In Vivo Model of Histologic Changes AfterTreatment With the Superpulsed CO 2 Laser,

Erbium:YAG Laser, and Blended Lasers:A 4- to 6-Month Prospective Histologic and

Clinical StudyDavid Greene, MD,1–3* Barbara M. Egbert, MD,4 David S. Utley, MD,1,2 and

R. James Koch, MD1,2

1Division of Otolaryngology / Head and Neck Surgery, Facial Plastic and Reconstructive Surgery,Stanford University Medical Center, Palo Alto, California 94305

2The Palo Alto Veteran’s Administration Health Care System, Palo Alto, California 943043Department of Pathology/Dermatopathology, Palo Alto Veteran’s Administration Health Care

System, Palo Alto, California 943044Department of Otolaryngology—Head and Neck Surgery, Cleveland Clinic Florida,

Fort Lauderdale, Florida 33309

Background and Objective: To compare the in vivo histologic effects ofthe pulsed carbon dioxide (CO2) and erbium:ytrium aluminum garnet(Er:YAG) lasers and to assess the effects of combining CO2 and Er:YAGlaser modalities during a single treatment session. We previously re-ported 10 patients treated with four laser regimens: CO2 alone, CO2/Er:YAG, Er:YAG alone, Er:YAG/CO2 with time points at 1 hour and 7days between laser treatment and histologic analysis. This study foundthat the optimal treatment consisted of limited CO2 laser passes fol-lowed by Er:YAG. This treatment produced less collagen injury, lessthermal necrosis, and more robust epithelial and dermal fibrous tissueregeneration in the acute phase of healing. The present study examinesthe histologic changes resulting from the host healing response to lasertreatment on long-term follow-up of 4–6 months.Study Design/Materials and Methods: The Stanford University Commit-tee on Human Subjects in Medical Research approved this study. Ninepatients with actinic damage and indications for rhytidectomy volun-teered for this interventional study in which each patient served asboth experimental and control. The right preauricular area wastreated at five sites with the following: (1) CO2, (2) CO2 followed byEr:YAG, (3) Er:YAG, (4) blended CO2/Er:YAG (Derma-K™), (5) phenol.Each was subjected to full-face or sub-unit treatment. Each patient wasfollowed up initially daily then weekly for healing of the full-face laserand for differences in healing of the five treatment areas. Five patientswere selected for histologic evaluation. At 4–6 months, these patientsunderwent rhytidectomy with immediate removal of laser-treated skin,which was evaluated histologically by the study dermatopathologist,who was blinded to the treatment at each site.

Presented at the American Society for Lasers in Medicine andSurgery Meeting, Orlando, Florida, April 16, 1999.

*Correspondence to: David Greene, MD, Division of Otolar-yngology/Head and Neck Surgery, Cleveland Clinic Florida,3000 West Cypress Avenue, Fort Lauderdale, FL 33309.E-mail: dgreen8095@aol.com

Accepted 1 February 2000

Lasers in Surgery and Medicine 27:362–372 (2000)

Published 2000 Wiley-Liss, Inc. †This article is a USGovernment work and, as such, is in the public domain in theUnited States of America.

Results: CO2 laser treatment produced the greatest thickness of neo-collagen (0.27 mm; P < 0.05), the highest neocollagen density (P < 0.05),the greatest decrease in elastosis (27%), but took the longest time forhealing and resolution of erythema and inflammation (up to 6 months).Er:YAG used alone produced the least collagen density, with the thin-nest band of neocollagen (0.08 mm), but the most rapid resolution oferythema and inflammation (within 10 days). Combined CO2/Er:YAGtreatments, including Derma-K™ and CO2 followed by Er:YAG producedhistologic changes that were intermediate, as well as recovery that wasintermediate (resolution of erythema within 1 month); the developmentof neocollagen was greater in CO2-containing modalities than Er:YAGused alone by a statistically significant margin (P = 0.001). These his-tologic findings were corroborated by clinical correlation by examina-tion of the five treatment spots in nine patients and in full-face treat-ments in 100 patients.Conclusion: Collagenesis is greatest with CO2 and least with Er:YAG.Elastosis decreased to the greatest degree with CO2, least with erbium,and to an intermediate extent with blended CO2/Er:YAG regimens (se-quential and Derma-K™). These changes from control are statisticallysignificant with all regimens (P < 0.05). Blended CO2/Er:YAG treatmentsprovide an optimal combination of the benefits of CO2 but with lessererythema and healing delay. Clinical and histologic findings changeover time for different treatments. Thus, long-term histology is criticalfor predicting results of treatment. Lasers Surg. Med. 27:362–372, 2000Published 2000 Wiley-Liss, Inc.†

Key words: collagen; dermis; epidermis; facial; laser ablation; resurfacing; skin

INTRODUCTION

Rapid advances in laser technologies haverequired histologic studies to evaluate their tissueeffects, such as depth of penetration, necroticzone, uniformity of ablation, and other character-istics of each laser [1–5]. Recent reports that usedhuman and animal in vivo models now providedata on the effects of laser in physiologically nor-mal living skin [2,6–8]. We have successfully useda preauricular in vivo model in which laser treat-ments were placed at multiple preauricular sites,with subsequent biopsy at rhytidectomy [9]. Thismodel seems to provide accurate data on thephysiologic effects of various laser treatments inhuman skin. The present research examines thelong-term clinical and histologic results of fiveprincipal laser and chemical resurfacing modali-ties.

The pulsed CO2 laser has an established rec-ord of safety and efficacy for the application offacial skin resurfacing [1,9,10]. The advantages ofthe CO2 laser are reported as thermal contractionof collagen [11,12], widespread familiarity of phy-sicians with CO2, improved technology with addi-tion of computerized pattern generator (CPG)handpiece [13], relatively lower equipment coststhan alternative lasers, and adequate delineationby previous authors of the histologic tissue effectsand clinical outcomes expected from this laser

[1–14]. In addition, Kuo et al. reported that addi-tional thermal damage causes increased collagensynthesis [15]. These findings suggest that thethermal damage with CO2 lasers might be advan-tageous and that a scanned continuous wattagecarbon dioxide laser might even produce bettercollagenesis than the pulsed laser.

Potential disadvantages of the CO2 laser in-clude the creation of a thermal injury zone deep tothe ablation zone, as well as a diminishing abla-tion depth and increasing thermal necrosis zonewith each successive pass [1,2,9,16–18]. Excessivethermal injury has been implicated as a possibleetiology for prolonged erythema and delayedwound healing sometimes observed after treat-ment with the CO2 laser [1,2,18,19].

Considerably less data are available for theEr:YAG laser. Er:YAG has a 20-fold higher coef-ficient of absorption in water (12,700 l/cm at 2.94microns) than CO2 (639 l/cm at 10.6 microns) anda reduced optical penetration in tissue (1 mm vs.20–30 mm) [1,2]. The Er:YAG laser is reported tohave less thermal necrosis and less tissue abla-tion per pass and, therefore, less erythema andwound healing complications [4,20–22]. The useof combined lasers synergistically represents anexciting frontier in laser science. The use of Er:YAG in conjunction with the CO2 laser to reducethe thermal necrosis zone while retaining the col-

In Vivo Model of Histologic Changes 363

lagen-tightening advantages of the CO2 has beenpromising [9,23]. In a previous study that used anin vivo preauricular skin model, we found lasertreatments combining CO2 followed by Er:YAG tobe superior in terms of healing and favorable his-tologic changes [9].

Since that time, the advent of blended lasershas provided yet another mode of combining er-bium and CO2, and additional controversy as tothe mode of physiologic effects. For instance, theDerma-K™ (ESC Medical Systems, Yokoam, Is-rael) system alternates pulsed Er:YAG laser withcontinuous wave CO2 laser. The Er:YAG accom-plishes the tissue ablation, and the subablativeCO2 provides sufficient thermal damage to stimu-late collagenesis and induce collagen tightening.These effects have not been confirmed by a long-term in vivo model (Koch, R.J., personal commu-nication).

Phenol peels comprise possibly the oldestmodality of skin resurfacing and should be in-cluded in in vivo studies of laser effects as a treat-ment control. Minoli and Barton (1998) have usedan in vivo pig model to compare phenol, glycolicacid, Jessner’s solution, and CO2 with 3-monthfollow-up [24]. They report that the effects of CO2were undetectable by 3 months, and that wounddepth of dermabrasion and laserbrasion were lostas the skin resumed its normal architecture at 3months. They found that phenol and trichloroace-tic acid (TCA) produce changes to the upper re-ticular dermis that persisted throughout their ob-servation period [24]. Such a comparison had yetto be attempted in a human in vivo model.

Criteria critical to assessment of differencesbetween different laser regimens include theamount of tissue removed with a given fluence,irregularity of ablation, thickness of the necroticzone, ablation shrinkage, and inflammation. Toassess the ability of each model to measure thesedifferences, we examined the histologic effects oflaser regimens incorporating CO2 and erbium, themost popular laser types for resurfacing, and phe-nol.

To date, few studies have examined differ-ences in the effects of pure and blended lasers andchemical peel in a long-term in vivo humanmodel. We define as long-term the interval of 4–6months to differentiate it from short-term modelsof a week or less and ex vivo models by usingtissue after resection. The present study was un-dertaken to systematically assess the effects ofeach treatment modality both histologically andclinically in the preauricular in vivo skin model.

The objective of this study is to examine the his-tologic differences between four laser treatmentregimens and phenol peel.

MATERIALS AND METHODS

One hundred patients underwent laser re-surfacing. From this pool, 19 patients were se-lected for test site treatment followed by clinicalevaluation of each treatment site. Ten patientswith areas removed at rhytidectomy at 7 days and1 hour were previously reported [14]. Nine addi-tional patients underwent treatment to five testsites, and facial resurfacing, after which theywere followed clinically for 4–6 months. The algo-rithm for patient selection and evaluation is de-picted in Figure 1. Five patients were selected forhistologic evaluation and were scheduled to un-dergo cervicofacial rhytidectomy. This study wasapproved by the Stanford University Committeeon Human Subjects in Medical Research, andthorough informed consent was obtained.

Two lasers were used. The Luxar LX-20SPNovapulse™ CO2 laser (10,600 nM) with Sure-Scan Computer Pattern Generator (CPG) (LuxarCorporation, Bothell, WA) was used in the super-pulse mode at 6 watts, 16 hertz (Hz), and a 730-msec pulse duration. The 7.6 mm by 6.4 mm CPGparallelogram was used. This laser delivered afluence of 4.7 joules(J)/cm2.

The ESC Derma-K™ laser was used both asthe pure Er:YAG laser (2,940 nM) and with theblended Er:YAG and CO2 source. For Er:YAG, a6-mm spot size at 14 watts and 8 Hz, with a 350-msec pulse duration was used to deliver 1.7J/pulse and a fluence of 4.7 J/cm2. For the blendedlaser sites, Er:YAG component settings were asabove, and CO2 with a duty cycle of 50% at 3 watts

Fig. 1. Clinical algorithm for patient selection and evalua-tion.

364 Greene et al.

was used. These parameters were selected to pro-vide identical fluences between the laser systemscompared in this study.

Patient Evaluation

Each patient was examined preoperativelyto determine Fitzpatrick skin types [25], and tograde the degree of facial skin actinic damage andrhytidosis by using a five-point scale.

Experimental Treatment

Single or combination laser energy was ap-plied at four separate targets and phenol to onetarget after injection of the patient with ringblock, V1-3 nerve blocks, and oral sedation. Carewas taken to inject subcutaneously, not intrader-mally, to prevent distortion of the tissues. Treat-ments were applied in a crescent area anterior tothe right ear, in an area typically removed atrhytidectomy. Treatment regimens were as fol-lows: (1) CO2 (four passes), (2) CO2 (two passes)followed by Er:YAG (four passes), (3) Er:YAG(eight passes), (4) blended CO2/erbium (Derma-K™) (five passes), (5) phenol 53% in H2O appliedafter delipidizing skin with ethanol 80%.

Wiping was performed after each pass withthe CO2 laser. The balance of the face or estheticunit was treated with blended laser (Derma-K™)with 1.7 J/pulse Er:YAG and 3 watts CO2 at dutycycle of 30–50%. Two to four passes were used.

Clinical Correlation Evaluation

To provide clinical correlations for histologicfindings, the patients were followed every otherday for 1 week, then weekly with serial examina-tions and photography over a 4–6 month follow-upperiod.

Harvest of Specimens at Rhytidectomy

After general anesthesia or local sedationwas performed, minimal infiltration of local anes-thetic was accomplished. The treatment siteswere located with a standardized template andphotographs from the laser application session.Each specimen was excised with a sharp new6-mm skin punch and was transferred immedi-ately to 10% formalin in individually labeled con-tainers. An additional untreated specimen wassubmitted as control. The containers were labeledwith a code.

Histologic Evaluation

One histopathology technician prepared allspecimens. Hematoxylin and eosin (H and E)

staining was used. Each slide was labeled with astudy code allowing later identification of eachspecimen as to the type of laser used and the timeinterval of treatment. The study dermatopatholo-gist (B.M.E.) remained blinded to this codethroughout the histology evaluation process. Thecode was not broken until all histopathologic datawere recorded and ready for analysis.

Control specimens were histologicallygraded by using a five-point scale (Table 1) fordegree of melanocyte atypia, melanocytic hyper-trophy, and melanocytic hyperplasia, as well askeratocytic atypia, polarity abnormalities, andparakeratosis. Dermal elastosis was graded. Anoptical micrometer was used to determine thethickness (mm) of the epidermis, papillary dermis,and reticular dermis. In addition, the severity andlocation of baseline inflammation was evaluated.

For each treatment site, the following factorswere assessed: degree of healing, inflammation,neocollagen density, melanocytic atypia, kera-tinocytic atypia, and elastosis. The optical mi-crometer was used to measure the thickness ofthe epidermis, dermis, and neocollagen thickness.Averages of each histologic finding for each treat-ment type were compared, and Student t-testwith the Bonferroni-Dunn factor correction incases in which multiple treatments were com-pared, and chi-square tests were used to comparethese data with those found in the acute phase at7 days.

RESULTSHistologic Evaluation

Five patients were selected for long-term his-tologic evaluation 4–6 months after laser treat-ment. Four patients were Fitzpatrick Sun-Reactive Skin Type II, and 1 was type IVaveraging 2.4. Three had a history of acne, all hadactinic damage ranging from 3–5 (on the 0–5scale) with an average of 4.0. Rhytidosis rangedfrom 2–5 with a mean of 3.8.

Histologic examination of the control speci-mens (Fig. 2A) revealed moderate melanocytic

TABLE 1. Scoring System Used to Grade FacialChanges Both Clinically and Histologically

0 0% or no findings1 20% or mild2 40% or mild-moderate3 60% or moderate4 80% or moderate-severe5 100% or severe

In Vivo Model of Histologic Changes 365

atypia (mean, 2.0), melanocytic hypertrophy(mean, 2.2), and minimal melanocytic hyperplasia(0.6). Epidermal thickness averaged 0.074 mm,with an average of 8.6 cell layers. The dermis av-eraged 2.1 mm in thickness. Keratinocytic atypiaand parakeratosis were moderate. Significantdermal solar elastosis, ranging from 3–4 with amean of 3.6 on the five-point scale, was encoun-tered. Baseline minimal to moderate inflamma-tion was seen in the perivascular area in all speci-mens.

All treated areas (five of five) were com-pletely healed at 4–6 months. There was no evi-dence of collagen ablation or injury in any speci-men. All specimens demonstrated re-growth ofdermal appendages and epidermis (Fig. 2B–E).

Comparison of each treatment’s induction ofneocollagen are compared in Figure 3A,B. Neocol-lagen thickness, determined by the change fromcontrol, showed some degree of increase in everyspecimen. Measurement of the thickness of theneocollagen band width with the optical microm-eter showed the following. (1) CO2 produced thegreatest thickness with mean thickness of 0.27mm. (2) CO2 followed by Er:YAG produced a meanneocollagen thickness of 0.20 mm and Derma-K™

produced a mean collagen thickness of 0.21 mm.These were intermediate in thickness. (3) Phenoland Er:YAG had the least average thickness ofneocollagen with micrometer measurements of0.15 mm and 0.08 mm, respectively.

Evaluation of the density of neocollagen,measured on a scale of 0–5, showed the following:(1) greatest neocollagen density in CO2 and CO2/Er:YAG at 3.4 and 3.6, respectively; (2) blendedlasers and phenol were intermediate in density at2.7 and 2.8 on this scale; and (3) Er:YAG was thelowest in collagen density at 2.2; however, defini-tive collagen formation was observed (Fig. 3A).

CO2 was found to produce more neocollagendensity by a statistically significant margin (P <0.05) on Student’s t-test (n 4 10). Similarly, allCO2-containing treatments including Derma-K™

(n 4 20) produced statistically significant in-creases in neocollagen density when comparedwith Er:YAG alone (P 4 0.001). Statistical analy-sis of changes in neocollagen thickness approachbut do not attain statistical significance with thesmall sample size for differences between CO2and Er:YAG (P 4 0.068) (Table 2).

Elastosis was found to be decreased in alltreated specimens (Fig. 4A,B). When comparedwith control by using the five-point scale, CO2 in-duced a one-point decrease in elastosis, compared

with 0.5-point decreases by Derma-K™ and phe-nol, 0.25 by Er:YAG, and 0.15 by CO2/Er:YAG(Fig. 4A). This finding corresponded with a 27%decrease in elastosis with CO2, 14% with Derma-K™ and phenol, 7% with Er:YAG, and 4% withCO2/Er:YAG (Fig. 4B). All treatments produced astatistically significant decrease in elastosis fromcontrol on t-test (P < 0.001, n 4 30). CO2 producedthe greatest change from control (27%) with P <0.05.

Examination of inflammation, melanocyticatypia, and keratocytic atypia in treated speci-mens (n 4 25) found pure CO2 to have the great-est effect on keratocytic atypia, Derma-K™ tohave the greatest effect on melanocytic atypia,and Derma-K™ to have the least inflammation(Fig. 5).

Overall, examination of histologic samplesdemonstrated differences among the varioustreatments. CO2 treatment resulted in long-termdevelopment of the greatest increase in collagen-esis and decrease in solar elastosis (Fig. 2B). CO2/Er:YAG and Derma-K™ seemed to be intermedi-ate in these effects (Fig. 2C,D). Phenol treatmentproduced similar neocollagen density and elasto-sis treatment, but a thinner band of neocollagen(Fig. 2E). Er:YAG alone produced the least visiblelong-term improvement in collagenesis (Fig. 2F).

Clinical Correlation: Observation of Healing Sitesand Treated Areas

All test sites were observed over a 4- to6-month period in nine patients. In all patients,the Er:YAG site was fully healed and without er-ythema by 1 week. The CO2 site was mostly epi-thelialized. However, CO2-treated sites were sig-nificantly erythematous with swelling anddesquamation at 1 week, and with residual ery-thema at 1 month, which did not completely re-solve until 4 months had passed. Combined CO2/Er:YAG regimens including the Derma-K™ andCO2 followed by Er:YAG were intermediate incharacter, with full re-epithelialization by 1 week,and complete resolution of erythema within 1month in all cases. Phenol initially becameblanched, then developed a waxy erythema dur-ing the first week. This condition resolved by 11month, but a waxy texture persisted through 3months. At 6 months, the CO2-treated sites werethe smoothest and, grossly, the most improved.Erbium had little long-term change from un-treated skin. Grossly, blended lasers, combinedEr:YAG and CO2, and phenol, had intermediate

366 Greene et al.

Figure 2.

In Vivo Model of Histologic Changes 367

amounts of tightening and improvement andcould be distinguished from untreated skin.

Clinical Correlations: Observations in 100 Cases

One hundred patients participated in fourtreatment groups as follows: CO2, 25 patients;Er:YAG, 25 patients; CO2/Er:YAG, 25 patients;blended lasers, 25 patients. These patients servedas the pool from which, as depicted in Figure 1, 19patients were selected for the test site protocol.The entire pool of 100 patients served as histori-cal controls for evaluation of the clinical effects ofthese treatments. Note that our program nolonger uses phenol as a peeling agent since 1994,

when we started by using laser for deeper resur-facing. Phenol is, thus, not represented in thiscontrol group.

These patients were observed for healing, re-epithelialization, and resolution of erythema. Inaddition, long-term satisfaction was measured bya survey at 4–6 months. Visual grading at 7 daysfound complete re-epithelialization in 100% of thepatients treated with Er:YAG alone and CO2/Er:YAG [26]. Treatment with blended lasers re-sulted in complete re-epithelialization at 7 days in90% of the patients studied, and CO2 in only 70%,with a mean time of 8 days for re-epithelialization[26].

Fig. 2. A: Photomicrographs of control skin specimen demon-strating marked dermal elastosis and mild-moderate melano-cyte atypia and hypertrophy representative of the 5 patientssampled. Original magnification, ×400. B: Photomicrographof CO2-treated skin 6 months after laser treatment. The de-gree of solar elastosis is substantially improved, and a 0.27-mm band of neocollagen is visible. Melanocytic and keratino-cytic atypia are likewise improved. Original magnification,×400. C: Photomicrograph of skin treated with CO2 followedby Er:YAG. A band of neocollagen has developed with a meanof 0.21 mm. Moderate improvement in elastosis is seen. Thesefindings are intermediate between CO2 and Er:YAG alone.Original magnification, ×400. D: Photomicrograph of skintreated with Derma-K™ laser (combined CO2 and Er:YAG). A

band of neocollagen has developed with a mean of 0.20 mm.Moderate improvement in elastosis is seen. These findingsare intermediate between CO2 and Er:YAG alone. Originalmagnification, ×400. E: Photomicrograph of skin treated withphenol 53%. A limited band of neocollagen has developed witha mean of 0.15 mm. Moderate improvement in elastosis isseen. These findings are intermediate between CO2 andEr:YAG alone. Original magnification, ×400. F: Photomicro-graph of representative specimen of skin treated withEr:YAG. A small but identifiable band of neocollagen has de-veloped with a mean of 0.08 mm is present. Minimal improve-ment in elastosis is seen. These changes are the smallest ofthe tested resurfacing modalities.

368 Greene et al.

At 1 month, none of the patients (0%) treatedwith Er:YAG or CO2 followed by Er:YAG had anyerythema [26]. At 1 month, erythema was stillpresent in 30% of the patients treated withblended lasers and 60% of the patients treatedwith CO2. By 4–6 months, all patients were fullyhealed. The most dramatic improvement in elas-ticity and rhytid resolution was found in the CO2cohort. This effect was quantitatively measuredat 6 months with a Cutometer, which found amean increase in skin elasticity of 18.2% [27]. Vi-sual assessment by the treating facial plastic sur-geon (R.J.K.) found an 80% improvement in theCO2 cohort, 75% improvement with CO2/Er:YAG,60% with blended lasers, and 45% with Er:YAG[26].

Patient satisfaction was evaluated at 6months by questionnaire response to, “Were yousatisfied with the results of your resurfacing pro-cedure” [26]. Ninety percent of the CO2 patients,

85% of the CO2/Er:YAG, 80% of the blended la-sers, and 45% of the Er:YAG were fully satisfiedwith the effects of their resurfacing treatment.

DISCUSSION

The present study uses an in vivo preauricu-lar human skin model to evaluate the effects ofthe different resurfacing modalities. Long-termhistologic evaluation demonstrates significantdifferences between the histologic effects of eachof these treatment modalities, which are corrobo-rated by our clinical observations of 45 test sitesin nine patients and long-term observation of 100full-face resurfacing patients.

Our observation of 100 resurfacing patientsfound that carbon dioxide treatment clearly hasthe longest recovery period but the most dramaticlong-term results. At the other end of the spec-trum, Er:YAG has the briefest recovery period,but the most limited long-term benefits. Com-bined regimens including CO2 followed byEr:YAG and blended laser, which combines thetwo modalities, are intermediate in effect. Combi-nation therapy produces more marked improve-ments in skin elasticity and rhytid ablation thanEr:YAG, but significantly more rapid recoverythan found with CO2 used alone.

Examination of 5 test spots in each of 9 pa-tients at regular intervals over 4–6 months con-trolled for differences between patients. Thesefindings correspond closely to those found in the100 historical control patients treated with full-face resurfacing. These data were corroborated byour long-term histology obtained by histologicanalysis of 30 specimens from five patients at 4–6months. The CO2 laser produced the most dra-matic increase in neocollagen density and thick-ness, Er:YAG the least, with combined CO2/Er:YAG and blended lasers intermediate ineffects.

A similar pattern was evident in changes insolar elastosis. CO2 produced the most dramaticand statistically significant improvement fromcontrol with a 27% decrease in elastosis, com-pared with 14% with blended lasers and phenol.Er:YAG and CO2 followed by Er:YAG were lessimpressive, with 7% or less elastosis improve-ment.

It is interesting to note that blended lasersexceeded the effects of CO2/ErYAG, despite thatboth treatments combine CO2 and erbium. Thisphenomenon suggests that the CO2 component ismostly responsible for removal of elastosis. By us-

Fig. 3. A: Neocollagen thickness (in millimeters). B: Neocol-lagen density at 4–6 months measured on a visual analogscale (0–5).

In Vivo Model of Histologic Changes 369

ing Er:YAG to remove the thermal necrosis zonecaused by CO2 may lead to improved healing, butit may also attenuate the beneficial effects of theCO2 laser on elastosis. Further study will beneeded to answer this question.

Comparison of our long-term (4- to 6-month)data to our 7-day data [14] demonstrated signifi-cant changes over time. The short-term (1–7 days)histology reflects the direct effects of the laser ab-lation, thermal necrosis, and acute inflammation.

The 4- to 6-month study examines the skin’s heal-ing response to laser treatment. Histology duringthe acute injury period (1 hour) found that speci-mens treated with CO2 laser as the sole or lasttreatment (i.e., after erbium) had higher levels ofinjury than specimens treated with Er:YAG laseras the sole or final treatment (i.e., after CO2).These differences were statistically significant (P< 0.05) [14]. By 7 days (i.e., the subacute period),no difference between laser treatment was seen.By 4–6 months, however, differences were againseen between each of the treatments. This findingreflects long-term tissue response to the lasertreatment, which are not present at 7 days. Thesefindings suggest that a long-term in vivo model isnecessary to determine the histologic differencesbetween different resurfacing treatments. In ad-dition, the characteristics of the tissue responseand changes in neocollagen over time at 7 daysand 6 months are consistent with our clinical ob-servations.

These findings also support the thesis thatlong-term in vivo testing of lasers should be thegold standard for laser testing. To date, the vastmajority of these studies have been done on skinthat has already been removed from the body,that is, in the ex vivo model [9,13]. More recentlypublished reports that used in vivo models (hu-man and animal) seem to provide more accuratedata on laser effects [2,6–8]. Previously, we have

TABLE 2. Clinical Examination Findings for 45 Test Spots Over Time

Treatment 1 week 1 month 4–6 months

CO2 Erythema Erythema Slightly tighter with maximal removal of rhytidsCO2/Er:YAG Minimal erythema Barely visible Smooth with moderate removal of rhytidsEr:YAG Barley visible Normal Unchanged from controlBlended CO2/Er:YAG Minimal erythema Barely visible Smooth with moderate removal of rhytidsPhenol Waxy Waxy Smooth with moderate removal of rhytids

Fig. 4. A: Decrease in elastosis at 4–6 months as change fromcontrol measured on a visual analog scale (0–5). B: Post-resurfacing residual elastosis as percentage of control.

Fig. 5. Changes in inflammation, melanocyte atypia, andkaratocytic atypia compared among skin treatments.

370 Greene et al.

demonstrated that ex vivo laser treatment pro-duces significantly different histologic findingswhen compared with in vivo lasered skin from thesame patient [28].

It is not surprising that the pulsed CO2 laserdemonstrated so many beneficial effects at the 4-to 6-month time point. It was, however, the pro-longed erythema and healing time caused by theCO2 laser that prompted many clinicians to incor-porate Er:YAG into their resurfacing practices.Considering the present data, and patient con-cerns over erythema and healing, combined CO2/Er:YAG modalities (sequential and Derma-K)seem to be intermediate in these extremes, and anoptimal compromise between degree of histologicimprovement, and rapidity of patient recovery.

CONCLUSION

Collagenesis is greatest with CO2, and leastwith Er:YAG. (P < 0.05). Elastosis decreases tothe greatest degree with CO2, least with erbium,and to an intermediate extent with blended CO2/Er:YAG regimens (sequential and Derma-K).These changes from control are statistically sig-nificant with all regimens (P < 0.05). BlendedCO2/Er:YAG treatments provide an optimal com-bination of the benefits of CO2 but with lessererythema and healing delay. Clinical and histo-logic findings change over time for different treat-ments. Thus, long-term histology is critical forpredicting results of treatment. The preauricularin vivo model has proven to be a valuable methodfor comparing laser and chemical resurfacing mo-dalities.

REFERENCES

1. Stuzin JM, Baker TJ, Baker TM, Kligman AM. Histologiceffects of the high-energy pulsed CO2 laser on photoagedfacial skin. Plast Reconstr Surg 1997;99:2036–2050.

2. Cotton J, Hood AF, Gonin R, Beeson WH, Hanke W. His-tologic evaluation of preauricular and postauricular skinafter high-energy short-pulsed carbon dioxide laser. ArchDermatol 1996;132:425–428.

3. Rubach BW, Schoenrock LD. Histologic and clinicalevaluation of facial resurfacing using a carbon dioxidelaser with the computer pattern generator. Arch Oto-laryngol Head Neck Surg 1997;123:929–934.

4. Kauvar ANB, Grossman MC, Bernstein LJ, Kovacs SO,Wuintana AT, Geronemus RG. Comparison of tissue ef-fects of carbon dioxide, erbium:YAG, and Novel infraredlasers for skin resurfacing. Presented at the 18th AnnualMeeting of the American Society for Laser Medicine andSurgery, April 6, 1998.

5. Kauvar ANB, Waldorf HA, Geronemus RG. A histopath-ological comparison of “char-free” carbon dioxide lasers.Dermatol Surg 1996;22:343–348.

6. Yang C, Chai C. Animal study of skin resurfacing usingthe ultrapulse carbon dioxide laser. Ann Plast Surg 1995;35:154–158.

7. Smith KJ, Skelton HG, Graham JS, Hamilton TA, Hack-ley BE Jr, Hurst CG. Depth of morphologic skin damageand viability after one, two, and three passes of a high-energy, short-pulse CO2 laser (Tru-Pulse) in pig skin. JAm Acad Dermatol 1997;37:204–210.

8. Fitzpatrick RE, Tope WD, Goldman MP, Satur NM.Pulsed carbon dioxide laser, trichloroacetic acid, Baker-Gordon phenol, and dermabrasion: a comparative clinicaland histologic study of cutaneous resurfacing in a porcinemodel. Arch Dermatol 1996;132:469–471.

9. Utley D, Koch RJ, Egbert BM. Histological analysis of thethermal effect on epidermal and dermal structures fol-lowing treatment with the superpulsed CO2 laser and theerbium:YAG laser. Lasers Surg Med 1999;24:93–102.

10. Weinstein C. Carbon dioxide laser resurfacing: long-termfollow-up in 2123 patients. Clin Plast Surg 1998;25:109–130.

11. Gardner ES, Reinisch L, Stricklin GP, Ellis DL. In vitrochanges in non-facial human skin following CO2 laserresurfacing: a comparison study. Lasers Surg Med 1996;19:379–387

12. Fulton JE, Barnes T. Collagen shrinkage (selective der-maplasty) with the high-energy pulsed carbon dioxide la-ser. Dermatol Surg 1998;24:37–41.

13. Rubach BW, Schoenrock LD. Histologic and clinicalevaluation of facial resurfacing using a carbon dioxidelaser with the computer pattern generator. Arch Oto-laryngol Head Neck Surg 1997;123:929–934.

14. Alster TS, Garg S. Treatment of facial rhytides with ahigh energy pulsed carbon dioxide laser. Plast ReconstrSurg 1996;98:791–794.

15. Kuo T, Speyer MT, Ries WR, Reinisch L. Collagen ther-mal damage and collagen synthesis after cutaneous laserresurfacing. Lasers Surg Med 1998;23:66–71.

16. Goldman MP, Manuskiatte W. Combined laser resurfac-ing with the 950-microsecond pulse CO2 + Er:YAG lasers.Dermatol Surg 1999;25:160–163.

17. Trelles MA, David LM, Rigau J. Penetration depth ofUltrapulse carbon dioxide laser in human skin. DermatolSurg 1996;22:863–865.

18. Trelles MA, Mordon S, Svaasand LO, Mellor TK, RiugauJ, Garcia L. The origin and role of erythema after carbondioxide laser resurfacing: a clinical and histologicalstudy. Dermatol Surg 1998;24:25–29.

19. Ruiz-Esparza J, Barba-Gomez JM, Gomez de la TorreOL, David L. Erythema after laser skin resurfacing. Der-matol Surg 1998;24:113–117.

20. Teikemeier G, Goldberg DJ. Skin resurfacing with theerbium:YAG laser. Dermatol Surg 1997;23:685–687.

21. Kaufman R, Hibst R. Pulsed erbium:YAG laser ablationin cutaneous surgery. Lasers Surg Med 1996;19:324–330.

22. Hohenleutner U, Hohenleutner S, Baumler W, Landtha-ler M. Fast and effective skin ablation with an Er:YAGlaser: determination of ablation rates and thermal dam-age zones. Lasers Surg Med 1997;20:242–247.

23. Goldman MP, Fitzpatrick RE, Tse Y, Manuskiatte W.Combined laser resurfacing with the Ultra Pulse CO2laser followed by the erbium:YAG laser. Presented at the18th Annual Meeting of the American Society for LaserMedicine and Surgery, April 5, 1998.

24. Minoli JJ, Barton FE: A comparison of the histologic ef-

In Vivo Model of Histologic Changes 371

fects of chemabrasion, dermabrasion, and laserbrasion inthe minipig. Aesthetic Surg J 1998;12:11–18.

25. Fitzpatrick TB. The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 1988;124:869–871.

26. Koch RJ. Laser skin resurfacing with the Erbium-YAGlaser. Facial Aesthetic Comm Eur (FACE) (in press).

27. Koch RJ. Initial clinical evaluation of the NovaScan car-

bon dioxide laser handpiece. Dermatol Surg 1998;24:369–371.

28. Greene D, Koch RJ, Utley D, Egbert B: The validity ofex-vivo laser treatment for histological analysis. Arch Fa-cial Plast Surg 1999;1;159–164.

29. Koch RJ, Cheng E. Quantification of skin elasticitychanges associated with laser skin resurfacing. Arch Fa-cial Plast Surg 1999;1:272–275.

372 Greene et al.