Microdermabrasion has become an extremely

Microdermabrasion with and without aluminum oxide crystal abrasion: A comparative molecular analysis of dermal remodeling Darius J. Karimipour, MD, Sewon Kang, MD, Timothy M. Johnson, MD, Jeffrey S. Orringer, MD, TedHamilton,MS,CraigHammerberg,PhD,JohnJ.Voorhees,MD,andGaryFisher,PhD Ann Arbor, Michigan Background: Microdermabrasion is a popular method of superficial skin resurfacing with effects on dermal remodeling. Objective: The purpose of this study was to evaluate the relative importance of the two components of microdermabrasion, negative pressure and abrasion, in stimulating expression of key genes involved in dermal remodeling. Methods: Ten subjects were treated with a microdermabrasion machine using focal crystal abrasion and negative pressure or negative pressure alone for 3 seconds. Serial biochemical analyses were performed. Reverse transcriptase real-time polymerase chain reaction assays were used to evaluate changes in transcription factor activator protein-1, primary cytokines (interleukin 1b, tumor necrosis factor-a), and matrix metalloproteinases (MMP-1, MMP-3, MMP-9). Results: Significant increases in gene expression of the c-jun component of activator protein-1, interleukin 1b, tumor necrosis factor-a, MMP-1, MMP-3, and MMP-9 were found with crystal abrasion combined with negative pressure. Negative pressure alone resulted in increased gene expression of MMP-1 and MMP-3 but of a quantitatively reduced magnitude when compared with negative pressure with crystal abrasion. Limitations: It is unclear that molecular changes seen with these treatments can result in clinical effect. Conclusion: The abrasive component of microdermabrasion is necessary for stimulating expression of key genes involved in dermal remodeling. ( J Am Acad Dermatol 2006;54:405-10.) From the Department of Dermatology, University of Michigan Medical School. Funding sources: Research for this article supported by the Skin Surgical Research Fund at the University of Michigan, Bella Products, and a Dermatology Foundation Clinical Career Development Award (to D. J. K.). Bella Products donated the microdermabrasion machine, hand pieces, and crystals. Bella Products had no knowledge of the outcomes of this study. As a condition of acceptance of the donated equipment from Bella Products, the study was designed to be performed, analyzed, and reported solely by the faculty and staff of the Department of Dermatology at the University of Michigan. Conflicts of interest: None identified. Accepted for publication November 27, 2005. Reprint requests: Darius J. Karimipour, MD, University of Michigan, 1500 E Medical Center Dr, Department of Dermatology, 1910 Taubman Center, Ann Arbor, MI 48109-0314. E-mail: dariusjk@ umich.edu. Published online January 24, 2006. 0190-9622/$32.00 ª 2006 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2005.11.1084 Abbreviations used: AP-1: activator protein-1 IL: interleukin MMP: matrix metalloproteinase NF: nuclear factor RT-PCR: real-time polymerase chain reaction TNF: tumor necrosis factor Microdermabrasion has become an extremely popular method of superficial skin resurfacing. Approximately 860,000 cases of microdermabrasion were performed in the United States in 2004. 1 Microdermabrasion may improve many skin problems including acne scars, 2 acne, 3 skin texture irregularities, mottled pigmentation, 4 and fine wrinkles. 5,6 To date, there are many different types of microdermabrasion systems. The basic aluminum oxide/sodium chloride-based 405

406 Karimipour et al JAM ACAD DERMATOL MARCH 2006 microdermabrasion system is a closed loop consisting of the following: vacuum suction distributed through a hand piece that sucks the skin into the hand piece, a compression system that propels crystals (corundum) at a programmable pressure when the skin is sucked into the hand piece, a reservoir for unused corundum, and a waste receptacle which contains used corundum and skin debris that have been aspirated back through the vacuum. We recently reported that microdermabrasion induces epidermal signal transduction pathways associated with remodeling of the dermal matrix. 7 Emerging evidence indicates that internal or externally applied mechanical tension can substantially alter the activity of key pathways that regulate connective tissue homeostasis. 8-10 Therefore we have investigated the efficacy of negative pressure, which stretches the skin, and compared it with microdermabrasion in its ability to induce expression of key genes involved in dermal matrix remodeling. METHODS This study was approved by the University of Michigan Medical School Institutional Review Board for Human Subjects Research. All subjects provided written informed consent. Ten subjects (7 male, 3 female), aged 21-55 years, received focal, brief (3-second) exposure of forearm skin to the microdermabrasion hand piece at a suction setting of 15 mm Hg, using the Bellamed Microdermabrasion unit (Bella Products, Foothill Ranch, Calif). The hand piece was placed on an area of skin that was the size of the hand piece aperture (0.6 3 0.6 cm). Each subject experienced treatment with negative pressure with crystal abrasion and negative pressure alone. This method of treatment resulted in a mild pink erythema similar to what is seen in day-to-day clinical use of microdermabrasion. A second group of 6 subjects (4 male, 2 female), aged 25 to 60 years, were treated similarly; however, negative pressure was applied to buttock skin in 10-, 20-, and 30-second increments, whereas negative pressure with crystal abrasion was applied to skin for 3 seconds. Punch biopsy specimens (4 mm) were obtained from treated and untreated (control) skin at 4, 8, and 24 hours after treatment. Skin specimens were snap-frozen in liquid nitrogen. Each subject served as his or her own control. Snap-frozen tissue was used to quantify mrna levels using reverse transcriptase/real time quantitative polymerase chain reaction (RT-PCR). 11 Total RNA 100 ng was reverse transcribed with random primers by using a Taqman Reverse Transcription Kit (Applied Biosystems, Foster City, Calif). A 7700 sequence detector (Applied Biosystems) and Taqman Universal PCR Master Mix Reagents (Applied Biosystems) were used for RT-PCR. Appropriate PCR primers and probes for the analyzed molecular markers were produced by the custom oligonucleotide synthesis service (Applied Biosystems). Transcript levels for primary cytokines (interleukin 1b [IL-1b], tumor necrosis factor-a [TNF-a]), extracellular matrix-degrading metalloproteinases (MMP- 1, MMP-3, and MMP-9), and c-jun component of activator protein-1 (AP-1), were determined. Statistical methods Baseline/control transcript levels were assigned a value of 1, and fold change above baseline was measured for each subject for each of the biochemical end points. Arithmetic means for the fold changes in biochemical end points were calculated across all subjects. Repeated measures analysis of variance was used to statistically evaluate changes in biochemical end points over the time course of the study. The Dunnett test was used to make individual pairwise comparisons of values at each subsequent time point and with baseline levels. The type I error rate was set at 0.05. Logarithmic transformations of the data were made to achieve normality before analysis, if necessary. All figures depict the data on the untransformed scale. Summary statistics include means and standard errors. The data were analyzed with SAS statistical software (SAS Institute, Inc. Cary, NC). RESULTS Negative pressure with crystal abrasion, but not negative pressure alone, induces c-jun Expression of the c-jun component of AP-1 was elevated after negative pressure with crystal abrasion was applied (P \.05). There was an approximate 3- and 5-fold increase in c-jun gene expression at 4 and 8 hours after treatment with negative pressure and crystal abrasion, respectively. Negative pressure alone failed to cause statistically significant increases in c-jun mrna levels (Fig 1). Negative pressure with crystal abrasion, but not negative pressure alone, induces primary cytokines IL-1b and TNF-a A single microdermabrasion treatment with negative pressure and crystal abrasion resulted in substantial elevations in both IL-1b and TNF-a gene expression. Similar increases were not noted in the negative pressure alone group. TNF-a mrna expression increased 4-fold after treatment with negative pressure and crystal abrasion at 4 hours. This statistically significant effect persisted for at least 8 hours after treatment (P \.05). Small increases in TNF-a mrna expression occurred after treatment

JAM ACAD DERMATOL VOLUME 54, NUMBER 3 Karimipour et al 407 Fig 1. Negative pressure with crystals induces c-jun mrna at 4 and 8 hours, whereas negative pressure alone results in minimal induction as quantified by RT-PCR (N = 10; each subject served as his or her own control). Asterisk, P \.05 versus no treatment control. Fig 2. Elevation of TNF-a gene expression at 4 and 8 hours after treatment with negative pressure and abrasion. Minimal induction occurs with negative pressure alone as quantified by RT-PCR (N = 10; each subject served as his or her own control). Asterisk, P \.05 versus no treatment control. with negative pressure alone at 4, 8, and 24 hours; however, these elevations were not statistically significant (Fig 2). IL-1b gene expression demonstrated a 27-fold increase at 4 hours (P \.05) followed by a decline over the ensuing 24 hours in forearm skin exposed to negative pressure with abrasion. IL-1b mrna levels tended to increase at 4 and 8 hours following treatment with negative pressure alone; however, these alterations failed to reach statistical significance (Fig 3). Negative pressure alone and with crystal abrasion both induce matrix degrading enzymes We examined expression of 3 MMPs known to be inducible by microdermabrasion in human skin. 7 Interstitial collagenase (MMP-1) gene expression was elevated 200-fold above pretreatment levels at 4 hours, 1800-fold at 8 hours, and 37-fold at 24 hours after treatment (P \.05) when using negative pressure with abrasive crystals. Statistically significant elevations in interstitial collagenase were also noted following treatment with negative pressure alone, where 18-fold and 90-fold increases were noted at 4 and 8 hours after treatment, respectively (P \.05) (Fig 4). MMP-1 induction was significantly less with negative pressure alone when compared to negative pressure in combination with crystal abrasion at each of these time points (P \.05). Similarly, stromelysin- 1 (MMP-3) gene expression was elevated 400- fold at 4 hours and 1750-fold at 8 hours after treatment with negative pressure and crystal abrasion before rapidly decreasing at 24 hours (P \.05). Similar increases of reduced magnitude, 150-fold and 26-fold, were noted at 4 and 8 hours, respectively, in skin treated with negative pressure alone (P \.05) (Fig 5). The Fig 3. Elevation of IL-1b gene expression at 4 hours after treatment with negative pressure and abrasion. Minimal induction occurs with negative pressure alone as quantified by RT-PCR (N = 10; each subject served as his or her own control. Asterisk, P \.05 versus no treatment control. difference between treatment with negative pressure alone and negative pressure with crystal abrasion was statistically significant at the 8-hour evaluation (P \.05). Induction of gelatinase-b (MMP-9) gene expression displayed kinetics similar to MMP-1 and MMP-3. Significant increases in MMP-9 gene expression were noted at 8 and 24 hours with 8-fold and 4-fold elevations, respectively, in forearm skin treated with negative pressure and abrasion (P \.05). Negative pressure alone failed to demonstrate any statistically significant increases in MMP-9 expression (Fig 6). Increasing time of exposure to negative pressure does not replicate the results seen when using abrasive crystals Skin treated with negative pressure alone for 10, 20, and 30 seconds did demonstrate elevations in

408 Karimipour et al JAM ACAD DERMATOL MARCH 2006 Fig 4. Elevation of MMP-1 gene expression after treatment with negative pressure and crystal abrasion and negative pressure alone (N = 10; each subject served as his or her own control) Asterisk, P \.05 versus no treatment control. Quantitatively higher gene expression is noted after treatment with negative pressure and crystal abrasion when compared with negative pressure alone (double asterisk, P \.05 vs time-matched control). Fig 5. Elevation of MMP-3 gene expression after treatment with negative pressure and abrasion and negative pressure alone (N = 10; each subject served as his or her own control). Asterisk, P \.05 versus no treatment control. Quantitatively higher gene expression is noted after treatment with negative pressure and crystal abrasion when compared with negative pressure alone (double asterisk, P \.05 vs time-matched control). gene expression of MMP-1 and MMP-3 (P \.05). However, increasing exposure time to negative pressure did not result in a simulation of the effect on gene expression noted when negative pressure and crystal abrasion were combined for 3 seconds. That is, negative pressure combined with crystal abrasion demonstrated statistically higher levels of gene expression for MMP-1, MMP-3, and MMP-9 despite increases in negative pressure application time up to 30 seconds (P \.05) (data not shown). DISCUSSION We recently elucidated a dermal remodeling cascade that may account for the beneficial histologic and aesthetic effects following microdermabrasion noted by several authors. 7 This cascade involves components known to be important in the repair seen after ultraviolet damage and wound healing that occurs after carbon dioxide laser resurfacing. 12,13 Our previous work demonstrated a predictable timeline by which microdermabrasion treatment induced activation of transcription factors AP-1 and nuclear factor-kb (NF-kB), which, in turn, stimulated gene expression of TNF-a and IL-1b These cytokines are known to induce certain MMPs that then degrade collagen and other dermal connective tissue components. 14 Interstitial collagenase, MMP-1, is the first to be expressed and it degrades mature collagen into smaller fragments that can then be further degraded by stromelysin-1 (MMP-3) and gelatinase-b (MMP- 9). 14 This should set the stage for new collagen production to maintain a homeostatic balance. The purpose of this study was to determine the relative effects of the abrasive component of the Fig 6. Elevation of MMP-9 gene expression after treatment with negative pressure and crystal abrasion at 8 and 24 hours compared with minimal induction with negative pressure alone (N = 10; each subject served as his or her own control). Asterisk, P \.05 versus no treatment control. microdermabrasion procedure when compared with the negative pressure component alone. Using negative pressure alone would be beneficial since it could decrease procedure expense. Additionally, safety concerns have arisen related to microdermabrasion. Although microdermabrasion is a very safe procedure, complications related to the corundum may occur. Ocular complications, including eye irritation and adherence of crystals to the cornea, have been reported. 15 Aluminum oxide dust has also been associated with pulmonary fibrosis. 16 Aluminum exposure has also been implicated in the pathophysiology of Alzheimer s disease. 17 Although some of these assertions regarding the deleterious effects of aluminum need further

JAM ACAD DERMATOL VOLUME 54, NUMBER 3 Karimipour et al 409 evaluation, crystal-free microdermabrasion might be a welcome addition to a dermatologist s armamentarium. In fact, certain entrepreneurs have noticed this opportunity and developed machines based on a crystal-free system where an abrasive hand piece replaces spraying corundum. There is also precedent to suggest that negative pressure alone might induce skin remodeling. Endermologie, a machine-assisted massage system using positive and negative pressure, is a technique that is used with some success in the remodeling of skin and subcutaneous tissue. 18 In our previous article, we suggested that microdermabrasion may use minimal epidermal barrier disruption and mechanotransduction to stimulate a dermal remodeling cascade. 7 Indeed, evidence exists to suggest that mechanical forces can regulate extracellular matrix homeostasis through its effects on collagen and matrix degrading enzymes produced by fibroblasts. 19 Although negative pressure alone demonstrates some downstream activation of the cascade alluded to in our previous article (eg, MMPs), it activates gene expression of the cascade components to quantitatively lower levels. Hence it may be expected to result in less significant molecular remodeling and, necessarily, clinical effect. We did not see activation of the full cascade described in our previous work with negative pressure alone, and there may be several reasons for this. When we looked at AP-1 activation and cytokine production, negative pressure alone showed merely a trend toward statistical significance. A variety of different physical and physiological stimuli activate transcription factor AP-1. This transcription factor plays a critical role in regulation of expression of many genes involved in inflammation, wound healing, growth, differentiation, and apoptosis. AP-1 is composed of two families of gene products, Jun and Fos. The major Fos family member, c-fos, is abundant, whereas the major Jun family member c-jun is limiting for AP-1 function in normal epidermis. 20,21 c-jun gene expression was not elevated to statistically significant levels when using negative pressure alone, but it was elevated when crystal abrasion was added. IL-1b and TNF-a may be derived from preformed precursor proteins in keratinocytes or may be produced de novo. These cytokines induce AP-1 and elicit increased gene expression of MMPs. 14 Although gene expression of MMP-1 and MMP-3 was elevated to statistically significant levels after treatment with the microdermabrasion hand piece using negative pressure alone, these responses were statistically less when compared with the combined effect of microdermabrasion with aluminum oxide. Large increases in the negative pressure exposure time could not replicate the effects on gene expression seen when negative pressure and crystal abrasion are combined. This does not appear to be related to differences in treatment location (forearm vs buttock) as our previous work demonstrated that a single microdermabrasion treatment to buttock skin elicits changes in c-jun, primary cytokines, and MMPs that parallel the changes seen in forearm skin in this study. 7 It is possible that we did not see statistically significant elevations in gene expression of c-jun and the inflammatory cytokines because their baseline levels in the skin are not as low as baseline cutaneous MMP expression. Hence small changes in MMP levels could be resolved from their very low baseline levels, allowing detection of a statistically significant change. Additionally, small upstream elevations in AP-1, IL-1b and TNF-a might result in sufficient amplification of the signal downstream to elicit significant increased expression of MMP levels in the skin treated with negative pressure alone. That is, the proximal pathway amplified the distal pathway to detectable levels. Finally, NF-kB is another proinflammatory transcription factor that might participate in induction of cytokines or MMPs. NF-kB was not measured in our study so its role in stimulating responses to negative pressure alone through the microdermabrasion hand piece is unknown. Research on pulmonary tissue has shown that NF-kB is sensitive to mechanical stretch 22 ; hence it may have been operating to elicit low-level cytokine responses in skin treated with negative pressure alone. Indeed, small elevations in TNF-a and IL-1b gene expression were noted in the skin treated with negative pressure alone. It is important to note that subjects received only focal treatment for 3 to 30 seconds with negative pressure alone and 3 seconds with crystal abrasion; this was different from our previously published work where microdermabrasion was performed with successive passes over a larger area of skin, as would be performed in a clinical setting. 7 The focal nature of the treatment may account for any quantitative differences when compared with our previous work. Although we did not evaluate changes in collagen gene expression in this study, it is doubtful that any changes would have been seen. Procedures that strongly elicit new collagen production (carbon dioxide laser resurfacing, dermabrasion) cause significant epidermal disruption. One would expect that the relatively nondisruptive nature of microdermabrasion without corundum would induce little, if any, collagen production. Indeed, we observed collagen production in only 2 of 11 subjects in our previous work after one aggressive microdermabrasion treatment. 7

410 Karimipour et al JAM ACAD DERMATOL MARCH 2006 In conclusion, aluminum oxide microdermabrasion appears to have significant advantage over nonabrasive microdermabrasion using only negative pressure as evaluated by alteration of molecular markers of dermal remodeling. We thank Kenne Currie and Suzan Rehbine for their work in this project. REFERENCES 1. ASAPS. 2004 ASAPS Statistics: 9.2 million cosmetic plastic surgery procedures in 2004: up 5% growth paces U.S. economy despite reality TV fad. [American Society for Plastic Surgeons website]. Available at: http://www.plasticsurgery. org/new_room/press_releases/2004-overall statistics.cfm Accessed Aug 26, 2005. 2. Tsai R, Wang C, Chan H. Aluminum oxide crystal microdermabrasion: a new technique for treating facial scarring. Dermatol Surg 1995;21:539-42. 3. Lloyd J. The use of microdermabrasion for acne: a pilot study. Dermatol Surg 2001;27:329-31. 4. Shim E, Barnette D, Hughes K, Greenway H. Microdermabrasion: a clinical and histopathologic study. Dermatol Surg 2001; 27:524-30. 5. Hernandez-Perez E, Ibiett E. Gross and microscopic findings in patients undergoing microdermabrasion for facial rejuvenation. Dermatol Surg 2001;27:637-40. 6. Tan M, Spencer J, Pires L, Ajmeri J, Skover G. The evaluation of aluminum oxide crystal microdermabrasion for photodamage. Dermatol Surg 2001;27:943-9. 7. Karimipour DJ, Kang S, Johnson TM, Orringer JS, Hamilton T, Hammerberg C, et al. Microdermabrasion: a molecular analysis following a single treatment. J Am Acad Dermatol 2005; 52:215-23. 8. Eckes B, Krieg TR. Regulation of connective tissue homeostasis in the skin by mechanical forces. Clin Exp Rheumatol 2004; 22(Suppl 33):S73-6. 9. Silver FH, Siperko LM, Seehra GP. Mechanobiology of force transduction in dermal tissue. Skin Res Technol 2003;9:3-23. 10. Bershadsky AD, Balaban NQ, Geiger B. Adhesion-dependent cell mechanosensitivity. Annu Rev Cell Dev Biol 2003;19: 677-95. 11. Quan T, He T, Kang S, Voorhees J, Fisher G. Connective tissue growth factor: expression in human skin in vivo and inhibition by ultraviolet irradiation. J Invest Dermatol 2002; 118:402-8. 12. Fisher G, Wang Z, Datta S, Varani J, Kang S, Voorhees J. Pathophysiology of premature skin aging induced by ultraviolet light. N Engl J Med 1997;337:1419-28. 13. Orringer JS, Kang S, Johnson TM, Karimipour DJ, Hamilton T, Hammerberg C, et al. Connective tissue remodeling induced by carbon dioxide laser resurfacing of photodamaged human skin. Arch Dermatol 2004;140:1326-32. 14. Herouy Y. Matrix metalloproteinases in skin pathology. Int J Mol Med 2001;7:3-12. 15. Lu CF, Chan HL, Yu HS. The physical effect and repair process of skin after aluminum oxide crystal microdermabrasion [abstract]. Presented at the 19th Annual Meeting of the Chinese Dermatological Society, Taipei, November 13-14, 1993. 16. Jederlinic PJ, Abraham JL, Churg A, Himmelstein JS, Epler GR, Gaensler EA. Pulmonary fibrosis in aluminum oxide workers. Investigation of nine workers, with pathologic examination and microanalysis in three of them. Am Rev Respir Dis 1990; 142:1179-84. 17. Candy JM, Oakley AE, Klinowski J, Carpenter TA, Perry RH, Atack JR, et al. Aluminosilicates and senile plaque formation in Alzheimer s disease. Lancet 1986;1:354-7. 18. Chang P, Wiseman J, Jacoby T, Salisbury AV, Ersek RA. Noninvasive mechanical body contouring: (Endermologie) a one-year clinical outcome study update. Aesthetic Plast Surg 1998;22:145-53. 19. Lambert CA, Soudant EP, Nusgens BV, Lapiere CM. Pretranslational regulation of the extracellular matrix macromolecules and collagenase expression in fibroblasts by mechanical forces. Lab Invest 1992;66:444-51. 20. Fisher G, Datta S, Talwar H, Wang ZQ, Varani J, Kang S, et al. Molecular basis of sun-induced premature skin ageing and retinoid antagonism. Nature 1996;379:335-9. 21. Fisher G, Talwar H, Lin J, Lin P, McPhillips F, Wang Z, et al. Retinoic acid inhibits induction of c-jun protein by ultraviolet radiation that occurs subsequent to activation of mitogenactivated protein kinase pathways in human skin in vivo. J Clin Invest 1998;101:1432-40. 22. Kumar A, Lnu S, Malya R, Barron D, Moore J, Corry DB, et al. Mechanical stretch activates nuclear factor-kappab, activator protein-1, and mitogen-activated protein kinases in lung parenchyma: implications in asthma. FASEB J 2003;17: 1800-11.