Methoxsalen

Received 13 December 2002; accepted 16 April 2003

ABSTRACT

We compared the characteristics of psoralen and ultraviolet A (PUVA) erythema in skin photosensitized by bath or oral methoxsalen in 20 subjects. Erythema was assessed visually and with a reflectance instrument at 24 h intervals for 7 days. In addition, narrowband ultraviolet B (TL-01 UVB) erythema was examined in 19 of these subjects at 4, 8, 12, 24, 48 and 72 h and in another nine subjects at 12, 15, 18, 21 and 24 h. Both bath and oral PUVA exhibited broad erythemal peaks beyond 72 h. For topical PUVA the lowest minimal phototoxic dose (MPD) occurred at 120 and 144 h (P = 0.01 and 0.03 compared with 72 h). Oral PUVA erythema peaked earlier at 96 h: the MPD was significantly lower at 96, 120 and 144 h compared with 72 h (P = 0.001, 0.01 and 0.02, respectively). At 120 h, bath PUVA had a significantly steeper slope compared with oral PUVA. The TL-01 UVB minimal erythema dose was significantly lower at 12 h compared with 24 h (P = 0.019). The majority of subjects were at maximal erythema at 12 h (22 of 28) and 15 h (eight of nine). Our results suggest that peak erythema for bath PUVA, oral PUVA and TL-01 UVB occurs at 120, 96 and 12-15 h, respectively.

Abbreviations: BB UVB, broadband ultraviolet B; MED, minimal erythema dose; MPD, minimal phototoxic dose; PUVA, psoralen and ultraviolet A; TL-01 UVB, narrowband ultraviolet B.

INTRODUCTION

Photochemotherapy (psoralen and ultraviolet A [PUVA]) is a well-established second-line treatment for psoriasis (1,2) and is effective for a range of other dermatoses (3,4). In recent years concern over PUVA carcinogenesis has lead to the increased use of narrowband ultraviolet B (TL-01 UVB) phototherapy. This form of therapy is now considered to be the treatment of choice for patients with moderate to severe psoriasis by the majority of dermatologists (5).

During any form of ultraviolet light therapy, the dose administered is dependent on the development of erythema. Erythema is the most frequent adverse effect encountered in PUVA therapy (2,4), and, in addition, it may influence the long-term carcinogenic risk because PUVA carcinogenesis is related to the cumulative ultraviolet A (UVA) dose, which in turn is dependent on the erythemal response. Greater understanding of the characteristics of PUVA and TL-01 UVB erythema would therefore permit prescription of a more accurate starting- and incremental-dose regimen, which would improve short-term, and possibly long-term, safety.

Previously, maximal PUVA erythema was assumed to occur 48-72 h after UVA irradiation in psoralen-sensitized skin (6-9). This formed the basis of minimal phototoxic dose (MPD) determination at 72 h and a 72 h interval between treatments (10,11). Contrary to this unconfirmed assumption, recent studies have shown independently that peak PUVA erythema after photosensitization with topical and oral methoxsalen occurs at or beyond 96 h (12-16), suggesting that the current MPD assessment time underestimates the phototoxic effect of PUVA. It is not known if the time-course and dose-response relationship of erythema induced by topical and oral methoxsalen within individuals are similar.

The characteristics of UVB-induced erythema have not been studied in detail. In one study, broadband UVB (BB UVB) erythema peaked between 8 and 24 h after irradiation (17). Limited data exist that suggest that the erythemal characteristics of BB UVB and TL-01 UVB (311-313 nm) are similar (18,19).

We conducted this study to compare the characteristics of erythema induced by bath or oral PUVA and by TL-01 UVB within subjects.

MATERIALS AND METHODS

Ethical approval was obtained from the Tayside Research Ethics Committee, Dundee, Scotland.

Subjects. Twenty healthy volunteers (12 women; median age 32 [range 21-56] years) with skin types I (n = 2), II (n = 13) and III (n = 5) were enrolled for the PUVA study. TL-01 UVB erythema was determined in 19 of these subjects and in another nine volunteers (skin types II [n = 8 ] and III [n = 1]). None of the subjects had a history of photosensitivity, was receiving photoactive medication, or had sunbathed within 3 months of participating in the study.

Photoirradiation apparatus. The UVA irradiation source was an MPD-testing UVA unit equipped with 10 x 2 ft Waldmann F15W/T8 fluorescent tubes, constructed by our medical physics department. The spectral output ranged between 320 and 400 nm (peak 350 nm; 0.008 of the emission spectrum was between 280 and 320 nm, and 0.98 was between 320 and 400 nm, when normalized to a 250-400 nm bandwidth [1.00]). The skin surface irradiance was 5.7 mW/cm^sup 2^ at a distance of 20 cm, measured using a Waldmann UV meter.

UVB irradiation was performed using 8 x 4 ft Philips TL 40W/01 fluorescent tubes, with spectral output between 311 and 313 nm (0.80 of the emission spectrum was between 280 and 320 nm, and 0.20 was between 320 and 400 nm, when normalized to a 250-400 nm bandwidth [1.00]). The skin surface irradiance was 1.76 mW/cm^sup 2^ at a distance of 20 cm (measured using an IL 1400A radiometer with diffuser TD #20737, filter UVB 1#17944).

TL-01 UVB phototesting. Eight 1.5 cm^sup 2^ test sites on each subject's untanned, upper-back skin were irradiated with a TL-01 UVB dose series (50, 70, 100, 140, 200, 280, 390 and 550 mJ/cm^sup 2^; 1.4 incremental factor).

PUVA phototesting. Before skin photosensitization, 12 x 1 cm^sup 2^ test sites on the contralateral upper-back skin were exposed to a series of UVA doses up to 12 J/cm^sup 2^. This was to ensure that erythema did not occur in the absence of psoralen.

One forearm of each subject was assigned to receive bath methoxsalen (1.2% methoxsalen, Crawfords Pharmaceuticals, Milton Keynes, UK). After 15 min immersion (3 mL of 1.2% methoxsalen in 15 L of water at 37[degrees]C; final concentration 2.4 mg/L), skin was gently dried, and 12 x 1 cm^sup 2^ test sites on the flexor forearm were immediately exposed to increasing UVA doses (0.06, 0.08, 0.1, 0.15, 0.22, 0.28, 0.33, 0.56, 0.82, 1.2, 1.8, 2.2 and 2.7 J/cm^sup 2^; 1.2-1.7 incremental factor; the first or last 12 doses were used).

Following this procedure, on the same day, an oral dose of methoxsalen (Meladinine, Galderma, UK) was administered at a standard dose of 0.6 mg/kg. Two hours later, 12 x 1 cm^sup 2^ test sites on the contralateral flexor forearm were irradiated with a UVA dose series (1.0, 1.2, 1.5, 1.8, 2.2, 2.7, 3.3, 3.9, 4.7, 5.6, 6.8, 8.2, 10 and 12 J/cm^sup 2^; 1.2 incremental factor; the first or last 12 doses were used).

The last 12 doses were used initially, but because of blistering reactions at the higher doses, doses were subsequently reduced to the first 12 doses.

Because topical and oral PUVA phototesting were performed on different test sites, no interaction between the methods of psoralen delivery would have occurred.

Erythema assessment. TL-01 UVB erythema was assessed at 4, 8, 12, 24, 48 and 72 h after irradiation in 19 subjects. Because of practical limitations, TL-01 UVB erythema was examined at 12, 15, 18, 21 and 24 h in another nine subjects. PUVA erythema was examined at 24 h intervals for 7 days. At each time point the minimal erythema dose (MED) for UVB or the MPD for PUVA was recorded, and we used the widely accepted definition of the dose of radiation required to induce just perceptible erythema in this study. In addition, the intensity of erythema was measured using a reflectance instrument (Dia-Stron(R) Limited, Andover, Hampshire, UK) (20). Triplicate erythema meter readings from irradiated and adjacent nonirradiated test sites were taken. At each test site the mean increase in erythema index was calculated and plotted against the logarithm of the UVA or UVB dose (21). A sigmoidal dose-response curve was constructed by computer fit (22). For each curve the dose of irradiation required to induce an increase in erythema index of 0.025 (D^sub 0.025^), an objective measurement of the MED and MPD, and the maximum slope were determined. Test sites were examined carefully for pigmentation at each time point, and measurements were interpreted with caution if pigment was detected.

Statistical analysis. Values were expressed as median (range). The lowest MPD and MED, D^sub 0.025^ and the maximum slope of the erythemal dose-response curves were compared with the data at the clinically relevant time points of 24 h for TL-01 UVB and 72 h for PUVA using Wilcoxon matched-pairs signed-rank test because geometric dose series were used. Correlation of data was analyzed using Spearman's rank correlation coefficient.

RESULTS

TL-01 UVB erythema

In the 19 subjects where TL-01 UVB erythema was assessed from 4 to 72 h, erythema was detectable at the first assessment time 4 h after irradiation, and the median time to reach peak response was 12 (8-24) h. In all 28 subjects the lowest MED occurred at 12 h (170 [70-390] mJ/cm^sup 2^), which was significantly lower than the conventional 24 h MED assessment time (200 [100-550] mJ/cm^sup 2^; P = 0.019) (Fig. 1a). The majority of the subjects were at maximal erythema at 12 h (22 of 28) and 15 h (eight of nine). If the MED had been determined at 24 h (MED^sub 24h^), peak erythema would have been missed in 15 of the 28 (54%) subjects. In addition, if, as is commonly practiced, 70% of the MED^sub 24h^ had been given as the first treatment dose, 10 of the 28 (36%) subjects would have received a dose equal to or exceeding the lowest MED, compared with only two (7%) if 50% of the MED^sub 24h^ had been given (P = 0.0078; 95% CI for difference in percentages, 10-46.5%).

Objectively, from the dose-response data, D^sub 0.025^ at 12 h (184 [108-650] mJ/cm^sup 2^) was significantly lower than at 24 h (252 [130-571] mJ/cm^sup 2^; P = 0.0019) (Fig. 1b). The slope of the erythemal dose-response curve remained constant from 8 to 48 h.

Pigmentation was not evident in any of the test sites up to 72 h (n = 28).

Topical and systemic PUVA erythema

No erythema was evident 24 h after radiation in unsensitized skin (n = 20).

PUVA erythema was examined in 20 subjects. The time-course of PUVA erythema after photosensitization by bath and oral methoxsalen followed a similar trend with broad erythemal peaks beyond 72 h, as determined by the lowest MPD and D^sub 0.025^. However, oral PUVA peaked earlier at 96 h compared with 120 h for bath PUVA.

Visually, for bath PUVA the lowest median MPD occurred at 120 and 144 h (0.25 [0.1-0.82] J/cm^sup 2^), and this was significant when compared with 72 h (0.33 [0.15-1.8] J/cm^sup 2^) (P = 0.01 and 0.03, respectively) (Fig. 2a). For oral PUVA the median MPD was significantly lower at 96 (1.65 [1.0-5.6] J/cm^sup 2^), 120 (1.65 [1.0-6.8] J/cm^sup 2^) and 144 h (1.5 [1.0-5.6] J/cm^sup 2^) compared with 72 h (2.2 [1.2-6.8] J/cm^sup 2^) (P = 0.001, 0.01 and 0.02, respectively) (Fig. 2b). The median time to reach peak erythema was 96 (48-144) h for both types of PUVA. For bath PUVA, 13 (65%) and 14 (70%) subjects had reached and remained at peak erythema at 120 and 144 h, respectively. Similarly, the majority of subjects were at maximal response at 96 h (75%), 120 h (70%) and 144 h (80%) for oral PUVA. Only six (30%) and seven (35%) subjects were at peak response at 72 h for bath and oral PUVA, respectively (Fig. 3).

Dose-response data were available only for 19 subjects because insufficient erythema developed in one subject to permit construction of accurate dose-response curves. Mild pigmentation was present in one subject at 120 and 144 h and in two subjects by 168 h for bath PUVA. For oral PUVA, two volunteers had developed faint pigmentation at 96 and 120 h, five at 144 h and six by 168 h. Dose-response data at and beyond 144 h were interpreted with caution because of the development of pigmentation.

Dose-response data corresponded to the visual MPD results (Fig. 4a,b).

Fourteen and 15 subjects had the lowest D^sub 0.025^ at or beyond 96 h for bath and oral PUVA, respectively. For bath PUVA the slope of the dose-response curves was steeper at each successive time point, reaching a maximum between 120 and 144 h (Fig. 5a). In contrast, the slopes for oral PUVA remained constant from 48 to 168 h (Fig. 5b). Bath PUVA had a significantly steeper slope (238 [69-449]) than oral PUVA (105 [28-319]) at 120 h (P = 0.003).

Comparison between topical PUVA, oral PUVA and TL-01 UVB

Erythema induced by bath PUVA, oral PUVA and TL-01 UVB was compared intraindividually (n = 19). At the current MED and MPD assessment limes of 24 and 72 h, respectively, there was a trend toward a steeper curve for TL-01 UVB (169 [37-628]), but this curve was not significantly different from that of bath (142 [43-404]) (P = 0.07) or oral (121 [36-731]) (P = 0.52) PUVA. However, at the time of maximal erythema, i.e. 96 h for oral PUVA, 120 h for bath PUVA and 12 h for TL-01 UVB, bath PUVA (238 [80-449]) had a significantly steeper slope compared with oral PUVA (128 [24-378]) and TL-01 UVB (189 [49-564]) (P = 0.01 and 0.028, respectively). There was a nonsignificant (P = 0.29) trend toward a steeper slope for TL-01 UVB compared with oral PUVA.

Comparison of the two types of PUVA for erythemal sensitivity, as determined by MPD (r^sub s^ = -0.16, P = 0.51), and for the time to reach peak erythema (r^sub s^ = 0.25, P = 0.32) showed no significant correlation. Similarly comparison of bath and oral PUVA with TL-01 UVB failed to identify an association in the erythemal sensitivity (r^sub s^ = 0.17, P = 0.49 and r^sub s^ = 0.13, P = 0.51, respectively) or in the time to achieve maximal response (r^sub s^ = 0.003, P = 0.99 and r^sub s^ = -0.16, P = 0.53, respectively).

DISCUSSION

Our results are in agreement with recent reports and provide further evidence to contradict previous assumptions that peak PUVA erythema occurs 48-72 h after UVA exposure (6-9). We showed that UVA-induced erythema in skin photosensitized by bath and oral methoxsalen exhibits a broad erythemal peak beyond 72 h, as determined by the lowest MPD and D^sub 0.025^. Oral PUVA erythema peaks earlier at 96 h compared with 120 h for bath PUVA.

Current recommendations suggest assessment of the MPD 72 h after UVA (10,11). On the basis of our findings, if the MPD had been determined at this time point, maximal erythema would have been missed in up to 70% and 65% of subjects for bath and oral PUVA, respectively. If either 40% of the bath PUVA MPD or 70% of the oral PUVA MPD had been administered as first treatment, based on existing guidelines (10,11), 30% of subjects would have developed perceptible erythema. Although clinical observation suggests that the incidence of burning after the first PUVA exposure is low, overall, it is the most common adverse effect encountered during therapy (2,4,23). Repeated PUVA administered 72 h apart, i.e. before maximal erythema has developed, may account for this increased risk.

Unlike PUVA, TL-01 UVB displayed a narrow erythemal peak. The time-course of TL-01 UVB erythema is comparable with that of BB UVB (17,18). In a previous report (17), BB UVB erythema was present by 4 h, and maximal reaction occurred between 8 and 24 h after UVB irradiation in eight subjects. The time-course of TL-01 UVB erythema had been examined as part of a study comparing the efficacy of BB UVB and TL-01 UVB phototherapy in the treatment of psoriasis. In this study (18), erythema was visible between 4 and 6 h and peaked between 12 and 24 h for both light sources. Our findings were consistent with this publication. Erythema induced by TL-01 UVB was evident at the first assessment time 4 h after exposure, and the majority of subjects were at peak erythema at 12 and 15 h. Only 45% were at peak response at the current MED measurement time of 24 h. The starting dose of TL-01 UVB varies among phototherapy units, and in our department 70% of the MED read at 24 h is given. If this dose had been administered as a first treatment dose to the volunteers in this study, 39% would have received a dose equal to or exceeding their lowest MED. In clinical practice, however, erythema after the initial treatment is rarely documented. This could be explained by the fact that erythema may have subsided by the time patients attended their subsequent treatment 48 h later. Nevertheless, if treatment is started with a dose producing a transient erythema, a cumulative effect could result and increase the risk of burning later during therapy.

An individual's susceptibility to develop burning during phototherapy and photochemotherapy is related to the steepness of the erythema dose-response curves (24). There are conflicting data with regard to the incidence of erythema during topical and oral methoxsalen PUVA (25-27), but topical PUVA is generally considered to be more phototoxic, especially for trioxalen (28,29). Although this may be related to the different formulations, another plausible explanation is that topical PUVA has a steeper dose-response curve. To test this hypothesis, we compared the slopes of bath and oral PUVA within subjects. We demonstrated that although the slope of the bath PUVA dose-response curve increased in a stepwise fashion at each successive time point, the slope of the oral PUVA curve remained constant. It is of clinical importance that we did not detect a significant difference between the two types of PUVA at 72 h, although at 120 h the slope of response for topical PUVA was significantly steeper than that for oral PUVA by a factor of 2.3.

The historical view that PUVA has a steeper dose-response curve than BB UVB has been refuted (12,30). In one study (30) the mean slope for BB UVB erythema was approximately four-fold steeper compared with oral PUVA erythema determined at 48 h. Ibbotson and Fair (12) also observed that oral PUVA had an approximately two-fold shallower dose-response curve compared with UVB (305 + or - 5 nm). In this study, although there was a trend toward a steeper slope for UVB compared with both types of PUVA, at 24 h for TL-01 UVB and 72 h for PUVA, it was not significant. The reason for this inconsistency is not clear, although it may be due to the different light sources used: a monochromatic source (350 + or - 30 nm) was used in the former study and a broadband fluorescent UVA source in the current study (0.98 emission between 320 and 400 nm and 0.008 emission between 280 and 320 nm). It has been shown previously that this small percentage of UVB emission from UVA sources induces the majority of pyrimidine dimers in keratinocytes in vitro, and, if extrapolated to PUVA phototoxicity in vivo, the differences in UVB emission between the sources may explain these slight differences in the findings (31). However, a comparative study of TL-01 (narrow-band) and TL-12 (broadband) UVB erythema showed no significant difference in dose-response characteristics or persistence of erythema at 24 and 72 h (19). Interestingly, we did demonstrate that at the time of peak erythema, i.e. 12-15 h for TL-01 UVB, 120 h for bath and 96 h for oral PUVA, bath PUVA had a significantly steeper slope than oral PUVA or TL-01 UVB.

In a previous study (30) no significant association was detected between skin type and oral methoxsalen MPD or BB UVB MED. In keeping with this observation, our within-subject comparisons showed no significant correlation between bath and oral PUVA or between PUVA and TL-01 UVB in the threshold erythema dose or in the time to develop maximal erythema. It is thus not possible to predict an individual's bath PUVA MPD from the oral MPD or similarly from the TL-01 UVB MED or vice versa.

Our findings have important clinical implications. Revision of the current MPD protocol appears necessary. These studies were performed predominantly in skin phototype-I and -II subjects, and because of concerns regarding the replacement of erythema by pigmentation, especially with extrapolation to higher skin types, it would seem appropriate to avoid MPD measurement at or beyond 120 h. To accurately determine PUVA erythemal sensitivity, the MPD reading should therefore be conducted at 96 h for both bath and oral PUVA, i.e. at 4 days. In an attempt to account for those individuals who were not at maximal erythema at this time point, adjustment of the MPD-based starting dose, determined at 96 h (MPD^sub 96h^), should be undertaken. On the basis of current recommendations, if 40% and 70% of the MPD determined at 72 h was given as the first treatment dose for bath and oral PUVA, respectively, six of the 20 subjects (30%) would receive a dose that would be equal to or exceed the lowest MPD. If the same percentages were given but with the MPD determined at 96 h, only two (10%) and none (0%) of the subjects for bath and oral PUVA, respectively, would be at risk of erythema. Hence, we recommend that 40% and 70% of the MPD^sub 96h^ should be given as the first treatment dose for bath and oral methoxsalen PUVA, respectively. In addition, increasing the time interval between PUVA treatments to 96 h and a lower UVA incremental dose regimen for bath PUVA may further reduce the risk of burning. However, the impact of these adjustments on the effectiveness of treatment will need to be determined in a controlled clinical study. Further work examining the characteristics of erythema in higher skin types should also be conducted.

Our data for TL-01 UVB suggest that the optimal time to conduct MED measurements is 12-15 h, although this is impractical. One realistic solution is to alter the percentage of MED administered for the initial exposure dose. Our results indicate that a significantly lower proportion of patients would be at risk of developing erythema if the first treatment dose was 50% compared with 70% of the MED determined at 24 h (MED^sub 24h^).

Acknowledgements-We thank Dee Watson, Lynn Fullerton and Chesarea McGeoghie for providing medical physics support and ensured accurate dosimetry. This study was supported by the Anonymous Trust Fund, Tayside University Hospitals, Dundee.

[para]Posted on the website on 28 April 2003.

REFERENCES

1. Williams, R. E. A. (1991) Guidelines for management of patients with psoriasis. Workshop of the Research Unit of the Royal College of Physicians of London. Br. Med. J. 303, 829-835.

2. Spuls, P. I., L. Witkamp, P. M. M. Bossuyt and J. D. Bos (1997) A systematic review of five systemic treatments for severe psoriasis. Br. J. Dermatol. 137, 943-949.

3. Ortel, B. and H. Honigsmann (1998) Phototherapy and photochemotherapy. In Photodermatology, Edited by J.L.M. Hawk pp. 227-240. Chapman & Hall, London.

4. Luftl, M., K. Degitz, G. Plewig and M. Rocken (1997) Psoralen bath plus UV-A therapy: possibilities and limitations. Arch. Dermatol. 133, 1597-1603.

5. Ferguson, J. (2002) What is the role of narrowband UVB in the treatment of psoriasis? Photodermatol. Photoimmunol. Photomed. 18, 42-43.

6. Fitzpatrick, T. B., C. E. Hopkins, D. D. Blickenstaff and S. Swift (1955) Augmented pigmentation and other responses of normal human skin to solar radiation following oral administration of 8-methoxypsoralen. J. Investig. Dermatol. 25, 187-190.

7. Parrish, J. A. (1976) Methoxsalen-UV-A therapy of psoriasis. J. Investig. Dermatol. 67, 669-671.

8. Pathak, M. A., T. B. Fitzpatrick and J. A. Parrish (1976) Pharmacologic and molecular aspects of psoralen photochemotherapy. In Psoriasis: Proceedings of the Second International Symposium (Edited by E. M. Farber and A. J. Cox), pp. 262-271. Yorke Medical Books, New York, USA.

9. Frain-Bell, W. (1985) Treatment. In Cutaneous Photobiology, pp. 170-202. Oxford University Press, Oxford.

10. British Photodermatology Group (1994) British Photodermatology Group guidelines for PUVA. Br. J. Dermatol. 130, 246-255.

11. Halpern, S. M., A. V. Anstey, R. S. Dawe, B. L. Diffey, P. M. Farr, J. Ferguson, J. L. Hawk, S. Ibbotson, J. M. McGregor, G. M. Murphy, S. E. Thomas and L. E. Rhodes (2000) Guidelines for topical PUVA: a report of a workshop of the British Photodermatology Group. Br. J. Dermatol. 142, 22-31.

12. Ibbotson, S. H. and P. M. Farr (1999) The time-course of psoralen ultraviolet A (PUVA) erythema. J. Investig. Dermatol. 113, 346-350.

13. Behrens-Williams, S., C. Gruss, M. Grundmann-Kollmann, R. U. Peter and M. Kerscher (2000) Assessment of minimal phototoxic dose following 8 methoxypsoralen bath: maximal reaction on average after 5 days. Br. J. Dermatol. 142, 112-115.

14. Grundmann-Kollmann, M., U. Leiter, S. Behrens, P. Gottlober, G. Mooser, G. Krahn and M. Kerscher (1999) The time course of phototoxicity of topical PUVA: 8-methoxypsoralen cream-PUVA vs. 8-methoxypsoralen gel-PUVA. Br. J. Dermatol. 140, 988-990.

15. Man, I., Y. K. Kwok, R. S. Dawe, J. Ferguson and S. H. Ibbotson (2003) The time course of topical 8 methoxypsoralen PUVA erythema following 15 minutes and 5 minutes immersion. Arch. Dermatol. 139, 331-334.

16. Man, I., J. McKinlay, R. S. Dawe, J. Ferguson and S. H. Ibbotson (2003) An intra-individual comparative study of PUVA erythema induced by bath 8 methoxypsoralen and 4, 5', 8 trimethylpsoralen. J. Am. Acad. Dermatol. 49, 59-64.

17. Farr, P. M., J. E. Besag and B. L. Diffey (1988) The time course of UVB and UVC erythema. J. Investig. Dermatol. 91, 454-457.

18. Coven, T. R., L. H. Burack, P. Gilleaudeau, M. Keogh, M. Ozawa and J. G. Krueger (1997) Narrowband UV-B produces superior clinical and histopatholocal resolution of moderate to severe psoriasis in patients compared with broadband UV-B. Arch. Dermatol. 133, 1514-1522.

19. Das, S., J. J. Lloyd and P. M. Farr (2001) Similar dose-response and persistence of erythema with broad-band and narrow-band ultraviolet B lamps. J. Investig. Dermatol. 117, 1318-1321.

20. Diffey, B. L., R. J. Oliver and P. M. Farr (1984) A portable instrument for quantifying erythema induced by ultraviolet radiation. Br. J. Dermatol. 111, 663-672.

21. Diffey, B. L. and P. M. Farr (1991) Quantitative aspects of ultraviolet erythema. Clin. Phys. Physiol. Meas. 12, 311-325.

22. Davidson, A. (1999) Calculation of erythema dose-response curves using a spreadsheet. Photodermatol. Photoimmunol. Photomed. 15, 104-108.

23. Morison, W. L., S. Marwaha and L. Beck (1997) PUVA-induced phototoxicity: incidence and causes. J. Am. Acad. Dermatol. 36, 183-185.

24. Sakuntabhai, A., G. R. Sharp and P. M. Farr (1993) Response of psoriasis to twice weekly PUVA. Br. J. Dermatol. 128, 166-171.

25. Lowe, N. J., D. Weingarten, T. Bourget and L. S. Moy (1986) PUVA therapy for psoriasis: comparison of oral and bath-water delivery of 8-methoxypsoralen. J. Am. Acad. Dermatol. 13, 754-760.

26. Collins, P. and S. Rodgers (1992) Bath-water compared with oral delivery of 8-methoxypsoralen PUVA therapy for chronic plaque psoriasis. Br. J. Dermatol. 127, 392-395.

27. Calzavara-Pinton, P. G., B. Ortel, H. Honigsmann, C. Zane and G. De Panfilis (1994) Safely and effectiveness of an aggressive and individualized bath-PUVA regimen in the treatment of psoriasis. Dermatology 189, 256-259.

28. Turjanmaa, K., H. Salo and T. Reunala (1985) Comparison of trioxsalen bath and oral methoxsalen PUVA in psoriasis. Acta Derm. Venereol. (Stockh.) 65, 86-88.

29. Koulu, L. M. and C. Jansen (1984) Skin phototoxicity variations during repeated bath PUVA exposures to 8-methoxypsoralen and trimethylpsoralen. Clin. Exp. Dermatol. 9, 64-69.

30. Cox, N. H., P. M. Farr and B. L. Diffey (1989) A comparison of the dose-response relationship for psoralen-UVA erythema and UVB erythema. Arch. Dermatol. 125, 1653-1657.

31. Woollons, A., C. Kipp, A. R. Young, C. Petit-Frere, C. F. Arlett, M. H. L. Green and P. H. Clingen (1999) The 0.8% ultraviolet B content of an ultraviolet A sunlamp induces 75% of cyclobutane pyrimidine dimers in human keratinocytes in vitro. Br. J. Dermatol. 140, 1023-1030.

Irene Man, Robert S. Dawe, James Ferguson and Sally H. Ibbotson*

Photobiology Unit, Department of Dermatology, Ninewells Hospital and Medical School, University of Dundee, Dundee, UK

*To whom correspondence should be addressed at: Photobiology Unit, Department of Dermatology, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, Scotland, UK. Fax: 1382-646047; e-mail: s.h.ibbotson@dundee.ac.uk

Copyright American Society of Photobiology Jul 2003
Provided by ProQuest Information and Learning Company. All rights Reserved