A cross-sectional pilot study evaluating the histopathology of atrophic acne scars with a focus on the vertical depth of ice pick, boxcar, and rolling scars and its implications in skin of colour

Introduction

Atrophic acne scars are common sequelae of acne, consequent to varied immunological and cytokine alterations which eventually lead to alteration of collagen and elastin fibres. They have profound emotional and psychological consequences, affecting the quality of life.^1,2

Jacob et al.³ had classified atrophic post-acne scars into three subtypes: ice-pick, rolling, and boxcar scars. Notably, each type of scar has a unique architecture and depth, necessitating a tailored approach to treatment, unlike the “shot gun” approach used by many clinicians. Often, a patient can exhibit several scar morphologies, requiring a multi-modal approach to therapy. The existing therapies range from surgical excision, chemical peels, subcision, fillers, and energy devices. Fractional ablative lasers are especially useful as they can address, though not uniformly, all types of atrophic acne scars.⁴

In vitro, in vivo, and ex vivo data with fractional lasers have addressed their depth of penetration, confirming that rolling and boxcar scars are amenable to this intervention, though ice-pick scars do not consistently respond to fractional lasers. The tissue effect of fractional lasers is achieved both by collagen remodelling and scar relieving,^5,6 although a lack of knowledge of acne scar depth makes its dosimetry inaccurate. To the best of our knowledge, no study has yet assessed the histopathology of atrophic acne scars prior to laser intervention. The paucity of data on the histopathology of atrophic acne scars leaves an important lacuna in their management.^7,8 There is some data on the effect of interventions on collagen remodelling,^9,10 but it is unclear whether such interventions are able to ameliorate the depth of atrophic scars.

Thus, the primary objective of our study was to assess the histological changes in atrophic acne scars with a focus on the vertical depth of each scar type. The existing published data on the depth achieved by fractional energy devices was collated to ascertain if these devices could achieve scar-specific results.

Methods

This cross-sectional study was conducted in a tertiary hospital in New Delhi from April 2023 to August 2024. Patients > 18 years, of either sex, with atrophic acne scars diagnosed clinically with an acne-free period of a minimum of 6 months, and without history of use of topical or systemic drugs for the same period, were selected. Patients who were previously treated with fillers, or any skin resurfacing procedures within the preceding year were excluded from the study. The sample size was 49 based on a previous scar-specific study by Sardana K et al.¹¹

Results

Histopathology

We biopsied boxcar scars in 32 patients (65%), followed by ice-pick scars in 10 (20.4%) and rolling scars in 7 (14.28%). The mean histological depth was 1933.4 ± 1117.8 µm, 1327.88 ± 571.34 µm, and 1357.14 ± 578.3 µm for ice-pick, boxcar, and rolling scars, respectively [Figures 1a, 1b, Table 2, and Supplement 1]. There was a statistically significant difference in the depth of ice-pick scars and boxcar scars (p=0.02) [Supplement 1].

Supplement 1

Download

Close

Figure 1a:

Image showing V-shaped icepick scar (Verhoeff-van Gieson, 40x).

Export to PPT

Close

Figure 1b:

Tissue section showing Boxcar atrophic acne scar with a vertical depth of 724.86 µm measured from the stratum basale to the base of the scar (Masson’s trichrome, 100x).

Export to PPT

Table 2: Histological characteristics and vertical depth of atrophic acne scars

Histological parameters	Frequency (n=49 biopsies)
Loss of pilosebaceous units	47 (96%)
Appendageal tilting	44 (90%)
Epidermal atrophy	44 (90%)
Collagen grading
Very loose, haphazard	35 (71.4%)
Relatively compact	13 (26.5%)
Compact	1 (2%)
Elastin in scar
Absent	33 (67.3%)
Grade 1	14 (28.5%)
Grade 2	2 (4%)
Elastin arrangement
Haphazard	13 (81.2%)
Horizontal	3 (18.7%)
Surrounding dermal elastin
Absent	2 (4%)
Grade 1	24 (49%)
Grade 2	19 (38.8%)
Grade 3	4 (8%)
Dermal inflammation
Absent	6 (12.2%)
Grade 1	31 (63.2%)
Grade 2	11 (22.4%)
Grade 3	1 (2%)
Location of inflammation
Absent	6 (12.2%)
Perivascular	38 (77.5%)
Perifollicular	5 (10%)
Scar vascularisation
Absent	3 (6%)
Grade 1	11 (22.4%)
Grade 2	33 (67.3%)
Grade 3	2 (4%)
Scar vascular ectasia
Absent	31 (63.2%)
Grade 1	12 (24.5%)
Grade 2	6 (12.2%)
Surrounding dermal vascular ectasia
Absent	20 (40.8%)
Grade 1	25 (51%)
Grade 2	4 (8%)
Depth of scar (µm)	Mean ± SD (µm) ɸ
Ice-pick scars (n=10)	1933.4 ± 1117.8 (range: 626-4323 µm)
Boxcar scars (n=32)	1327.88 ± 571.34 (range: 448-2477 µm)
Rolling scars (n=7)	1357.14 ± 578.34 (range: 560-2076 µm)
	P value*
Ice-pick versus rolling	0.23 (ns)
Rolling versus boxcar	0.90 (ns)
Ice-pick versus boxcar	0.02 (s)
Ancillary findings
Scar inflammation
Mild	7 (14.3%)
Moderate	4 (8.2%)
Severe	0
Solar elastosis	2 (4.1%)
Calcinosis	3 (6.1%)
Procedural dermal hemorrhage	3 (6.1%)
Granulomatous perifolliculitis	4 (8.2%)
Demodex	2 (4.1%)
Pulled up appendages	8 (16.3%)
Pigment laden macrophages	2 (4.1%)

ɸ Shapiro-Wilk test: Ice-pick (p=0.191), Boxcar (p=0.203), Rolling (p=0.369); p >0.05 indicates normal distribution, ^* Unpaired t-test; ns: non-significant; s: significant

Certain histological findings were consistently noted including, loose, haphazardly arranged collagen fibres and complete absence or significant reduction in elastin fibres [Table 2]. The collagen and elastin changes have been depicted in Figures 1c, 1d, and 2a.

Close

Figure 1c:

Complete loss of elastic fibres in scar tissue (left), compared to adjoining non-scarred dermis (right) (Verhoeff-van Gieson, 400x).

Export to PPT

Close

Figure 1d:

Horizontally arranged elastin fibres (black arrow) in the scar tissue (Verhoeff-van Gieson, 200x).

Export to PPT

Close

Figure 2a:

Presence of loosely arranged fine fibrillar collagen (black arrow) and ectatic capillary channels (blue arrow) in the scar tissue (Masson’s trichrome, 200x) (Inset: Broad compact collagen in normal dermis, Masson’s trichrome, 400x).

Export to PPT

Mild to moderate mononuclear inflammatory infiltrates was seen in the adjoining dermis of 43 biopsies, with 77.5% showing a perivascular distribution (n= 38) and the rest around the follicles (n=5) [Figure 2b]. Other findings included pigment laden macrophages (n=2) [Figure 2c], granulomatous perifolliculitis (n=4) [Figure 3a], pulled up eccrine glands (n=8) [Figure 3b], calcinosis (n=3) [Figure 3c], mild to moderate mononuclear inflammatory infiltrate in the scar tissue (n=11), demodex mite (n=2), and solar elastosis (n=2) [Table 2].

Close

Figure 2b:

Photomicrograph of acne scar showing minimal vascularisation (blue arrow), ectatic capillary channels (black arrow), and scattered inflammatory cells in the scar (Haematoxylin & eosin, 100x).

Export to PPT

Close

Figure 2c:

Presence of pigment-laden macrophages in the scar tissue (Haematoxylin & eosin, 200x).

Export to PPT

Close

Figure 3a:

Epithelioid cell granuloma (black arrow) and hair shaft (black star) in the scarred tissue (Haematoxylin & eosin, 200x).

Export to PPT

Close

Figure 3b:

Figure showing pulled up eccrine glands higher up in the dermis (black arrow) (Haematoxylin & eosin, 40x).

Export to PPT

Close

Figure 3c:

Calcinosis, absence of elastin fibres in the scar and tilting of appendages towards the scar (black arrow) (Verhoeff -van Gieson,100x) (Inset: Calcinosis Haematoxylin & eosin, 100x).

Export to PPT

Close

Figure 3d:

Superficial boxcar (black star) with tilted adnexae and ectatic vascular channels (Haematoxylin & eosin, 100x).

Export to PPT

Appendageal tilting (either towards or away from the scar), loss of pilosebaceous units, and vascular ectasia were also noted in the atrophic scars [Figure 2b, 3d]. Procedural and histopathological artefacts like tissue shrinkage, scar pinching, and procedural dermal hemorrhage were seen in three tissue specimens.

Discussion

Our study found that the most common scars were boxcar (32/49, 65%), the patients were in the age group of 18 to 35 years (two-thirds of patients were 18-25 years old), with a male to female ratio of 1.8:1. The clinical scar severity ranged from moderate to severe, with median duration of acne scars and acne free period being five years and one year, respectively.

Acne scarring is consequent to the evolution of non-inflammatory comedones into inflammatory lesions, which rupture through the infra-infundibular section of the follicle, resulting in perifollicular abscesses.^13,14 This is normally repaired without scarring in seven to ten days due to the intrinsic reparative process that originates from the epidermis and appendageal structures to encapsulate the inflammatory reaction. However, if this perifollicular abscess ruptures or there is an inadequate encapsulation or deeper extension into subcutis and deep dermis, a robust wound healing process is initiated, which leads to the formation of multi-channelled fistulous tracts.¹⁴

A study by Holland et al.¹⁵ noted that in patients with scarring, there was a large infiltration by macrophages, blood vessels, and vascular adhesion molecules, thereby implying that excessive inflammation is predictive of scarring. In addition, varied cytokines and chemokines have been implicated (Toll like receptor-4, Interleukin-2 (IL-2), IL-10, tissue inhibitor of metalloproteinase-2 (TIMP-2), JUN gene)¹⁶ with a predominant role of transforming growth factor-β1(TGF-β1), which in conjunction with interleukin-6 (IL-6) drives a predominant T helper 17 (Th17) mediated inflammatory response.¹⁷ Myofibroblast-rich areas are seen in hypertrophic scars with mild B-cell infiltration.⁸

The existing histologic data on atrophic acne scars is scant and does not focus on scar depth, which is a key factor determining therapeutic response to energy devices. The reported findings include flattened-out, thin epidermis, numerous ectatic lymph and venous vessels in the dermis, fine horizontally arranged collagen bundles, numerous fibroblasts, and irregular foci of lymphohistiocytic cells. Irregular and thinned elastic fibres in early scars to a complete absence of elastic fibres in mature atrophic scars have also been observed.^17-19 Loss of pilosebaceous units, calcification, foreign body reaction, and presence of mast cells have also been noted by some authors.^8,19 Notably, these are not scar specific and it is unclear what type of scars were studied.

The salient features noted in our study included atrophy of epidermis and appendageal tilting (both towards and away from the scar) seen in 90% of the biopsies, suggesting the presence of adjacent scar. The other prominent findings were loose and haphazardly arranged fine fibrillar collagen with absent or minimal elastin fibres which has been described previously.^17-19 Loss of pilosebaceous units in the scar tissue was a consistent finding in almost all the biopsies with complete absence of appendageal structures seen in five biopsy specimens. We also noted scar vascularisation and ectatic channels both in the scar, as well as in the surrounding papillary dermis.

Pursuant to the aim of our study, we estimated the depth of the varied atrophic scars, which was 1933.4 ± 1117.8 µm (range: 626-4323 µm), 1327.88 ± 571.34 µm (range:448-2477 µm), and 1357.14 ± 578.3 µm (range: 560-2076 µm) for ice-pick, boxcar, and rolling scars, respectively. The various types of atrophic scars include ice-pick scars, which are narrow (<2 mm), punctiform, deep, sharply marginated epithelial tracts that extend vertically to the deep dermis or subcutaneous tissue and are difficult to treat due to their deeper extent. Rolling scars consequent to dermal tethering of the dermis to the subcutis are wider than 4 to 5 mm. The boxcar scars are round or oval depressions with sharply demarcated vertical edges, similar to varicella scars, and can be shallow (0.1-0.5mm) or deep (≥0.5 mm) and are most often 1.5 to 4.0 mm in diameter. However, this classification was devised for interventions and was not preceded by a formal histological study of acne scars. It is obvious that any intervention that does not reach the depth of the scar being treated will not succeed, and this is especially relevant for energy-based devices including lasers.²⁰

The histopathology of atrophic scars has direct relevance to the use of fractional laser, the salutary effect of which is based on the formation of microscopic thermal zones (MTZ).^5,21 MTZs are zones of coagulative necrosis with a zone of denaturation of epidermis and dermis, which are subsequently replaced by new collagen and elastin within 3-6 months.^5,20,21 While there is data on the effect of fractional laser on collagen and elastin remodelling, a more crucial parameter is the depth and width achieved by fractional devices. Fractional Erbium-doped Yttrium-Aluminium-Garnet (Er:YAG) results in a superficial and broad MTZ with little thermal collateral damage, whereas fractional CO₂ (Fr:CO₂) results in a narrow and deep “cone”-like MTZ and fractional radiofrequency (Fr:RF) causes a superficial and broad “crater”-like MTZ.²²

While very few good head-to-head studies have been done on dose penetration¹⁹ (especially on facial skin), it is believed that ablative fractional laser achieves better depth than non-ablative fractional laser devices. As acne scars are usually a mix of ice-pick, boxcar, and rolling scars, the final effect of fractional lasers would largely depend on the predominant scars and the type of laser used. However, most studies do not report individual scar improvement.

The tissue depth and width achieved by fractional energy devices would suggest that superficial boxcar scars would respond to Fr:RF and Fr:CO₂, while deep boxcar and ice-pick scars would respond best to Fr:CO₂ laser.^2,20,22 Despite ex vivo and in vivo (facial/abdominal) data, ice-pick scars do not respond consistently to energy devices,¹¹ which is possibly because the existent models for dose penetration have not accounted for the histological depth of acne scars. Another reason is the variations in simulated models and in vivo depth of MTZ created by energy devices.^20,23-25 This is evident by studies that have analysed depth achieved by fractional lasers in pig skin,^26,27 abdominoplasty specimens,^24,25 and not on facial skin and also variations due to tissue shrinkage by formalin fixation.

Limitations

The external validity of our study is that we have histologically assessed the average depth of individual atrophic scars which can be used by clinicians to set the appropriate dose of Fr:CO2 lasers. However, small sample size and technical difficulties in histological sectioning of the scars, especially ice-pick scars, remain major limitations of our study. Also, there is a need for in vivo facial skin depth penetration data of various energy devices, which needs corroborative clinical validation to ensure that device energies can be set to achieve the ideal depth to effectively treat atrophic acne scars, which would also help in avoiding adverse effects due to inappropriate dosages.

Conclusion

The clinical advantage of knowledge of the depth of atrophic acne scars is that this in vivo, facial skin histological assessment, in conjunction with the existent data on depth penetration of fractional energy devices, can be used to set the appropriate dose for effective treatment of acne scars. Our study shows that the dose of the energy device should be set to achieve an approximate depth of 1933.4 µm (the mean depth of ice-pick scars), which would eventually treat all atrophic scars. There is thus a need to correlate the depth penetration data of Fr:CO₂ with respect to the depth of atrophic scars and study the consequent clinico-histological improvement so that this can be translated to achieve meaningful outcomes.