Table 2 Prediction of melanoma metastasis based on models derived from multivariate logistic regression

	Model 1		Model 2		Model 3
	OR	95%CI	OR	95%CI	OR	95%CI
Breslow thickness (Cont.)	-	-	1.37	1.24–1.54	1.21	1.10–1.35
Histopathologic ulceration (Present)	-	-	2.78	1.78–4.34	2.14	1.27–3.59
Pigmentation
Absent	Ref.				Ref.
25	1.07	0.45–2.54	-	-	1.12	0.45–2.73
50	0.50	0.20–1.22	-	-	0.59	0.23–1.50
75	0.14	0.05–0.33	-	-	0.20	0.08–0.51
100	0.07	0.03–0.18	-	-	0.10	0.04–0.27
Dermatoscopic ulceration
Absent	Ref.		-	-	Ref.
1-49	1.46	0.87–2.42	-	-	1.10	0.63–1.89
>50	3.45	1.63–7.69	-	-	1.68	0.74–3.97
Regression structures
Absent	Ref.		-	-	Ref.
1-49	0.75	0.44–1.27	-	-	0.85	0.49–1.47
>50	0.42	0.19–0.89	-	-	0.36	0.15–0.81
Blue – white veil	6.43	3.87–11.06	-	-	5.91	3.48–10.40

N number of patients, OR Odds ratio, CI confidence interval, Ref. Reference level.
*Breslow thickness was included as a continuous variable, and histopathologic ulceration as binary (present vs absent). Pigmentation, dermatoscopic ulceration, and regression structures are expressed as percentages of feature extent in the lesion. Variables included in the first step of multivariable logistic regression for model 1: brown, blue/gray, red, white, black color, pigmentation, dermatoscopic ulceration, regression structures, white shiny streaks, blue-white veil, eccentric blotch, parallel ridge, vessels. Significant variables included in the last step of backward elimination are presented. Variables included in the first step of multivariable logistic regression for model 2: Breslow thickness, histopathologic ulceration. Variables included in the first step of multivariable logistic regression for model 3: Breslow thickness, histopathologic ulceration, pigmentation, dermatoscopic ulceration, regression structures, blue-white veil. All models were adjusted for age, sex, and anatomic location.

Quick links

Search