Comparative Study of CO₂ Laser and Curettage-Electrodesiccation for Superficial Basal Cell Carcinoma Treatment: A Focus on Cosmetic Results

Abstract

Introduction

Basal cell carcinoma (BCC) is the most common type of skin cancer worldwide, and the incidence is rising. While surgical excision remains the gold standard, it can be time-consuming and resource-intensive. Therefore, there is a growing need for simpler and more cost-effective treatment options for low-risk tumors.

Objectives

This study compared the scar quality and the recurrence rate from continuous wave CO₂ laser and curettage and electrodesiccation treatments. Scar quality and the recurrence rate from the treatments were assessed.

Methods

A single-blind randomized prospective interventional trial was conducted at Odense University Hospital, Denmark, from 2017 to 2023. Patients with histologically confirmed superficial BCC were randomized 1:1 to Curettage-Electrodesiccation (CE) or continuous wave CO₂ (CW CO₂) laser treatment. Scar quality was assessed using the Patient and Observer Scar Assessment Scale (POSAS) and a modified visual-only scale (VSAS). Recurrence within one year was a secondary outcome.

Results

Thirty-two patients were treated for 32 tumors: 17 with CE and 15 with CO₂ laser. There was no significant difference in scar quality or recurrence rate between the CE and CO₂ laser groups (PSAS: P = 0.422; OSAS: P = 0.747; VSAS: P = 0.522). After one year, neither group showed tumor recurrence.

Conclusion

CO₂ laser used in continuous wave setting offers a treatment for superficial BCCs with cosmetic outcomes and recurrence rates comparable to CE. However, a larger patient sample and longer follow-up are needed for definitive conclusions.

Introduction

Basal cell carcinoma (BCC) represents the predominant type of skin cancer globally, and the incidence is rising [1]. In order to address the socioeconomic implications, it is crucial to provide simple and cost-effective treatment options while ensuring optimal patient satisfaction. Although excision results in low recurrence rates, it is a time-intensive procedure and incurs high costs [2]. Consequently, exploring alternative treatment modalities for low-risk tumors becomes relevant. In recent years, the CO₂ laser has become a staple in dermatological practices due to its versatility. Therefore, the cost of obtaining one is often not a concern, as most dermatology clinics already consider it standard equipment. CO₂ laser offers advantages such as precise energy delivery, minimal thermal impact on surrounding tissue, and reduced bleeding during the procedure [3].

Curettage-electrodesiccation (CE) is a well-established and widely used procedure, conceptually similar to continuous wave CO₂ laser treatment, as both aim to destroy tumors while minimizing damage to surrounding tissue in a time- and cost-effective manner. Therefore, we found it relevant to compare these two treatments in terms of the quality of resulting scars and, as a secondary outcome, their recurrence rates.

Methods

This study was a randomized prospective interventional trial with single-blind evaluation carried out between 2017 and 2023 at Odense University Hospital in Denmark. The primary objective was to compare the scars resulting from CE versus continuous wave (CW) CO₂ laser in the treatment of superficial basal cell carcinomas.

The secondary outcome was to compare recurrence rates. Only superficial basal cell carcinomas were included. This decision was based on the presumption that the continuous wave CO₂ laser may not be effective for nodular tumors, as indicated by a previous study[4].

The study was conducted over four appointments; treatment visit and follow up one, six, and 12 months after.

Patients suspected of having superficial BCC underwent a biopsy for confirmation, making them eligible for this study. After documentation of superficial BCC, patients were informed about both treatment modalities, and consent was obtained for enrollment. Randomization with a 1:1 ratio was performed by picking a sealed envelope.

Exclusion criteria included pregnancy, dementia, having a pacemaker, local anesthesia intolerance, tumor location in the scalp or near the eyes, psoriasis (because of the Koebner phenomenon), immunosuppressing medication, and suspected poor healing potential.

Inclusion criteria included biopsy-verified superficial BCC, presence of only one tumor at the time of inclusion, tumor accessibility for both electro surgery and CO₂ laser, and age above 18.

The study was approved by the Danish Data protection Agency and the Regional Committees on Health Research Ethics (project ID: S-20170097)

Interventions

We assessed the apparent margin using a dermatoscope and demarcated 3 mm of normal skin around the tumor. The lesion was anesthetized with an injection of lidocaine and adrenaline (10 mg/5 μg per mL)

Curettage was performed with disposable dermal curettes of two different sizes (4 and 7 mm) depending on the tumor size. The sharp side of the instrument was applied in different directions across the lesion until all fragile or abnormal tissue had been removed. Afterwards, the surface of the lesion was cauterized using a monopolar ball-shaped electrode with power set to 24 watts. In a second cycle, curettage and electrodesiccation were then repeated in a more superficial manner across the lesion. CO₂ laser ablation of tumors was performed with Lumenis, UltraPulse CO₂ laser in continuous wave mode with fluences ranging from 2–6 watts/cm² with spot size 1 millimeter. The laser beam was applied across the lesion, removing tissues in layers with slight overlap. Debris tissue was removed between each laser passage using a gauze. Inspection of the treated area was done between each pass to determine whether residual tumor was present.

CE procedures were performed by either certified dermatologists or resident physicians, while the CO₂ laser treatments were performed by two certified dermatologists within our department with expertise in the field of lasers. A photograph was taken prior to the treatment and at each of the follow up visits.

Outcomes

For the assessment of scar quality, the Patient and Observer Scar Assessment Scale (POSAS) was used [5]. POSAS consists of two separate scales, the observer (OSAS) and the patient scale (PSAS). It is based on clinically relevant scar characteristics. The patient scale includes the following six characteristics: pain, pruritus, color, thickness, relief, and pliability. The observer scale (OSAS) consists of the following five characteristics: vascularization, pigmentation, thickness, surface roughness, and pliability. Although the observer scale originally includes surface area, it was excluded to focus the score on potential differences in scar quality rather than on size, as scar size is likely influenced by tumor size, not by the treatment modality.

All included items are scored on the same 10-point scale, in which a score of 1 is given when the scar characteristic is comparable to “normal skin”, and a score of 10 reflects the “worst imaginable scar”. All items are summed to give a total scar score. The OSAS score ranges from 5 to 50, and the PSAS score from 6 to 60.

In this study, the observer scale (OSAS) was not performed in a blinded setting, and consequently, photographs taken during the follow-up visits were evaluated after the conclusion of the trial in a blinded setting. In doing so, a modified version of the observer scale was used, including only the four categories that we found suitable for visual assessment of photographs (vascularity, pigmentation, thickness, and surface roughness), with a total score ranging from 4 to 40. We called this scale VSAS (visual scar assessment scale).

The assessment of scars at follow-up visits was done by one of the five doctors who are the authors of this paper. The assessment of photographs was performed by two of the authors (RL, TV), who are certified specialists in dermatology. A mean of these two scores was used.

The secondary outcome was recurrence. Treated areas were evaluated clinically and with dermoscopy at follow-up-visits. If recurrence was suspected, a punch biopsy was performed.

Statistics

For our primary endpoint of scar quality, we considered a minimum clinically significant difference of 1 on a 1–10 scale. With a significance level of 5% and a statistical power of 80%, and given the non-normal distribution in our patient population, we used a non-parametric approach for sample size determination. The sample size was calculated using STATA 17, Statacorp, resulting in 20 patients per group, totaling 40–45 patients overall. Potential differences between the treatment conditions (CE vs. CO₂) across patient characteristics, tumor characteristics, and the three outcomes (PSAS, OSAS, and VSAS) were tested using Mann-Whitney U tests.

For the three outcomes, Hedges g with 95% confidence intervals (CI) were computed as effect sizes[6]. Subsequently, multiple linear regression analyses were performed with PSAS, OSAS, and VSAS as dependent variables, with treatment condition (CE vs. CO₂) as the independent variable, adjusting for tumor size, age, sex, and length of follow-up.

Heteroscedasticity was tested using the White test, and collinearity was assessed using variance inflation factors (VIF). All data analyses were performed using STATA 17, Statacorp.

Results

A total of 36 tumors in 36 patients were included and treated. However, four patients were lost to follow-up: one cancelled the final follow-up, two patients only filled out parts of the patient assessment scale, which invalidated the sum of scores, and one patient was excluded because the photo was lost. Therefore, 32 tumors are included in the data set. The majority of patients had tumors on the trunk (n=22/69%), while three patients had tumors in a lower limb (9.4%), four in an upper limb (12.5%), and three had tumors in the head or neck area (9.4%). The treatment conditions did not differ in terms of the area of the tumors. Also, they did not differ with regard to tumor size, age, sex, or the exact length of follow-up (Table 1). The Mann-Whitney U tests did not suggest any significant difference between the means of PSAS (P=0.422) or OSAS (P=0.747) between the two groups (CO₂ laser and CE) (Table 2). Comparing the means of VSAS showed the biggest difference in means but was still not significant (P=0.522). An outlier was detected among the PSAS and OSAS scores. Removing this outlier did not change the results of the between-group comparisons on these outcomes.

Table 1.

Descriptive Statistics of Patient and Tumor Characteristics.

	Curettage (n=17)	CO2 (n=15)	P-values
Age (mean ± SD)	62.94 ± 11.52	59.60 ± 11.66	.449
Female	11(65%)	11(73%)
Diameter (mean ± SD)	12.76 ± 6.69	10.12 ± 5.30	.282
Location, n (%)
• Head and neck	3(20%)	0 (0%)	NaN
• Trunk	11(73%)	11(65%)	.599
• Upper limbs	2 (13%)	2(12%)	.893
• Lower limbs	1(7%)	2(12%)	.471
Follow-up length, months (mean ± SD)	12.67 ± 0.75	13.02 ±1.56	.992

Note: NaN = could not be computed due to lack of variance in one of the treatment conditions

Table 2.

Comparison between Curettage and CO2 in Scar Quality (Mann-Whitney U-test).

Outcomes	Curettage	CO2-laser	Z	P	g (95% CI)
	Mean + SD	Mean + SD
PSAS	9.24 ± 3.07	11.33 ± 7.63	.82	.422	.37 (−.33, 1.07)
OSAS	10.82 ± 3.32	11.40 ± 5.59	.32	.747	.13 (−.57, .82)
VSAS	8.62 ± 1.10	8.14 ± 1.55	.66	.522	.25 (−.96, .46)

Abbreviations: PSAS: Patient Scar Assessment Scale, OSAS: Observer Scar Assessment Scale, VSAS: Visual Scar Assessment Scale, g: Hedges g, CI: confidence intervals (lower and upper bound). Range in scores: PSAS (6–60), OSAS (5–50), VSAS (4–40), Lower is better.

Due to the lower statistical power of non-parametric tests, differences between the groups were also checked with Welch t-test, the parametric counterpart to Mann-Whitney U tests, producing similar results for the PSAS (P= 0.332) the OSAS (P=0.730), and the VSAS (P=0.475).

While the groups did not differ significantly on any of the demographic or tumor characteristics (Table 1), it is still possible that they, in combination, could have influenced the differences between the treatment conditions on the outcomes. However, the multivariate regression analyses did not suggest that this was the case (Table 3), as the results for the treatment conditions remained largely the same. The variables pertaining to the location of the tumor did not contain sufficient variance in one of the areas (neck and head) and too small cell sizes in other areas (lower and upper limbs) for them to be considered relevant covariates and were therefore not included in the multivariate analyses. The Breusch-Pagan tests did not suggest heteroscedasticity, and the VIF were all close to one, indicating the absence of multicollinearity.

Table 3.

Multivariate Regression Analyses with Treatment Condition as Predictor of Scar Quality at End of Follow-Up while Adjusting for Demographics, Tumor size, and Follow-Up Length.

	PSAS			OSAS			VSAS
	Beta	95% CI	P	Beta	95% CI	P	Beta	95% CI	P
Treatment Modality	2.63	−1.44, 6.71	.196	.83	−2.40, 4.07	.602	−.36	−1.85, 1.13	.625
Sex	−2.79	−7.06, 1.48	.191	−2.64	−6.03, .74	.121	−1.35	−2.85, .14	.074
Age	.05	−.12, .23	.546	.05	−.09, .19	.512	−.02	−.08, .04	.475
Tumor size	−.21	−.55, .13	.220	−.216	−.49, .05	.112	.03	−.11, .16	.665
Follow-up length	−1.92	−3.63, −.22	.028	−1.28	−2.63, .07	.062	−.01	−.62, .59	.962

Abbreviations: CI: confidence intervals (lower and upper bound), PSAS: Patient Scar Assessment Scale, OSAS: Observer Scar Assessment Scale, VSAS: Visual Scar Assessment Scale

No tumor recurrence was observed in either of the two treatment groups during follow-up.

Discussion

In an effort to elucidate any differences in scar quality and recurrence rates between the CO₂ laser and curettage treatments, no significant differences were found. Both treatments resulted in relatively low PSAS, OSAS, and VSAS scores, reflecting that both resulted in relatively good scar qualities. Furthermore, no recurrence was found after one year of observation.

Many studies have focused on the use of lasers in the treatment of basal cell carcinomas, but only a few studies have described the use of continuous wave mode CO₂ laser. In 1979, Adams et al. [4] treated 25 basal cell carcinomas (both superficial and nodular) with a single, non-overlapping continuous mode of a CO₂ laser, using variable exposure times. They found recurrence in 50% of tumors. Wheeland et al [7] used a defocused mode of continuous wave CO₂ laser coupled with curettage in the treatment of 370 superficial BCCs in 52 patients. The overall result was satisfactory, with no recurrence of tumors. Similar to our method, they also allowed for several treatment cycles if inspection of the treated area showed residual tumor. No other study has exclusively used continuous mode CO₂ lasers as in our research; recent studies generally employ a pulsed setting for CO₂ lasers instead [3,5,6,10]. These studies employ the CO₂ with different settings and techniques and have varying results regarding both cosmetic outcomes and cure rates. CO₂ lasers are also used in conjunction with topical chemotherapeutic agents[8,11,13,14], but these methods are less comparable to our technique, although generally showing favorable results.

Continuous wave lasers emit a constant beam, effectively a long pulse, which increases thermal damage to surrounding tissue as a function of pulse length[15], leading to more nonspecific damage and potentially worse scarring. These characteristics may explain why we found that scar quality was similar to that produced by CE, as the thermal impact of the electrodesiccation tool is likely comparable to that of a continuous wave CO₂ laser. Nevertheless, the nonspecific tissue damage also contributes to the clearance of tumor.

CO₂ laser can be directed in a precise manner in an almost bloodless surgical field, making it a good choice for patients with bleeding disorders or those on anticoagulants. Additionally, patients equipped with pacemakers are not precluded from this treatment. Effective use of the CO₂ laser demands physicians experienced in the field of lasers, but compared to other modes, the continuous wave mode requires very few adjustments of settings.

A key limitation to both CE and CW CO₂ is the absence of margin control, which makes it unsuitable for tumors with a high risk of recurrence. This also underlines the importance of delineating tumor margins accurately preoperatively[16]

Immediate distinction between tumor and healthy skin is important to ensure clearance of malignant cells[2]. CE gives tactile feedback as the clinician maneuver the curette across the tumor, a feature that is appreciated by many. CO₂ lasers offers a bloodless surgical field, and wiping with a wet cloth between cycles also provides visualization of any potential residual tumor.

Limitations

Initially, the study was designed with three follow up visits at one, six, and 12 months after treatment to follow the development of scars. However, due to many cancelled or rescheduled appointments within the first year, we opted to include only the 12-month follow-up data. We believe that this does not change the overall result or aim of this study as we also believe that the final assessment of scars at 12 months post-treatment is the most important. This study’s statistical power is limited due to the modest number of participants.

After the initiation of this study, our clinic narrowed the intake of referrals to primarily include organ transplant recipients, whom we had to exclude due to their use of immunosuppressive medication. Furthermore, the COVID-19 pandemic led to the cancellation of several follow-up visits. It would be preferable if we had had a longer timeline for the follow-up, as BCC is a slow-growing tumor and recurrence can occur years after treatment.

Conclusion

The present findings indicate that both CE and CW CO₂ laser are effective options for the treatment of superficial basal cell carcinomas; both yielded a relatively good quality of scars, although their significance is limited by the low number of treated patients and the short follow-up period. We believe that CW CO₂ lasers offer a relatively simple method for ablating tumors without the need to adjust multiple settings, allowing practitioners with limited expertise in the field of lasers to carry out this treatment. It offers an alternative to CE in patients with pacemaker, but the choice between them should be guided by factors such as patient characteristics, tumor characteristics, cost considerations, and the expertise available.