding patchy erythematous flare (Fig. 91-12), often with clear demarcation at lines of clothing. Rarely, there is sparing of the face and hands, perhaps as a result of UVR-induced tolerance. In some patients, specific sites may be affected repeatedly.

Laboratory Tests

Histology

There is dermal vasodilation, edema, and predominantly perivascular neutrophil and eosinophil infiltration at 5 minutes and at 2 hours, but not 24, hours.23 Endothelial cell swelling occurs early on, with mononuclear cell infiltration later following higher irradiation doses. Extensive eosinophil granule major basic protein deposition is also present in the dermis at 2 and 24 hours, which suggests eosinophil degranulation.113 Histologic features do not distinguish SU from other causes of urticaria.

Blood Tests

Tests for an ANA and ENA should be employed to exclude cutaneous LE, as should the blood, urine, and stool porphyrins to exclude cutaneous porphyria.

Phototesting

Phototesting with a monochromator, broad-spectrum source, or sunlight allows confirmation of the diagnosis of SU and identification of the inducing wavelengths. However, negative phototest results do not exclude the disorder, because the ease of SU induction, particularly when mild, may vary. If no monochromator is available, appropriately filtered broadband sources can be used, and minimal urticarial dose estimation may help in assessing treatment efficacy. Patients with SU may have a biphasic response to phototesting whereby wheals develop in response to one action spectrum but are inhibited by another action spectrum.114 Most commonly, shorter wavelengths induce wheals on phototesting while longer wavelengths may inhibit wheal formation.114

Differential Diagnosis

(See Box 91-5)

Complications

Severe primary SU may lead to anaphylactic shock, which is rarely fatal. The rare secondary SU related to drug or chemical photosensitivity, cutaneous porphyria, or cutaneous lupus may be associated with the complications of the primary conditions.

Course and Prognosis

SU often persists indefinitely, sometimes with deterioration but also sometimes with improvement, with the probability of clinical resolution at 5, 10, and 15 years of 12%, 26%, and 46%, respectively.105,106

Treatment

Restricting sun exposure and using high protection-factor broad-spectrum sunscreens and appropriate clothing may be helpful in preventing SU. Sunscreens with UVA and UVB protection effectively increase the minimal urticarial dose on phototesting.115 Antihistamines have been demonstrated to suppress wheal and itch formation in patients with SU, and, when combined with sunscreens, the increase in UV tolerance may be remarkable.115 Phototherapy may be helpful in those patients who commonly develop SU tolerance as summer advances and also sometimes in those with persistent disease. Unfortunately, phototherapy usually needs to be continued to maintain its effect, and, consequently, carries the usual risks of long-term phototherapy. In addition, phototherapy should be undertaken with caution early on to avoid the risk of anaphylaxis, particularly in severely affected individuals.113 Multiple UVA exposures with increasing doses during the same day (so-called rush hardening) have helped some patients.116 Others with recalcitrant disease have been reported to respond to plasma exchange, or plasmapheresis, particularly if they are found to have circulating SU-associated serum factors; these remissions may be long-lived.117,118 Intravenous immunoglobulin has also been helpful on occasion,119 as has oral cyclosporine. Partial improvement has recently been reported with omalizumab, a monoclonal antibody directed against IgE.120

Several dermatoses that are not caused by UVR may be worsened by it (Table 91-3). Mechanisms of this phenomenon, termed photoexacerbation, have rarely been studied. The initial condition may be severely worsened even if it was originally only mild or subclinical.121,122 These disorders are relatively common and account for a significant percentage of all photodermatoses. Such conditions, especially the eczemas, psoriasis, and acne, improve with UVR exposure in most patients, perhaps because cutaneous reactivity is reduced, but in a small proportion of individuals, it is instead aggravated. If photoexacerbation does occur, the new eruption generally develops or worsens initially at sites typical of the basic disorder (Fig. 91-13), followed at times by extension to other areas. In photoexacerbated seborrheic eczema, however, an unpleasant sensation at the exposed sites may be the first or only feature. Treatment consists of minimizing UVR exposure, protection with suitable clothing, application of high protection-factor broad-spectrum sunscreens, and careful treatment of the underlying disorder. Taking these steps alone, frequently, if perhaps surprisingly, even may abort the photosensitivity.121 If these actions are inadequate, low-dose phototherapy, as for PMLE, can sometimes help, for example, in seborrheic or atopic eczema and psoriasis, but its use is contraindicated in cutaneous LE or dermatomyositis, in which aggravation of the systemic disease is a risk. Photoexacerbated acne commonly requires treatment with oral isotretinoin.

Individual diseases for which photoexacerbation may occur are discussed in more detail in the online sections of this chapter include the disorders many of the disorders in Table 91-3.

Acne

Darier Disease

Disseminated Superficial Actinic Porokeratosis

Herpes Simplex

Lichen Planus Actinicus

In addition to being able to trigger lichen planus as a Koebner reaction to sunburn injury, UVR in suberythemogenic doses may also induce a lichenoid reaction known as lichen planus actinicus.134 The condition most commonly affects darker skinned persons, particularly those from the Middle East, but affected individuals have also been reported in India, Europe, and the United States. Several clinical patterns occur, the most common consisting of annular, hyperpigmented plaques predominantly on the face and dorsa of the hands. Others include pigmented melasma-like patches of the face and neck and skin-colored, closely aggregated, pinhead papules, particularly of the face and dorsa of the hands.134,135 The peak age of onset is the third decade, and the lesions are generally only mildly pruritic, developing mainly on exposed sites during spring and summer and improving or remitting in winter. Experimental reproduction of lesions was successful with UVB but not with UVA radiation in one patient.136 On the other hand, phototherapy has been used successfully to treat conventional lichen planus.137,138

Eczema

Lupus Erythematosus

Pellagra

Pemphigus

Pemphigus foliaceus and pemphigus erythematosus may be exacerbated or induced by sun exposure, and lesions with the characteristic histologic and immunofluorescent features have been produced in patients with both variants with artificial UVR exposure. Pemphigus vulgaris is not associated clinically with photosensitivity.

Bullous Pemphigoid

Psoriasis

Clinical Features

Patients with abnormal photosensitivity present in three ways: (1) sporadic or (2) persistent eruptions in sunlight-exposed areas, or, infrequently, (3) erythroderma. When sporadic, the patient usually considers sunlight exposure to be responsible; when persistent, the physician often must identify the association. However, careful history taking is essential, first to confirm that sunlight exposure is responsible and then to make a diagnosis. Information of considerable importance are age at disease onset, gender, family history, previous sunlight sensitivity, occupation, leisure pursuits, and systemic and topical drug (or chemical) use. Additional relevant details include distribution of lesions, effects of season, exposure times required for induction, time between exposure and the appearance of lesions, duration of the eruption after exposure ceases, effects of sunlight received through window glass (implicating UVA and visible light), presence of systemic symptoms, and patient-assessed morphologies (progression of the disease before the clinic visit).

In terms of age and sex, young woman are more likely to develop PMLE; women or girls more commonly develop AP; children of either gender may have HV, xeroderma pigmentosum (XP), or EPP; elderly men or younger individuals with a history of eczema most often develop CAD.

A family history of sunlight sensitivity may be present in patients with PMLE, AP, XP, and the porphyrias. CAD is more common in outdoor enthusiasts exposed to both sunlight and airborne allergens, although exacerbations of disease, despite sunscreen use, invoke the possibility of sunscreen allergy. On the other hand, exacerbations with sunscreens may even occur in the absence of allergy in PMLE. An eruption appearing in minutes and remitting within 2 hours suggests SU or occasionally photosensitivity to drugs, such as amiodarone. Onset within 20 minutes to several hours, with resolution over days suggests PMLE, HV, EPP, cutaneous LE or other photoexacerbated dermatoses, or other drug photosensitivities, such as to thiazides. Systemic malaise is uncommon in PMLE, HV, and SU. Development of lesions after exposure through window glass suggests an inducing spectrum that includes UVA, although it may occur in virtually all of the photodermatoses.

The eruption described by patients with PMLE is generally that of small or large, elevated, pruritic, red or skin-colored, and often clumped spots of blisters, sometimes confluent, that usually involve several, but not all exposed sites. In HV, blistering with scar formation occurs, and in SU, elevated pruritic wheals are often confluent. In EPP and amiodarone drug photosensitivity, a marked burning sensation, without visible change, has been reported. In EPP, relatively lengthy exposure may lead to firm, colorless or pink, diffuse swelling, rarely with scattered blisters. In most drug photosensitivity reactions and in XP, an exaggerated sunburn-like reaction is possible, which is often maximal in XP at 2–3 days. Finally, in photoexacerbated dermatoses, the eruption resembles that of the primary disorder.

Photosensitivity eruptions are usually present on some, and occasionally all, of the forehead, nose, upper cheeks, tip of the chin, rims of the pinnae, back and sides of the neck, upper chest, backs of the hands and feet, and extensor aspects of the limbs. Covered areas may also be involved, but to a lesser extent. On the other hand, portions of the face protected by hair or customarily shown in shadow such as upper eyelids, finger webs, skin creases and skin under the nose, lower lip, chin, and earlobes are frequently unaffected, except when there is associated airborne contact dermatitis. Excoriated papules suggest AP, whereas eczematous lesions, or very rarely light-associated erythroderma, suggest CAD or photoexacerbated atopic or seborrheic eczema. Finally, skin fragility, bulla formation, and atrophic superficial scarring suggest hepatic porphyria or pseudoporphyria, especially if there has been drug or excessive alcohol intake.

Clinical appraisal along with the history usually results in a diagnosis, although for complete certainty, several of the studies listed below may be appropriate.

Laboratory Studies

If the diagnosis is not certain, appropriate additional studies include an assessment of the ANA and ENA. If present at significant titers, cutaneous LE should be considered. In addition, examination of blood, urine, and stools for porphyrins should be considered.

Biopsies may be helpful. Lesional histologic features are characteristic in several photodermatoses, especially PMLE, HV and CAD. However, with the exception of HV, histopathologic changes in photodermatoses are rarely entirely diagnostic. These are reviewed in the preceding disease descriptions.

Phototesting of normal back skin with a monochromator in CAD and SU often produces the papules or wheals of the condition itself, frequently at low irradiation doses, and this also may identify the action spectrum. Phototesting also helps to confirm XP through the delayed development of erythema over 2–3 days, with an abnormally low-dose threshold, often eventuating in blister formation (Table 91-4). In eczematous photosensitivity, patch and photopatch testing are also essential to identify relevant allergens. Finally, special techniques such as the assessment of DNA excision repair or of RNA synthesis recovery rate in cultured fibroblasts after UVR exposure are essential for the diagnoses of certain genophotodermatoses.

Phototesting

Techniques of phototesting vary greatly from country to country and from center to center. Although it is the investigational technique of choice for photodermatoses when the diagnosis is uncertain or when details of the inducting action spectrum are required, it remains primarily a research tool employed in a limited number of clinical centers. The cost of the equipment and its infrequent use in most clinical practices means that patients should be referred for consultation to such centers whenever indicated.

Phototesting falls into two categories: (1) Monochromatic phototesting, usually of the upper back with selected wavelengths and selected doses to identify the action spectrum for the disorder and (2) photoprovocation with a broad-spectrum source to induce the eruption for its clinical appearance and subsequent biopsy if indicated. Table 91-4 lists the disorders for which monochromatic testing may be helpful.

For precise characterization of the wavelength dependency of a disorder, monochromatic testing, preferably with a xenon arc irradiation monochromator, should be employed. For photoprovocation, the favored device is a solar simulator, usually a xenon arc-filtered source that produces a spectrum that resembles the terrestrial sunlight spectrum at noon on a midsummer's day in temperate regions of the world. Keep in mind that the terrestrial spectrum at noon in June varies considerably between Iceland and Kenya, as it also does between high elevations and sea level. Several suitable protocols have also been described for using simple broad-spectrum metal halide or fluorescent light sources with filters if necessary. In some parts of the world, sunlight with filters has also been used, although this method is generally too unpredictable for clinical use.148,149

The mainstay of phototesting is a monochromator. It is composed of a high-pressure xenon arc source that emits radiation along a pathway incorporating a diffraction grating angled to produce the required waveband at the exit slit. Such equipment needs regular calibration of output and wavelength. Because even large centers cannot always afford such equipment, lesser alternatives have been created, for example, metal halide or fluorescent light sources of sufficient output intensity. With such sources, the UVB, UVA, and visible light components of patient photosensitivity can be studied, based on deviation from normal erythemal reactions throughout the UVR spectrum.

Monochromatic phototesting is preferably performed on unaffected skin of the upper back, lateral to the paravertebral groove whereas lesion induction, except when done relatively easily with the monochromator, as in SU and CAD, is best undertaken using broadband sources with output directed over larger areas of skin known to be susceptible to the eruption. PMLE, AP, and HV are conditions in which repeated irradiation with UVA- or UVB-emitting or combined sources is often required to reproduce the disease.

It is important that the use of potent topical and systemic steroids be avoided when possible for at least several days before phototesting to prevent false-negative results. It is not certain how much the other oral immunosuppressive agents affect testing, but they should be stopped whenever possible, as well. False-positive results may also occur in patients with widespread disease, and the eruption should first be well controlled whenever possible, if necessary by keeping the patient in a reduced-light environment. However, it is often difficult to fulfill these requirements if the eruption is active, and in such circumstances, testing may need to be undertaken with knowledge of its limitations.

All phototesting should be undertaken at carefully standardized sequential doses (often a geometric series) and wavelengths, and the results read at consistent times after exposure in carefully controlled conditions of light and temperature. Furthermore, because testing involves UVR exposure, potentially noxious to both skin and eyes, the patient and the investigator should be protected with appropriate clothing, shielding and goggles.

Photopatch Testing

(See Chapter 92.)

Photopatch testing is an established investigational tool designed to identify photoallergic contact dermatitis, although it can also be employed to help identify phototoxic agents. It is essentially a more complex version of patch testing, and it is used in patients with exposed-site eczema, whether or not they also have another photodermatoses, to determine whether photoallergy is also present.

The methodology of photopatch testing has received less attention than allergen testing or phototesting, as it resides between the two specialty areas of photodermatology and contact dermatology. However, consensus methodology is now available.150

Using this approach, test materials (usually sunscreens, topical nonsteroidal and anti-inflammatory agents, and other possible causative agents) are applied in duplicate for 24–48 hours to normal skin. One set of test sites is then uncovered and irradiated with a broad-spectrum UVA source, usually at 5 J/cm2 from fluorescent PUVA lamps, and the results read 24 and 48 hours later. Strongly positive reactions at sites exposed to both chemical agent and UVA, with no reactions at the covered control sites, confirms a diagnosis of photoallergy. Occasionally, however, contact irritation or contact allergy occurs in both sites, making a diagnosis of photoallergy uncertain. One should also be alert to the possibility that all irradiated sites may become positive, suggesting that underlying widespread UVA photosensitivity is responsible. Furthermore, the identification of potential photoallergens is still primitive, often with separation of phototoxicity from photoallergy uncertain. Once again, testing for photoallergy is best conducted in regional centers or by physicians with appropriate experience.

The current authors (Travis W. Vandergriff and Paul R. Bergstresser) are grateful to the authors of this chapter in the previous edition for leaving behind an outstanding framework that we employed as our starting point for this updated edition. Special thanks go to John L. M. Hawk and James Ferguson.