- Email: [email protected]
Comparison of suberythemogenic and maximally aggressive ultraviolet B therapy for psoriasis
Comparison of suberythemogenic and maximally aggressive ultraviolet B therapy for psoriasis Louis D. Eells, D.O., Jeffrey M. Wolff, M.D., Joan Garloff, R.N., and William H. Eaglstein, M.D, Pittsburgh, PA We prospectively compared the maximally aggressive erythemogenic ultraviolet B (UVB) schedule (LeVine et al, i979) and a suberythemogenic UVB schedule (Frost et al, 1979) in the treatment of psoriasis. Patients were treated with either the maximally aggressive erythemogenic UVB (MAEUVB) schedule or the suberythemogenic UVB (SEUVB) schedule. Each patient received daily topical treatment with 2% crude coal tar ointment on half of the body and the ointment vehicle on the other half of the body. All patients were hospitalized and treated 7 days a week. The treatment results were determined daily by grading preselected reference lesions for erythema, scaling, and thickness. With both UV schedules, no difference in response could be detected between the ointment vehicle side and the tar ointment side. In a comparison of the two UV schedules, patients treated with MAEUVB had no statistically significant difference in the resolution of their psoriasis, and the length of remission was similar. The average energy needed for the SEUVB-treated patients was 189 mjoules/cm~, whereas the average for the MAEUVB-treated patients was 984 mjoule~/cm2. Our findings suggest that neither crude coal tar nor MAEUVB is needed for the optimal UVB treatment of psoriasis in the hospital. (J AM ACADDERMATOL11:105-110, 1984.)
Combinations of topically applied tar preparations and erythemogenic ultraviolet (UV) radiation are called the Goeckerman or modified Goeckerman treatment. 1 Psoriasis is the disease for which the Goeckerman treatment is most frequently used. Although the combination is usually successful, the most desirable method of using ultraviolet B (UVB) irradiation and the importance of tar preparations are still being determined. In 1979 LeVine et alz described the maximally aggressive erythemogenic ultraviolet B From the Department of Dermatology, University of Pittsburgh. Supported by the Skin and Cancer Foundation of Pittsburgh. Accepted for publication Nov. 18, I983. Reprint requests to: Dr. Louis D. Eells, Department of Dermatology, University of Pittsburgh, 3601 Fifth Ave., Pittsburgh, PA 15213.
Table I. Predetermined SEUVB schedule Increment increase Time of previous treatment (rain)
Time
(sec)
Energy (mjoules/cm 2)
0-2 2-5 5-10 > 10
13 26 52 104
1.5 3,0 6.0 12.0
(MAEUVB) treatment for psoriasis. This schedule employs irradiation with UV doses intended to keep the nonlesional skin erythematous throughout the course of therapy. They found that tar was not needed, since MAEUVB combined with either tar ointment or the ointment alone produced equal therapeutic effects. LeVine et al concluded that 105
106
Journal of the American Academyof Dermatology
E e l l s et al
Table II. Summary o f results: MAEUVB Lesion severity score Patient
v
Age
No.
Sex
(yr)
(initial/final)
1 2 3 4 5 6 7
F F M F M F F
40 29 32 40 32 74 23
24/3 25/5 32/20 31/1 32/4 38/4 35/3
1
V&T
Cumulative dose
(initial/final)
(mjoules/era :)
24/3 25/4 31/19 31/0 32/4 38/4 35/0
1,174 261 102 674 1,227 599 2,849 Avg = 984
Total No. of treatments
24 15 tl 17 22 19 22
V: Vehicle treated; V & T: vehicle and tar treated; initial/final: initial severity score/final severity score.
Table I I I . Summary of results: SEUVB Lesion severity score Age
V
No.
Sex
(yr)
(initial/final)
(initial/final)
1 2 3 4 5 6
F F F M F F
30 32 83 30 26 23
31/4 24/4 21/3 33/0 19/3 35/16
30/0 25/4 21/3 33/0 19/0 38/20
Patient
[
V&T
Cumulative dose (mjoules/era 2)
Total No. of treatments
225 205 140 210 286 47 Avg = 189
23 22 15 21 17 11
V: Vehicle treated; V & T: vehicle and tar treated; initial/final: initial severity score/final severity score.
M A E U V B was preferable to mildly erythemogenic, or " s t a n d a r d , " U V dosage schedules because M A E U V B did not require tar. In 1979 Frost et aP reported that UV irradiation in doses that did not cause erythema of nonlesional skin, suberythemogenic UVB (SEUVB), in combination with a tar gel was effective for psoriasis. Recently Lowe et al 4 found that outpatients treated with an aggressive UVB schedule did not benefit from tar preparations. Patients receiving the Lowe SEUVB schedule did benefit from the use of a tar preparation. In this study we compared the effects of the UV schedule described by Frost et aP (SEUVB) and the UV schedule described by LeVine et aI2 (MAEUVB) in hospitalized psoriatic patients treated with 2% crude coal tar ointment on half of their body and the ointment alone on the other half.
MATERIALS AND METHODS Patients. Patients with the chronic form of plaque psoriasis involving greater than 20% of their body surface (as measured by the rule of nines) ~ were admitted to the dermatology service of the University of Pittsburgh-Presbyterian Hospital. Patients who are sensitive to coal tar products and who have a history of unusual reactions to UV irradiation or of taking drugs with photosensitizing potential were excluded from this study. Patients who had received Goeckerman therapy, psoralens with UVA (PUVA), methotrexate, or mycophenolic acid in the preceding 4 months were aiso excluded. UV treatment schedules. Each patient's minimal erythema dose (MED) was determined. The patients were randomly assigned to one of the following treatment schedules: 1, MAEUVB, as described by LeVine et al. 2 The patient's first UV treatment is 0.8 of the MED, If no erythema is present, the dose is increased by 1.2 times
Volume 11 Number I Iuly, 1984
Subel),themogenic and maximally aggressive UVB for psoriasis
3035[~ (2 Iii rr O
MAEUV o VEHICLE ~ VEHICLE AND TAR O
0 25 >" I..-
35 [
SEUV ,', VEHICLE & VEHICLE ANIbTAFt
30 o |
107
UJ
rr aAa o 25 ~ . . . 03
O
O~
> 2o m
.o
Z O
9
,,,'~ I5
t'w W > 20 LI.I m Z O
~ ~176 ~ o
<
z~
~'
_J LLI ~,,,~, o
Oo o
LI./
5
%
5
to 15 20 TREATMENTDAYS
25
30
Cl
I
VO
5
I0
15
20
TREATMENTDAYS
I
I
25
50
Fig, 1. Results from MAEUVB-treated group. Descending regression lines represent mean response to t ~ and vehicle treatment and to vehicle treatment alone (p > 0.05).
Fig. 2. Results from SEUVB-treated group. Descending regression lines represent mean response to the vehicle and to the vehicle with tar (p > 0,05).
the previous day's treatment. If erythema is present, the previous day's dose is repeated. If pain or bright erythema occurs, the U V treatment is not given that day but is given the next day at the last dose used. 2. S E U V B , as described by Frost et al? The patient's first UV treatment is 0.5 of his MED. The dose is not increased so long as there is an improvement in lesion severity score of 1 point per day. If improvement does not occur, the UV dose is increased according to a predetermined schedule (Table I). For both groups the treatments are given every day of the week. Treatments are discontinued when there is complete flattening of the plaques, including the borders. U V s o u r c e . The UV source was a square cabinet 90 cm wide. It contained sixteen 120-cm-long fluorescent sunlamp bulbs (Westinghouse FS40) and sixteen 60cm-long fluorescent lamp bulbs (Westinghouse FS20) mounted vertically on polished metal reflectors. The output of U V energy in the light cabinet was measured with an International Phototherapy Radiometer (IL 500A) fitted with a sensor (SEE 240 No. ] 193), dif-
fuser (W 1359), and filter (UVB No. 129). Energy determinations were made with the sensor held in the center of the cabinet and directed at the center point of each wall and each corner of the cabinet. Determinations were made at 46 cm, 92 cm, and 137 cm from the floor of the cabinet. The U V energy of the cabinet was calculated as the average of these determinations. Evaluations. In each patient, four pairs of reference lesions (eight lesions) were selected for evaluation. The paired lesions were bilateral and similar in size, morphologic features, and site. The paired reference lesions were on nonexposed areas of the trunk or thighs. Each day the reference lesions were graded for erythema, scale, and thickness. These three characteristics were graded on the following scale: 4 = severe; 3 = moderate; 2 = mild; I = trace; and 0 = absent. The total daily score obtained by evaluating the four lesions on each side of the body was called the lesion severity score. The lesion severity score was determined once daily throughout the course of therapy. Topical treatments. The patients were carefully instructed as to the proper methods of applying their
108
Journal of the American Academyof Dermatology
E e t l s et al
RESULTS 35Io 30 1.1.1 tit"
o o 60 25
>-
z~ SEUV 9 MAEUV
9 ~A D 9
rr
~
20
,~o
O3 Z o
UJ 15 _.1 UJ < rr IO nl
,xe
<
,"
o
o
5
i
5
10 15 20 TREATMENT DAYS
25
50
Fig. 3. Comparison of SEUVB- and MAEUVB-treated groups. Regression lines are represented for the mean response of the vehicle-treated psoriatic lesions (p > 0.05). medications and were allowed to use only those preparations provided. The tar preparation was 2% crude coal tar in white petrolatum USP. Each patient received the tar treatment on one side and the vehicle (white petrolatum USP) on the other. The side to receive each was selected randomly. Routine. Each patient was required to follow a specific routine: 1. 2. 3. 4.
Awake; remove tar preparation with mineral oil. Shower with warm water and soap. After shower, receive total UV exposure. After UV exposure, appIy the specified topical agents. 5. In the evening, bathe. 6. Reapply specified topical agents. 7. Wear same pajamas for 3 days.
Statistical methods. The slope of the regression line for each respective treatment group was analyzed for covariance.6 Follow-up. In order to determine the length of remission obtained from each treatment schedule, we conducted a telephone interview of each patient approximately 9 months after the study was completed.
Thirteen patients, three men and ten women, were entered into and completed this study. The average age was 34.8 years (range, 23-74). The sex, age, initial and final lesion severity scores for the vehicle-treated side and the tar-treated side, cumulative UV dose administered, and total number of treatments for each patient are presented in Tables II and III. Six female and one male patient, with an average age of 36.3 years (range, 21-74), received the MAEUVB schedule. No statistically significant difference was found in the response of the vehicle-treated and the tar-treated sides. The daily average severity score for each treatment side on each day and time and the least squares line derived from these points are seen in Fig. 1. The slopes of the tar-treated and the vehicle-treated lines are - 1.033 and - 1.054, respectively (p > 0.05). Two males and four females, with an average age of 32.5 years (range, 23-83), received the SEUVB schedule. No statistically significant difference was found in the response of the vehicletreated and the tar-treated sides. The daily average severity score for each treatment side and the least squares line derived from these points are seen in Fig. 2. The slopes of the tar-treated and the vehicle-treated lines are - 1.090 and - 1.114, respectively (p > 0.05). Comparison of SEUVB and M A E U V B schedules As seen in Fig. 3, the response to these two treatment schedules was similar, the slopes for the MAEUVB Iine and the SEUVB line being - 1.054 and - 1.114, respectively (p > 0.05). The average cumulative UV dose for the two groups was 189 mjoules/cm ~ for the SEUVB schedule and 984 mjoules/cm 2 for the MAEUVB schedule (Fig. 4). Follow-up As presented in Table IV, eleven of thirteen patients were contacted. Four of the eleven (two in each UV treatment group) remained completely clear for 9 months after completing the study. The other patients had slight recurrences of their disease, which has been controlled with topical ap-
Volume 11 Number 1 July, 1984
Suberythemogenic and maximally aggressive UVB for psoriasis
MAEUV
Table I V . Follow-up data
PatientNo.I
MAEUVB
group
3000
SEUV
Av~ean~r~'~,nm . . . . . . . . . . . . . . .n. .~.
300(]
~500
2500
g>._ z0o0
20oo
~_ ~5oo
15oo
~ i0oo _J
,o0o
0
500
SEINB group
109
Average Cure, DoSe 189 mJ/crn2
.~
l 2 3 4 5 6 7
Minimal recurrence 3 mo after discharge Minimal scalp 2 mo after discharge Clear to date* Minimal recurrence 2 mo after discharge Clear to date Minimal recurrence 7 mo after discharge Lost to follow-up
Fl~e at 2 wk following fever and URI Minimal recurrence 6 mo after discharge Clear to date Clear to date Minimal recurrence at 4 mo after discharge Lost to follow-up
URk Upper respiratory infection. *Patient left hospital after eleven treatments. He was treated as an outpatient three times weekly to clearing.
plication o f crude coal tar ointment. None have returned to their pretreatment condition. Special attention should be noted regarding Patient 1 of the S E U V B group, who was reported to have a flare of psoriasis associated with a fever and upper respiratory infection 2 weeks after completing the study. This patient's condition subsequently cleared completely following treatment with topical steroids and tar ointment. DISCUSSION
To our knowledge this is the first comparison of the Frost SEUVB schedule and the LeVine M A E U V B schedule for treating psoriasis with or without tar. Our results confirm the efficacy of SEUVB reported by Frost et ai? A recent controlled study by Halprin et aF also supports the conclusion that constant low UV doses are effective therapy for psoriasis. In the Halprin low-dose U V schedule, UVB therapy was given twice a day, the dose was increased each day during the first week of treatment, and after a dose of approximately 1 MED was reached, daily treatments with this dose were continued. This schedule differs from that of Frost et al, which was used in our comparison. Although the skin of two of the eight patients burned after the Halprin low-dose UV schedule was given (one because of an error in dose), the schedule is generally suberythemogic.
500 E z
3
4
5
6 7 I PATI ENT NUMBER
2
5 4
5
6
Fig. 4. Total UV energy (rnjoules/cm2) necessary to treat each patient in this study.
It does, however, emphasize the effectiveness of lower doses of U V . The Halprin study was a bilateral paired comparison, and the constant low UV dose results equaled the standard erythemogenic (not MAEUVB) control results. Since UV erythema carries over for at least 12 hours, the wisdom of a twice-daily schedule has been questioned. 8 In earlier studies, Fisher 9 found 0.5 MED doses of narrow wavelength bands of approximately 313 nm UV to be somewhat effective; however, his control patients, who received higher, 1.0 MED doses of 313 nm, had better responses. Most recently, Lowe et al 4 found their SEUVB schedule combined with tar treatment to be as effective as their aggressive U V schedule combined with petrolatum treatment. In our study using the Frost SEUVB schedule, hospitalized patients were treated daily with an almost constant UV dose. The Lowe SEUVB schedule was used for outpatients who were treated with UV 3 days a week, and the dose was increased at each treatment. The differences in therapeutic results probably relate to these different SEUVB schedules. Some part of the difference may also be due to different responses between hospitalized patients and outpatients. In our study there was an equal rate of improvement for the SEUVB group and the MAEUVB group (p > 0.05). In the Frost study, a six of seven patients in group B had a greater decrease in their lesional severity score with the SEUVB schedule than with the standard schedule, although the difference did not reach statistical
110
Journal of the American Academy of Dermatology
Eells et al
significance. Both our patients and those in Frost's group B '~ were hospitalized during the trials. It is possible that the results seen with SEUVB in these studies is related to the hospitalized patient population. Patients whose condition requires hospitalization may be those whose psoriasis is either inherently or at that time easily irritated or worsened by irritants, This population might be somewhat irritated by MAEUVB, Alternatively, the condition o f patients receiving M A E U V B might not improve or might worsen because lesions "koebn e r i z e d , " whereas the surrounding normal skin might be protected by tanning. Many studies have found the combination of a tar preparation and erythemogenic U V superior to either alone. 1'1~ However, LeVine et ala could not detect a therapeutic effect of tar when maximal doses of UV were given, and they suggested that the tar and UV combination was effective with mildly erythemogenic UV doses. The results of Lowe et ai ~2 support this conclusion. We were unable to detect a therapeutic effect of tar with either MAEUVB or SEUVB. Our inability to detect a tar effect even in the SEUVB group might be related to the variability of the crude coal tar products. 12 Frost et aP found that SEUVB combined with a tar containing gel was more effective than in combination with gel alone. The speculation that emollients enhance the therapeutic effect of UV by altering psoriatic skin's optical properties may explain the better response we obtained with SEUVB and petrolatum 2 than Frost et al obtained with a gel. Our results are consistent with the suggestion of LeVine et al 2 that tar is not a necessary ingredient in the "modified Goeckerman r e g i m e n . " In addition, our findings support the conclusion of Frost el al :~ that SEUVB is effective, and in this trial SEUVB and M A E U V B induced remissions at equal rates. In telephone interviews we determined that these two schedules were associated with similar lengths o f remission. Telephone interview is not the optimaI method o f follow-up. However, we believe it unlikely that severely affected patients would report their skin to be clear or that patients
whose condition had flared would be unaware of their generally worsened state. These observations suggest the possibility of improving the "modified Goeckerman treatment" by (1) eliminating the tar, which is messy, malodorous, a follicular irritant, and has potential toxic and cmcinogenic effects, and (2) decreasing the UV dose, thus eliminating the frequent UVinduced " s u n b u r n " and possible koebnerization and decreasing the UV carcinogenic potential by using only about 20% of the usual UV dose. Studies of variables such as the effect of hospitalization and treatment intervals may help to determine which modified Goeckerman regimen is best or to define which modified Goeckerman regimen is best for which subset of psoriasis patients. REFERENCES
1. Goeckerman WH: Treatment of psoriasis. Northwest Med 24:229-231, 1925. 2. LeVine NJ, White HAD, Parrish JA: Componentsof the Goeckerman regimen. J Invest Dermatol 73:170-173, 1979. 3. Frost P, Horowitz SN, Caputo RV: Tar geI: Phototherapy for psoriasis. Arch Dermatol 115:840-846, I979. 4. Lowe NJ, Wortzman MS, Breeding J, et al: Coal tin" phototherapy for psoriasis re-evaluated: Erythemogenic versus suberythemogenicultraviolet with a tar extract in oil and crude coal tar. J AM ACADDERMATOL8:781789, 1983. 5. Sabiston DC Jr, editor: Davis-Christopher Textbook of surgery. Philadelphia, 1972, W. B. Saunders Co., p. 275. 6. Brownlee KA: Comparison of several regression lines: Simple analysis of covariance, in Statistical theory and methodology in science and engineering. New York, 1960, John Wiley & Sons, Inc. 7. Halprin KM, Comerford M, Taylor JR: Constant lowdose ultravioletlight therapy for psoriasis. J AM ACAD DERMATOL 7:614-619, 1982. 8. ColskyLC, Halprin KM, Taylor JR, et al: UVB is additive when repeated withinan 8-hour interval. J AMACAD DERMATOL8:760-761, 1983. (Letter to Editor.) 9. Fischer T: UV-light treatment for psoriasis. Acta Derm Venereol (Stockh) 56:473-479, 1976. I0. Marisco AR, Eaglstein WH, Weinstein GD: Ultraviolet light and tar in the Goeckerman treatment of psoriasis. Arch Dermatol 112: I249-1250, 1976. l 1. Petrozzi JW, Barton JO, Kaidbey KK, Kligman AM: Updating the Goeckerman regimen for psoriasis. Br J DermatoI 98:437-444, I978. 12. Lowe NJ, Breeding J, Wortzman MS: The pharmacological variability of crude coal tar. Br J Dermatot 107:475-479, 1982.
Combinations of topically applied tar preparations and erythemogenic ultraviolet (UV) radiation are called the Goeckerman or modified Goeckerman treatment. 1 Psoriasis is the disease for which the Goeckerman treatment is most frequently used. Although the combination is usually successful, the most desirable method of using ultraviolet B (UVB) irradiation and the importance of tar preparations are still being determined. In 1979 LeVine et alz described the maximally aggressive erythemogenic ultraviolet B From the Department of Dermatology, University of Pittsburgh. Supported by the Skin and Cancer Foundation of Pittsburgh. Accepted for publication Nov. 18, I983. Reprint requests to: Dr. Louis D. Eells, Department of Dermatology, University of Pittsburgh, 3601 Fifth Ave., Pittsburgh, PA 15213.
Table I. Predetermined SEUVB schedule Increment increase Time of previous treatment (rain)
Time
(sec)
Energy (mjoules/cm 2)
0-2 2-5 5-10 > 10
13 26 52 104
1.5 3,0 6.0 12.0
(MAEUVB) treatment for psoriasis. This schedule employs irradiation with UV doses intended to keep the nonlesional skin erythematous throughout the course of therapy. They found that tar was not needed, since MAEUVB combined with either tar ointment or the ointment alone produced equal therapeutic effects. LeVine et al concluded that 105
106
Journal of the American Academyof Dermatology
E e l l s et al
Table II. Summary o f results: MAEUVB Lesion severity score Patient
v
Age
No.
Sex
(yr)
(initial/final)
1 2 3 4 5 6 7
F F M F M F F
40 29 32 40 32 74 23
24/3 25/5 32/20 31/1 32/4 38/4 35/3
1
V&T
Cumulative dose
(initial/final)
(mjoules/era :)
24/3 25/4 31/19 31/0 32/4 38/4 35/0
1,174 261 102 674 1,227 599 2,849 Avg = 984
Total No. of treatments
24 15 tl 17 22 19 22
V: Vehicle treated; V & T: vehicle and tar treated; initial/final: initial severity score/final severity score.
Table I I I . Summary of results: SEUVB Lesion severity score Age
V
No.
Sex
(yr)
(initial/final)
(initial/final)
1 2 3 4 5 6
F F F M F F
30 32 83 30 26 23
31/4 24/4 21/3 33/0 19/3 35/16
30/0 25/4 21/3 33/0 19/0 38/20
Patient
[
V&T
Cumulative dose (mjoules/era 2)
Total No. of treatments
225 205 140 210 286 47 Avg = 189
23 22 15 21 17 11
V: Vehicle treated; V & T: vehicle and tar treated; initial/final: initial severity score/final severity score.
M A E U V B was preferable to mildly erythemogenic, or " s t a n d a r d , " U V dosage schedules because M A E U V B did not require tar. In 1979 Frost et aP reported that UV irradiation in doses that did not cause erythema of nonlesional skin, suberythemogenic UVB (SEUVB), in combination with a tar gel was effective for psoriasis. Recently Lowe et al 4 found that outpatients treated with an aggressive UVB schedule did not benefit from tar preparations. Patients receiving the Lowe SEUVB schedule did benefit from the use of a tar preparation. In this study we compared the effects of the UV schedule described by Frost et aP (SEUVB) and the UV schedule described by LeVine et aI2 (MAEUVB) in hospitalized psoriatic patients treated with 2% crude coal tar ointment on half of their body and the ointment alone on the other half.
MATERIALS AND METHODS Patients. Patients with the chronic form of plaque psoriasis involving greater than 20% of their body surface (as measured by the rule of nines) ~ were admitted to the dermatology service of the University of Pittsburgh-Presbyterian Hospital. Patients who are sensitive to coal tar products and who have a history of unusual reactions to UV irradiation or of taking drugs with photosensitizing potential were excluded from this study. Patients who had received Goeckerman therapy, psoralens with UVA (PUVA), methotrexate, or mycophenolic acid in the preceding 4 months were aiso excluded. UV treatment schedules. Each patient's minimal erythema dose (MED) was determined. The patients were randomly assigned to one of the following treatment schedules: 1, MAEUVB, as described by LeVine et al. 2 The patient's first UV treatment is 0.8 of the MED, If no erythema is present, the dose is increased by 1.2 times
Volume 11 Number I Iuly, 1984
Subel),themogenic and maximally aggressive UVB for psoriasis
3035[~ (2 Iii rr O
MAEUV o VEHICLE ~ VEHICLE AND TAR O
0 25 >" I..-
35 [
SEUV ,', VEHICLE & VEHICLE ANIbTAFt
30 o |
107
UJ
rr aAa o 25 ~ . . . 03
O
O~
> 2o m
.o
Z O
9
,,,'~ I5
t'w W > 20 LI.I m Z O
~ ~176 ~ o
<
z~
~'
_J LLI ~,,,~, o
Oo o
LI./
5
%
5
to 15 20 TREATMENTDAYS
25
30
Cl
I
VO
5
I0
15
20
TREATMENTDAYS
I
I
25
50
Fig, 1. Results from MAEUVB-treated group. Descending regression lines represent mean response to t ~ and vehicle treatment and to vehicle treatment alone (p > 0.05).
Fig. 2. Results from SEUVB-treated group. Descending regression lines represent mean response to the vehicle and to the vehicle with tar (p > 0,05).
the previous day's treatment. If erythema is present, the previous day's dose is repeated. If pain or bright erythema occurs, the U V treatment is not given that day but is given the next day at the last dose used. 2. S E U V B , as described by Frost et al? The patient's first UV treatment is 0.5 of his MED. The dose is not increased so long as there is an improvement in lesion severity score of 1 point per day. If improvement does not occur, the UV dose is increased according to a predetermined schedule (Table I). For both groups the treatments are given every day of the week. Treatments are discontinued when there is complete flattening of the plaques, including the borders. U V s o u r c e . The UV source was a square cabinet 90 cm wide. It contained sixteen 120-cm-long fluorescent sunlamp bulbs (Westinghouse FS40) and sixteen 60cm-long fluorescent lamp bulbs (Westinghouse FS20) mounted vertically on polished metal reflectors. The output of U V energy in the light cabinet was measured with an International Phototherapy Radiometer (IL 500A) fitted with a sensor (SEE 240 No. ] 193), dif-
fuser (W 1359), and filter (UVB No. 129). Energy determinations were made with the sensor held in the center of the cabinet and directed at the center point of each wall and each corner of the cabinet. Determinations were made at 46 cm, 92 cm, and 137 cm from the floor of the cabinet. The U V energy of the cabinet was calculated as the average of these determinations. Evaluations. In each patient, four pairs of reference lesions (eight lesions) were selected for evaluation. The paired lesions were bilateral and similar in size, morphologic features, and site. The paired reference lesions were on nonexposed areas of the trunk or thighs. Each day the reference lesions were graded for erythema, scale, and thickness. These three characteristics were graded on the following scale: 4 = severe; 3 = moderate; 2 = mild; I = trace; and 0 = absent. The total daily score obtained by evaluating the four lesions on each side of the body was called the lesion severity score. The lesion severity score was determined once daily throughout the course of therapy. Topical treatments. The patients were carefully instructed as to the proper methods of applying their
108
Journal of the American Academyof Dermatology
E e t l s et al
RESULTS 35Io 30 1.1.1 tit"
o o 60 25
>-
z~ SEUV 9 MAEUV
9 ~A D 9
rr
~
20
,~o
O3 Z o
UJ 15 _.1 UJ < rr IO nl
,xe
<
,"
o
o
5
i
5
10 15 20 TREATMENT DAYS
25
50
Fig. 3. Comparison of SEUVB- and MAEUVB-treated groups. Regression lines are represented for the mean response of the vehicle-treated psoriatic lesions (p > 0.05). medications and were allowed to use only those preparations provided. The tar preparation was 2% crude coal tar in white petrolatum USP. Each patient received the tar treatment on one side and the vehicle (white petrolatum USP) on the other. The side to receive each was selected randomly. Routine. Each patient was required to follow a specific routine: 1. 2. 3. 4.
Awake; remove tar preparation with mineral oil. Shower with warm water and soap. After shower, receive total UV exposure. After UV exposure, appIy the specified topical agents. 5. In the evening, bathe. 6. Reapply specified topical agents. 7. Wear same pajamas for 3 days.
Statistical methods. The slope of the regression line for each respective treatment group was analyzed for covariance.6 Follow-up. In order to determine the length of remission obtained from each treatment schedule, we conducted a telephone interview of each patient approximately 9 months after the study was completed.
Thirteen patients, three men and ten women, were entered into and completed this study. The average age was 34.8 years (range, 23-74). The sex, age, initial and final lesion severity scores for the vehicle-treated side and the tar-treated side, cumulative UV dose administered, and total number of treatments for each patient are presented in Tables II and III. Six female and one male patient, with an average age of 36.3 years (range, 21-74), received the MAEUVB schedule. No statistically significant difference was found in the response of the vehicle-treated and the tar-treated sides. The daily average severity score for each treatment side on each day and time and the least squares line derived from these points are seen in Fig. 1. The slopes of the tar-treated and the vehicle-treated lines are - 1.033 and - 1.054, respectively (p > 0.05). Two males and four females, with an average age of 32.5 years (range, 23-83), received the SEUVB schedule. No statistically significant difference was found in the response of the vehicletreated and the tar-treated sides. The daily average severity score for each treatment side and the least squares line derived from these points are seen in Fig. 2. The slopes of the tar-treated and the vehicle-treated lines are - 1.090 and - 1.114, respectively (p > 0.05). Comparison of SEUVB and M A E U V B schedules As seen in Fig. 3, the response to these two treatment schedules was similar, the slopes for the MAEUVB Iine and the SEUVB line being - 1.054 and - 1.114, respectively (p > 0.05). The average cumulative UV dose for the two groups was 189 mjoules/cm ~ for the SEUVB schedule and 984 mjoules/cm 2 for the MAEUVB schedule (Fig. 4). Follow-up As presented in Table IV, eleven of thirteen patients were contacted. Four of the eleven (two in each UV treatment group) remained completely clear for 9 months after completing the study. The other patients had slight recurrences of their disease, which has been controlled with topical ap-
Volume 11 Number 1 July, 1984
Suberythemogenic and maximally aggressive UVB for psoriasis
MAEUV
Table I V . Follow-up data
PatientNo.I
MAEUVB
group
3000
SEUV
Av~ean~r~'~,nm . . . . . . . . . . . . . . .n. .~.
300(]
~500
2500
g>._ z0o0
20oo
~_ ~5oo
15oo
~ i0oo _J
,o0o
0
500
SEINB group
109
Average Cure, DoSe 189 mJ/crn2
.~
l 2 3 4 5 6 7
Minimal recurrence 3 mo after discharge Minimal scalp 2 mo after discharge Clear to date* Minimal recurrence 2 mo after discharge Clear to date Minimal recurrence 7 mo after discharge Lost to follow-up
Fl~e at 2 wk following fever and URI Minimal recurrence 6 mo after discharge Clear to date Clear to date Minimal recurrence at 4 mo after discharge Lost to follow-up
URk Upper respiratory infection. *Patient left hospital after eleven treatments. He was treated as an outpatient three times weekly to clearing.
plication o f crude coal tar ointment. None have returned to their pretreatment condition. Special attention should be noted regarding Patient 1 of the S E U V B group, who was reported to have a flare of psoriasis associated with a fever and upper respiratory infection 2 weeks after completing the study. This patient's condition subsequently cleared completely following treatment with topical steroids and tar ointment. DISCUSSION
To our knowledge this is the first comparison of the Frost SEUVB schedule and the LeVine M A E U V B schedule for treating psoriasis with or without tar. Our results confirm the efficacy of SEUVB reported by Frost et ai? A recent controlled study by Halprin et aF also supports the conclusion that constant low UV doses are effective therapy for psoriasis. In the Halprin low-dose U V schedule, UVB therapy was given twice a day, the dose was increased each day during the first week of treatment, and after a dose of approximately 1 MED was reached, daily treatments with this dose were continued. This schedule differs from that of Frost et al, which was used in our comparison. Although the skin of two of the eight patients burned after the Halprin low-dose UV schedule was given (one because of an error in dose), the schedule is generally suberythemogic.
500 E z
3
4
5
6 7 I PATI ENT NUMBER
2
5 4
5
6
Fig. 4. Total UV energy (rnjoules/cm2) necessary to treat each patient in this study.
It does, however, emphasize the effectiveness of lower doses of U V . The Halprin study was a bilateral paired comparison, and the constant low UV dose results equaled the standard erythemogenic (not MAEUVB) control results. Since UV erythema carries over for at least 12 hours, the wisdom of a twice-daily schedule has been questioned. 8 In earlier studies, Fisher 9 found 0.5 MED doses of narrow wavelength bands of approximately 313 nm UV to be somewhat effective; however, his control patients, who received higher, 1.0 MED doses of 313 nm, had better responses. Most recently, Lowe et al 4 found their SEUVB schedule combined with tar treatment to be as effective as their aggressive U V schedule combined with petrolatum treatment. In our study using the Frost SEUVB schedule, hospitalized patients were treated daily with an almost constant UV dose. The Lowe SEUVB schedule was used for outpatients who were treated with UV 3 days a week, and the dose was increased at each treatment. The differences in therapeutic results probably relate to these different SEUVB schedules. Some part of the difference may also be due to different responses between hospitalized patients and outpatients. In our study there was an equal rate of improvement for the SEUVB group and the MAEUVB group (p > 0.05). In the Frost study, a six of seven patients in group B had a greater decrease in their lesional severity score with the SEUVB schedule than with the standard schedule, although the difference did not reach statistical
110
Journal of the American Academy of Dermatology
Eells et al
significance. Both our patients and those in Frost's group B '~ were hospitalized during the trials. It is possible that the results seen with SEUVB in these studies is related to the hospitalized patient population. Patients whose condition requires hospitalization may be those whose psoriasis is either inherently or at that time easily irritated or worsened by irritants, This population might be somewhat irritated by MAEUVB, Alternatively, the condition o f patients receiving M A E U V B might not improve or might worsen because lesions "koebn e r i z e d , " whereas the surrounding normal skin might be protected by tanning. Many studies have found the combination of a tar preparation and erythemogenic U V superior to either alone. 1'1~ However, LeVine et ala could not detect a therapeutic effect of tar when maximal doses of UV were given, and they suggested that the tar and UV combination was effective with mildly erythemogenic UV doses. The results of Lowe et ai ~2 support this conclusion. We were unable to detect a therapeutic effect of tar with either MAEUVB or SEUVB. Our inability to detect a tar effect even in the SEUVB group might be related to the variability of the crude coal tar products. 12 Frost et aP found that SEUVB combined with a tar containing gel was more effective than in combination with gel alone. The speculation that emollients enhance the therapeutic effect of UV by altering psoriatic skin's optical properties may explain the better response we obtained with SEUVB and petrolatum 2 than Frost et al obtained with a gel. Our results are consistent with the suggestion of LeVine et al 2 that tar is not a necessary ingredient in the "modified Goeckerman r e g i m e n . " In addition, our findings support the conclusion of Frost el al :~ that SEUVB is effective, and in this trial SEUVB and M A E U V B induced remissions at equal rates. In telephone interviews we determined that these two schedules were associated with similar lengths o f remission. Telephone interview is not the optimaI method o f follow-up. However, we believe it unlikely that severely affected patients would report their skin to be clear or that patients
whose condition had flared would be unaware of their generally worsened state. These observations suggest the possibility of improving the "modified Goeckerman treatment" by (1) eliminating the tar, which is messy, malodorous, a follicular irritant, and has potential toxic and cmcinogenic effects, and (2) decreasing the UV dose, thus eliminating the frequent UVinduced " s u n b u r n " and possible koebnerization and decreasing the UV carcinogenic potential by using only about 20% of the usual UV dose. Studies of variables such as the effect of hospitalization and treatment intervals may help to determine which modified Goeckerman regimen is best or to define which modified Goeckerman regimen is best for which subset of psoriasis patients. REFERENCES
1. Goeckerman WH: Treatment of psoriasis. Northwest Med 24:229-231, 1925. 2. LeVine NJ, White HAD, Parrish JA: Componentsof the Goeckerman regimen. J Invest Dermatol 73:170-173, 1979. 3. Frost P, Horowitz SN, Caputo RV: Tar geI: Phototherapy for psoriasis. Arch Dermatol 115:840-846, I979. 4. Lowe NJ, Wortzman MS, Breeding J, et al: Coal tin" phototherapy for psoriasis re-evaluated: Erythemogenic versus suberythemogenicultraviolet with a tar extract in oil and crude coal tar. J AM ACADDERMATOL8:781789, 1983. 5. Sabiston DC Jr, editor: Davis-Christopher Textbook of surgery. Philadelphia, 1972, W. B. Saunders Co., p. 275. 6. Brownlee KA: Comparison of several regression lines: Simple analysis of covariance, in Statistical theory and methodology in science and engineering. New York, 1960, John Wiley & Sons, Inc. 7. Halprin KM, Comerford M, Taylor JR: Constant lowdose ultravioletlight therapy for psoriasis. J AM ACAD DERMATOL 7:614-619, 1982. 8. ColskyLC, Halprin KM, Taylor JR, et al: UVB is additive when repeated withinan 8-hour interval. J AMACAD DERMATOL8:760-761, 1983. (Letter to Editor.) 9. Fischer T: UV-light treatment for psoriasis. Acta Derm Venereol (Stockh) 56:473-479, 1976. I0. Marisco AR, Eaglstein WH, Weinstein GD: Ultraviolet light and tar in the Goeckerman treatment of psoriasis. Arch Dermatol 112: I249-1250, 1976. l 1. Petrozzi JW, Barton JO, Kaidbey KK, Kligman AM: Updating the Goeckerman regimen for psoriasis. Br J DermatoI 98:437-444, I978. 12. Lowe NJ, Breeding J, Wortzman MS: The pharmacological variability of crude coal tar. Br J Dermatot 107:475-479, 1982.