- Email: [email protected]
Topical erythromycin with zinc in acne
Topical erythromycin with zinc in acne A double-blind
COil trolled
study
Christopher L. Feucht, M.D., Bruce S. Allen, M.D., Dan K. Chalker, M.D., and J. Graham Smith, Jr., M.D. Augusta, GA This double-blind controlled study investigated the effectiveness of 4% topical erythromycin liquid with 1.2% zinc .acetate and 4% topical erythromycin gel with 1.2% zinc octoate applied twice daily in comparison with 250 mg of oral tetracycline twice daily and placebo. One hundred forty-nine patients began and one hundred forty-one completed the study. Three judges recorded acne measurements of severity grade, papule counts, pustule counts, comedo grades, and comedo counts without mutual consultation on visits at 0, 2, 4, 6, 8, and 10 weeks. Analysis of the reduction in the acne severity grade and papule count showed that the erythromycinlzinc liquid and gel were statistically.significantly better than placebo and as effective as the oral tetracycline. Analysis of comedo grades showed that at weeks 8 and 10 the 4% topical erythromycin/zinc liquid showed a reduction statistically better than placebo. (J AM ACAD DERMATOL 3:483-491, 1980.)
For the past quarter of a century, systemic antibiotics have been the mainstay of inflammatory acne therapy. Usually patients were treated with either oral tetracycline or erythromycin and often for protracted periods of time. I During the past several years, there has been an increasing interest among dermatologists and patients alike to find topical antibiotic preparations with comparable effectiveness to oral medicines but without the problems and restrictions inherent to the oral antibiotics. In 1961, Robinson et aF found topical neomycin to be no more effective in acne than the vehicle alone. Two years later, Miroliubov" found chloramphenicol to be an effective topical antiacne From Ihe Department of Dermatology, Medical College of Georgia. Reprinl requests' to: Dr. Christopher L. Feucht, Garden City Medical Ar Bldg., 601 West Spruce, Missoula, t-IT 59801.
0190-0622/80/110483+09$00.9010 © 1980 Am Acad Dermatol
agent, but because of the potential toxicity, this treatment did not gain widespread acceptance. Beginning in 1957 with Hollander et al," various preparations of tetracycline, with and without penetration-enhancing agents such as dimethylsulfoxide, were tried in acne. Work with this antibiotic has continued, and a topical tetracycline (Topicycline) is now available for acne treatment. 5,6 Fulton and Pablo, t in 1974, observed a 49% decrease in fatty acid levels of surface lipids and clinical improvement with 2% erythromycin topically. In 1976, Van Ketel" found that erythromycin stearate had demonstrable topical sensitizing potential. As late as 1977, Fisher? was unable to document any cases of sensitization to topical erythromycin base in treatment of stasis ulcers,a condition frequently associated with the development of allergic contact dermatitis. Since 1974, 483
484
lournal of the American Academy of Dermatology
Feucht et (I!
Table I. Participants in the study
4
40 39 39 31
Totals
149
I
2 3
Withdrawn or did not finish
Completed
Group
38
2
35
4 I I
38 30
8 (5%)
141 (95%)
PATIENTS
Table II. Number of missed visits
Group
Participants included in stud)'
Missed one visit
Missed two visits
I
38
2
o
2 3 4
35
4
2
38
3 3
o
30
I
cal efficacy in wound healing led to the addition of zinc to a topical erythromycin preparation. In preliminary investigations, several experimental topical preparations were evaluated. We found the. erythromycin-zinc preparation to produce greater improvements than the others*; therefore, we undertook a large-scale double-blind study comparing it to placebo and oral tetracycline in the treatment of acne.
Total number of vlslts missed (%)
2 8 5 3
(0.9%) (3.8%) (2.2%) (1.7%)
there have been few well-controlled clinical studies on the efficacy of topical erythromycin for acne. From the time of ancient Egyptians, zinc oxide and other derivatives have been used to protect and promote wound healing. 10 Only in the last few years has zinc been investigated for the treatment of acne. In several double-blind controlled studies, results and conclusions have varied. In 1977, Michaelsson and co-workers reported that zinc orally was statistically better than placebo after 4 weeks in the treatment of acne. 11 In another study of 12 weeks' duration, they found no statistical difference between zinc and tetracycline groups since both showed significant improvement." In other studies in 1977, Hillstrom et al 13 found statistically significant improvement over placebo at 4, 8, and 12 weeks of treatment, while Weismann et ap-t found no statistical difference over an entire 12-week study. In 1978, Orris et aps and Weimar et al, 16 in two separate studies, found no significant differences between zinc- and placebo-treated patients, while Goransson et al '? in a 6-week study found zinc to be slightly but statistically significantly better than placebo. Zinc has not been used topically previously in the treatment of acne. The possible benefits, although controversial, of zinc in acne and its topi-
All patients were male students at the University of Georgia and were between the ages of 18 and 26. They were informed as to procedures, medications, benefits, and risks, and written consent was obtained. None gave a history of diabetes mellitus, cardiac, renal, hepatic, or other systemic diseases. None had a history of atopic dermatitis or allergy to the medications we employed. No patients were included in the study who had been on systemic tetracycline or erythromycin for 6 weeks before the start of the study. Each patient was required at the time of recruitment (0-5 weeks before the study) to discontinue any other acne therapy and to avoid antibacterial soaps or antiseborrheic (medicated) shampoos for the duration of the study. MEDICATIONS Patients were assigned consecutive numbers and randomly distributed by computer to one of four groups. The groups and their medications were as follows: Group I: Topical 4% erythromycin base with 1.2% zinc acetate in a liquid vehicle Oral placebo (capsule containing cornstarcht) Group 2: Topical 4% erythromycin base with 1.2% zinc octoate in a gel vehicle Oral placebo Group 3: Topical liquid or gel vehicle without the erythromycin or zinc acetate/octoate Oral tetracycline (capsule containing 250 mg of tetracycline) Group 4: Topical liquid or gel vehicle without the erythromycin or zinc acetate/octoate Oral placebo Each patient applied the topical medications (liquid
* Unpublished data,
Department of Dermatology. Medical College of Georgia. Augusta, GA 30912. t'Ihe cornstarch was dyed yellow with F.D. and C.No 5 to match the color of tetracycline.
Volume 3 Number 5 Novemb er , 1980
Erythromycin and zinc in acne
485
Table III. Average severity grades by groups Weeks (anrage number) Groups
-2
1. Erythromycin zinc liquid
3.86 34 3.63 33 3.63 36 3.39 25
2. Erythromycin zinc gel 3. Tetracycline 4. Placebo
10
4.05 38 3.90 35 4.08 38 3.67 30
or gel) and took a capsule (tetracycline or cornstarch) twice daily. All patients were given a bland soap and nonmedicated shampoo to use during the study.
2.96 38 2.74 31 3.25 36 3.30 29
8
GRADING All patients were evaluated on each visit by three graders according to the grading scale as described by Cook et al. 18 Grades were recorded for overall acne severity on a scale of 0-8. A similar scale was used in recording a comedonal grade. These grades and their corresponding clinical descriptions or criteria are listed below. Intermediate grades such as I, 3, 5, and 7 were possible and were used.
Grading scale for overall severity Grade
o
2
4
6
Description Skin need not be perfectly clear. A few small scattered comedones or papules may be present, but these should be visible only upon close examination . About one fourth of the facial area is involved with small papules (about 6-12) and comedones (may be a few large comedones or as many as 20-30 small closed comedones) . Very few pustules or large prominent papules are present. About one half of the facial area is involved with small papules and large or small comedones. A few pustules or large prominent papules are · usually present. (If lesions are generally large, a subject may have a "grade 4 " severity even though less than half of the facial area is involved.) About three fourths of the facial area is involved with papules and/or large open comedones . (A lesser facial area involvement is permissible if inflammatory lesions are large .)
2.72 38 2.67 33 3.06 37 3.14 30
2.48 37 2.43 35 2.59 36 3 .07 29
2.35 37 2.67 33 2.87 38 3.34 29
2.25 38 2.47 35 2.62 38 3.19 30
Numerou s pustules are usually present, some of which may be large. Practically all of the facial area is involved with lesions. Large, prominent pustules are usually quite visible. Lesions are usually of a highly inflammatory type. Other types of acne (such as conglobata, including sinus and cystic types) may be present.
Grading scale for comedones Grade
o
Subjective and objective criteria
No comedones Mild comedonal acne 2 About 10 lesions per side of face 4 Obvious comedonal acne About 25 lesions per side of face 6 Severe comedonal acne About 50 lesions per side of face 8 Worst possible comedonal acne About 100 or more lesions per side of face Counts were made of papules and comedones (open and clo sed) on the right side of the face from the hairline to the mandibular ridge . Pustules were counted on both sides of the face. STUDY DESIGN The study lasted a total of 10 weeks. Patients were evaluated by the same three graders on weeks 0, 2, 4, 6, 8, and 10. Two weeks priorto the onset of the study, a preadmission evaluation was performed on a majority of participants. There was no collaboration between graders at the time of the grades , and counts were recorded. Both graders and patients were unaware of the medications they were using. Pictures were taken on weeks 0, 4, 8, and 10. Medications and directions for their use were given on week O. New supplies of medications were given on weeks 4 and 8. On weeks 4, 8,
486
Journal of the Am er ican Academy of Dermatology
Felicia et at
Table IV. Reduction and percent reduction in overall severity grade by groups Weeks Groups
I
Reduction and % reduction
2
Reduction % reduction Reduction % reduction Reduction % reduction Reduction % reduction
1.09 (0.00 I) 25.69 (0.001) 1.10 (0.001) 27.42 (0.001) 0.85 (0.05) 18.35 (0.05) 0.37 6.04
I. Topical erythromycin/zinc liquid
2. Topical erythromycin/zinc gel 3. Oral tetracycline 4. Placebo
Statistic ally significant reduction in comparison with placebo by the Wilcoxon Rank Sum Test with p significant .
<
4
1.33 (0.001) 30.80 (0.01) 1.24 (0.01) 29.32 (0.005) 1.05 (NS) 20.76 (NS) 0.52 11.51
the value in parentheses; NS: not
Table V. Average papule counts by groups Weeks (average number) . Groups
I. Topical erythromycin/zinc liquid
2. Topical erythromycin/zinc gel 3. Oral tetracycline 4. Placebo
-2
14.27 34 12.97 33 12.97 36 11.24 25
and 10, the quantities of unused medications were recorded. The 4% erythromycin/zinc in the liquid vehicle was distributed in squeeze containers along with 5-cm adsorbant pads. Patients were instructed to saturate the pads and to apply the liquid to the entire face. The 4% erythromycin/zinc in the gel vehicle was distributed in plastic squeeze tubes with instructions for it to be applied with the tips of the fingers, again to the entire face. The precounted capsules were distributed in plastic containers, and patients were instructed to take them only with water I hour before or 2 hours after eating . ST ATISTICAL l\IETHODS In the statistical analysis of the data (acne severity grades, papule counts, etc.), the response to therapy in this paper was defined as reduction or percent reduction. This reduction or percent reduction was calculated by comparing the treated condition at weeks 2, 4, 6, 8, and 10 with the pretreatment condition at week O. The reduction or percent reduction of each active group was then compared to the placebo group at each visit to
10
17.41 38 15.96 35 17.35 38 14.44 30
9.93 38 8.52 3I . 10.36 36 11.91 29
8.12 38 7.80 33 9.42 37 10.13 30
7.13 37 7.54 35 7.51 36 9.64 29
7.17 37 7.57 33 8.35 38 9.93 29
6.47 38 7.50 35 7.64 38 9.39 30
determine statistical significance . The comparisons between the active groups and the placebo group were made using the Wilcoxon Rank Sum Test. 19.20 RESULTS One hundred forty-nine participants began the study. The data from 141 arc included. The data from three participants are not included in this paper since during the course of the study they were given antibiotics by their own physicians for other illnesses. One participant was given erythromycin for a skin infection of the right ankle. The other two participantswere treated for upper respiratory infections with tetracycline and erythromycin, respectively. Five students did not finish the study. Of these, one moved, three did not return, and one complained of irritation and decided to stop the study. The one student who complained of irritation was on topical 4% erythromycin/zinc liquid and oral
Volume 3 Number 5 November, 1980
Erythromycin and zinc in acne
C)
0.00
.....
"g
~
Weeks
6
1.60 39.24 1.47 33.93 1.54 35.02 0.64 13.98
(0.001) (0.001) (0.01) (0.01) (0.01) (0.01)
8
1.61 (0.001) 40 .87 (0.001) 1.23 (0.01) 27.62 (0.05) 1.21 (0.01) 26.77 (0.05) 0.37 7.24
10
1.81 (0.001) 46.07 (0.001) 1.43 (0.01) 33.02 (0.01) 1.46 (0.01) 32.05 (0.01) 0.48 6.88
~f-l
'~
0.50
~
.~
C)
~
C/)
~'>1~ TT..J r' t4~ .'"
'.,-,;
1.00
.s
.2
1.50
PLACEBO . - - - ETM LIQUID ..--- ElM GEL D-ORAL TeN 0--
U j
"0
~
487
2.00
"',
- - .....
<,
~
I
Bars are ± 10 S.E.
2.50 ' - - - - . l - - - - ' - - _ - ' - - _ - - L - _ - - L _ - - - - ' - _ - . J -2 a 2 4 6 8 10 Time on Study (weeks)
Fig. lAo Reduction in severity grade.
placebo. He had few signs of an eruption other than some erythema which disappeared within 24 to 48 hours. He refused patch tests, but by history, clinical findings, and length of problem , it was judged not to be a contact dermatitis. The number of participants who began, finished, and were included in the study in each group are listed in Table I. Participants were not dropped from the study if they missed a visit for an evaluation as long as they returned on subsequent visits . No participant whose data were included in the results missed more than two visits or failed to return for a new supply of medications. In Table II are the number of visits missed by participants . An average individual grade or count was determined from the evaluation of the three graders. An average group grade or count for each group was determined by averaging these individual scores of the group at each visit. Table III shows the results for the average overall severity grade for each group at each visit. The reduction for a group is calculated by determining the reduction of each individual in the group and then averaging those reductions. The percent reduction of each individual in the group is averaged together with the others for the percent reduction of the group. Table IV shows the reduction and percent reduction of the overall severity grade in each group at weeks 2, 4, 6, 8, and 10 in comparison with week O. Statistically significant differences from placebo are followed by the p value in parentheses . Figs. IA and lB graphically demonstrate the
~
00.00
,
~-~=---=-------T
'~-~~
c
l5 20.00 >. II>
<, .Ii':....
~ 40.00
o
~ 60.00 j
PLACEBO . - - - ElM LIQUID .---- ETM GEL o-ORALTC
"0
Bars are .1 1.0 S.E.
0--
C/)
c
C)
~ 80.00
iJ:
T
=t -~.::.:..... ~~~"6
.~
~
--'<>---Q
I
1.
r---I--_I I
'---_--'--_---'.L_-L_ _' - -_--'---_---'---J
-2
o
2 4 6 8 Time on Study (weeks)
10
Fig. lB. Percent reduction in severity grade .
average reduction and average percent reduction in severity grade, respectively, versus time. Table V shows the results of the average papule counts (taken from the right side of the face) for each group at each visit. Table VI shows the reduction and percent reduction of the papule counts in each group at weeks 2, 4, 6, 8 , and 10 in comparison with week O. Again, statistically significant differences from placebo arc noted by the p value in parentheses. Figs. 2A and 2B show graphically the reduction and percent reduction in papule counts, respectively, versus time. Pustule counts showed no statistically significant difference between groups throughout the study. Table VII shows average comedo grades and counts obtained for each group at each visit. The only statistically significant differences from placebo occurred in the topical erythromycin group at
Journal of the American Academy of Dermatology
488 Feucht et at
Table VI. Reduction and percent reduction papule counts by groups Weeks Groups
7.48 31.69 6.82 28.38 7.28 35.70 2.56 7.23
Reduction
I. Topical erythromycin/zinc liquid
% reduction
Reduction % reduction Reduction % reduction Reduction % reduction
2. Topical erythromycin/zinc gel 3. Oral tetracycline 4. Placebo
I
2
Reduction and % reduction
4
(0.01) (0.001) (0.001) (0.001) (0.001) (0.001)
9.29 46.16 8.20 38.97 8.01 36.89 4.31 23.15
(0.01) (0.01) (0.01) (0.05) (0.05) (NS)
Statistically significant reduction in comparison with placebo by the Wilcoxon Rank Sum Test with p < the value in parentheses; NS: not significant.
Table VII. Average comedo grades and counts by groups
Groups
I. Topical erythromycin/
zinc liquid 2. Topical erythromycin/ zinc gel 3. Oral tetracycline 4. Placebo
I
Parameter
Grade Count No. Grade Count No. Grade Count No. Grade Count No.
I
Weeks
-2
0
2
4
6
8
10
2.51 '29.35 34 2.46 29.50 33 2.38 28.44 36 2.17 24.57 25
2.11 31.50 38 2.02 32.28 35 2.16 34.83 38 2.07 31.74 30
2.18 28.82 38 2.27 28.66 31 2.40 33.32 36 2.32 29.76 29
2.55 30.81 38 2.31 29.68 33 2.64 34.32 37 2.49 31.90 30
2.27 25.99 37 2.24 24.70 35 2.46 28.41 36 2.39 26.57 29
2.05 22.78 37 2.29 25.03 33 2.46 27.04 38 2.75 29.45 29
1.89 20.42 38 2.21 23.10 35 2.36 26.09 38 2.59 27.70 30
weeks 8 and to. Table VIII shows these differences at weeks 8 and to. Note that the average comedo grade for the topical erythromycin/zinc liquid group decreased from week 0 while in all other groups these grades increased. All groups showed some slight decrease in average comedo counts from week O. However, the topical erythromycin/zinc liquid group showed a much larger decrease. Table IX shows correlations among the three judges for each of the five acne measurements performed. All of these correlation coefficients are statistically significant with a p value of less than 0.01. In the grading of acne, though it is not possible to decide a right or wrong grade given for a particular patient at a particular visit, it is possible
through these correlations to see how close the judges were in their measurements. The amount of antibiotic usage was determined for each group. This was determined from the amount of medication that was returned. The group with 4% erythromycin in a liquid vehicle used 2.32 mIlday or 86.3 mg/day of erythromycin. This does not take into account the amount which is absorbed into the pads and discarded after each usage. The group with 4% erythromycinin a gel vehicle used 670 mg/day of the gel or 26.8 mg/day of erythromycin. This does not take into account what remairis on the fingertips and is not used. The groups taking tetracycline orally took an average of 1.85 capsules per day or 462.5 mg of tetracycline per day.
Volume 3
Erythromycin and zinc ill acne
Number 5 November, 1980
Weeks 6
10.41 51.87 8.42 39.06 10.22 54.94 5.08 23.34
(0.01) (0.001) (0.05) (0.05) (0.01) (0.001)
10
8
9.83 53.36 8.35 39.93 9.00 48 .33 4.62 23. I I
(0.01) (0.001) (0.05) (0.05) (0.01) (0.001)
10.94 58.11 8.46 44.65 9.71 51.17 5.06 .25.43
(0.01) (0.001) (0.05) (0.05) (0.01) (0.01)
C ::I
~
•
04.00
go 06.00
a,
'f
,~
~rN
\
,~
\,~I
.:
s 08.00
PLACEBO . - - - ETM LIQUID .---- ETM GEL o-ORAL TC
0--
CONCLUSIONS AND .DISCUSSION
l
r '" k!+~?.t
10.00 12.00
'
l.,t~--t--j----l
U ::I
-g 0::
489
..........
Bcrs ore :t 1.0 SE .
Evaluation of the preceding results supports several conclusions. First, the particular topical erythromycin/zinc liquid and gel formulations that were used in the treatments described in the preceding paragraphs produced statistically significant decreases in both overall severity grades and papule counts compared to placebo. Further, the erythromycin/zinc liquid and gel are as effective as 500 mg tetracycl ine orally per day. Also, comedo grades and counts were statistically significantly decreased by use of the topical erythromycin liquid but by none of the other formulations. It is of interest to note that this improvement only became manifest after 8 to 10 weeks. The exact explanation for the delayed improvement is not known, but one possible consideration might be that the topical erythromycin or zinc acetate exerts some as yet unknown inhibitory effect at some stage in the process of comedo formation. Whether this effect is due to antibacterial, chemotactic, keratoplastic, or some other totally unrelated mechanism is as yet undetermined. The placebo group throughout the study showed a slight but definite improvement. Although there was some dryness noted in these patients probably secondary to the topical placebo, liquid or gel, these placebo medications are not known to have any acne-improving qualities. This slight improvement is what is considered by most investigators as the placebo effect which in this study is evident. Despite the improvement in papule and severity grades, pustule counts were not improved by any
14.00 L . - _ - ' - - _ - - - ' - - - _- ' -_ _ ~_ _ '--_.l_______J -2 o 2 4 6 8 10 Time on Study (weeks)
Fig. 2A. Reduction in papule count.
COo.OO~\
s
i
~\
2000
~ 40.00
~ ~
cp
0::
;;.e
60.00
~
\l: t---l-"- ...
y---1
~~-+---+--".I '!-.. ~ -"-i
o--PLACEBO _---ElM LIQUID •• • ETM GEL 0 - ORAL TC Bers ore ± 1.0 S.E.
80.00 ' - - _ - ' - _ - - ' - _ - - L _ - - - ' _ _-'--_--'----' -2 024 6 8 ID Time on Study (weeks)
Fig. 28. Percent reduction in papule count.
of the utilized preparations. This might be because so few pustules (compared to papules) were seen in our patients, and these numbers were too small for significant results to be obtained. The usage data displayed in this study show that the 4% erythromycin liquid and gel patients used on an average no more than 86.3 mg per day and 21.6 mg per day, respectively, of the erythromycin to achieve the previously noted improvement. When compared to the 500 mg per day dosages usually required for oral ·treatment of mild or moderate acne, it becomes evident that the liquid and gel erythromycin used in this study enable the patients to experience significant improvement with the use of a minimum amount of active in-
490
Journal of the Ameri can Academy of Dermatology
F cue/It et of
Table VIII. Reduction and percent reduction in comedo grades and counts between placebo and topical erythromycin liquid groups at weeks 8 and 10 Weeks Groups
Grades and count (reduction and % reduction)
8
Grades: Reduction % reduction Count: Reduction % reduction Grades: Reduction % reduction Counts: Reduction % reduction
0.09 (0.01) 0.86 (0.01) 8.80 (0.05) 10.93 (0.01) -0.64 -41.78 2.53 -17.43
I. Topical erythromycin/zinc liquid
2. Placebo
Stati stically significant reduction in comparison with placebo by the Wilcoxon Rank Sum Test with p
I
10
0.23 (0.01) 8.71 (0.01) 11 .08 (0.05) 14.83 (0.05) -0.52 -33.80 4.04 -14.09
< the value in parentheses.
Table IX. Correlation coefficient of judges* Weeks on study Measurements
Judges
0
Severity grade
I vs 2 I vs 3 2 vs 3 I vs 2 1 vs 3 2 vs 3 I vs 2 I vs 3 2 vs 3 1 vs 2 I vs 3 2 vs 3 1 vs 2 1 vs 3 2 vs 3
0.82 0.87 0.87 0.81 0.88 0.83 0.68 0.65 0.84 0.73 0.78 0.85 0.69 0.70 -0.74
Papule count Pustule count Comedo count Comedo grade
·AII corr elation coefficients are statistically significant (p
I
2
0.81 0.86 0.83 0.83 0.82 0.88 0.70 0.74 0.80 0.87 0.85 0.89 0.78 0.79 0.84
I
4
0.83 0.82 0.85 0.83 0.85 0.88 0.82 0.81 0.80 0.84 0.84 0.84 0.82 0.75 0.79
I
6
0.81 0.86 0.82 0.85 0.87 0.89 0.84 0.87 0.85 0.88 0.89 0.88 0.81 0.77 0.83
I
8
0.86 0.84 0.86 0.86 0.90 0.89 0.78 0.86 0.79 0.88 0.86 0 .89 0.84 0.76 0.83
I
10
0.88 0.88 0.88 0.91 0.86 0.88 0.83 0.81 0.85 0.86 0.86 0.90 0.85 0.78 0.81
< 0.01).
gredient. Further, patients using the liquid and gel erythromycin preparation actually received less than the stated milligram per day dosages of erythromycin in view of the fact that a certain percentage (unable to be exactly determined) of the medication is lost at each application, either by being absorbed into the pad used for liquid application or by being washed off the fingers in the case of the gel preparation . From Table IX, our correlation coefficients showed that three independent judges can evaluate acne with good correlation, especially in regard to severity grades and papule count. This speaks highly for the Cook grading scale and supports the
fact that objective data can be obtained from the usage of this grading system." In regard to the zinc component of this topical preparation, little is known or established about its mechanism of action. Erythromycin and zinc form a complex when in solution in the vehicle. Studies in rats, utilizing radiolabeled erythromycin and zinc, have demonstrated that the penetration of erythromycin into the skin is enhanced by the presence of zinc in the formulation. Also, special staining technics applied to excised skin from hairless mice and pigs have demonstrated that when erythromycin is present in the formulation the zinc penetrates the stratum corneum into the epidermis
Volume 3 Number 5 November, 1980
and perifollicular apparatus. * Though these studies give some idea of its localization and penetration, only speculation can be made about its mechanism of action. In summary, we have presented a double-blind placebo-controlled study evaluating a topical erythromycin liquid and a topical erythromycin gel compared to liquid and gel placebos and 500 mg per day oral tetracycline . The erythromycin liquid and gel tested gave statistically significant improvement of acne when judged by papule counts and by severity grades with the Cook method. A further unexpected finding was the late but definite reduction in comedo grades and counts in patients using the erythromycin liquid.
Erythromycin and zinc ill acne
7. Fulton JE Jr, Pablo G: Topical antibacterial therapy for acne . Arch Dennatol 110:83-86 , 1974.
8. Van Ketel WG: Immediate and delayed type allergy to erythromycin. Contact Dermatitis 2:363-364, 1976.
9. Fisher A: Erythromycin "free base " -a nonstinging top-
10.
II.
12.
13.
14.
'Unpublished data. on file, Procter and Gamble, Cincinnati. OH . REFERENCES
I . Crounse RG: The respons e of acne to placebos and antibiotics. JAMA 193:906-910, 1965. 2. Robinson HM Jr. et al: Topical acne therapy. South Med J s.t:1105. 1961. 3. Miroliubov VI: Chloromphenicol in the treatment of folliculitis, acne vulgaris and rosace a. Vestn Dennatol Ven erol 37 :68-71, 1963. 4 . Hollander L, Shelton JM, Hardy SM : Acrornycin f lotion an adjunct in treating acne vulgaris. American Prac titioner Digest of Treatment 8:1602-1605 , 1957. 5. Blaney OJ, Cook CH : Topical use of tetracycline in the treatment of acne: A double-blind study comparing topical and oral tetracycline therapy and placebo" Arch Dermatol 112:971-973, 1976 . 6. Smith JG Jr, Chalker OK, Wehr RF: The effectiveness of topical and oral tetracycline in acne. South Med J 69:695-697. 1976.
491
15. 16. 17.
18.
19.
20 .
21.
ical antibiotic for infected dermatosi s and acne vulgaris. Cutis 20:17,26.30 . 1977. Lee PWR , Green MA. Long \VB III, and Gill W: Zinc and wound healing. Surg Gynecol Ob stet 143:549-554, 1976. Michaelsson G, Juhlin L , Vahlquist A: Effects of oral zinc and vitamin A in acne. Arch Dermatol 113:31·36, 1977. Michaelsson G, Juhlin L, Ljunghall K: A double blind study of the effect of zinc and oxytetracycline in acne vulgaris. Br) Derrnatol 97:679·684, 1977. Hill strom L, et al: Comparison of oral treatment with zinc sulphate and placebo in acne vulgaris . Br J Dermatol 97:679-684, 1977. Weismann K, Wadsker S, Sondergaard J: Oral zinc sulphate therapy for acne vulg aris . Act a Denn Venereol (Stockh) 57:357-360 , 1977. Orris L. et al: Oral zinc therapy of acne . Arch Dermatol 114:1018-1020, 1978. Weimar VM, et al: Zinc sulfate in acne vulgaris. Arch Dennatol 114:1776-1778, 1978. Goran sson K. Liden S, Odsell L: Oral zinc in acne vulgaris-a clinical and methodological study. Acta Denn Venereol (Stockh) 58:443-448, 1978 . Cook CH , Centner RL, Mich aels SF: An acne grading method using photographic standards. Arch Derrnatol 115:571-575, 1979. Hollander M, Wolfe DA: Nonparametric statistical methods. New York, 1973, John Wiley & Sons, Inc., pp . 68-74. Brown BW, Hollander M: Statistics: A biomedical introduct ion. New York, 1977 , John Wiley & Sons, Inc., pp. 134-136. Allen BS, Smith JG Jr: A study of various parameters for grading acne vulgaris. (To be published.)
COil trolled
study
Christopher L. Feucht, M.D., Bruce S. Allen, M.D., Dan K. Chalker, M.D., and J. Graham Smith, Jr., M.D. Augusta, GA This double-blind controlled study investigated the effectiveness of 4% topical erythromycin liquid with 1.2% zinc .acetate and 4% topical erythromycin gel with 1.2% zinc octoate applied twice daily in comparison with 250 mg of oral tetracycline twice daily and placebo. One hundred forty-nine patients began and one hundred forty-one completed the study. Three judges recorded acne measurements of severity grade, papule counts, pustule counts, comedo grades, and comedo counts without mutual consultation on visits at 0, 2, 4, 6, 8, and 10 weeks. Analysis of the reduction in the acne severity grade and papule count showed that the erythromycinlzinc liquid and gel were statistically.significantly better than placebo and as effective as the oral tetracycline. Analysis of comedo grades showed that at weeks 8 and 10 the 4% topical erythromycin/zinc liquid showed a reduction statistically better than placebo. (J AM ACAD DERMATOL 3:483-491, 1980.)
For the past quarter of a century, systemic antibiotics have been the mainstay of inflammatory acne therapy. Usually patients were treated with either oral tetracycline or erythromycin and often for protracted periods of time. I During the past several years, there has been an increasing interest among dermatologists and patients alike to find topical antibiotic preparations with comparable effectiveness to oral medicines but without the problems and restrictions inherent to the oral antibiotics. In 1961, Robinson et aF found topical neomycin to be no more effective in acne than the vehicle alone. Two years later, Miroliubov" found chloramphenicol to be an effective topical antiacne From Ihe Department of Dermatology, Medical College of Georgia. Reprinl requests' to: Dr. Christopher L. Feucht, Garden City Medical Ar Bldg., 601 West Spruce, Missoula, t-IT 59801.
0190-0622/80/110483+09$00.9010 © 1980 Am Acad Dermatol
agent, but because of the potential toxicity, this treatment did not gain widespread acceptance. Beginning in 1957 with Hollander et al," various preparations of tetracycline, with and without penetration-enhancing agents such as dimethylsulfoxide, were tried in acne. Work with this antibiotic has continued, and a topical tetracycline (Topicycline) is now available for acne treatment. 5,6 Fulton and Pablo, t in 1974, observed a 49% decrease in fatty acid levels of surface lipids and clinical improvement with 2% erythromycin topically. In 1976, Van Ketel" found that erythromycin stearate had demonstrable topical sensitizing potential. As late as 1977, Fisher? was unable to document any cases of sensitization to topical erythromycin base in treatment of stasis ulcers,a condition frequently associated with the development of allergic contact dermatitis. Since 1974, 483
484
lournal of the American Academy of Dermatology
Feucht et (I!
Table I. Participants in the study
4
40 39 39 31
Totals
149
I
2 3
Withdrawn or did not finish
Completed
Group
38
2
35
4 I I
38 30
8 (5%)
141 (95%)
PATIENTS
Table II. Number of missed visits
Group
Participants included in stud)'
Missed one visit
Missed two visits
I
38
2
o
2 3 4
35
4
2
38
3 3
o
30
I
cal efficacy in wound healing led to the addition of zinc to a topical erythromycin preparation. In preliminary investigations, several experimental topical preparations were evaluated. We found the. erythromycin-zinc preparation to produce greater improvements than the others*; therefore, we undertook a large-scale double-blind study comparing it to placebo and oral tetracycline in the treatment of acne.
Total number of vlslts missed (%)
2 8 5 3
(0.9%) (3.8%) (2.2%) (1.7%)
there have been few well-controlled clinical studies on the efficacy of topical erythromycin for acne. From the time of ancient Egyptians, zinc oxide and other derivatives have been used to protect and promote wound healing. 10 Only in the last few years has zinc been investigated for the treatment of acne. In several double-blind controlled studies, results and conclusions have varied. In 1977, Michaelsson and co-workers reported that zinc orally was statistically better than placebo after 4 weeks in the treatment of acne. 11 In another study of 12 weeks' duration, they found no statistical difference between zinc and tetracycline groups since both showed significant improvement." In other studies in 1977, Hillstrom et al 13 found statistically significant improvement over placebo at 4, 8, and 12 weeks of treatment, while Weismann et ap-t found no statistical difference over an entire 12-week study. In 1978, Orris et aps and Weimar et al, 16 in two separate studies, found no significant differences between zinc- and placebo-treated patients, while Goransson et al '? in a 6-week study found zinc to be slightly but statistically significantly better than placebo. Zinc has not been used topically previously in the treatment of acne. The possible benefits, although controversial, of zinc in acne and its topi-
All patients were male students at the University of Georgia and were between the ages of 18 and 26. They were informed as to procedures, medications, benefits, and risks, and written consent was obtained. None gave a history of diabetes mellitus, cardiac, renal, hepatic, or other systemic diseases. None had a history of atopic dermatitis or allergy to the medications we employed. No patients were included in the study who had been on systemic tetracycline or erythromycin for 6 weeks before the start of the study. Each patient was required at the time of recruitment (0-5 weeks before the study) to discontinue any other acne therapy and to avoid antibacterial soaps or antiseborrheic (medicated) shampoos for the duration of the study. MEDICATIONS Patients were assigned consecutive numbers and randomly distributed by computer to one of four groups. The groups and their medications were as follows: Group I: Topical 4% erythromycin base with 1.2% zinc acetate in a liquid vehicle Oral placebo (capsule containing cornstarcht) Group 2: Topical 4% erythromycin base with 1.2% zinc octoate in a gel vehicle Oral placebo Group 3: Topical liquid or gel vehicle without the erythromycin or zinc acetate/octoate Oral tetracycline (capsule containing 250 mg of tetracycline) Group 4: Topical liquid or gel vehicle without the erythromycin or zinc acetate/octoate Oral placebo Each patient applied the topical medications (liquid
* Unpublished data,
Department of Dermatology. Medical College of Georgia. Augusta, GA 30912. t'Ihe cornstarch was dyed yellow with F.D. and C.No 5 to match the color of tetracycline.
Volume 3 Number 5 Novemb er , 1980
Erythromycin and zinc in acne
485
Table III. Average severity grades by groups Weeks (anrage number) Groups
-2
1. Erythromycin zinc liquid
3.86 34 3.63 33 3.63 36 3.39 25
2. Erythromycin zinc gel 3. Tetracycline 4. Placebo
10
4.05 38 3.90 35 4.08 38 3.67 30
or gel) and took a capsule (tetracycline or cornstarch) twice daily. All patients were given a bland soap and nonmedicated shampoo to use during the study.
2.96 38 2.74 31 3.25 36 3.30 29
8
GRADING All patients were evaluated on each visit by three graders according to the grading scale as described by Cook et al. 18 Grades were recorded for overall acne severity on a scale of 0-8. A similar scale was used in recording a comedonal grade. These grades and their corresponding clinical descriptions or criteria are listed below. Intermediate grades such as I, 3, 5, and 7 were possible and were used.
Grading scale for overall severity Grade
o
2
4
6
Description Skin need not be perfectly clear. A few small scattered comedones or papules may be present, but these should be visible only upon close examination . About one fourth of the facial area is involved with small papules (about 6-12) and comedones (may be a few large comedones or as many as 20-30 small closed comedones) . Very few pustules or large prominent papules are present. About one half of the facial area is involved with small papules and large or small comedones. A few pustules or large prominent papules are · usually present. (If lesions are generally large, a subject may have a "grade 4 " severity even though less than half of the facial area is involved.) About three fourths of the facial area is involved with papules and/or large open comedones . (A lesser facial area involvement is permissible if inflammatory lesions are large .)
2.72 38 2.67 33 3.06 37 3.14 30
2.48 37 2.43 35 2.59 36 3 .07 29
2.35 37 2.67 33 2.87 38 3.34 29
2.25 38 2.47 35 2.62 38 3.19 30
Numerou s pustules are usually present, some of which may be large. Practically all of the facial area is involved with lesions. Large, prominent pustules are usually quite visible. Lesions are usually of a highly inflammatory type. Other types of acne (such as conglobata, including sinus and cystic types) may be present.
Grading scale for comedones Grade
o
Subjective and objective criteria
No comedones Mild comedonal acne 2 About 10 lesions per side of face 4 Obvious comedonal acne About 25 lesions per side of face 6 Severe comedonal acne About 50 lesions per side of face 8 Worst possible comedonal acne About 100 or more lesions per side of face Counts were made of papules and comedones (open and clo sed) on the right side of the face from the hairline to the mandibular ridge . Pustules were counted on both sides of the face. STUDY DESIGN The study lasted a total of 10 weeks. Patients were evaluated by the same three graders on weeks 0, 2, 4, 6, 8, and 10. Two weeks priorto the onset of the study, a preadmission evaluation was performed on a majority of participants. There was no collaboration between graders at the time of the grades , and counts were recorded. Both graders and patients were unaware of the medications they were using. Pictures were taken on weeks 0, 4, 8, and 10. Medications and directions for their use were given on week O. New supplies of medications were given on weeks 4 and 8. On weeks 4, 8,
486
Journal of the Am er ican Academy of Dermatology
Felicia et at
Table IV. Reduction and percent reduction in overall severity grade by groups Weeks Groups
I
Reduction and % reduction
2
Reduction % reduction Reduction % reduction Reduction % reduction Reduction % reduction
1.09 (0.00 I) 25.69 (0.001) 1.10 (0.001) 27.42 (0.001) 0.85 (0.05) 18.35 (0.05) 0.37 6.04
I. Topical erythromycin/zinc liquid
2. Topical erythromycin/zinc gel 3. Oral tetracycline 4. Placebo
Statistic ally significant reduction in comparison with placebo by the Wilcoxon Rank Sum Test with p significant .
<
4
1.33 (0.001) 30.80 (0.01) 1.24 (0.01) 29.32 (0.005) 1.05 (NS) 20.76 (NS) 0.52 11.51
the value in parentheses; NS: not
Table V. Average papule counts by groups Weeks (average number) . Groups
I. Topical erythromycin/zinc liquid
2. Topical erythromycin/zinc gel 3. Oral tetracycline 4. Placebo
-2
14.27 34 12.97 33 12.97 36 11.24 25
and 10, the quantities of unused medications were recorded. The 4% erythromycin/zinc in the liquid vehicle was distributed in squeeze containers along with 5-cm adsorbant pads. Patients were instructed to saturate the pads and to apply the liquid to the entire face. The 4% erythromycin/zinc in the gel vehicle was distributed in plastic squeeze tubes with instructions for it to be applied with the tips of the fingers, again to the entire face. The precounted capsules were distributed in plastic containers, and patients were instructed to take them only with water I hour before or 2 hours after eating . ST ATISTICAL l\IETHODS In the statistical analysis of the data (acne severity grades, papule counts, etc.), the response to therapy in this paper was defined as reduction or percent reduction. This reduction or percent reduction was calculated by comparing the treated condition at weeks 2, 4, 6, 8, and 10 with the pretreatment condition at week O. The reduction or percent reduction of each active group was then compared to the placebo group at each visit to
10
17.41 38 15.96 35 17.35 38 14.44 30
9.93 38 8.52 3I . 10.36 36 11.91 29
8.12 38 7.80 33 9.42 37 10.13 30
7.13 37 7.54 35 7.51 36 9.64 29
7.17 37 7.57 33 8.35 38 9.93 29
6.47 38 7.50 35 7.64 38 9.39 30
determine statistical significance . The comparisons between the active groups and the placebo group were made using the Wilcoxon Rank Sum Test. 19.20 RESULTS One hundred forty-nine participants began the study. The data from 141 arc included. The data from three participants are not included in this paper since during the course of the study they were given antibiotics by their own physicians for other illnesses. One participant was given erythromycin for a skin infection of the right ankle. The other two participantswere treated for upper respiratory infections with tetracycline and erythromycin, respectively. Five students did not finish the study. Of these, one moved, three did not return, and one complained of irritation and decided to stop the study. The one student who complained of irritation was on topical 4% erythromycin/zinc liquid and oral
Volume 3 Number 5 November, 1980
Erythromycin and zinc in acne
C)
0.00
.....
"g
~
Weeks
6
1.60 39.24 1.47 33.93 1.54 35.02 0.64 13.98
(0.001) (0.001) (0.01) (0.01) (0.01) (0.01)
8
1.61 (0.001) 40 .87 (0.001) 1.23 (0.01) 27.62 (0.05) 1.21 (0.01) 26.77 (0.05) 0.37 7.24
10
1.81 (0.001) 46.07 (0.001) 1.43 (0.01) 33.02 (0.01) 1.46 (0.01) 32.05 (0.01) 0.48 6.88
~f-l
'~
0.50
~
.~
C)
~
C/)
~'>1~ TT..J r' t4~ .'"
'.,-,;
1.00
.s
.2
1.50
PLACEBO . - - - ETM LIQUID ..--- ElM GEL D-ORAL TeN 0--
U j
"0
~
487
2.00
"',
- - .....
<,
~
I
Bars are ± 10 S.E.
2.50 ' - - - - . l - - - - ' - - _ - ' - - _ - - L - _ - - L _ - - - - ' - _ - . J -2 a 2 4 6 8 10 Time on Study (weeks)
Fig. lAo Reduction in severity grade.
placebo. He had few signs of an eruption other than some erythema which disappeared within 24 to 48 hours. He refused patch tests, but by history, clinical findings, and length of problem , it was judged not to be a contact dermatitis. The number of participants who began, finished, and were included in the study in each group are listed in Table I. Participants were not dropped from the study if they missed a visit for an evaluation as long as they returned on subsequent visits . No participant whose data were included in the results missed more than two visits or failed to return for a new supply of medications. In Table II are the number of visits missed by participants . An average individual grade or count was determined from the evaluation of the three graders. An average group grade or count for each group was determined by averaging these individual scores of the group at each visit. Table III shows the results for the average overall severity grade for each group at each visit. The reduction for a group is calculated by determining the reduction of each individual in the group and then averaging those reductions. The percent reduction of each individual in the group is averaged together with the others for the percent reduction of the group. Table IV shows the reduction and percent reduction of the overall severity grade in each group at weeks 2, 4, 6, 8, and 10 in comparison with week O. Statistically significant differences from placebo are followed by the p value in parentheses . Figs. IA and lB graphically demonstrate the
~
00.00
,
~-~=---=-------T
'~-~~
c
l5 20.00 >. II>
<, .Ii':....
~ 40.00
o
~ 60.00 j
PLACEBO . - - - ElM LIQUID .---- ETM GEL o-ORALTC
"0
Bars are .1 1.0 S.E.
0--
C/)
c
C)
~ 80.00
iJ:
T
=t -~.::.:..... ~~~"6
.~
~
--'<>---Q
I
1.
r---I--_I I
'---_--'--_---'.L_-L_ _' - -_--'---_---'---J
-2
o
2 4 6 8 Time on Study (weeks)
10
Fig. lB. Percent reduction in severity grade .
average reduction and average percent reduction in severity grade, respectively, versus time. Table V shows the results of the average papule counts (taken from the right side of the face) for each group at each visit. Table VI shows the reduction and percent reduction of the papule counts in each group at weeks 2, 4, 6, 8 , and 10 in comparison with week O. Again, statistically significant differences from placebo arc noted by the p value in parentheses. Figs. 2A and 2B show graphically the reduction and percent reduction in papule counts, respectively, versus time. Pustule counts showed no statistically significant difference between groups throughout the study. Table VII shows average comedo grades and counts obtained for each group at each visit. The only statistically significant differences from placebo occurred in the topical erythromycin group at
Journal of the American Academy of Dermatology
488 Feucht et at
Table VI. Reduction and percent reduction papule counts by groups Weeks Groups
7.48 31.69 6.82 28.38 7.28 35.70 2.56 7.23
Reduction
I. Topical erythromycin/zinc liquid
% reduction
Reduction % reduction Reduction % reduction Reduction % reduction
2. Topical erythromycin/zinc gel 3. Oral tetracycline 4. Placebo
I
2
Reduction and % reduction
4
(0.01) (0.001) (0.001) (0.001) (0.001) (0.001)
9.29 46.16 8.20 38.97 8.01 36.89 4.31 23.15
(0.01) (0.01) (0.01) (0.05) (0.05) (NS)
Statistically significant reduction in comparison with placebo by the Wilcoxon Rank Sum Test with p < the value in parentheses; NS: not significant.
Table VII. Average comedo grades and counts by groups
Groups
I. Topical erythromycin/
zinc liquid 2. Topical erythromycin/ zinc gel 3. Oral tetracycline 4. Placebo
I
Parameter
Grade Count No. Grade Count No. Grade Count No. Grade Count No.
I
Weeks
-2
0
2
4
6
8
10
2.51 '29.35 34 2.46 29.50 33 2.38 28.44 36 2.17 24.57 25
2.11 31.50 38 2.02 32.28 35 2.16 34.83 38 2.07 31.74 30
2.18 28.82 38 2.27 28.66 31 2.40 33.32 36 2.32 29.76 29
2.55 30.81 38 2.31 29.68 33 2.64 34.32 37 2.49 31.90 30
2.27 25.99 37 2.24 24.70 35 2.46 28.41 36 2.39 26.57 29
2.05 22.78 37 2.29 25.03 33 2.46 27.04 38 2.75 29.45 29
1.89 20.42 38 2.21 23.10 35 2.36 26.09 38 2.59 27.70 30
weeks 8 and to. Table VIII shows these differences at weeks 8 and to. Note that the average comedo grade for the topical erythromycin/zinc liquid group decreased from week 0 while in all other groups these grades increased. All groups showed some slight decrease in average comedo counts from week O. However, the topical erythromycin/zinc liquid group showed a much larger decrease. Table IX shows correlations among the three judges for each of the five acne measurements performed. All of these correlation coefficients are statistically significant with a p value of less than 0.01. In the grading of acne, though it is not possible to decide a right or wrong grade given for a particular patient at a particular visit, it is possible
through these correlations to see how close the judges were in their measurements. The amount of antibiotic usage was determined for each group. This was determined from the amount of medication that was returned. The group with 4% erythromycin in a liquid vehicle used 2.32 mIlday or 86.3 mg/day of erythromycin. This does not take into account the amount which is absorbed into the pads and discarded after each usage. The group with 4% erythromycinin a gel vehicle used 670 mg/day of the gel or 26.8 mg/day of erythromycin. This does not take into account what remairis on the fingertips and is not used. The groups taking tetracycline orally took an average of 1.85 capsules per day or 462.5 mg of tetracycline per day.
Volume 3
Erythromycin and zinc ill acne
Number 5 November, 1980
Weeks 6
10.41 51.87 8.42 39.06 10.22 54.94 5.08 23.34
(0.01) (0.001) (0.05) (0.05) (0.01) (0.001)
10
8
9.83 53.36 8.35 39.93 9.00 48 .33 4.62 23. I I
(0.01) (0.001) (0.05) (0.05) (0.01) (0.001)
10.94 58.11 8.46 44.65 9.71 51.17 5.06 .25.43
(0.01) (0.001) (0.05) (0.05) (0.01) (0.01)
C ::I
~
•
04.00
go 06.00
a,
'f
,~
~rN
\
,~
\,~I
.:
s 08.00
PLACEBO . - - - ETM LIQUID .---- ETM GEL o-ORAL TC
0--
CONCLUSIONS AND .DISCUSSION
l
r '" k!+~?.t
10.00 12.00
'
l.,t~--t--j----l
U ::I
-g 0::
489
..........
Bcrs ore :t 1.0 SE .
Evaluation of the preceding results supports several conclusions. First, the particular topical erythromycin/zinc liquid and gel formulations that were used in the treatments described in the preceding paragraphs produced statistically significant decreases in both overall severity grades and papule counts compared to placebo. Further, the erythromycin/zinc liquid and gel are as effective as 500 mg tetracycl ine orally per day. Also, comedo grades and counts were statistically significantly decreased by use of the topical erythromycin liquid but by none of the other formulations. It is of interest to note that this improvement only became manifest after 8 to 10 weeks. The exact explanation for the delayed improvement is not known, but one possible consideration might be that the topical erythromycin or zinc acetate exerts some as yet unknown inhibitory effect at some stage in the process of comedo formation. Whether this effect is due to antibacterial, chemotactic, keratoplastic, or some other totally unrelated mechanism is as yet undetermined. The placebo group throughout the study showed a slight but definite improvement. Although there was some dryness noted in these patients probably secondary to the topical placebo, liquid or gel, these placebo medications are not known to have any acne-improving qualities. This slight improvement is what is considered by most investigators as the placebo effect which in this study is evident. Despite the improvement in papule and severity grades, pustule counts were not improved by any
14.00 L . - _ - ' - - _ - - - ' - - - _- ' -_ _ ~_ _ '--_.l_______J -2 o 2 4 6 8 10 Time on Study (weeks)
Fig. 2A. Reduction in papule count.
COo.OO~\
s
i
~\
2000
~ 40.00
~ ~
cp
0::
;;.e
60.00
~
\l: t---l-"- ...
y---1
~~-+---+--".I '!-.. ~ -"-i
o--PLACEBO _---ElM LIQUID •• • ETM GEL 0 - ORAL TC Bers ore ± 1.0 S.E.
80.00 ' - - _ - ' - _ - - ' - _ - - L _ - - - ' _ _-'--_--'----' -2 024 6 8 ID Time on Study (weeks)
Fig. 28. Percent reduction in papule count.
of the utilized preparations. This might be because so few pustules (compared to papules) were seen in our patients, and these numbers were too small for significant results to be obtained. The usage data displayed in this study show that the 4% erythromycin liquid and gel patients used on an average no more than 86.3 mg per day and 21.6 mg per day, respectively, of the erythromycin to achieve the previously noted improvement. When compared to the 500 mg per day dosages usually required for oral ·treatment of mild or moderate acne, it becomes evident that the liquid and gel erythromycin used in this study enable the patients to experience significant improvement with the use of a minimum amount of active in-
490
Journal of the Ameri can Academy of Dermatology
F cue/It et of
Table VIII. Reduction and percent reduction in comedo grades and counts between placebo and topical erythromycin liquid groups at weeks 8 and 10 Weeks Groups
Grades and count (reduction and % reduction)
8
Grades: Reduction % reduction Count: Reduction % reduction Grades: Reduction % reduction Counts: Reduction % reduction
0.09 (0.01) 0.86 (0.01) 8.80 (0.05) 10.93 (0.01) -0.64 -41.78 2.53 -17.43
I. Topical erythromycin/zinc liquid
2. Placebo
Stati stically significant reduction in comparison with placebo by the Wilcoxon Rank Sum Test with p
I
10
0.23 (0.01) 8.71 (0.01) 11 .08 (0.05) 14.83 (0.05) -0.52 -33.80 4.04 -14.09
< the value in parentheses.
Table IX. Correlation coefficient of judges* Weeks on study Measurements
Judges
0
Severity grade
I vs 2 I vs 3 2 vs 3 I vs 2 1 vs 3 2 vs 3 I vs 2 I vs 3 2 vs 3 1 vs 2 I vs 3 2 vs 3 1 vs 2 1 vs 3 2 vs 3
0.82 0.87 0.87 0.81 0.88 0.83 0.68 0.65 0.84 0.73 0.78 0.85 0.69 0.70 -0.74
Papule count Pustule count Comedo count Comedo grade
·AII corr elation coefficients are statistically significant (p
I
2
0.81 0.86 0.83 0.83 0.82 0.88 0.70 0.74 0.80 0.87 0.85 0.89 0.78 0.79 0.84
I
4
0.83 0.82 0.85 0.83 0.85 0.88 0.82 0.81 0.80 0.84 0.84 0.84 0.82 0.75 0.79
I
6
0.81 0.86 0.82 0.85 0.87 0.89 0.84 0.87 0.85 0.88 0.89 0.88 0.81 0.77 0.83
I
8
0.86 0.84 0.86 0.86 0.90 0.89 0.78 0.86 0.79 0.88 0.86 0 .89 0.84 0.76 0.83
I
10
0.88 0.88 0.88 0.91 0.86 0.88 0.83 0.81 0.85 0.86 0.86 0.90 0.85 0.78 0.81
< 0.01).
gredient. Further, patients using the liquid and gel erythromycin preparation actually received less than the stated milligram per day dosages of erythromycin in view of the fact that a certain percentage (unable to be exactly determined) of the medication is lost at each application, either by being absorbed into the pad used for liquid application or by being washed off the fingers in the case of the gel preparation . From Table IX, our correlation coefficients showed that three independent judges can evaluate acne with good correlation, especially in regard to severity grades and papule count. This speaks highly for the Cook grading scale and supports the
fact that objective data can be obtained from the usage of this grading system." In regard to the zinc component of this topical preparation, little is known or established about its mechanism of action. Erythromycin and zinc form a complex when in solution in the vehicle. Studies in rats, utilizing radiolabeled erythromycin and zinc, have demonstrated that the penetration of erythromycin into the skin is enhanced by the presence of zinc in the formulation. Also, special staining technics applied to excised skin from hairless mice and pigs have demonstrated that when erythromycin is present in the formulation the zinc penetrates the stratum corneum into the epidermis
Volume 3 Number 5 November, 1980
and perifollicular apparatus. * Though these studies give some idea of its localization and penetration, only speculation can be made about its mechanism of action. In summary, we have presented a double-blind placebo-controlled study evaluating a topical erythromycin liquid and a topical erythromycin gel compared to liquid and gel placebos and 500 mg per day oral tetracycline . The erythromycin liquid and gel tested gave statistically significant improvement of acne when judged by papule counts and by severity grades with the Cook method. A further unexpected finding was the late but definite reduction in comedo grades and counts in patients using the erythromycin liquid.
Erythromycin and zinc ill acne
7. Fulton JE Jr, Pablo G: Topical antibacterial therapy for acne . Arch Dennatol 110:83-86 , 1974.
8. Van Ketel WG: Immediate and delayed type allergy to erythromycin. Contact Dermatitis 2:363-364, 1976.
9. Fisher A: Erythromycin "free base " -a nonstinging top-
10.
II.
12.
13.
14.
'Unpublished data. on file, Procter and Gamble, Cincinnati. OH . REFERENCES
I . Crounse RG: The respons e of acne to placebos and antibiotics. JAMA 193:906-910, 1965. 2. Robinson HM Jr. et al: Topical acne therapy. South Med J s.t:1105. 1961. 3. Miroliubov VI: Chloromphenicol in the treatment of folliculitis, acne vulgaris and rosace a. Vestn Dennatol Ven erol 37 :68-71, 1963. 4 . Hollander L, Shelton JM, Hardy SM : Acrornycin f lotion an adjunct in treating acne vulgaris. American Prac titioner Digest of Treatment 8:1602-1605 , 1957. 5. Blaney OJ, Cook CH : Topical use of tetracycline in the treatment of acne: A double-blind study comparing topical and oral tetracycline therapy and placebo" Arch Dermatol 112:971-973, 1976 . 6. Smith JG Jr, Chalker OK, Wehr RF: The effectiveness of topical and oral tetracycline in acne. South Med J 69:695-697. 1976.
491
15. 16. 17.
18.
19.
20 .
21.
ical antibiotic for infected dermatosi s and acne vulgaris. Cutis 20:17,26.30 . 1977. Lee PWR , Green MA. Long \VB III, and Gill W: Zinc and wound healing. Surg Gynecol Ob stet 143:549-554, 1976. Michaelsson G, Juhlin L , Vahlquist A: Effects of oral zinc and vitamin A in acne. Arch Dermatol 113:31·36, 1977. Michaelsson G, Juhlin L, Ljunghall K: A double blind study of the effect of zinc and oxytetracycline in acne vulgaris. Br) Derrnatol 97:679·684, 1977. Hill strom L, et al: Comparison of oral treatment with zinc sulphate and placebo in acne vulgaris . Br J Dermatol 97:679-684, 1977. Weismann K, Wadsker S, Sondergaard J: Oral zinc sulphate therapy for acne vulg aris . Act a Denn Venereol (Stockh) 57:357-360 , 1977. Orris L. et al: Oral zinc therapy of acne . Arch Dermatol 114:1018-1020, 1978. Weimar VM, et al: Zinc sulfate in acne vulgaris. Arch Dennatol 114:1776-1778, 1978. Goran sson K. Liden S, Odsell L: Oral zinc in acne vulgaris-a clinical and methodological study. Acta Denn Venereol (Stockh) 58:443-448, 1978 . Cook CH , Centner RL, Mich aels SF: An acne grading method using photographic standards. Arch Derrnatol 115:571-575, 1979. Hollander M, Wolfe DA: Nonparametric statistical methods. New York, 1973, John Wiley & Sons, Inc., pp . 68-74. Brown BW, Hollander M: Statistics: A biomedical introduct ion. New York, 1977 , John Wiley & Sons, Inc., pp. 134-136. Allen BS, Smith JG Jr: A study of various parameters for grading acne vulgaris. (To be published.)