By Abby Van Voorhees, MD, July 01, 2011
In this month’s Acta Eruditorum column, Physician Editor Abby S. Van Voorhees, MD, talks with primary care physician Daniel G. Federman, MD, about his recent Archives in Dermatology article, “Outcomes of Referral to Dermatology for Suspicious Lesions: Implications for Teledermatology.”
Dr. Van Voorhees: What are the best features of teledermatology? What are its limitations?
Dr. Federman: Clearly, teledermatology has a potential role in the care of patients afflicted with skin problems. It may be very helpful in parts of the world where there are no dermatologists and in parts of the U.S. where patients may not have access to dermatologists. Teledermatology has already been shown to be associated with decreased time to diagnosis and surgical treatment when compared to the standard referral system. However, before this is widely implemented, we must be cognizant of specific limitations. Our findings suggest that teledermatology should not substitute for a full body skin examination (FBSE) when a single lesion is suspected of being potentially cancerous, if access to dermatologists is possible.
It is important to mention that our study did not test teledermatology directly. The study reviewed face-to-face in-office consults that took place with a dermatologist following referral by a primary care provider, but it shows the limitations of a lesion-specific approach to a teledermatology consult.
Dr. Van Voorhees: What patient population was studied? What diagnoses were you studying? How many patients were reviewed? How many of these patients had a history of prior non-melanoma skin cancer or melanoma?
Dr. Federman: We looked at a convenience sample of 400 patients referred to the dermatology clinic for a lesion suspected of being cancer. The consultation of a dermatologist was requested electronically through the VA’s system. Since the study was conducted at a VA medical center and since we only looked at lesions suspected of being cancerous, 98 percent of the patients were white males. The average age of our population was almost 78 years old. It is possible that the high number of “incidental cancers” found might be lower in a younger or more heterogeneous population. Since patients can receive care within the VA system or outside of the VA system, we could not ascertain how many patients had a previous history of malignancy due to the nature of our retrospective chart review. It would be interesting to see in future studies how these variables affect the outcome.
Dr. Van Voorhees: Tell us about your findings. What percentages of the index lesions warranted biopsies? What percentage of these were found to be malignant? What kinds of malignancies were identified?
Dr. Federman: A majority of the index lesions (56 percent) were thought to be benign by the dermatologist and were not biopsied. Of the 176 lesions that were biopsied by dermatologists, half (88) proved to be malignant, including 61 basal cell carcinomas, 21 squamous cell carcinomas, and five melanomas. Overall, we found nearly 40 percent of skin cancers were incidentally found, and were not the lesion of concern, including six melanomas. We were shocked to find that almost 10 percent of all incidentally found cancers were melanoma.[pagebreak]
Dr. Van Voorhees: Is there a pattern as to where these incidental lesions were located? What does this finding suggest?
Dr. Federman: A pattern seemed to emerge where the overwhelming majority of suspicious lesions found by non-dermatologists were on the face, head/neck area, or extremities, and incidental cancers were distributed more widely. This suggests, although it does not prove, that non-dermatologists may not be performing FBSEs with the patient completely undressed. Given the multitude of responsibilities that a primary care provider has, chronic disease management, health promotion, vaccinations, cancer screening, etc., it is quite possible that a FBSE may be overlooked, especially since there is a paucity of data proving that FBSEs lead to a decreased mortality from skin cancer.
However, our findings suggest that teledermatology should not substitute for a FBSE. Addressing a single lesion is only that — you’re looking at one lesion. Patients who have one lesion suspicious of cancer probably have risks for malignancies and can have other lesions in other parts of the body that could be suspicious. The genesis for this project was from my own clinical practice; I’d send patients to the dermatologist to have actinic keratoses frozen off the dorsum of their hand and the next time I’d see them they’d have new scars on their back or forehead. These were cancers that were found when I referred the patient.
Dr. Van Voorhees: You found yourself addressing one lesion and not realizing that another might be more pressing.
Dr. Federman: Yes, and I consider myself aware. One patient had a squamous cell carcinoma of the thumbnail; another had a melanoma detected incidentally. I realized this was potentially a big problem.
Dr. Van Voorhees: Do you think the problem is that primary care doctors are not performing a FBSE and just looking at the lesion the patient brings to their attention, or do they not recognize lesions in the first place?
Dr. Federman: I think it’s a combination. We’ve done studies through multiple modalities asking if primary care providers do FBSE; all of them showed very low frequency of performance by primary care providers. Given the lack of confidence of primary care providers about detecting skin lesions and lack of adequate teaching about potentially dangerous skin findings I can understand that.[pagebreak]
Dr. Van Voorhees: Were these patients considered at high risk for either non-melanoma skin cancer or melanoma?
Dr. Federman: They were elderly and white, so that is a risk, but due to the retrospective chart review nature of our study we weren’t able to ascertain how many people had other risk factors such as family history, outdoor sun exposure, or tanning bed use. We didn’t look for information about previous cancers, which would have elevated their risk category, because that information might not appear in the VA medical record system.
Dr. Van Voorhees: What were other limitations of this study? How would you like to extend this work going forward?
Dr. Federman: Again, our study, which was not a study of teledermatology per se, was conducted in an elderly population of white males and was restricted to lesions suspected of being cancerous. We don’t know if we can generalize about younger populations, women, or anything other than cutaneous malignancies. We can’t comment on other skin processes for which teledermatology might be used. We cannot determine whether teledermatology has a role if the non-dermatologist is mandated to perform a FBSE and include the images of ALL suspicious lesions. We hope future studies look at this question and the role of teledermatology in non-cancerous lesions in this population and others.
Dr. Federman is a professor of medicine at Yale University School of Medicine and firm chief for primary care in the VA Connecticut Healthcare System. His article was published in Archives of Dermatology, 147(5), 556560 (May 2011). doi:10.1001/archdermatol.2011.108.