Long-Term Results of Carbon Dioxide Laser Resurfacing of the Face

ORIGINAL ARTICLE Long-term Results of Carbon Dioxide Laser Resurfacing of the Face

P. Daniel Ward, MD, MS; Shan R. Baker, MD

Objective: To evaluate the long-term effectiveness of, of treatment, hypopigmentation was the only long-term and the complications associated with, carbon dioxide adverse effect. This complication was present in 6 pa- laser treatment of rhytidosis and solar aging. tients (13%). The patients who developed hypopigmentation were more likely to have a greater response to treat- Methods: This retrospective report describes our ex- ment. perience with 47 patients who underwent entire facial carbon dioxide laser resurfacing. Conclusion: Our findings show that carbon dioxide laser resurfacing is a safe and effective treatment for facial Results: The mean improvement in facial rhytid score rhytids. at long-term follow-up was 45%. This improvement was consistent in all facial subsites. With the exception of 1 case of hyperpigmentation, which resolved within 2 years Arch Facial Plast Surg. 2008;10(4):238-243

HE LASER IS A WELL-ESTAB- The experience with carbon dioxide la- lished and often used tool for ser resurfacing at our institution indi- the treatment of facial cates that nearly all patients experience rhytids. Use of the laser al- some degree of hypopigmentation after la- lows precise treatment, giv- ser resurfacing. However, the degree of ingT the surgeon more control over the re- postprocedure hypopigmentation is highly surfacing procedure than is possible with variable, as is the level of concern that this other techniques such as chemical peels and problem causes for patients.1 The inci- dermabrasion. The carbon dioxide laser, dence of hypopigmentation is thought to which is commonly used in facial resurfac- be related to the degree of thermal in- ing, exerts its effects by vaporizing intra- jury,2 a theory that is supported by the find- cellular and extracellular water molecules, ing that hypopigmentation is associated causing thermal damage to the surround- with the degree of improvement in rhyti- ing tissue. This insult prompts fibroblasts dosis.3 Furthermore, hypopigmentation to increase collagen production and re- has also been associated with a greater de- sults in the observed clinical effects. gree of preprocedure rhytidosis, perhaps suggesting that patients who receive more See also page 244 aggressive treatment are at increased risk of developing hypopigmentation. In addition to structural changes, the The primary purpose of this study was healing process frequently leads to pig- to assess the long-term results after full- mentary changes. These changes in skin face carbon dioxide laser resurfacing. pigmentation may be desirable, such as Long-term efficacy and complications were when patients wish to remove solar evi- examined, with particular attention di- dence of aging (Figure 1 and Figure 2); rected at determining the incidence of hy- however, changes in pigmentation after popigmentation that was evident on pho- treatment can often be a troubling ad- tographic review by an independent Figure 3 observer. The population of patients who Author Affiliations: verse effect ( ). Both hyperpig- Department of mentation and hypopigmentation are com- developed hypopigmentation was fur- Otolaryngology–Head and Neck plications that are associated with laser ther analyzed to determine whether there Surgery, University of Michigan, resurfacing, with the latter being more are any other factors that are associated Ann Arbor. common. with hypopigmentation.

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 10 (NO. 4), JULY/AUG 2008 WWW.ARCHFACIAL.COM 238

Figure 1. Patient who underwent carbon dioxide laser resurfacing because of concerns about her freckles. Patient’s appearance before (A) and 5 years after (B) the procedure.

METHODS was performed to assess any differences between the treated anatomical subsites. All complications were recorded. The presence of hypopig- After written informed consent was obtained from the pa- mentation was determined by review of postoperative photo- tients, all procedures were performed by the senior author graphs. The characteristics of the patients who developed hy- (S.R.B.) using a commercially available carbon dioxide laser (Co- popigmentation were compared with those of the patients who herent UltraPulse; Coherent Inc, Santa Clara, California) be- did not develop hypopigmentation. The mean age, preopera- tween December 1996 and December 2004. The settings used tive rhytid score, and percentage of improvement in rhytid for the resurfacing procedure were 300 mJ with a density of score were all recorded and analyzed to determine any differ- 5 to 6 for the facial skin and 250 mJ with a density of 5 for the ence between the groups. One-way analysis of variance was eyelids. Three passes were used for the face and the eyelids. again used to determine whether there was any statistical sig- Care was taken not to use the laser in areas where skin vascu- nificance between the means of the parameters of the patients larity was potentially at risk, such as on skin flaps elevated dur- with hypopigmentation and those of the patients without hy- ing concurrent rhytidectomy. popigmentation. The patient’s sex, skin type, and age at the time of the procedure were all recorded. Any other procedures that the patient also underwent at the time of resurfacing were also docu- RESULTS mented. Preoperative and postoperative photographs were analyzed and graded to determine wrinkle scores before and after treatment. The scale used was described by Lemperle et Sixty-two patients who underwent entire facial laser re- al4 in 2001. It was initially designed and verified to be accurate surfacing were identified. Complete data were available for grading facial wrinkles before and after treatment with in- in 47 cases. There were 42 women and 5 men, with a mean jectable fillers; however, it has since been demonstrated to be age of 52 years. All 47 patients had Fitzpatrick skin type an effective method to use for grading facial rhytids that have I, II, or III, with the exception of 1 patient who had skin been treated with the carbon dioxide laser.1 The glabellar, peri- type IV. The mean and median duration of follow-up was orbital, and melolabial regions as well as the corners of the mouth 2.3 and 1.8 years, respectively (range, 0.2-8.1 years). Many and the upper lip were graded and assigned preprocedure and patients underwent other procedures along with facial postprocedure scores. Also, the average score for the entire face resurfacing, including dermabrasion, blepharoplasty, was calculated using the scores for each of the 5 subsites. Im- brow-lift, and rhytidectomy (Figure 4). provement in rhytid score was calculated in relation to the preoperative rhytid score. The improvement in rhytid score was The mean improvement in facial rhytid score was 45% determined by calculating the ratio of the difference between (95% confidence interval [CI], 40.6%-49.7%), and there the preprocedure and postprocedure scores and the preopera- was no difference in improvement between facial sub- tive score. Only the most recent photographs were used to de- sites (Figure 5). Complications encountered included termine the postoperative score. One-way analysis of variance milia or acne in 14 cases (30%), hyperpigmentation in

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 10 (NO. 4), JULY/AUG 2008 WWW.ARCHFACIAL.COM 239

C D

Figure 2. Patient who underwent carbon dioxide laser resurfacing because of rhytidosis and solar aging. Patient’s appearance before (A and B) and 8 years after (C and D) the procedure.

At the long-term follow-up visit, most of the compli- A B cations had resolved. The only complications present after 1 year were hypopigmentation and 1 instance of hyperpigmentation, which resolved by 2 years. Of the 14 patients with follow-up time of less than 1 year, 1 had hyperpigmentation that resolved after 5 months, and 3 had milia or acne flare-ups. None of the patients with follow-up of less than 1 year developed hypopigmentation. The patients with hypopigmentation had a mean age of 49.0 years (95% CI, 39.0-59.0) compared with a mean age of 52.7 years (95% CI, 48.7-56.7) in those without hypopigmentation (Figure 7). This difference was not statistically significant (P=.51). To determine whether the patient’s preoperative degree of rhytidosis may have led to a more aggressive treatment, the preoperative rhytid scores were calculated for patients who did and did not develop hypopigmentation (Figure 8). The average preoperative score for patients with hypopigmentation was 2.20 Figure 3. Left (A) and right (B) views of postoperative hypopigmentation in a (95% CI, 1.23-3.17). The average preoperative score patient who underwent carbon dioxide laser resurfacing. for patients without hypopigmentation was 2.15 (95% CI, 1.82-2.48). This difference was not significant 8 cases (17%), hypopigmentation in 6 cases (13%), in- (P=.92). fection in 1 case (2%), and ectropion in 1 case (2%) The response to treatment was assessed by compar- (Figure 6). Twenty-one patients (45%) had no com- ing the mean improvement in rhytid scores of the pa- plications. tients with and without hypopigmentation (Figure 9).

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 10 (NO. 4), JULY/AUG 2008 WWW.ARCHFACIAL.COM 240

60 25 50 20 40

15 30 Mean Age, y 20 No. of Patients 10 10

5 0 No Hypopigmentation Hypopigmentation

None Figure 7. Mean age of patients with and without hypopigmentation. Error Brow-lift Face-lift bars indicate 95% confidence intervals. Canthopexy Rhinoplasty Dermabrasion Submentoplasty Debulk Nasal Flap Submalar Implant

BlepharoplastyBlepharoplasty (Upper) (Lower) 3.5 Blepharoplasty (4 Quadrants) 3.0

Figure 4. Other procedures that were performed along with carbon dioxide 2.5 laser resurfacing. 2.0

1.5 80 1.0 70 0.5 60 Mean Preoperative Rhytid Score

50 0 No Hypopigmentation Hypopigmentation 40

30 Figure 8. Mean preoperative rhytid score in patients with and without Improvement, % 20 hypopigmentation. Error bars indicate 95% confidence intervals. 10

0 Glabellar Periorbital Melolabial Corner of Upper Overall 100 Region Region Region Mouth Lip 90 80 Figure 5. Percentage of improvement in facial rhytid score after carbon 70 dioxide laser resurfacing. Error bars indicate 95% confidence intervals. 60 50 40 50

Improvement, % 30 45 20 40 10 35 0 No Hypopigmentation Hypopigmentation 30

25 Figure 9. Percentage of improvement in rhytid score in patients with and

Patients, % 20 without hypopigmentation. Error bars indicate 95% confidence intervals. 15

10 5 improvement (95% CI, 33.9%-49.7%) in patients who did 0 not develop hypopigmentation. This difference was sig- None Infection nificant (P=.004). Milia/Acne Ectropion

Hypopigmentation COMMENT

Hyperpigmentation (Transient) The efficacy of treating facial rhytids with the carbon di- Figure 6. Percentage of patients with complications after carbon dioxide oxide laser is well established, and the short- and long- laser resurfacing. term utility of the carbon dioxide laser in treating solar facial aging has previously been documented.3,5-7 Our re- The patients with hypopigmentation had a greater results verify those of previous studies that found that car- sponse to treatment, with a 73.9% mean improvement bon dioxide laser resurfacing leads to long-term improve- (95% CI, 59.8%-88.0%), compared with a 41.8% mean ment in facial rhytidosis.1,3,5-7 The mean improvement in

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 10 (NO. 4), JULY/AUG 2008 WWW.ARCHFACIAL.COM 241

©2008 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 rhytid score in the present study was 45%. This degree ies from patient to patient, as does the degree of con- of improvement in facial rhytid score was consistent in cern that this problem elicits. Dijkema and van der Lei1 all 5 anatomical subsites. Prior studies have docu- found that 34% of patients who underwent laser resur- mented decreased efficacy in the perioral region, pre- facing of the upper lip used some sort of camouflage to sumably because of the increased muscular activity in this hide their hypopigmentation. region.1,5 Although the perioral region demonstrated the Our experience is that all patients who undergo car- least degree of improvement among the subsites in our bon dioxide laser resurfacing with our protocol have some study, the difference was not significant (P=.51). degree of hypopigmentation. We attempted to deter- The complications that were encountered in this study mine the incidence of hypopigmentation that was evi- are similar to those described in previous studies. Milia dent on photographic review. Previous authors have re- and acne were the most frequent adverse effects, fol- ported that the degree of thermal injury and the depth lowed by hyperpigmentation, hypopigmentation, infec- of penetration are directly related to the incidence of hy- tion, and ectropion. Milia and acne flares, which are rela- popigmentation.1-3 Patients who develop hypopigmen- tively common reactions, are thought to be caused by the tation have been shown to have a greater degree of solar use of postoperative dressings, ointments, and aberrant damage before laser resurfacing as well as a greater im- follicular epithelialization.2 Spontaneous resolution of acne provement in rhytids.3 Our study did not note a greater is the typical course, although short courses of oral an- degree of preprocedure rhytidosis in patients who de- tibiotics or, if the skin has fully reepithelialized, topical veloped hypopigmentation but did note a greater de- antibiotics may be helpful. For milia, the use of retinoic gree of improvement in rhytid score in these patients. An acid can be helpful in accelerating resolution. additional indicator of possible hypopigmentation de- One patient developed a postoperative viral erup- velopment could be increasing patient age, with its re- tion, and 1 patient developed postoperative ectropion. sultant decrease in skin thickness. Our study did not note The viral infection occurred despite the use of an anti- a difference in age between patients who did and did not viral agent before and after the procedure. Zoster-dose develop hypopigmentation; however, our small sample antiviral medications and supportive treatment were ad- size may have prevented us from finding a difference in ministered, and the patient had no further sequelae. The age between the 2 groups. patient with ectropion was treated with eye lubricants For patients who develop hypopigmentation, chemi- and had complete resolution within 1 year. cal peels using glycolic acid or tricholoroacetic acid may The only complications present after 1 year were hy- be used to help blend the lines of demarcation between perpigmentation (1 patient) and hypopigmentation (6 pa- treated and untreated skin. Efforts to increase melano- tients). All 8 patients who had postprocedure hyperpig- genesis, such as limited exposure to UV light and appli- mentation were treated with topical hydroquinone, and cation of methoxsalen, may also be helpful.2 The most 7 (88%) of them had resolution within 1 year. The pa- important treatment, however, is prevention. The risk of tient with persistent hyperpigmentation at the1-year fol- developing hypopigmentation can be minimized by cau- low-up visit had resolution within 2 years after the pro- tioning the patients to use sunscreen before and after the cedure. Our experience supports that of several previous procedure and by treating all skin within cosmetic units studies in which hyperpigmentation occurred in 0% to to decrease the incidence of observable lines of demar- 30% of patients but usually resolved within the first sev- cation. eral months.1-3,5-12 The incidence of this complication, In conclusion, carbon dioxide laser resurfacing is a safe which is more common in patients with darker skin types, and effective treatment for facial rhytidosis. The mean is decreased by avoidance of sun exposure for several improvement in facial rhytid score was 45%. Complica- weeks (at least 4) before the resurfacing procedure. Ad- tions encountered included hypopigmentation, milia, vising patients to continue sunscreen use postopera- acne, hyperpigmentation, infection, and ectropion. All tively is also important to help prevent further pigmen- complications resolved in all patients by 1 year after the tary abnormalities.2 One month after the procedure, procedure, with the exception of 1 patient with hyper- bleaching agents (eg, hydroquinone), retinoic acid, and pigmentation and 6 patients with hypopigmentation. Some glycolic acid peels may be used to hasten resolution of degree of hypopigmentation occurs in all patients who the pigmentary abnormalities.2 undergo carbon dioxide laser resurfacing with a proto- Hypopigmentation was photographically evident at the col similar to that used in our study. Patients who expe- long-term follow up visit in 6 of the 47 patients (13%). rience a greater improvement in rhytidosis are more likely This rate compares favorably with the rates quoted in the to develop hypopigmentation. Although increased age and literature, which range from 8% to 57%.1,3,5,6 The litera- preoperative level of rhytidosis were not found to be as- ture concerning hypopigmentation after laser resurfac- sociated with increased hypopigmentation in this study, ing is somewhat difficult to interpret owing to the lack they are likely to be associated with an increased risk of of a standard definition of hypopigmentation. Also, some hypopigmentation. authors make a distinction between true hypopigmentation and pseudohypopigmentation. True hypopigmen- Accepted for Publication: February 22, 2008. tation, which refers to skin lightening due to decreased Correspondence: P. Daniel Ward, MD, MS, Depart- melanogenesis, is rare, whereas pseudohypopigmenta- ment of Otolaryngology–Head and Neck Surgery, Uni- tion, which refers to areas of treated skin that are lighter versity of Michigan, 1904 Taubman Center, 1500 E Medi- than surrounding untreated skin,3 is much more com- cal Center Dr, Ann Arbor, MI 48109 (pdanielw@med mon. The degree of hypopigmentation that occurs var- .umich.edu).

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 10 (NO. 4), JULY/AUG 2008 WWW.ARCHFACIAL.COM 242

©2008 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 Author Contributions: Drs Ward and Baker had full ac- 3. Manuskiatti W, Fitzpatrick RE, Goldman MP. Long-term effectiveness and side cess to all of the data in the study and take responsibil- effects of carbon dioxide laser resurfacing for photoaged facial skin. J Am Acad ity for the integrity of the data and the accuracy of the Dermatol. 1999;40(3):401-411. 4. Lemperle G, Holmes RE, Cohen SR, Lemperle SM. A classification of facial wrinkles. data analysis. Study concept and design: Ward and Baker. Plast Reconstr Surg. 2001;108(6):1735-1752. Acquisition of data: Ward and Baker. Analysis and inter- 5. Schwartz RJ, Burns AJ, Rohrich RJ, Barton FE Jr, Byrd HS. Long-term assess- pretation of data: Ward and Baker. Drafting of the manu- ment of CO2 facial laser resurfacing: aesthetic results and complications. Plast script: Ward. Critical revision of the manuscript for impor- Reconstr Surg. 1999;103(2):592-601. tant intellectual content: Ward and Baker. Statistical 6. Bisson MA, Grover R, Grobbelaar AO. Long-term results of facial rejuvenation analysis: Ward and Baker. Obtained funding: Ward and by carbon dioxide laser resurfacing using a quantitative method of assessment. Br J Plast Surg. 2002;55(8):652-656. Baker. Administrative, technical, and material support: Ward 7. Senso¨z O, Nazmi Baran C, Sahin Alago¨z M, Cag˘ri Uysal A, Unlu¨ RE. Long-term and Baker. Study supervision: Baker. results of ultrapulsed carbon dioxide laser resurfacing of the Mediterranean face. Financial Disclosure: None reported. Aesthetic Plast Surg. 2004;28(5):328-333. Previous Presentation: This study was presented in part 8. Bernstein LJ, Kauvar AN, Grossman MC, Geronemus RG. The short- and long- at the Fall Meeting of the American Academy of Facial term side effects of carbon dioxide laser resurfacing. Dermatol Surg. 1997; Plastic Surgery; September 17, 2007; Washington, DC. 23(7):519-525. 9. Nanni CA, Alster TS. Complications of carbon dioxide laser resurfacing: an evalu- ation of 500 patients. Dermatol Surg. 1998;24(3):315-320. REFERENCES 10. Lowe NJ, Lask G, Griffin ME. Laser skin resurfacing: pre- and posttreatment guidelines. Dermatol Surg. 1995;21(12):1017-1019. 1. Dijkema SJ, van der Lei B. Long-term results of upper lips treated for rhytides 11. Alster TS. Cutaneous resurfacing with CO2 and erbium:YAG lasers: preopera- with carbon dioxide laser. Plast Reconstr Surg. 2005;115(6):1731-1735. tive, intraoperative, and postoperative considerations. Plast Reconstr Surg. 1999; 2. Alster TS, Lupton JR. Prevention and treatment of side effects and complica- 103(2):619-634. tions of cutaneous laser resurfacing. Plast Reconstr Surg. 2002;109(1):308- 12. Ratner D, Tse Y, Marchell N, Goldman MP, Fitzpatrick RE, Fader DJ. Cutaneous 318. laser resurfacing. J Am Acad Dermatol. 1999;41(3, pt 1):365-392.

Announcement

Visit www.archfacial.com. As an individual subscriber to Archives of Facial Plastic Surgery, you have full-text online access to the journal from 1999 forward. In addition, you can find abstracts to all JAMA & Archives Journals as far back as 1975.

(REPRINTED) ARCH FACIAL PLAST SURG/ VOL 10 (NO. 4), JULY/AUG 2008 WWW.ARCHFACIAL.COM 243