Kevin McKenna1 and Sally Ibbotson2
1Dermatology Department, Belfast Trust, Belfast, UK
2Photobiology Unit, University of Dundee, Ninewells Hospital and Medical School, Dundee, UK
Phototherapy (or ‘light therapy’) is a form of treatment for skin conditions involving the administration of non-ionizing radiation (most commonly within the ultraviolet part of the electromagnetic spectrum) in a controlled manner to the skin. Photodynamic therapy and laser treatment, which may be regarded as specialized forms of phototherapy, are discussed elsewhere (see Chapters 22 and 23) but a discussion of extracorporeal photochemotherapy (ECP), in which the target is white blood cells rather than skin, is included here. A wide variety of dermatoses are responsive to phototherapy, which is used most commonly for psoriasis and eczema.
The UVB part of the spectrum is usually defined as that between 280 and 320 nm (though the International Commission on Illumination (CIE) has set the upper limit of the range at 315 nm). It can be delivered with the ‘full spectrum’ (broad-band UVB lamps 270–350 nm, which include shorter wavelengths from the UVA spectrum) or with just a small part (narrow-band UVB [TL-01R] lamps 311–313 nm) (Figure 21.1). In recent years narrow-band UVB has largely replaced the use of broad-band UVB [1, 2].
The UVA spectrum is commonly defined as that between 320 and 400 nm and is usually used in combination with a psoralen photosensitizer, either orally or topically, when it is known as photochemotherapy (PUVA) [3, 4]. Skin diseases responsive to PUVA are similar to those responsive to UVB phototherapy, particularly psoriasis and eczema. ECP (photopheresis) involves the addition of psoralen to the patient's white blood cells after they have been separated from whole blood ex vivo.
The photoactivated white blood cells are then irradiated with UVA and reinfused back into the patient. Photopheresis is used to treat the erythrodermic stage of cutaneous T-cell lymphoma, graft-versus-host-disease and other conditions [5, 6]. The longer wavelengths of UVA are known as UVA-1 (340–400 nm) and have been shown to be beneficial in a number of chronic dermatoses including atopic eczema and sclerosing skin disorders [7, 8].
UVB phototherapy has anti-inflammatory, immunosuppressive and cytotoxic properties. The mechanisms of its action are unclear, but include the induction of cis-urocanic acid, Langerhans cell depletion, altered antigen presentation, decreased activity of natural killer (NK) cells and apoptosis of T lymphocytes and keratinocytes [9, 10, 11].
The mechanisms of action of PUVA include the cross-linking of DNA by psoralen photoadducts, the inhibition of DNA replication, Langerhans cell depletion, immunosuppressive effects on T-lymphocyte function and migration and the restoration of Th17/regulatory T-cell imbalance in psoriasis [12, 13, 14]. Photopheresis results in dendritic cells acquiring antigen from apoptotic lymphocytes, which elicit a specific immune response without causing systemic immunosuppression [15, 16, 17]. UVA-1 phototherapy penetrates deeper into the dermis and induces interstitial collagenase and several cytokines, resulting in a softening of sclerotic skin [18, 19]. In addition, UVA-1 phototherapy causes a reduction in tumour necrosis factor α (TNF-α) in the skin, amongst other mediator effects; it is also cytotoxic with T-cell apoptosis being a prominent feature [20].
The beneficial effects of sunlight or heliotherapy on the skin have been known since antiquity. The treatment of vitiligo with psoralen extract from seeds in combination with sunlight is recorded in ancient Egyptian, Indian and Chinese manuscripts [3, 21, 22]. The primary therapeutic component of sunlight, ultraviolet radiation, would not be discovered until 1801 by Johann Ritter [23]. The lethality of UV radiation to bacteria was demonstrated by Downes and Blunt in 1877. The father of modern phototherapy, Niels Finsen, demonstrated the effectiveness of UV therapy for lupus vulgaris at the end of the 19th century [24, 25]. Finsen went on to be awarded the Nobel Prize for his work in 1903.
It was suggested by Alderson in 1923 that a mercury quartz lamp be used to treat psoriasis [26]. The combination of tar and UVB therapy for psoriasis was promoted by Goeckerman in the USA in 1925, and the combination of UVB and dithranol (anthralin) for the same by Ingram in England in 1953 [27, 28]. A broad-band fluorescent UVB system for the treatment of psoriasis was developed by Wiskemann in the 1960s. These lamps had a much higher output than the mercury quartz lamp. The action spectrum for the clearance of psoriasis with a peak at 313 nm was defined in 1976 by Parrish and Jaenicke in the USA [29]. This work paved the way for the development of the more specific and efficient narrow-band UVB phototherapy for psoriasis in the 1980s [30, 31].
Photochemotherapy was revived in 1947 with the isolation of 8-methoxypsoralen and 5-methoxypsoralen from plant extracts [32, 33]. El Mofty et al. demonstrated the clinical benefit of these psoralens with natural sunlight or with UV lamps for the treatment of vitiligo [34]. Photochemotherapy or PUVA was revolutionized by Parrish et al. in 1974 with the introduction of a high-intensity fluorescent UVA tube for the treatment of psoriasis [35]. A decade later, extracorporeal photochemotherapy or photopheresis was introduced by Edelson for the treatment of cutaneous T-cell lymphoma (CTCL) [5]. ECP is recommended as first line treatment for erythrodermic CTCL. In 1981, Mutzhas and colleagues first reported the development of an irradiation device that emitted long-wave UV radiation (UVA-1; 340–400 nm) [36]. It was noted that it readily provoked pigmentation and polymorphic light eruption (PLE). It was thus initially used for PLE provocation testing. The therapeutic potential of UVA-1 was not to be recognized until the early 1990s when it was reported to be beneficial for atopic eczema [7].
Ultraviolet radiation (UVR) is that part of the solar electromagnetic spectrum that lies between X-rays and visible light. UVR is normally defined as UVA (320–400 nm), UVB (280–320 nm) and UVC (<280 nm), although the International Commission on Illumination uses 315 nm rather than 320 nm as the boundary between UVA and UVB [37].
UVC radiation is absorbed by the earth's atmosphere before it reaches the surface of the earth. UVB radiation is significantly more effective (100–1000 times more potent) than UVA at producing erythema, cellular effects and DNA damage. UVA radiation is less energetic but penetrates more deeply into the skin, particularly the longer UVA-1 wavelengths, and as such has been implicated in chronic skin photoageing [38].
The most common means of producing UVR artificially is by the passage of an electric current through mercury vapour enclosed in a fluorescent tube. The excited electrons of the mercury are absorbed by a phosphor coating on the inside of the tube, which results in re-emission of radiation of longer wavelengths by the process of fluorescence (Figure 21.2). By changing the phosphors used, a variety of different spectra in the UVA or UVB regions can be produced [39]. Tubes commonly used include narrow-band UVB: TL-01 (Waldmann/Cosmedico/Hybec/Lumenis); broad-band UVB: TL-12 (Philips) or the more selective UV-6 (Sylvania); and UVA: TL-09 (Philips) lamps for PUVA. UVA-1 can be produced either by low output fluorescent tubes, such as the TL-10 (Philips) or by the higher output metal halide UVA-1 sources (Sellamed, Dr Hönle, Dermalight), which require significant cooling [40].
A wide variety of narrow-band (TL-01) UVB units are available, including whole body cabins, whole body panels, small panel irradiators and point sources (Figure 21.3) [1]. Whole body cabins contain 1800 mm long fluorescent tubes that line the walls in front of reflective metal surfaces which enable greater dose uniformity and greater treatment efficiency to be achieved. Whole body panels necessitate rotation of the patient to provide uniform irradiation: careful attention to position and posture is required to avoid under- or overdosing. Furthermore, these units are a significant UV hazard to others, who must avoid passing in front of the panel. They may, however, be useful for home phototherapy. Small panel irradiators are used for the treatment of palmar and plantar skin. Point source devices avoid unnecessary irradiation of unaffected skin but care is needed to avoid under- or overdosing at overlap areas.
PUVA units are either whole body cabins or small panel irradiators for the treatment of the hands and feet or other localized areas of the body, such as the lower legs or scalp [37].
Space and expense may be saved by using a cabin with a combination of UVB and UVA tubes. However, as only half the number of tubes of each type can be accommodated, longer treatment times are required. They also pose a significant hazard if the wrong tubes are selected and so in general are not recommended.
To provide consistency and repeatability of treatment doses it is essential that the irradiance of the UV unit being used is determined at regular intervals. The dose is calculated from the formula below:
It is important that this is performed using calibrated radiometers that are traceable back to a national reference. This ensures that treatment delivered at one phototherapy centre is the same as that given at any other phototherapy centre. Some whole body cabins have built-in radiometers, but these are unreliable. They may give inaccurate readings when tubes are changed, if the radiometer is not cleaned or if shielding occurs with very obese patients.
To determine the mean UV irradiance to which a patient will be exposed in a whole body cabin, a designated patient irradiance (DPI) is determined [37]. This is for a subject of average height and build at chest, waist and knee level. The irradiance should be measured at the anterior, posterior and both lateral positions, that is, at 12 body sites. This can be determined in one of two ways, the direct or the indirect methods. In the direct method the irradiance at the specified sites must be measured while the investigator (usually from the nearest medical physics department) is standing in the unit. The skin and eyes must be protected with suitable protective clothing and goggles. In the indirect method the irradiance is measured whilst the unit is unoccupied. This can be done with the radiometer clamped to a stand whose position can be adjusted up or down (Figure 21.4). The mean value of the DPI is multiplied by a correction factor (typically 0.8–0.9) to take into account the occupancy effect of the patient in the cabin.
Psoriasis is the most frequent indication for both UVB and PUVA [2, 41, 42, 43]. Narrow-band (TL-01) UVB (NB-UVB) can be used to treat all variants except generalized pustular and erythrodermic psoriasis, for which PUVA can, however, be considered [44]. NB-UVB is more effective than broad-band UVB (BB-UVB) and is on average of similar efficacy to PUVA for the treatment of psoriasis (Figure 21.5). Thin plaques as seen in guttate and seborrhoeic psoriasis respond readily to UVB therapy. UVB is generally used in preference to PUVA: it has a lesser skin cancer risk than PUVA, is easier to administer and, with its shorter exposure times, facilitates a greater throughput of patients. PUVA should be considered in those patients who have failed to respond to UVB or whose duration of remission following UVB is consistently of short duration.
Atopic eczema has been treated effectively with BB-UVB [45], combined UVB and UVA [46], UVA-1 [7], NB-UVB [47] and PUVA (Figure 21.6) [48]. It is commonly stated that it is preferable that UVA-1, if available, should be used for acute flares and NB-UVB should be used for chronic disease, although there is no evidence to support this. A flare of disease is often seen in the early stages of treatment, usually due to the heat generated in the cabinets. Lower dose increments and more prolonged treatment courses than used for psoriasis are often required [49]. Clinical impression is that relapse also tends to be more frequent than in psoriasis, although there is no firm objective evidence for this. PUVA can be considered if UVB has failed or in severe atopic eczema in children associated with growth retardation.
Narrow-band UVB and PUVA have been used effectively for the treatment of CTCL (mycosis fungoides) [50, 51, 52]. UVB is particularly useful for patch stage CTCL, but PUVA is preferable for the plaque stage. Increased inflammation may be seen in the early stages of treatment but usually settles without stopping therapy. ‘Sanctuary sites’ such as the flexures, which are less accessible to phototherapy, may be areas where disease persists.
Vitiligo may respond both to NB-UVB and PUVA [53, 54, 55]. NB-UVB is more effective than PUVA but response is variable and treatment usually has to be prolonged, often over many months to a year or more. Affected areas that tend to be more photoresponsive include those on the face and those of recent onset, of limited extent, containing pigmented hairs or involving non-acral sites. Acral areas typically respond less well but are often the sites that bother the patient most. Tanning of normal skin, which exaggerates the contrast with vitiliginous skin, is more obvious in patients with darker skin types. Patients of skin type I and II are more at risk of burning. Patients should be carefully selected and the risks and benefits clearly discussed.
Narrow-band UVB and PUVA are equally effective for desensitization of PLE but the former is used more frequently nowadays as it has a better side effect profile [56, 57]. It is usually administered in spring time or before travel to sunny locations. Provocation of the rash, which occurs in approximately 50% of patients, can be managed with topical corticosteroid creams and reduction of dose. Patients are typically given 15 treatments (three times per week for 5 weeks) for three consecutive years prior to a year without treatment to assess whether there is any change in disease expression. Evidence as to the optimal regimen to use is lacking.
Many other skin conditions have been treated with UVB and PUVA therapy with variable efficacy. Representative examples of such conditions are listed in Table 21.1.
Table 21.1 Other conditions that may respond to UVB phototherapy or PUVA.
Condition | Comment |
Alopecia areata [58, 59] | Efficacy of PUVA is not clearly established; relapse is common |
Graft-versus-host disease [60, 61] | PUVA is helpful in extensive lichenoid disease |
Granuloma annulare [62] | Can be beneficial |
Lichen planus [63, 64] | Response is variable; PUVA is useful for palmar disease |
Necrobiosis lipoidica [65] | PUVA can be beneficial – in ulcer healing and camouflage effect |
Nodular prurigo [66, 67] | Can be beneficial; relapse is common |
Other photodermatoses (actinic prurigo, chronic actinic dermatitis, erythropoietic protoporphyria, solar urticaria) [68, 69, 70] | Can be effective but close supervision is required; consider referral to a centre with special interest |
Palmo-plantar pustulosis [71] | PUVA is more effective |
Pityriasis lichenoides chronica [72] | Can be beneficial |
Pityriasis rosea [73] | Rarely needed; used in severe symptomatic disease |
Pityriasis rubra pilaris [74, 75] | PUVA only |
Pompholyx eczema [76, 77] | PUVA is more effective |
Pruritus [78, 79] | Can be beneficial; UVB is effective for itch of renal failure or liver disease |
Seborrhoeic dermatitis [80] | Responsive to UVB but rapid relapse is common |
Subcorneal pustular dermatosis [81, 82] | Can be beneficial |
Systemic sclerosis/morphoea [83] | PUVA is beneficial |
Urticaria [84] | Good evidence to support a role for UVB |
Urticaria pigmentosa [85] | Can be beneficial; relapse is common |
Narrow-band UVB therapy is now much more commonly used than PUVA. This is primarily due to the well-documented cumulative risk of skin cancer associated with PUVA [86, 87]. In addition, many studies have shown the efficacy of NB-UVB for psoriasis and other dermatoses to be comparable with that of PUVA [1]. It should not be forgotten that PUVA penetrates more deeply into the skin than UVB radiation and is thus better suited for thick plaque disease and for treating the palms and soles (Box 21.1). The relapse rate of psoriasis following PUVA therapy is less than that of NB-UVB therapy. PUVA can also be highly effective for generalized pustular or erythrodermic psoriasis.
Patients should be assessed for their suitability for either UVB or PUVA therapy prior to starting therapy (Box 21.2). Individual risk and benefit must always be assessed.
The evidence base for using UVA-1 relates mainly to its use in atopic eczema, the sclerosing skin conditions, in particular morphoea and scleroderma, and various subtypes of lupus erythematosus. The application of UVA-1 has also been investigated for several other skin diseases (e.g. urticaria pigmentosa, disseminated granuloma annulare, CTCL and psoriasis in HIV-infected patients), although the evidence base is largely restricted to case series and case reports [40, 88]. In general, UVA-1 is probably most useful as an option to be considered for some of the rarer skin diseases where other treatment options are limited. It is not usually first line phototherapy and its availability at present is limited to specialist centres.
UVA-1 phototherapy can be of benefit, particularly for atopic prurigo, and, if it is available, should be considered for the patient who has failed to respond to NB-UVB or PUVA [7, 89, 90] Studies have shown that high- and medium-dose UVA-1 are of equivalent efficacy although medium-dose treatment is more effective than low-dose. UVA-1 has been shown to be superior to potent topical corticosteroid and broadband UVA/B. UVA-1 is of equivalent efficacy or possibly slightly less effective than NB-UVB and is less effective than PUVA.
There is evidence of utility of UVA-1 for the treatment of cutaneous sclerosis in morphoea and systemic sclerosis, particularly if the disease is progressive, symptomatic and/or restricting movement [91]. There is also limited evidence to support its use in the treatment of cutaneous graft-versus-host disease [92], nephrogenic fibrosing dermopathy [93] and both extragenital and genital lichen sclerosus [94, 95]. UVA-1 at medium dose has been shown to be superior to NB-UVB and even low-dose treatment can be effective, although prolonged treatment courses may be needed.
Paradoxically there is good evidence for the use of very low-dose (about 6 J/cm2 per individual dose) UVA-1 for systemic lupus erythematosus (LE) with demonstrable reduction in systemic disease activity [96, 97]. UVA-1 has also been used successfully for subacute cutaneous LE, tumid LE and, although generally less responsive, for chronic cutaneous (discoid) LE.
Extracorporeal photochemotherapy, or photopheresis, has evidence of benefit for the treatment of erythrodermic cutaneous T-cell lymphoma [5, 98, 99, 100] and for graft-versus-host disease [98, 101, 102, 103]. It has also been used for a wide range of skin diseases such as systemic sclerosis [104], some immunobullous disorders [105], psoriasis [106], atopic eczema [107], lichen planus [108], LE [109], scleredema [110] and dermatomyositis [111]. Evidence of benefit for any of these conditions is generally weak. It is of interest that the best evidence for the efficacy of ECP is in the prophylaxis of cardiac transplant rejection [112].
Extracorporeal photochemotherapy is licensed for the treatment of CTCL, particularly in patients with erythrodermic disease, including those with Sézary syndrome. Patients who respond best have a disease of shorter duration, have near normal CD8+ cell counts [113], are immunocompetent [114] and have circulating Sézary cells [115, 116]. Evidence for ECP for the treatment of non-erythrodermic mycosis fungoides is poor [117–119]. Added benefit has been claimed when ECP is combined with interferon [115, 120], bexarotene [121] or electron beam therapy [122].
Graft-versus-host disease complicating allogeneic bone marrow transplantation can be subdivided into acute and chronic phases. There is evidence that ECP is of benefit for the chronic phase, particularly for cutaneous or mucosal involvement. There is little evidence of benefit for associated hepatic disease [6, 98, 101, 102, 103].
Phototherapy is most frequently referred to as UVB therapy and can be delivered using either of two different sources: broad-band or narrow-band. It has been shown that NB-UVB is more effective than BB-UVB, and thus BB-UVB is no longer commonly used. It is recommended that the starting dose of UVB is based on the minimal erythema dose in order to reduce the risks of burning on the one hand or under-treatment on the other, and to detect unsuspected photosensitivity [1].
The minimal erythema dose (MED) is defined in the UK as the dose of radiation that produces minimal, just perceptible, erythema at 24 h post irradiation (in other countries it is defined as the dose that produces well-defined erythema) [123]. There is a poor correlation between MED and skin type [124]. The UVB MED gives an objective measure of a patient's cutaneous reactivity to UVB.
The MED can be determined by the use of a homemade template or by using automatic or semiautomatic UV exposure devices. The homemade template can be made of any UV-opaque material in which there are a number of apertures (typically eight) through which different doses of UVB can be delivered to the skin. The range of doses used are selected based on skin phototype and are a geometric series [125] such as: 25, 50, 70, 100, 140, 200, 280, 390, 550, 650 and 770 mJ/cm2 for NB-UVB (Figure 21.7). The usual site for testing with the template is on the mid-back, avoiding the midline. The UVB testing source uses a bank of UVB tubes identical to those used in the irradiation cabinet and whose irradiance is regularly monitored. In some units the actual irradiation cabinet is used with the patient standing or sitting within it, but this requires careful covering of the patient to avoid unnecessary erythema and to some patients this can be very claustrophobic. In addition to being time consuming it also means that this cabinet is not available for treatment whilst the procedure is undertaken. Alternatively a hand-held semiautomated UV device can be used which is placed directly on the skin being tested. The plate in contact with the skin is perforated with a number of apertures fitted with UV-attenuating mesh foil. Essentially, the finer the mesh, the more UV it blocks. One such device delivers 10 doses with an attenuation factor of 1.26 between apertures allowing a sequence of doses to be delivered (Figure 21.8) [126].
The evidence for optimal dose regimens mainly relates to psoriasis. The action spectrum for the clearance of psoriasis peaks in the UVB region around 311–313 nm and it was with this in mind that NB-UVB lamps were designed by Philips. To achieve a clearance of psoriasis it is not necessary to use erythemogenic dose regimens. The starting dose is commonly initiated at a percentage of the MED (usually 70% or 50%) to minimize the risk of developing significant erythema. Interestingly, one study showed increased erythemal episodes with a 50% MED starting dose regimen when compared with 70% MED or arbitrary skin phototype-based starting dose regimens, and showed no difference in efficacy between any of the three regimens [127]. However, undertaking a MED or a small area test dose is important to detect unsuspected abnormal photosensitivity.
As the skin acclimatizes to UVB therapy by epidermal thickening and pigmentation, it is necessary to increase the dose incrementally. Comparison of a low increment regimen (20%) with a high increment regimen (40%) showed the former to result in 50% fewer episodes of erythema requiring postponement of treatment [128]. A low-dose regimen is advocated with a reduction to 10% increments if significant erythema develops.
Treatment three times a week is preferred to a five times per week schedule for patients of skin phototypes I–III. This is because the advantage of a marginally more rapid clearance of psoriasis with the latter regimen is outweighed by the inconvenience to the patient of a higher frequency of treatment, and the higher cumulative UVB dose received [129]. Both twice and three times weekly regimens are effective for psoriasis, although three times a week treatment results in faster clearance.
Clearance of psoriasis can usually be achieved with 20–25 treatments, although more prolonged courses may be needed, particularly with stubborn disease. Atopic eczema also usually requires more prolonged treatment courses than those used for achieving clearance in psoriasis. A course of 15 treatments is usually used for PLE desensitization, although optimal treatment parameters have not been established.
UVB phototherapy is most commonly delivered in a cabinet in which the patient is surrounded by TL-01 fluorescent tubes (Figure 21.9). Hand and foot units can also be used for localized hand and foot disease.
Involved psoriatic skin has a much higher tolerance of UVB than the surrounding uninvolved skin. With UVB devices that can be targeted directly at individual psoriatic plaques, higher doses of UVB can therefore be used [130, 131]. Treatment with 2–6 multiples of MED can clear localized plaques of psoriasis after five or six treatments. Coherent UVB in the form of the 308 nm excimer laser can also be used for stubborn localized psoriatic plaques, although availability is limited [132].
Home phototherapy has been particularly advocated for patients who cannot attend hospital because of geographical, work, economic or other reasons [133, 134]. Such treatment has raised concerns of suboptimal therapy, greater risks and medicolegal implications, which have not been confirmed [135]. However, a recent study from the Netherlands, in which home phototherapy was compared with out-patient UVB, found that the treatments had similar efficacy and that the cumulative doses and rates of short-term side effects were comparable [136]. Cost analysis of home phototherapy versus hospital UVB therapy in Scotland concluded that home phototherapy was cost effective and that the development of this means of treatment delivery should be encouraged [134, 137].
The most frequently used psoralens for oral use are 8-methoxypsoralen (8-MOP) and 5-methoxypsoralen (5-MOP) [3, 49, 138]. The former is used preferentially as it is considered to be more effective and is less expensive. Nausea is a common side effect of 8-MOP and, if troublesome, switching to 5-MOP is helpful. The dose of psoralen used is most commonly determined on the basis of body weight: 0.6 mg/kg for 8-MOP and 1.2 mg/kg for 5-MOP. Some authorities advocate calculating the dose according to body surface area: 25 mg/m2 for 8-MOP and 50 mg/m2 for 5-MOP [139]. The latter method of dosing has been shown to improve the therapeutic effect of PUVA in psoriasis.
Oral 8-MOP is taken 2 h before treatment with a light meal and 5-MOP should be taken at 2–2.5 h before treatment.
Topical psoralens have been used for centuries for the treatment of vitiligo. Psoralens currently used for topical therapy include 8-MOP and trimethylpsoralen (TMP) [140]. In equivalent concentrations TMP is up to 30 times more phototoxic, although it is now rarely used [141]. Psoralen can be applied topically in a variety of ways: a bath solution for whole body treatment and soak, paint, cream or gel for hands and feet, scalp and other localized areas.
Topical therapy is preferable to oral therapy in patients with hepatic dysfunction, gastrointestinal disease, cataracts, poor compliance with eye protection and risk of drug interactions (e.g. warfarin), and to allow shorter irradiation times (e.g. for children, the elderly or those with claustrophobia).
A frequently used bath PUVA regimen in the UK involves dissolving 30 mL of a 1.2% 8-MOP lotion in 140 litres of water (final concentration 2.6 mg/L) [140]. The patient bathes in this for 15 min, followed by immediate exposure to UVA. When treating hands and feet with topical PUVA there should be a 30 min delay prior to irradiation to allow psoralen absorption into palmar and plantar skin [142].
The minimal photoxic dose (MPD) is determined in a similar way to MED except that it is measured 2 h (or 2.5 h for 5-MOP) after the patient has ingested a standard dose of psoralen (8-MOP or 5-MOP). Typical UVA test doses used include 0.5, 0.7, 1.0, 1.4, 2.0, 2.8, 3.7, 5.5, 7.7 and 10.8 J/cm2 but may vary between centres. The test sites are read after 72–96 h [143].
MPD testing allows the determination of the optimal starting dose, allows identification of under-dosage or poor absorption and may identify patients with abnormal photosensitivity.
The starting dose should ideally be based on the patient's MPD, and between 40% (topical PUVA) and 70% (oral PUVA) of the MPD is recommended [144, 145]. If MPD testing is unavailable or the patient's skin is too extensively involved to measure the MPD, the starting dose is based on skin phototype: I, 0.5 J/cm2; II, 1.0 J/cm2; III, 1.5 J/cm2; IV, 2.0 J/cm2. The starting dose of bath PUVA should always be based on MPD testing because of the risk of severe photosensitivity.
Doses are normally increased in 20–40% increments; lower increments may be indicated if significant erythema develops. If barely perceptible asymptomatic erythema develops, the dose can be kept the same until the erythema settles.
PUVA therapy is usually administered twice weekly as PUVA erythema is delayed [146]. It had been assumed that topical PUVA erythema was maximal at 72 h but more recent studies have shown the peak to occur at between 96 and 144 h [147, 148]. Some centres still use PUVA three times per week but this regimen can result in an increased incidence of burning and is therefore not advised, particularly given the cumulative carcinogenic risk of PUVA.
The number of exposures required for PUVA therapy of dermatoses is similar to that of TL-01 UVB therapy (see earlier in the chapter). As there is a significantly increased risk of skin cancer with increasing cumulative exposure to PUVA, the number of treatments should be kept to a minimum and maintenance therapy avoided.
As with UVB phototherapy, PUVA is usually delivered in a cabin with fluorescent tubes that surround the patient. PUVA can also be delivered locally with units that can be used to irradiate the hands and feet (Figure 21.10). Bath PUVA and localized PUVA are also available for patients when systemic PUVA is not recommended or is less appropriate.
Following the ingestion of psoralen tablets, patients are required to wear UVA-absorbing glasses before therapy and for at least 12 h after therapy (24 h for children and patients with pre-existing cataracts or atopic eczema) [49]. During therapy UV-blocking goggles must be worn. Eye protection with bath PUVA is only necessary in patients with very extensive disease when the risk of systemic absorption becomes significant.
Combination therapy (i.e. UVB or PUVA combined with topical or systemic therapy) is used to increase the clinical response of psoriasis and to decrease the number of phototherapy exposures and thus the cumulative dose.
A variety of topical agents, including emollients [149], tar [150], dithranol (anthralin) [151], tazarotene [152] and vitamin D analogues such as calcipotriol [153, 154], have been used in combination with phototherapy. Evidence for significant adjuvant or UV sparing effects is limited. Studies of both calcipotriol and tazarotene have shown faster clearance of psoriasis with significantly lower median cumulative UV exposure. Vitamin D analogues are best used after UV exposure as they are unstable if exposed to UV radiation [155]. Topical immunomodulatory agents such as tacrolimus and pimecrolimus should be avoided. A multicentre study has shown significant benefit when patients are treated with TL-01 UVB in the presence of Dead Sea salt solution [156]. In practice, if phototherapy is used optimally, the role of adjunctive therapies is less clear.
There is good evidence that the combination of PUVA or UVB with an oral retinoid (so-called Re-PUVA or Re-UVB, respectively) increases efficacy and has a significant dose sparing effect [157]. Retinoids are also beneficial due to their protective effects against skin cancer development [158]. Concomitant systemic immunosuppressive agents should be avoided in combination with phototherapy in order to avoid an augmented risk of skin cancer. Eruptive skin cancers have been reported when ciclosporin is used in patients who have received high cumulative doses of PUVA [159]. NB-UVB enhances the efficacy of a variety of biological agents including alefacept [160], etanercept [161], adalimumab [162] and ustekinumab [163]. However, most studies did not include a phototherapy-alone control group. The potential for enhanced skin carcinogenesis is a significant concern with such combinations and they should, in general, be avoided.
UVA-1 phototherapy uses long wavelength UVA radiation (340–400 nm), filtering out the erythemogenic UVA2 and UVB wavelengths (290–340 nm). UVA-1 phototherapy has been used at a wide variety of doses, typically described as high (>60 J/cm2), medium (30–60 J/cm2), low (10–20 J/cm2) and very low (<10 J/cm2). There is very little evidence to define what the optimal UVA-1 treatment regimen should be, although common practice is stated below [40, 86].
This is determined using a range of doses which reflect whether a low, medium or very high-dose schedule is being used.
The starting dose is based on the MED and is usually 50% of the latter.
Ten to 20% increments can be used at each treatment depending on whether or not erythema develops.
Treatment is administered 3–6 times per week.
In general, a trial of a minimum of 15 treatments would be recommended before deciding on whether to continue treatment or not.
UVA-1 can be delivered either via low irradiance lamps (such as Philips TL-10) or from metal halide portable or whole body UVA-1 units. The latter provide a much higher irradiance thus allowing increased dose delivery.
The photopheresis procedure takes place in three stages: leukapheresis, photoactivation and reinfusion [6, 164]. Whole blood is removed from the patient and then centrifuged to separate the red blood cells (RBCs) from the white blood cells (WBCs). The WBCs (along with some plasma and RBCs) form the buffy coat. The latter is then mixed with saline and 8-MOP (UVADEX®) and photoactivated by being passed through a plastic film which is irradiated by UVA lamps (Figure 21.11). The irradiated buffy coat is then reinfused into the patient.
Minimal photoxic dose testing is not necessary as irradiation occurs outside the body in a machine.
The dose of UVA delivered to the patient's lymphocytes is 1–2 J/cm2.
None, the above standard dose is used for each treatment.
It is administered on 2–3 consecutive days once per month (for graft-versus-host disease every 2–3 weeks). The frequency of treatment can be increased in non-responders.
Treatment is generally continued for 6 months before declaring treatment failure if there is a lack of response. In patients who respond to therapy this can be continued at a decreased frequency to provide maintenance control if necessary.
Extracorporeal photochemotherapy is most commonly delivered using the UVAR XTSR system (Therakos) as described previously. Although 8-MOP may be administered orally, plasma levels can be erratic, side effects such as nausea, vomiting and diarrhoea may be troublesome, and there is a risk of burning from exposure to ambient UV [165, 166].
Erythema and burning are the most common side effects of treatment. Erythema peaks at 12–24 h and can be associated with pain, swelling and blistering (Figure 21.12) [1, 167]. Burning is more likely in patients who are of skin phototype I/II, are obese, have inadvertent exposure of previously covered skin (e.g. after a hair cut), are taking phototoxic drugs or herbal medication or have unknown photosensitivity (e.g. lupus erythematosus). Burning may also be due to operator error in programming the cabinet timer or selecting the correct dose for the correct patient. Mild erythema is managed using topical corticosteroid creams and emollients. Further treatment is withheld until the erythema settles, after which a lower incremental dose regimen should be introduced. More severe reactions may require a short course of oral corticosteroids and/or non-steroidal anti-inflammatory drugs (NSAIDs).
This is usually due either to phototherapy-induced dryness of the skin or to the underlying disorder being treated. It can usually be controlled by emollients [168].
This most commonly affects the lip and occurs more frequently in those with a past history of such outbreaks [169]. The areas in question can be protected by a sunscreen or visor if necessary.
Blistering of psoriatic plaques is an uncommon side effect of TL-01 phototherapy and is usually asymptomatic [170]. It typically occurs mid-way through a treatment course and usually settles spontaneously or following dose reduction. It can occur in the absence of erythema or burning.
The precipitation of PLE is a not uncommon occurrence during phototherapy in those who are predisposed to it (Figure 21.13) [168]. This is managed by the use of topical corticosteroid creams and adjustment of the treatment regimen.
Premature cutaneous ageing may be induced or exacerbated by UVB phototherapy. The skin develops a leathery appearance, xerosis, wrinkling, pigmentary changes, loss of elasticity and increased fragility [2].
Broad-band UVB has a carcinogenic risk that is not well defined due to the many confounding variables; these include recreational sun exposure, exposure to sunbeds or PUVA, skin type and exposure to systemic immunosuppressive therapies. A meta-analysis of studies using BB-UVB estimated an excess risk of non-melanoma skin cancer of about 2% per year [171]. Stern's long-term follow-up study of PUVA recipients in the USA showed that the risk of developing genital squamous cell carcinoma was 4.6 times greater in men with psoriasis and a history of significant UVB exposure (>300 treatments) than in men with similar PUVA exposure but without UVB exposure [172]. Genital protection during phototherapy continues to be recommended to minimize this risk.
The carcinogenic risk of NB-UVB, which is being increasingly used for delivering phototherapy, is not fully defined in humans but is less than that of PUVA. Extrapolation from animal studies would suggest that NB-UVB is 2–3 times per MED more carcinogenic than BB-UVB [173]. It has been suggested that this risk is abrogated in clinical practice, as the number of MEDs to clear psoriasis with NB-UVB is less than a third that required with BB-UVB. Clinical studies to date have not identified a significant risk of skin cancer in patients treated with NB-UVB, but further long-term follow-up studies are required [174, 175, 176]. As the risk of skin cancer with NB-UVB is not yet quantified, this therapy should continue to be used with a degree of caution.
The erythema induced by PUVA is delayed, thus incurring a risk of cumulative injury if treatments are more frequent than twice per week. Peak erythema following the administration of oral and topical PUVA occurs between 72 and 120 h and between 96 and 144 h, respectively [177, 178]. Acute PUVA phototoxicity can range from mild erythema to severe pain with oedema, blistering and systemic upset including malaise and fever. PUVA erythema not only appears later than UVB erythema but lasts longer. A rarer form of phototoxic reaction to PUVA is nail damage with photo-onycholysis [179] or subungual haemorrhage [180]. Episodes of burning are more common with topical PUVA than with oral PUVA due to greater epidermal concentrations of psoralen and uneven absorption into the skin. The smaller risk of burning from oral PUVA can be lessened further by using 5-MOP rather than 8-MOP [181]. Predisposing factors for the development of erythema are similar to those discussed for UVB- induced erythema.
Management of phototoxic reactions depend on their severity and can range from dose increment adjustments to withdrawal of treatment. Mild to moderate erythema can be managed with moderate to potent topical corticosteroid creams and emollients. Severe reactions with blistering may require a potent topical corticosteroid such as clobetasol propionate supplemented by a short course of systemic corticosteroids.
Pruritus occurs in up to 25% of patients and is associated with dryness of the skin [180, 182]. This can usually be controlled with emollients. A rare but severe idiosyncratic side effect is PUVA pain that can last for weeks or months and does not appear to correlate with PUVA phototoxicity [183, 184]. PUVA pain is important to identify as management is difficult. Low-frequency electrotherapy [185], topical capsaicin [186] and oral gabapentin [187] or phenytoin [188] have all been advocated. The continuation of PUVA or further courses in the future are contraindicated as recurrence of pain is common.
Nausea is a common side effect in patients treated with oral 8-MOP PUVA. This can be lessened by taking the psoralen with a light meal or by using an antiemetic. Alternatively switching to 5-MOP or topical PUVA may help [49].
Subepidermal bullae have been reported to be induced by PUVA [3, 189]. Phototoxic reactions to PUVA can result in bullae that occur most commonly acrally. Induction of bullous pemphigoid or a flaring of this in a patient with pre-existing disease is recognized.
Polymorphic light eruption can be precipitated by PUVA as can undiagnosed lupus erythematosus [168].
As with UVB phototherapy, outbreaks of herpes can be precipitated by PUVA [190]. Management is no different.
Less commonly observed side effects include transient nail pigmentation, a facial dermatitis resembling seborrhoeic dermatitis, folliculitis, facial hypertrichosis [168, 180], disseminated superficial actinic porokeratosis [191], lichenoid eruptions [192], headaches, insomnia [193] and allergic reactions to 8-MOP [194, 195].
PUVA induces features resembling photoageing (dermatoheliosis) except that changes are not confined to sun-exposed sites. The clinical features are similar to those described for UVB therapy (see UVB chronic effects on the skin) but pigmentary changes and xerosis are more prominent. These changes are cumulative and dose related [168, 196].
These manifest as multiple hyperkeratotic papules, most commonly located on non-sun-exposed sites, for example the legs, trunk and sides of the hands, feet and digits [197]. They occur most frequently in those patients exposed to high cumulative doses and are associated with an increased risk of non-melanoma skin cancer.
These are large, irregular, stellate, darkly pigmented macular lesions (Figure 21.14) that histopathologically consist of large and atypical melanocytes. They occur most frequently in patients of skin type I/II who have received high numbers of PUVA treatments [198, 199]. There is no proven link between PUVA lentigines and the development of melanoma. Interestingly, however, a recent study has found that the T1799A BRAF mutation is commonly found in PUVA lentigines, suggesting premalignant potential [200].
There is now substantial evidence for the carcinogenic risk of PUVA therapy in large cohorts of patients followed up for prolonged periods of time, particularly from the USA and Sweden [201, 202, 203, 204, 205, 206, 207]. There is a marked and dose-dependent increased risk of squamous cell carcinoma (SCC): in the most recent report from the American cohort, the risk of developing one or more SCCs in a year was strongly associated with the total number of PUVA treatments (350–450 versus <50 treatments; incidence rate ratio (IRR) 6.01) [208]. The overall excess risk of SCC in the high exposure group was significantly greater than this (IRR = 20.92). A 16-fold increase in incidence of genital SCC was observed in males from the same cohort whose exposure to PUVA was high (>240 treatments) rather than low (<140 treatments) [172]. It is recommended that lifetime exposure should be limited to 150–200 treatment sessions or a cumulative dose of 1000–1500 J/cm.2 Maintenance therapy should be avoided [49].
Important risk factors for PUVA- induced skin cancer include skin types I/II, a previous history of skin cancer, previous exposure to radiotherapy or arsenic [209, 210, 211, 212], immunosuppressive therapy, especially ciclosporin [159, 204, 213, 214], UVB therapy [215] and the presence of PUVA keratosis [197] or lentigines [198].
PUVA is mutagenic [216, 217], immunosuppressive [218, 219, 220] and carcinogenic [221, 222]. p53 mutations have been detected in 65% of PUVA- induced SCCs. The mutational pattern consists not only of PUVA-specific mutations (T to A transversions) but also ‘fingerprint’ mutations, as seen following solar exposure and UVB therapy (i.e. C to T or CC to TT mutations) [223, 224, 225]. PUVA has various effects on the immune system including decreasing the number of CD3+, CD4+ and CD8+ T lymphocytes in both the epidermis and dermis and affecting Langerhans cell immune expression [218, 219]. In addition, PUVA has also been reported to reduce a variety of circulating lymphocyte subsets [220]. Decreased immune surveillance may contribute to the carcinogenic effects of PUVA. To date there is no evidence to support an increased risk of skin cancer occurring in patients who have 227received bath or topical PUVA [226].
PUVA can stimulate the growth of melanoma cells and induce melanocytic tumours in animal experiments [227, 228]. A fivefold increase over the expected incidence of melanoma 15 years after first treatment was reported from a US 16-centre PUVA cohort [229]. Follow-up of the same cohort (an average of 2.25 years later) has showed a further increase to nine times the expected incidence [230]. The melanomas were more common in patients of skin type I/II and in those who had received at least 250 PUVA exposures. European study groups, on the other hand, have not found an increased risk of melanoma with PUVA [207, 211, 231, 232]. In particular, the study of 4799 Swedish patients treated with PUVA with an average follow-up of 16 years did not detect an increased incidence of melanoma [207]. Due to the long latent period of melanoma, further follow-up of these cohorts will be required to clarify melanoma risk in patients treated by PUVA.
It is known that patients with psoriasis have a threefold increased risk of lymphoma compared with the general population [233]. As PUVA alters immune function and surveillance there is a concern that the risk of internal malignancy, particularly of lymphoproliferative neoplasms, may be increased. To date studies have not demonstrated a significant association of PUVA with the development of lymphoma [234], nor a consistent relationship with other internal malignancies [205, 207, 235].
Psoralens can penetrate the ocular lens, where 8-MOP has been detected in humans at 12 h and in rats at 12–24 h after systemic administration [236, 237]. Following exposure to UVA, psoralens can bind to proteins in the lens, where they can accumulate as a result of the lack of cell turnover [238].
Cataract development following PUVA has been reported in some animal experiments [3]. Several studies of patients who have received PUVA have not shown an association between PUVA and cataract development [239, 240, 241]. A study of 82 patients who refused to wear protective sunglasses after PUVA did not observe any lens abnormalities. Decreased lacrimation and conjunctival hyperaemia occurred in some cases, however [242]. It is recommended that protective eyewear be worn for 12 h after the ingestion of psoralen and for 24 h in individuals with pre-existing cataracts or who may be at increased risk of cataract (e.g. children and patients with atopic eczema).
UVA-1 is generally well tolerated. Expected acute adverse effects include erythema, pigmentation, induction of PLE and recrudescence of herpes simplex virus infection [40]. Erythema induced by UVA-1 has a biphasic time course, with an early peak at 15–60 min and a delayed peak that may be maximal as early as 8 h after exposure, although this latter peak may be broad and plateau between 8 and 24 h [243].
UVA-1 phototherapy, in common with NB-UVB and PUVA, is photogenotoxic, photomutagenic and carcinogenic in a mouse model. There is no evidence as yet of increased risk of skin cancer in humans. There are, however, concerns about its potential carcinogenic risk as it has been shown to induce cyclobutane pyrimidine dimers, particularly thymine dimers, and the basal layer is particularly vulnerable [244]. Oxidative DNA damage occurs at both genomic and nucleotide levels and there is also altered calcineurin signalling, which may impair tumour suppression [245].
Adverse events occurring with ECP are uncommon providing psoralen is added to the treatment bag and not given orally [246]. Side effects reported include transient hypotension, low-grade pyrexia, an increase in erythema of the skin and anaemia with long-term use [164, 247].
Patients requiring phototherapy should be appraised for factors relating to skin cancer risk (see Box 21.2) [49]. A referral form completed in the clinic should document the patient's risk factor profile and should include documentation of skin type, previous natural sunlight/UVB/PUVA/sunbed use, previous skin cancer or precursors, previous systemic immunosuppressive drug therapy and exposure to radiotherapy (Figure 21.15). In addition, a full skin examination should be performed to assess solar damage and to look for the presence of multiple freckles/moles, premalignant skin lesions and skin cancers.
It is also important to document all medication, especially any oral photoactive drug or herbal product being taken such as a thiazide diuretic or St John's wort (Hypericum perforatum). Most photoactive drugs absorb maximally in the UVA region and only some (including thiazides and quinine) have extension of their action spectrum into the UVB region. In one study, patients taking NSAIDs, calcium channel antagonists and phenothiazines were reported to have a lower NB-UVB MED, a factor which may need to be taken into consideration in patients being assessed for phototherapy [248]. Photosensitivity with UVA-1 phototherapy is more likely in patients taking photoactive drugs [40]. With PUVA, the overwhelming photosensitization by psoralen usually means that other photoactive drugs are unlikely to cause any additional clinically important photosensitization.
Topical therapies should also be assessed. Calcineurin inhibitors such as tacrolimus should be discontinued [249]. Patients with epilepsy should be asked whether seizures can be induced by light exposure. As PUVA therapy is mutagenic this treatment should be avoided in pregnancy. Oral PUVA should be used with caution if the patient has significant liver dysfunction.
Patients who are commencing phototherapy or PUVA should be counselled regarding both the acute and chronic side effects of treatment. This should be reinforced by providing a patient information leaflet. Patients should be advised regarding the appropriate use of emollients. The importance of reporting of any new medication and of avoiding fragrance-containing products, sunbathing and sunbeds during the course of treatment should be emphasized. Also, it is important to advise patients to maintain a consistent hairstyle and to explain the rationale for this advice. Males should be advised of the need to protect the genitalia. The importance of wearing protective eyewear should be stressed. The treatment cabin should be shown to the patient and the protocol explained, stressing the importance of regular attendance. A consent form should be signed by the patient before treatment is commenced.
Both patient and staff safety should be ensured. For patient safety, see Box 21.3.
Individuals who may be exposed to UV radiation include nurses and nursing assistants, medical physicists, doctors or, less commonly, physiotherapists. The most important risk to staff is exposure to UVB. Staff who conduct dosimetry should wear goggles, face shields, appropriate clothing and a sunscreen (SPF>30) if they have to enter the cabinet with the lamps on. When UV cabinets are being used, the lamps should be switched off before opening or entering the cabinet.
It is well recognized that systemic PUVA therapy is associated with a dose-related increased risk of non-melanoma skin cancer, particularly squamous cell carcinoma. It has been recommended that the maximum lifetime dose should not exceed 1000–1500 J/cm2 or 150–200 exposures [49]. The increased carcinogenic risk seen with systemic PUVA has not been seen with bath PUVA. Bath PUVA involves lower UVA doses but results in a much higher epidermal psoralen concentration than oral PUVA. Thus the potential for induction of mutagenic DNA lesions remains and ongoing vigilance in these patients is required. It is important to take into account total UVA dose in addition to treatment numbers when considering the carcinogenic risk of PUVA. It is recommended that patients who exceed these maximum thresholds of exposure should have an annual skin examination carried out to detect pre-malignant and malignant skin lesions. This is particularly important for those patients with other risk factors, such as those of skin type I/II.
Because NB-UVB was introduced only relatively recently, long-term follow-up data equivalent to those available for PUVA are as yet lacking. The preliminary data have not, however, suggested a significant increase in skin cancer risk from NB-UVB [174, 175, 176]. Further long-term data are required from follow-up of large cohorts of patients.
Mathematical models have been created in attempts to quantify the risk of skin cancer resulting from NB-UVB but the methodologies used are very variable. One group, on the premise that an increased risk of non-melanoma skin cancer of no more than 50% is ‘acceptable’, has recommended limiting lifetime exposure to 450 treatments [253]. However, there are concerns about defining a ceiling number of treatments in the absence of robust clinical data to support this and, more importantly, without assessing risk/benefit on an individual basis.
A phototherapy service should have a designated responsible person or lead clinician (usually a consultant dermatologist). This individual should ensure that high quality standards are maintained. Guidelines and protocols should be maintained and updated. Staff should be suitably trained and their knowledge should be updated through formal clinical professional development (CPD). The service should be delivered by a multidisciplinary team which should consist of the lead clinician, lead phototherapy nurse, medical physicist and other relevant assistant staff. Areas that require being audited regularly include documentation, adverse events, equipment maintenance and risk management.
In most of Europe and North America there is a legal requirement for documentation that should include patient records, dosimetry and equipment maintenance, adverse events and summary records of departmental activity.
Referral forms for phototherapy or PUVA must be completed by a dermatologist or authorized dermatology practitioner. The content of this form should include patient identity, the disease being treated, type of phototherapy requested, current systemic medications, previous phototherapy and details about the absence or presence of any contraindications or risk factors for phototherapy. A record of informed consent must be obtained and patients must receive a patient information leaflet. A formal nursing assessment should be recorded. Phototest results, treatment type (including details of psoralen dose), incremental regimen prescribed, all treatment visits, UV doses administered, response to treatment and adverse incidents must all be systematically documented.
The UV equipment used must be ‘CE’ marked as a medical device and be appropriately maintained by an approved engineer or medical physicist. Assessment of the electrical safety and inspection of the integrity of all UV units should form part of a regular maintenance schedule, careful records of which must be kept.
Records must be kept of all UV calibration readings, which must be carried out by an approved medical physicist. Any radiometer used must be calibrated with traceability to the UK National Physics Laboratory or equivalent authority elsewhere [37]. A policy for the replacement of low output or failed UV lamps should be in place.
All adverse incidents should be recorded. The circumstances should be clearly described and action taken. These incidents should be discussed at department meetings to minimize the risk of recurrence of such events.
These reflect the workload of the department and should include the total number of patients treated, total number of treatment sessions with a breakdown by treatment modality, number of patients on the waiting list and average waiting time from referral to treatment.
A formal risk assessment of potential hazards in the phototherapy department, especially from inadvertent exposure to UV radiation, should be undertaken at least annually. The waiting areas must be separate from treatment areas. Warning signs should be displayed on entering areas where UV is being used. An infection control and hygiene policy should be established, particularly with regard to cleaning the phototherapy equipment.
The phototherapy service should be regularly audited to ensure it is maintaining standards and adhering to guidelines [1]. Many aspects of the service can be audited, including: quality of phototherapy records; clinical outcome including adverse events; workload and activity statistics; waiting list management and access to the service; and discharge and follow-up procedures.
The setting up of a phototherapy service involves the participation of different groups – including medical and nursing staff, the medical physics service, local health service management and hospital engineers (with regard both to electrical supply and safety). The involvement of patient organizations may help persuade administrators and funders of the desirability of providing such a service.
It is vital that adequate space is provided. For a modest district-based service serving a population of 200 000 people there should be sufficient room to accommodate one NB-UVB cabinet, one PUVA cabinet, a set of UV units for the local treatment of hands and feet and UV units for phototesting. If topical bath PUVA is to be offered, bathing facilities are necessary. In addition to the treatment area, which requires adequate ventilation to cope with heat emitted by the UV equipment, space is also required for a waiting area with access to toilet facilities and for reception and nursing administration areas with appropriate IT facilities and document storage space.
Policies need to be established for ongoing UV cabinet maintenance, lamp replacement and radiometry. Staff should have a background of experience and training in dermatology and should have attended a core phototherapy course followed by a period of supervised practice in a unit where UV therapy is already provided. Protocols and evidence-based guidelines must be in place and staff training completed before the service is opened.
Phototherapy is a highly effective and safe treatment option for many skin diseases and continued awareness of the availability of this form of treatment is important.
One of the factors limiting further therapeutic advances in phototherapy sources has been our relative lack of understanding of the action spectra for the clearance of diseases other than psoriasis. Most of the developmental work relating to optimizing phototherapy regimens has been with psoriasis in mind. Future work to define the optimal wavelengths for the treatment of diseases such as eczema, mycosis fungoides and vitiligo is required in order to promote the development of light sources and treatment regimens that are disease-specific.
Advances in technology are enabling the development of compact, portable, efficient and inexpensive UV and visible light-emitting diode (LED) light sources. Taken together with modern methods of instant communication, these may facilitate the development of patient self-administered home phototherapy services with ready access to expert supervision and assistance coordinated from a central phototherapy department. Clinical governance is essential and the development of national networks to centralize this is ongoing. The future of phototherapy is encouraging. As new insights into skin biology are gained, there will be opportunities to refine treatment further to improve its already good efficacy/risk profile.