Remote Monitoring of Cryosurgery Response Using a Smartphone App: Prospective Study

TL;DR

A smartphone app was used to monitor patients' skin responses after cryosurgery, showing that patient self-imaging can effectively track treatment outcomes over time.

Contribution

The study introduces a reliable method for remote monitoring of cryosurgery outcomes using smartphone-based imaging and patient-reported metrics.

Findings

Local skin response peaked 3 days after cryosurgery, with substantial inter-rater agreement on lesion resolution and erythema.

Good-quality photos showed higher agreement in image-derived metrics compared to poor-quality photos.

Peak skin response was moderately associated with lesion resolution at 90 days.

Abstract

Cryosurgery is an effective treatment for benign lesions, although current unstandardized approaches may result in inadequate responses and unwanted adverse reactions. Monitoring treatment characteristics, lesion responses, and patient-reported outcomes using patient-derived mobile imaging may facilitate longitudinal treatment assessment.

This study aimed to determine the reliability of metrics for assessing the response to cryosurgery in patients with actinic and seborrheic keratoses using remote photographic monitoring.

Patients who were recommended cryosurgery by their physician for treating seborrheic and/or actinic keratoses (22 patients with 31 lesions) were enrolled. After treatment, participants took “overview” and “close up” photos of their lesion(s) and rated appearance, pain, and degree bothered on a custom smartphone app at eight posttreatment time points (days 0, 3, 7, 10, 14, 30, 60, and 90). After study completion, independent raters scored the images for local skin response (eg, erythema, scaling, crust, swelling, vesiculation, and erosion), cosmetic outcome (eg, hyperpigmentation, hypopigmentation, scarring, and atrophy), and lesion resolution.

The local skin response peaked 3 days after cryosurgery, with 26% (7/27) of patients reporting pain. There was substantial agreement between raters for lesion resolution (κ=0.71, 95% CI 0.62‐0.79), erythema (κ=0.66, 95% CI 0.57-0.74), and the local skin response index (κ=0.69, 95% CI 0.61-0.77) as measured using the quadratic-weighted Cohen κ. Overall, 77% (151/195) of submitted photos were good quality, and most image-derived metrics showed higher agreement in good-quality photos (8/14, 57% metrics had moderate-substantial κ) compared to poor-quality photos (4/14, 29% metrics had moderate-substantial κ). The peak local skin response had a moderate positive association with the lesion response at 90 days (Spearman ρ=0.556, P=.01).

This study demonstrates the utility of patient self-imaging for longitudinal assessment of the response to cryosurgery.

Figures