Development and validation of the dermoscopy haemangioma activity index (DHAI) – A pilot study

Introduction

Infantile haemangiomas (IHs) are the most common soft-tissue tumours of infancy, occurring in 4% to 10% of children under one year of age, with a clear female to male predominance of 2.5 – 4:1.¹ Haemangiomas are classified based on depth (superficial, deep, or mixed), morphology, and their stage in the growth cycle (early proliferative, late proliferative, plateau, and involuting stages) but the duration of these phases can vary among subtypes.²

Dermoscopy, often referred to as the third eye of the dermatologist, offers advantages beyond unaided visual inspection. While a skilled clinician can discern much with the naked eye, dermoscopy enhances diagnostic accuracy through magnification and the use of polarised light, enabling visualisation of subsurface structures and patterns not otherwise visible. Various dermoscopy patterns and structures have been described for IH by various authors, by which it is possible to differentiate IH from capillary malformations as well as comment upon the depth of haemangiomas.^3-7 However, no attempt has been made to use dermoscopy to score IH. Herein, using a simple set of parameters, we have attempted to formulate an objective scoring system for IH; Dermoscopy Haemangioma Assessment Index (DHAI). Anybody having basic training in dermoscopy can use it to assess the activity of IH and depending on change in the activity of IH, the dermatologist can assess the effect of treatment. We aimed to formulate and validate an easy to use and non-time-consuming scoring system for IH based on dermoscopy (DHAI), to compare it with the Haemangioma Activity Score (HAS) for accuracy.

Methods

This was a prospective observational study, conducted over a period of one year (February 2019 to January 2020) in the Department of Dermatology, Venereology and Leprology, Postgraduate Institute of Medical Education and Research, Chandigarh. Since this was a pilot study, a formal sample size calculation was not done, and consecutive patients of IH were recruited from the outpatient Department of Dermatology of our institute. They had to meet the following inclusion criteria:

Inclusion Criteria: Patients with superficial or mixed IH

1. Colour of the haemangioma (E)^,

Colour of haemangioma greatly varies with changes in the activity of IH. ^3,4

• Bright red is seen in the very active phase (score 3) [Figure 1a]

• Purple colour (score 2) [Figure 1b]

• Light pink/Milky red areas (score 1) [Figure 1c]

• Residual pigmentation/White or brown structureless area (score 0) – It corresponds to fibrosis, and brown structureless area or residual pigmentation corresponds to hemosiderin deposition.

Close

Figure 1a:

Dermoscopy showing bright red colour of IH (score 3) (contact polarised mode, 10x; Dermlite DL2 Hybrid M).

Export to PPT

Close

Figure 1b:

Dermoscopy showing matte red colour (score 2) (contact polarised mode, 10x; Dermlite DL2 Hybrid M).

Export to PPT

Close

Figure 1c:

Dermoscopy showing light pink colour (score 1) (polarised mode; Dermlite DL2) (Contact polarised mode, 10x; Dermlite DL2 Hybrid M).

Export to PPT

2. Patterns of vessels (P)

Pattern of vessels also varies with the activity of IH. 5

• Red linear curved -Globular/circular/comma-like/wavy vessels - disclosing superficial ectatic vertical papillary vessels seen in active superficial or mixed haemangiomas, which are the first to respond to treatment (score 3) [Figure 2a]

• Red linear vessels - indicative of corresponding deeper dilated horizontal subpapillary capillaries, seen in the regressing phase of haemangiomas when most of the superficial component has resolved. (Score 2) [Figure 2b]

• Red diffuse amorphous areas - spaces of red colors which cannot be attributed to any pattern, perhaps indicative of a much deeper vascular component (score 1)

• None (score 0)

Close

Figure 2a:

Dermoscopy showing red globular vessel (blue colour circle) and red wavy vessels (black arrow) (score 3) (contact polarized mode, 10x; Dermlite DL2 Hybrid M).

Export to PPT

Close

Figure 2b:

Dermoscopy showing red linear vessel (blue arrow) (score 2) (contact polarised mode, 10x; Dermalite DL2 Hybrid M).

Export to PPT

4. Presence of follicular or eccrine openings (O)

This is a marker of regression of haemangiomas. When the haemangiomas are thick and active, the eccrine/follicular openings are hardly visible [Figure 3]. However, as they begin to flatten, the openings become more obvious. Thus, the presence of openings has been graded as

• >50% of the area of haemangioma in a particular field of dermoscopy shows the presence of openings = 0

• <50% of the area of haemangioma in a particular field of dermoscopy shows the presence of openings = 1

Close

Figure 3:

Dermoscopy showing eccrine openings (black arrow) (contact polarized mode, 10x; Dermlite DL2 Hybrid M).

Export to PPT

Results

Reliability

• a)

  Inter-rater reliability: The results of HAS and DHAI as scored by three observers at presentation have been given in Table 1. In the pairwise analysis, there were no statistically significant differences between HAS, DHAI, and each component of DHAI [Table 2].

  Among the three observers, the Cronbach’s alpha coefficient of 0.986 (95% confidence interval [CI] 0.97-0.99) was obtained for HAS, indicating good reliability and consistency. Similarly, reliability analysis showed satisfactory results for DHAI with a Cronbach’s alpha coefficient of 0.983 (95% confidence interval [CI] 0.97-0.99), further confirming good inter-rater reliability. Reliability was also found for each item included in calculating DHAI.

• b)

  Internal consistency: Inter-item correlation between all three observers was found to be positive, and the correlation was statistically significant (p-value <0.001 for all the scores) [Table 3].

• c)

  Clinical utility: The average time is 1 min 30 seconds to complete the questionnaire of DHAI.

Discussion

IH is present in an estimated 5% of neonates and infants. It primarily requires treatment during the proliferative phase to prevent complications.⁷ However, with the availability of safer and more efficacious treatment modalities, there is a decrease in the rate of complications. Dermoscopy can act as an essential tool in assessing IH, especially by evaluating its vascular morphology. However, there are no validated objective scoring systems based on dermoscopy to assess the efficacy of these drugs in clinical practice, as well as in research settings, even though a few scores have been published.^8-11 There are only a few case series based on the dermoscopy of IH.^3-5,12

The various clinical scores available at present, like the Haemangioma Severity Scale (HSS) and the Haemangioma Activity and Severity Index (HASI), suffer from the drawback of being long, complicated, time consuming, and have one or more subjective elements. Also, these scoring systems involve items that need physical examination of the lesion and, as such, have limited efficacy when being applied on images and in retrospective studies. Some authors have also used global assessment and the visual analogue scale (VAS), but they are very subjective. Further, VAS was originally used for scoring pain. Of all the scoring systems available, HAS has advantages over other scoring systems in being short, concise, time-saving, and having the ability to be used in person prospectively, as well as on images, retrospectively, besides its having one subjective component, the percentage regression of the deep component.^8,13 However, a good, reliable scoring system for IH using dermoscopy is lacking in the literature. As HAS is better in terms of accuracy, and sensitivity, we have compared DHAI with it.

In our study, IH under treatment or regressing IH were excluded, as treatment modality or type of pharmacologic intervention was not included as part of the inclusion criteria because the primary objective of the study was to develop a dermoscopy-based scoring system to assess the activity of infantile haemangiomas (IH), irrespective of the treatment status. Dermoscopy features such as colour, vascular patterns, and surface changes are reflective of biological activity and vascular architecture, which can be objectively assessed independent of the type of therapy. Including treatment status or regressing IH as a variable could have introduced confounding without improving the validity of the scoring model.

By using an objective dermoscopy scoring system, we can assess the activity of IH and compare the improvement of lesions during the follow-up visits. Activity of the lesion depends on the colour, pattern of vessels, and visibility of eccrine openings. The bright red colour of IH represents the active and proliferative phase. If the colour is matte red or purple, then it is a stable phase. Hence, we have selected colour as one of the prime factors for this scoring system. Similarly, the pattern of vessels and the presence of follicular or eccrine openings may also suggest the activity of the lesion. Red globular vessels represent superficial vessels seen in the active phase.⁵ It may also represent increased cellular activity of the vessels, as similar red globular vessels are also seen in superficial basal cell carcinoma and intra-epidermal carcinoma.^14,15 Further, on starting patients on treatment, red globules were the first dermoscopy component to disappear. This shows that with treatment, cellular activity reduces, and there is a disappearance of red globules.^4,5

Limitations

The main limitation is the small sample size. Larger studies with long-term follow-up are needed. The scoring system was not assessed longitudinally, limiting evaluation of changes over time, and with treatment. It also could not measure IH depth accurately, as clinical appearance may not reflect the subcutaneous component, which requires ultrasound for precise assessment.

The main advantage of DHAI is that we can assess an approximate depth of IH, and completion of the score can be done relatively quickly, as the assessor is only required to examine a small area. DHAI can be used both on the patients (prospectively) and the images (retrospectively) if dermoscopy images are available.

Conclusion

Thus, DHAI is a reliable and valid tool that can be used in day-to-day clinical practice. The use of a dermoscope decreases observer variation, resulting in a more objective score. DHAI was designed to simplify our clinical studies and to provide more objective information using a dermoscope to assess the activity of IH. This scoring system can be helpful in further research to assess drug efficacy. However, DHAI alone is not sufficient; analysis using dermoscopy along with ultrasonography is needed to completely analyse the stage and regression of IH.