Pain in Naïve and Non-naïve Subjects Undergoing Nonablative Skin Tightening Dermatologic Procedures: A Nested Randomized Control Trial ROHIT KAKAR, MD,* OMER IBRAHIM, MD,† WAREEPORN DISPHANURAT, MD,‡ NATALIE PACE, BS,§ DENNIS P. WEST, PHD,§ MARY KWASNY, SCD,¶ SONIA SHAH, BA,§ AND MURAD ALAM, MD, MSCI§**††
BACKGROUND Pain is expected during noninvasive skin tightening and can be anxiety provoking, especially for those who have not had prior treatments. OBJECTIVE To compare pain reported by patients naı¨ve to nonablative skin tightening energy devices with those who were not naive. METHODS AND MATERIALS The non-naı¨ve group at least three nonablative laser procedures or one nonablative skin tightening procedure, and the naı¨ve group no previous treatments. Four sites at each of two anatomic locations (periorbital and midface or cheek) were treated in each subject with needle prick, pulsed dye laser, radiofrequency, and ultrasound with the order of the interventions randomized. All interventions except ultrasound were also applied to three abdominal sites. The difference in mean pain scores between naı¨ve and nonnaı¨ve subjects were averaged over the anatomic sites. RESULTS Ten naı¨ve and 10 non-naı¨ve subjects completed study procedures. Mean pain scores ranged from 1.3 to 4.9. The mean for all naı¨ve conditions was 2.3 1.0, vs 2.2 1.4 for non-naı¨ve conditions. There was no overall difference according to group, device, or anatomic area. CONCLUSIONS There was no significant difference in pain between naı¨ve and non-naı¨ve subjects undergoing cutaneous energy treatments. Individual devices may elicit more pain at specific anatomic locations. The authors have indicated no significant interest with commercial supporters.
N
onablative, noninvasive, energy-based skin tightening offers the promise of mitigation of visible skin loosening and sagging without prolonged recovery, pain, scarring, or risk of serious adverse events. Current methods for such tightening include laser, light, radiofrequency, and ultrasound modalities, as well as combinations of these.1 Noninvasive skin tightening also has limitations, namely less effectiveness than excisional techniques such as face lifting, as well as some intratreatment
pain and discomfort. Pain is expected, because device-based skin tightening works, in part, through thermal contraction of the dermis and subcutis. Although the pain is limited, is not associated with sequelae, and remits after treatment cessation, it can be anxiety provoking for patients. Additionally, anecdotal reports have suggested that patients who have not had prior laser or energy-device treatments may be particularly prone to experience higher levels of pain. At least one published study mentions this propensity.2
*Washington Hospital Center, Georgetown University, Washington, District of Columbia; †Department of Dermatology, Cleveland Clinic Foundation, Cleveland, Ohio; ‡Dermatology Unit, Department of Medicine, Thammasat University, Patumthani, Thailand; §Department of Dermatology, Northwestern University, Chicago, Illinois; ¶ Department of Preventive Medicine, Northwestern University, Chicago, Illinois; **Department of Otolaryngology— Head and Neck Surgery, Northwestern University, Chicago, Illinois; ††Department of Surgery, Northwestern University, Chicago, Illinois © 2014 by the American Society for Dermatologic Surgery, Inc. Published by Wiley Periodicals, Inc. ISSN: 1076-0512 Dermatol Surg 2014;40:398–404 DOI: 10.1111/dsu.12440 398
KAKAR ET AL
The purpose of this study was to compare the degree of intraoperative treatment pain reported by patients naı¨ve to selected nonablative lasers and skin tightening energy devices with pain reported by patients who had previously experienced laser or skin tightening procedures.
Methods Setting and Study Design This was a randomized controlled trial (within each subject, allocation ratio of 4:4:3 to different devices at different anatomic sites) nested within a prospective cross-sectional cohort study of equal numbers of naı¨ve and non-naı¨ve laser and skin tightening subjects in a university-affiliated, urban dermatology department in Chicago, Illinois. The Northwestern University Institutional Review Board approved this study, which was posted on clinicaltrials.gov (NCT01285947) before start of data collection. Study funding was solely from the Northwestern University Department of Dermatology. Patient Selection Equal numbers of laser naı¨ve- and non-naı¨ve subjects were enrolled through recruitment in a dermatology clinic and by recruitment materials placed in the clinic and in nearby neighborhood sites. “Non-naı¨ve” subjects were defined as those who had undergone at least three nonablative laser procedures or at least one skin tightening procedure using an energy device (laser, light, radiofrequency, ultrasound, or some combination thereof). Naı¨ve subjects were defined as all subjects who were not “nonnaı¨ve”. Other inclusion criteria were age 18 to 65 and general good health. Exclusion criteria were history of photophobia and active skin disease or skin infection that, in the opinion of the investigator, would interfere with the study procedures. Equipment and Devices Equipment consisted of 30-G half-inch needles (30G1/2 PrecisionGlide, Becton Dickinson, Franklin
Lakes, NJ), a long-pulse 595-nm pulsed dye laser (Vbeam, Candela/Syneron, Wayland, MA), a radiofrequency skin tightening device (Thermage Thermacool, Thermage/Solta, Hayward, CA), and an ultrasound skin tightening device (Ultherapy system, Ulthera, Mesa, AZ). Randomization Protocol For each subject, symmetric areas of facial anatomic locations were identified at the periorbital and midface or cheek (Figure 1A). Areas of similar size were also identified on the bilateral sides of the abdomen (Figure 1B). In each subject, for each of the three anatomic locations (periorbital, cheek, abdomen), four distinct nonoverlapping 3- by 2-cm sites were marked and assigned values of 1, 2, 3, and 4 from left top to right bottom. Because each anatomic study location in each subject was comprised of bilaterally symmetric areas, in each case, two of the marked sites were on the left side of the body and two on the right. Using random number tables, the order of the four interventions (see Equipment and Devices) in each of the two facial locations and of the three interventions (all except the ultrasound device) in the abdominal location was randomized for each subject. Each of 20 such randomizations of two blocks of four numbers plus one block of three numbers was sealed in a separate opaque envelope and shuffled. Envelopes were then consecutively unsealed for each subject and used to assign interventions to each of the 11 treatment sites. One of the authors enrolled participants and defined and marked areas and sites (RK), another generated the random allocation sequences (OI), and a third assigned participants to interventions (NP). Blinding Although participants were not told what type of device or treatment they were receiving at each site, strict blinding was not achievable because of the need to position devices, differences in the feel of
40:4:APRIL 2014
399
PAIN IN NONABLATIVE SKIN TIGHTENING DERMATOLOGIC PROCEDURES
(A)
(B)
Figure 1. Treatment locations on face (A) and abdomen (B). In each case, right and left treatment areas were separately marked and divided into two portions each, for a total of four sites associated with each anatomic location: periorbital (left 1,2), (right 3,4); midface and cheek (left 5,6), (right 7,8); abdomen (left 9, 10), (right 11, 12). For the abdomen, the site randomized to ultrasound treatment was not treated.
device handpieces, and different sensations upon treatment. For similar reasons, care providers could not be blinded. Because the only outcome measure was patient pain ratings, there were no other raters. The data analyst (MK) was blinded. Study Procedures At the screening visit, subjects who qualified and provided written informed consent were asked to refrain from applying topical products such as makeup and lotion to their face and abdomen on the day of their participation and to use sunscreen and avoid significant sun exposure for 2 weeks before and after the experimental interventions. A brief questionnaire was administered to elicit demographic information and medical history. After randomization and marking of anatomic areas and sites (see Randomization Protocol), 11 experimental interventions were delivered in one study visit. Needle prick, pulsed dye laser, radiofrequency, and ultrasound were individually applied to distinct sites, according to the randomization procedure, at the periorbital and midface/cheek locations of each subject; all interventions except ultrasound were
400
DERMATOLOGIC SURGERY
also applied in a similar manner to distinct abdominal sites. For the energy devices, three pulses were used at each site. Subjects were asked to score the degree of pain they experienced with each of these experimental interventions from 0 (no pain) to 10 (maximal pain) on a visual analog scale (VAS). Device Settings Needle: Intradermal prick sustained for 3 s with constant pressure to reach the subcutaneous layer. Pulsed dye laser: 7.5 J/cm2 fluence, 6-ms pulse duration, 10-mm spot size, dynamic cooling device (DCD) 30/20. Radiofrequency skin tightening device: periorbital, 58 to 140 J/cm2; midface, 81 to 124 J/cm2; abdomen, 95 to 181 J/cm2 (specific setting in each area contingent on skin thickness). Ultrasound skin tightening device: transducer treatment frequency, 7 MHz; treatment depth, 3.0 mm; image depth, 0–8 mm.
KAKAR ET AL
Outcome Measures The primary outcome measure was pain as reported by subjects. This measure was elicited for each of the 11 experimental conditions for each subject. Statistical Analysis Because the primary objective was to determine whether pain inflicted using four methods varied according to procedure naivete, the statistical methods included mixed models with fixed effects of site and procedure naivete, accounting for the correlated measurements from each subject for each of the four insults, not adjusting for multiple comparisons because of the small sample sizes. When the models failed to converge, naivete groups averaged over site were compared for each of the four insults using twosided Mann-Whitney tests. Because preliminary data were not available for a power calculation using a mixed model, using a Mann-Whitney test, with a sample size of 20 would have had 80% power to detect an effect size of 1.4 between the groups. Because mixed models are more powerful, it is expected to be able to detect smaller effect sizes. Data analysis was completed on an intention-to-treat basis.
Results Twenty subjects, 10 laser naı¨ve and 10 laser nonnaı¨ve, were enrolled and treated in this crosssectional study with no follow-up period. Dates of enrollment and treatment were from March to May 2011. All completed the study procedures. For each subject, 11 sites were randomized and treated using one of four interventions for a total of 220 randomized and treated sites. All treated sites were analyzed for the primary outcome. Overall, mean pain scores for various interventions at various anatomic areas ranged from 1.3 to 4.9 (Table 1). The mean was 2.3 1.0 for all naı¨ve conditions and 2.2 1.4 for all non-naı¨ve conditions. In general, exploratory analyses showed that there was no appreciable difference between the groups
TABLE 1. Pain Ratings for Na€ıve and Non-na€ıve Subjects Na€ıve Site and Device Periorbital Needle Pulsed dye laser Radiofrequency Ultrasound Cheek Needle Pulsed dye laser Radiofrequency Ultrasound Abdomen Needle Pulsed dye laser Radiofrequency Average Needle Pulsed dye laser Radiofrequency Ultrasound Overall average
Mean Standard Deviation
Nonna€ıve
T-Test p-Value
1.7 1.5 3.2 2.0
2.7 2.2 1.5 1.2
.24 .04
0.9 2.0 2.5 2.0
1.3 2.2 3.7 3.1
.67 .09
1.4 1.3 2.4 1.6
1.7 1.5 1.7 1.6
.64 .34
0.4 0.8 4.0 2.4
0.9 1.1 3.7 2.9
.27 .80
0.8 1.2 2.1 2.3
1.5 1.3 1.2 1.5
.23 .32
2.9 2.4
4.2 3.4
.34
1.3 0.8 2.6 1.7
2.0 1.3 1.5 1.0
.19 .10
1.4 1.2 4.9 2.0 2.3 1.0
2.1 1.8 3.7 2.6 2.2 1.4
.32 .26 .81
overall according to device or at any anatomic area according to device (Figure 2). The only exception was that the pulsed dye laser was significantly more painful for naı¨ve subjects (3.2 2.0) than nonnaı¨ve subjects (1.5 1.2) (p = .04). However, even this statistically significant difference may not have been clinically significant, as there was only a modest absolute difference in pain scores. No unexpected harms or unintended effects were detected in any of the experimental arms.
Discussion This study was designed to compare the pain that naı¨ve and non-naı¨ve subjects undergoing cutaneous laser and energy device treatments experienced, in particular noninvasive skin tightening procedures. No significant difference was detected. In some
40:4:APRIL 2014
401
PAIN IN NONABLATIVE SKIN TIGHTENING DERMATOLOGIC PROCEDURES
Figure 2. Mean pain scores associated with different interventions and average scores for all conditions (error bars represent 95% confidence intervals).
cases, naı¨ve subjects reported nominally more pain but only in one specific condition of 11 (pulsed dye laser in the periorbital area) was this statistically significant, and overall mean pain for all naı¨ve subjects was not significantly more than that of nonnaı¨ve subjects. This study was not designed to compare pain with different types of energy devices. To perform such a comparison, we would first have had to conduct effectiveness studies to ensure that the settings used with each energy device (and especially the two skin tightening devices) were comparable in terms of the degree of beneficial skin change obtained. Only then could the relevant equivalent settings have been compared as to degree of pain. We had no data regarding therapeutic equivalence of the settings we used, and no side-by-side comparison data were available from the literature. Therefore, it is possible that more-painful treatments were merely those associated with greater effectiveness. The needle prick intervention was included to provide a clinical benchmark. Patients and investigators understand the degree of pain associated with such a prick well, so it is a useful metric by
402
DERMATOLOGIC SURGERY
which to interpret the pain elicited in the laser and energy device arms. Overall, we found that mean pain scores for any of the specific conditions (a given device at a given anatomic area) never exceeded 5, and overall mean scores were comparable with those associated with a needle prick. When pain is associated with laser and energy devices, it appears sufficiently modest to be easily managed using topical and oral analgesics. This study has several limitations. For each intervention, we used only three pulses of the relevant treatment. As such, we did not capture any increase in pain that would have been associated with prolonged treatment or gradually increasing subject fatigue. From anecdotal clinical experience, patients may experience significantly more pain when many hundreds of pulses are delivered continuously to one site. Therefore, our experimental design limited external validity, with the levels of pain that we elicited probably being less than pain that might have occurred if complete routine clinical treatments were delivered, but our methodology was necessary to ensure internal validity of the data by maintaining strict comparability between the naı¨ve and nonnaive treatment arms and to avoid excessively fatiguing
KAKAR ET AL
the subjects, who would probably not have been able to tolerate 11 separate complete treatments or to differentiate the pain associated with such prolonged treatments. Second, we did not use the ultrasound device on the abdomen because, at the time of this study, the Food and Drug Administration had not approved it for indications off the face, but based on unpublished cohort data involving thousands of patients, ultrasound skin tightening appears safe and effective off the face. Because this was a prospective intervention study, consent was obtained from participants. Although this can be considered potentially biasing for patients, prospective research is more powerful than retrospective studies, and conducting a human intervention study without eliciting informed consent is not ethical. The consent form did not clarify what we expected to find regarding pain levels in different conditions or what the pain levels of others had been, so it is unlikely that patients would have had a systematic bias in favor of one of the four arms or one of the three anatomic locations. There is a dearth of studies comparing the pain and discomfort associated with different cutaneous procedures. Most pain studies in dermatology compare the use of a topical or other anesthetic agent with placebo.3–5 Although these studies are useful, it is not surprising that most active agents are found to be superior to no pain management at all. Prior comparative studies examining two intervention arms, rather than one intervention arm and placebo, include investigations of electrodesiccation for dermatosis papulosa nigra,6 injectable botulinum toxin for facial rhytides,7–9 tumescent anesthesia for liposuction,10 different injectable anesthetics for Mohs surgery,11 and varying wavelengths and fluences for activating 5-aminolevulinic acid during photodynamic therapy.12–14 In the laser and energy device realm, there is an acute dearth of such studies. Although serious adverse events are unlikely with cutaneous laser and energy devices, we need more work in this area to better describe the minor
inconveniences, such as pain, that may guide patient or device selection. The current study is an early step in better understanding the character of intraoperative pain associated with nonablative skin tightening devices that have few visible sequelae. That the pinprick intervention elicited similar pain results at the facial sites and slightly less pain on the abdomen is consistent with expectations and suggests that experimental pain levels were correctly measured. The significant variation between patients indicates that interpatient differences are an important element that cannot be discounted, regardless of whether patients are device naı¨ve or not. In this controlled, comparative study, we had good internal validity that suggests comparability of our results; additional, more-naturalistic studies that may examine more-extensive treatment may be further able to confirm the external generalizability of our results.
References 1. Bogle MA, Dover JS. Tissue tightening technologies. Dermatol Clin 2009;27:491–9. 2. Alam M, White LE, Martin N, Witherspoon J, et al. Ultrasound tightening of facial and neck skin: a rater-blinded prospective cohort study. J Am Acad Dermatol 2010;62:262–9. 3. Monheit GD, Campbell RM, Neugent H, Nelson CP, et al. Reduced pain with use of proprietary hyaluronic acid with lidocaine for correction of nasolabial folds: a patient-blinded, prospective, randomized controlled trial. Dermatol Surg 2010;36:94–101. 4. Shaikh FM, Naqvi SA, Grace PA. The influence of a eutectic mixture of lidocaine and prilocaine on minor surgical procedures: a randomized controlled double-blind trial. Dermatol Surg 2009;35:948–51. 5. Levy PM, De Boulle K, Raspaldo H. Comparison of injection comfort of a new category of cohesive hyaluronic acid filler with preincorporated lidocaine and a hyaluronic acid filler alone. Dermatol Surg 2009;35(Suppl 1):332–7. 6. Carter EL, Coppola CA, Barsanti FA. A randomized, doubleblind comparison of two topical anesthetic formulations prior to electrodesiccation of dermatosis papulosa nigra. Dermatol Surg 2006;32:1–6. 7. Kranz G, Sycha T, Voller B, Gleiss A, et al. Pain sensation during intradermal injections of three different botulinum toxin preparations in different doses and dilutions. Dermatol Surg 2006;32:886–90.
40:4:APRIL 2014
403
PAIN IN NONABLATIVE SKIN TIGHTENING DERMATOLOGIC PROCEDURES
8. van Laborde S, Dover JS, Moore M, Stewart B, et al. Reduction in injection pain with botulinum toxin type B further diluted using saline with preservative: a double-blind, randomized controlled trial. J Am Acad Dermatol 2003;48:875–7. 9. Alam M, Dover JS, Arndt KA. Pain associated with injection of botulinum A exotoxin reconstituted using isotonic sodium chloride with and without preservative: a double290 blind, randomized controlled trial. Arch Dermatol 2002;138:510–4. 10. Yang CH, Hsu HC, Shen SC, Juan WH, et al. Warm and neutral tumescent anesthetic solutions are essential factors for a less painful injection. Dermatol Surg 2006;32:1119–23. 11. Morganroth PA, Gelfand JM, Jambusaria A, Margolis DJ, et al. A randomized, double-blind comparison of the total dose of 1.0% lidocaine with 1:100,000 epinephrine versus 0.5% lidocaine with 1:200,000 epinephrine required for effective local anesthesia during Mohs micrographic surgery for skin cancers. J Am Acad Dermatol 2009;60:444–52. 12. Babilas P, Knobler R, Hummel S, Gottschaller C, et al. Variable pulsed light is less painful than light-emitting diodes for topical
404
DERMATOLOGIC SURGERY
photodynamic therapy of actinic keratosis: a prospective randomized controlled trial. Br J Dermatol 2007;157:111–7. 13. Ericson MB, Sandberg C, Stenquist B, Gudmundson F, et al. Photodynamic therapy of actinic keratosis at varying fluence rates: assessment of photobleaching, pain and primary clinical outcome. Br J Dermatol 2004;151:1204–12. 14. Wiegell SR, Haedersdal M, Philipsen PA, Eriksen P, et al. Continuous activation of PpIX by daylight is as effective as and less painful than conventional photodynamic therapy for actinic keratoses; a randomized, controlled, single-blinded study. Br J Dermatol 2008;158:740–6.
Address correspondence and reprint requests to: Murad Alam, MD, MSCI, Department of Dermatology, Northwestern University, 676 North Saint Clair Street, Suite 1600, Chicago, Illinois 60611, or e-mail: [email protected]
BACKGROUND Pain is expected during noninvasive skin tightening and can be anxiety provoking, especially for those who have not had prior treatments. OBJECTIVE To compare pain reported by patients naı¨ve to nonablative skin tightening energy devices with those who were not naive. METHODS AND MATERIALS The non-naı¨ve group at least three nonablative laser procedures or one nonablative skin tightening procedure, and the naı¨ve group no previous treatments. Four sites at each of two anatomic locations (periorbital and midface or cheek) were treated in each subject with needle prick, pulsed dye laser, radiofrequency, and ultrasound with the order of the interventions randomized. All interventions except ultrasound were also applied to three abdominal sites. The difference in mean pain scores between naı¨ve and nonnaı¨ve subjects were averaged over the anatomic sites. RESULTS Ten naı¨ve and 10 non-naı¨ve subjects completed study procedures. Mean pain scores ranged from 1.3 to 4.9. The mean for all naı¨ve conditions was 2.3 1.0, vs 2.2 1.4 for non-naı¨ve conditions. There was no overall difference according to group, device, or anatomic area. CONCLUSIONS There was no significant difference in pain between naı¨ve and non-naı¨ve subjects undergoing cutaneous energy treatments. Individual devices may elicit more pain at specific anatomic locations. The authors have indicated no significant interest with commercial supporters.
N
onablative, noninvasive, energy-based skin tightening offers the promise of mitigation of visible skin loosening and sagging without prolonged recovery, pain, scarring, or risk of serious adverse events. Current methods for such tightening include laser, light, radiofrequency, and ultrasound modalities, as well as combinations of these.1 Noninvasive skin tightening also has limitations, namely less effectiveness than excisional techniques such as face lifting, as well as some intratreatment
pain and discomfort. Pain is expected, because device-based skin tightening works, in part, through thermal contraction of the dermis and subcutis. Although the pain is limited, is not associated with sequelae, and remits after treatment cessation, it can be anxiety provoking for patients. Additionally, anecdotal reports have suggested that patients who have not had prior laser or energy-device treatments may be particularly prone to experience higher levels of pain. At least one published study mentions this propensity.2
*Washington Hospital Center, Georgetown University, Washington, District of Columbia; †Department of Dermatology, Cleveland Clinic Foundation, Cleveland, Ohio; ‡Dermatology Unit, Department of Medicine, Thammasat University, Patumthani, Thailand; §Department of Dermatology, Northwestern University, Chicago, Illinois; ¶ Department of Preventive Medicine, Northwestern University, Chicago, Illinois; **Department of Otolaryngology— Head and Neck Surgery, Northwestern University, Chicago, Illinois; ††Department of Surgery, Northwestern University, Chicago, Illinois © 2014 by the American Society for Dermatologic Surgery, Inc. Published by Wiley Periodicals, Inc. ISSN: 1076-0512 Dermatol Surg 2014;40:398–404 DOI: 10.1111/dsu.12440 398
KAKAR ET AL
The purpose of this study was to compare the degree of intraoperative treatment pain reported by patients naı¨ve to selected nonablative lasers and skin tightening energy devices with pain reported by patients who had previously experienced laser or skin tightening procedures.
Methods Setting and Study Design This was a randomized controlled trial (within each subject, allocation ratio of 4:4:3 to different devices at different anatomic sites) nested within a prospective cross-sectional cohort study of equal numbers of naı¨ve and non-naı¨ve laser and skin tightening subjects in a university-affiliated, urban dermatology department in Chicago, Illinois. The Northwestern University Institutional Review Board approved this study, which was posted on clinicaltrials.gov (NCT01285947) before start of data collection. Study funding was solely from the Northwestern University Department of Dermatology. Patient Selection Equal numbers of laser naı¨ve- and non-naı¨ve subjects were enrolled through recruitment in a dermatology clinic and by recruitment materials placed in the clinic and in nearby neighborhood sites. “Non-naı¨ve” subjects were defined as those who had undergone at least three nonablative laser procedures or at least one skin tightening procedure using an energy device (laser, light, radiofrequency, ultrasound, or some combination thereof). Naı¨ve subjects were defined as all subjects who were not “nonnaı¨ve”. Other inclusion criteria were age 18 to 65 and general good health. Exclusion criteria were history of photophobia and active skin disease or skin infection that, in the opinion of the investigator, would interfere with the study procedures. Equipment and Devices Equipment consisted of 30-G half-inch needles (30G1/2 PrecisionGlide, Becton Dickinson, Franklin
Lakes, NJ), a long-pulse 595-nm pulsed dye laser (Vbeam, Candela/Syneron, Wayland, MA), a radiofrequency skin tightening device (Thermage Thermacool, Thermage/Solta, Hayward, CA), and an ultrasound skin tightening device (Ultherapy system, Ulthera, Mesa, AZ). Randomization Protocol For each subject, symmetric areas of facial anatomic locations were identified at the periorbital and midface or cheek (Figure 1A). Areas of similar size were also identified on the bilateral sides of the abdomen (Figure 1B). In each subject, for each of the three anatomic locations (periorbital, cheek, abdomen), four distinct nonoverlapping 3- by 2-cm sites were marked and assigned values of 1, 2, 3, and 4 from left top to right bottom. Because each anatomic study location in each subject was comprised of bilaterally symmetric areas, in each case, two of the marked sites were on the left side of the body and two on the right. Using random number tables, the order of the four interventions (see Equipment and Devices) in each of the two facial locations and of the three interventions (all except the ultrasound device) in the abdominal location was randomized for each subject. Each of 20 such randomizations of two blocks of four numbers plus one block of three numbers was sealed in a separate opaque envelope and shuffled. Envelopes were then consecutively unsealed for each subject and used to assign interventions to each of the 11 treatment sites. One of the authors enrolled participants and defined and marked areas and sites (RK), another generated the random allocation sequences (OI), and a third assigned participants to interventions (NP). Blinding Although participants were not told what type of device or treatment they were receiving at each site, strict blinding was not achievable because of the need to position devices, differences in the feel of
40:4:APRIL 2014
399
PAIN IN NONABLATIVE SKIN TIGHTENING DERMATOLOGIC PROCEDURES
(A)
(B)
Figure 1. Treatment locations on face (A) and abdomen (B). In each case, right and left treatment areas were separately marked and divided into two portions each, for a total of four sites associated with each anatomic location: periorbital (left 1,2), (right 3,4); midface and cheek (left 5,6), (right 7,8); abdomen (left 9, 10), (right 11, 12). For the abdomen, the site randomized to ultrasound treatment was not treated.
device handpieces, and different sensations upon treatment. For similar reasons, care providers could not be blinded. Because the only outcome measure was patient pain ratings, there were no other raters. The data analyst (MK) was blinded. Study Procedures At the screening visit, subjects who qualified and provided written informed consent were asked to refrain from applying topical products such as makeup and lotion to their face and abdomen on the day of their participation and to use sunscreen and avoid significant sun exposure for 2 weeks before and after the experimental interventions. A brief questionnaire was administered to elicit demographic information and medical history. After randomization and marking of anatomic areas and sites (see Randomization Protocol), 11 experimental interventions were delivered in one study visit. Needle prick, pulsed dye laser, radiofrequency, and ultrasound were individually applied to distinct sites, according to the randomization procedure, at the periorbital and midface/cheek locations of each subject; all interventions except ultrasound were
400
DERMATOLOGIC SURGERY
also applied in a similar manner to distinct abdominal sites. For the energy devices, three pulses were used at each site. Subjects were asked to score the degree of pain they experienced with each of these experimental interventions from 0 (no pain) to 10 (maximal pain) on a visual analog scale (VAS). Device Settings Needle: Intradermal prick sustained for 3 s with constant pressure to reach the subcutaneous layer. Pulsed dye laser: 7.5 J/cm2 fluence, 6-ms pulse duration, 10-mm spot size, dynamic cooling device (DCD) 30/20. Radiofrequency skin tightening device: periorbital, 58 to 140 J/cm2; midface, 81 to 124 J/cm2; abdomen, 95 to 181 J/cm2 (specific setting in each area contingent on skin thickness). Ultrasound skin tightening device: transducer treatment frequency, 7 MHz; treatment depth, 3.0 mm; image depth, 0–8 mm.
KAKAR ET AL
Outcome Measures The primary outcome measure was pain as reported by subjects. This measure was elicited for each of the 11 experimental conditions for each subject. Statistical Analysis Because the primary objective was to determine whether pain inflicted using four methods varied according to procedure naivete, the statistical methods included mixed models with fixed effects of site and procedure naivete, accounting for the correlated measurements from each subject for each of the four insults, not adjusting for multiple comparisons because of the small sample sizes. When the models failed to converge, naivete groups averaged over site were compared for each of the four insults using twosided Mann-Whitney tests. Because preliminary data were not available for a power calculation using a mixed model, using a Mann-Whitney test, with a sample size of 20 would have had 80% power to detect an effect size of 1.4 between the groups. Because mixed models are more powerful, it is expected to be able to detect smaller effect sizes. Data analysis was completed on an intention-to-treat basis.
Results Twenty subjects, 10 laser naı¨ve and 10 laser nonnaı¨ve, were enrolled and treated in this crosssectional study with no follow-up period. Dates of enrollment and treatment were from March to May 2011. All completed the study procedures. For each subject, 11 sites were randomized and treated using one of four interventions for a total of 220 randomized and treated sites. All treated sites were analyzed for the primary outcome. Overall, mean pain scores for various interventions at various anatomic areas ranged from 1.3 to 4.9 (Table 1). The mean was 2.3 1.0 for all naı¨ve conditions and 2.2 1.4 for all non-naı¨ve conditions. In general, exploratory analyses showed that there was no appreciable difference between the groups
TABLE 1. Pain Ratings for Na€ıve and Non-na€ıve Subjects Na€ıve Site and Device Periorbital Needle Pulsed dye laser Radiofrequency Ultrasound Cheek Needle Pulsed dye laser Radiofrequency Ultrasound Abdomen Needle Pulsed dye laser Radiofrequency Average Needle Pulsed dye laser Radiofrequency Ultrasound Overall average
Mean Standard Deviation
Nonna€ıve
T-Test p-Value
1.7 1.5 3.2 2.0
2.7 2.2 1.5 1.2
.24 .04
0.9 2.0 2.5 2.0
1.3 2.2 3.7 3.1
.67 .09
1.4 1.3 2.4 1.6
1.7 1.5 1.7 1.6
.64 .34
0.4 0.8 4.0 2.4
0.9 1.1 3.7 2.9
.27 .80
0.8 1.2 2.1 2.3
1.5 1.3 1.2 1.5
.23 .32
2.9 2.4
4.2 3.4
.34
1.3 0.8 2.6 1.7
2.0 1.3 1.5 1.0
.19 .10
1.4 1.2 4.9 2.0 2.3 1.0
2.1 1.8 3.7 2.6 2.2 1.4
.32 .26 .81
overall according to device or at any anatomic area according to device (Figure 2). The only exception was that the pulsed dye laser was significantly more painful for naı¨ve subjects (3.2 2.0) than nonnaı¨ve subjects (1.5 1.2) (p = .04). However, even this statistically significant difference may not have been clinically significant, as there was only a modest absolute difference in pain scores. No unexpected harms or unintended effects were detected in any of the experimental arms.
Discussion This study was designed to compare the pain that naı¨ve and non-naı¨ve subjects undergoing cutaneous laser and energy device treatments experienced, in particular noninvasive skin tightening procedures. No significant difference was detected. In some
40:4:APRIL 2014
401
PAIN IN NONABLATIVE SKIN TIGHTENING DERMATOLOGIC PROCEDURES
Figure 2. Mean pain scores associated with different interventions and average scores for all conditions (error bars represent 95% confidence intervals).
cases, naı¨ve subjects reported nominally more pain but only in one specific condition of 11 (pulsed dye laser in the periorbital area) was this statistically significant, and overall mean pain for all naı¨ve subjects was not significantly more than that of nonnaı¨ve subjects. This study was not designed to compare pain with different types of energy devices. To perform such a comparison, we would first have had to conduct effectiveness studies to ensure that the settings used with each energy device (and especially the two skin tightening devices) were comparable in terms of the degree of beneficial skin change obtained. Only then could the relevant equivalent settings have been compared as to degree of pain. We had no data regarding therapeutic equivalence of the settings we used, and no side-by-side comparison data were available from the literature. Therefore, it is possible that more-painful treatments were merely those associated with greater effectiveness. The needle prick intervention was included to provide a clinical benchmark. Patients and investigators understand the degree of pain associated with such a prick well, so it is a useful metric by
402
DERMATOLOGIC SURGERY
which to interpret the pain elicited in the laser and energy device arms. Overall, we found that mean pain scores for any of the specific conditions (a given device at a given anatomic area) never exceeded 5, and overall mean scores were comparable with those associated with a needle prick. When pain is associated with laser and energy devices, it appears sufficiently modest to be easily managed using topical and oral analgesics. This study has several limitations. For each intervention, we used only three pulses of the relevant treatment. As such, we did not capture any increase in pain that would have been associated with prolonged treatment or gradually increasing subject fatigue. From anecdotal clinical experience, patients may experience significantly more pain when many hundreds of pulses are delivered continuously to one site. Therefore, our experimental design limited external validity, with the levels of pain that we elicited probably being less than pain that might have occurred if complete routine clinical treatments were delivered, but our methodology was necessary to ensure internal validity of the data by maintaining strict comparability between the naı¨ve and nonnaive treatment arms and to avoid excessively fatiguing
KAKAR ET AL
the subjects, who would probably not have been able to tolerate 11 separate complete treatments or to differentiate the pain associated with such prolonged treatments. Second, we did not use the ultrasound device on the abdomen because, at the time of this study, the Food and Drug Administration had not approved it for indications off the face, but based on unpublished cohort data involving thousands of patients, ultrasound skin tightening appears safe and effective off the face. Because this was a prospective intervention study, consent was obtained from participants. Although this can be considered potentially biasing for patients, prospective research is more powerful than retrospective studies, and conducting a human intervention study without eliciting informed consent is not ethical. The consent form did not clarify what we expected to find regarding pain levels in different conditions or what the pain levels of others had been, so it is unlikely that patients would have had a systematic bias in favor of one of the four arms or one of the three anatomic locations. There is a dearth of studies comparing the pain and discomfort associated with different cutaneous procedures. Most pain studies in dermatology compare the use of a topical or other anesthetic agent with placebo.3–5 Although these studies are useful, it is not surprising that most active agents are found to be superior to no pain management at all. Prior comparative studies examining two intervention arms, rather than one intervention arm and placebo, include investigations of electrodesiccation for dermatosis papulosa nigra,6 injectable botulinum toxin for facial rhytides,7–9 tumescent anesthesia for liposuction,10 different injectable anesthetics for Mohs surgery,11 and varying wavelengths and fluences for activating 5-aminolevulinic acid during photodynamic therapy.12–14 In the laser and energy device realm, there is an acute dearth of such studies. Although serious adverse events are unlikely with cutaneous laser and energy devices, we need more work in this area to better describe the minor
inconveniences, such as pain, that may guide patient or device selection. The current study is an early step in better understanding the character of intraoperative pain associated with nonablative skin tightening devices that have few visible sequelae. That the pinprick intervention elicited similar pain results at the facial sites and slightly less pain on the abdomen is consistent with expectations and suggests that experimental pain levels were correctly measured. The significant variation between patients indicates that interpatient differences are an important element that cannot be discounted, regardless of whether patients are device naı¨ve or not. In this controlled, comparative study, we had good internal validity that suggests comparability of our results; additional, more-naturalistic studies that may examine more-extensive treatment may be further able to confirm the external generalizability of our results.
References 1. Bogle MA, Dover JS. Tissue tightening technologies. Dermatol Clin 2009;27:491–9. 2. Alam M, White LE, Martin N, Witherspoon J, et al. Ultrasound tightening of facial and neck skin: a rater-blinded prospective cohort study. J Am Acad Dermatol 2010;62:262–9. 3. Monheit GD, Campbell RM, Neugent H, Nelson CP, et al. Reduced pain with use of proprietary hyaluronic acid with lidocaine for correction of nasolabial folds: a patient-blinded, prospective, randomized controlled trial. Dermatol Surg 2010;36:94–101. 4. Shaikh FM, Naqvi SA, Grace PA. The influence of a eutectic mixture of lidocaine and prilocaine on minor surgical procedures: a randomized controlled double-blind trial. Dermatol Surg 2009;35:948–51. 5. Levy PM, De Boulle K, Raspaldo H. Comparison of injection comfort of a new category of cohesive hyaluronic acid filler with preincorporated lidocaine and a hyaluronic acid filler alone. Dermatol Surg 2009;35(Suppl 1):332–7. 6. Carter EL, Coppola CA, Barsanti FA. A randomized, doubleblind comparison of two topical anesthetic formulations prior to electrodesiccation of dermatosis papulosa nigra. Dermatol Surg 2006;32:1–6. 7. Kranz G, Sycha T, Voller B, Gleiss A, et al. Pain sensation during intradermal injections of three different botulinum toxin preparations in different doses and dilutions. Dermatol Surg 2006;32:886–90.
40:4:APRIL 2014
403
PAIN IN NONABLATIVE SKIN TIGHTENING DERMATOLOGIC PROCEDURES
8. van Laborde S, Dover JS, Moore M, Stewart B, et al. Reduction in injection pain with botulinum toxin type B further diluted using saline with preservative: a double-blind, randomized controlled trial. J Am Acad Dermatol 2003;48:875–7. 9. Alam M, Dover JS, Arndt KA. Pain associated with injection of botulinum A exotoxin reconstituted using isotonic sodium chloride with and without preservative: a double290 blind, randomized controlled trial. Arch Dermatol 2002;138:510–4. 10. Yang CH, Hsu HC, Shen SC, Juan WH, et al. Warm and neutral tumescent anesthetic solutions are essential factors for a less painful injection. Dermatol Surg 2006;32:1119–23. 11. Morganroth PA, Gelfand JM, Jambusaria A, Margolis DJ, et al. A randomized, double-blind comparison of the total dose of 1.0% lidocaine with 1:100,000 epinephrine versus 0.5% lidocaine with 1:200,000 epinephrine required for effective local anesthesia during Mohs micrographic surgery for skin cancers. J Am Acad Dermatol 2009;60:444–52. 12. Babilas P, Knobler R, Hummel S, Gottschaller C, et al. Variable pulsed light is less painful than light-emitting diodes for topical
404
DERMATOLOGIC SURGERY
photodynamic therapy of actinic keratosis: a prospective randomized controlled trial. Br J Dermatol 2007;157:111–7. 13. Ericson MB, Sandberg C, Stenquist B, Gudmundson F, et al. Photodynamic therapy of actinic keratosis at varying fluence rates: assessment of photobleaching, pain and primary clinical outcome. Br J Dermatol 2004;151:1204–12. 14. Wiegell SR, Haedersdal M, Philipsen PA, Eriksen P, et al. Continuous activation of PpIX by daylight is as effective as and less painful than conventional photodynamic therapy for actinic keratoses; a randomized, controlled, single-blinded study. Br J Dermatol 2008;158:740–6.
Address correspondence and reprint requests to: Murad Alam, MD, MSCI, Department of Dermatology, Northwestern University, 676 North Saint Clair Street, Suite 1600, Chicago, Illinois 60611, or e-mail: [email protected]
