- Email: [email protected]
The impact of total body photography on biopsy rate in patients from a pigmented lesion clinic
The impact of total body photography on biopsy rate in patients from a pigmented lesion clinic Jessica Risser, MD, MPH,a Zakiya Pressley, MD,a Emir Veledar, PhD,b Carl Washington, MD,a and Suephy C. Chen, MD, MSa,c Atlanta, Georgia Background: Total body cutaneous photography is increasingly being used by dermatologists to monitor patients at risk for the development of melanoma, but limited evidence exists regarding the impact of such photography on melanoma and melanoma-related outcomes. Objective: We sought to compare biopsy number in patients with multiple atypical nevi in their first year of care at our pigmented lesion clinic (PLC) between those who received total body skin examination alone and those who received total body skin examination and total body digital photography (TBDP). We sought to identify predictors of biopsy number and number of dysplastic nevi diagnosed in patients with multiple atypical nevi. Methods: A chart review was performed of patients attending the PLC during the years 1998 to 2003 to identify the number of biopsies performed in the first year of care. Patient demographics, melanoma risk factors, and melanoma outcome events were also abstracted from the charts. Results: The mean number of biopsies performed in patients in their first year of care at the PLC in those who did not receive TBDP was equal to the mean number of biopsies performed in patients who did receive TBDP (0.82 and 0.8, respectively). Linear regression analysis revealed that the interaction term between a lack of both personal history of melanoma and severe dysplastic nevi ( 0.930, P = .005) has a significant protective effect on the number of biopsies. Similar regression analysis also showed that the interaction term between a lack of both personal history of melanoma and of severe dysplastic nevi ( 1.209, P \.0001), increasing provider experience ( 0.047, P = .029), and increased number of biopsies before the initial PLC ( 0.028, P = .050) have a statistically significant protective effect on the number of dysplastic nevi diagnosed in the first year of PLC. TBDP did not have an effect on the number of biopsies or on the number of dysplastic nevi diagnosed in the first year of care at the PLC. Limitations: This study is limited by being retrospective in nature, having a small sample size, and having a short follow-up period. Conclusion: Overall, this small retrospective study does not provide evidence that would suggest that TBDP changes provider behavior in caring for patients at high risk for melanoma. Rather, our study supports the fact that a patient’s positive history of melanoma and a history of severe dysplastic nevi have the most significant impact on provider biopsy behavior, resulting in a lower threshold to biopsy suggestive lesions. ( J Am Acad Dermatol 2007;57:428-34.)
T
he incidence of melanoma continues to increase at an alarming rate in this country. Early diagnosis and treatment of melanoma
remain the best means of decreasing the number of deaths attributable to this potentially fatal malignancy. Several populations have been deemed to be
From the Department of Dermatologya and Center for Outcomes Research,b Emory University; and Department of Health Services Research and Development,c Atlanta Department of Veterans Affairs Medical Center. Drs Chen and Veledar are supported in part by a Mentored Patient Oriented Career Development Award (No. K23AR02185-01A1). Conflicts of interest: None declared. An abstract of this work was presented at the American Academy of Dermatology Annual Meeting, San Francisco, Calif, on March 4, 2006 and published in the poster abstract edition.
Accepted for publication February 25, 2007. Reprint requests: Suephy C. Chen, MD, MS, Department of Dermatology, Emory University, 101 Woodruff Circle, Atlanta, GA 30322. E-mail: [email protected]. Published online July 12, 2007. 0190-9622/$32.00 ª 2007 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2007.02.036
428
J AM ACAD DERMATOL
Risser et al 429
VOLUME 57, NUMBER 3
at increased risk for developing melanoma including patients with multiple nevi, especially those with multiple atypical nevi.1-3 Patients with multiple nevi have a relative risk of 5 to 12 of developing melanoma whereas those with multiple atypical nevi have a relative risk of 7 to 27.2 Given the elevated risk of developing melanoma in persons with multiple atypical nevi, increased regular skin cancer screening for this population is recommended.3-7 Total body skin examination (TBSE) by physicians has been the standard of care for performing routine skin cancer screening for such patients at high risk for melanoma. It is increasingly common practice to augment TBSE with photographic surveillance to monitor patients at high risk by increasing the objectivity of the TBSE.8,9 With the aid of baseline photographs, physicians are able to detect subtle changes in nevi that may have otherwise gone unnoticed and detect new pigmented lesions more easily. The use of total body photography in patients with multiple atypical nevi has been suggested by several studies6,10-12 and it has been suggested that photography may increase the sensitivity and specificity of melanoma diagnoses.12-14 Using total body photographs in conjunction with traditional TBSE may also increase the threshold at which physicians excise suggestive lesions in patients with multiple clinically atypical nevi, leading to decreased numbers of perhaps unnecessary biopsies in patients with multiple atypical nevi.10,12 However, limited data exist testing such a hypothesis and the converse may also be true: the use of total body photographs may result in increased numbers of biopsies in patients. This study seeks to determine the effect of total body digital photography (TBDP) on biopsy rates in patients with multiple atypical nevi presenting to our pigmented lesion clinic (PLC). Given that TBDP was introduced at our PLC in December 2000, this study compares biopsy rates in the first year of care between patients who received TBSE alone before December 2000, and those who received both TBSE and TBDP after December 2000.
METHODS Study population A chart review was performed of patients who attended our PLC between the years 1998 to 2003. To identify patients who would be eligible for the study, a database of the patients who had attended the PLC between 1998 and 2003 was generated (n = 3632). Patients were considered for the study if they had attended the PLC at least 3 times (n = 946). Patients who met these initial criteria were then stratified into a pre-TBDP group composed of patients who
presented to the PLC before December 2000 (n = 565) and a posttotal digital photography group composed of those patients who presented to the PLC after December 2000 (n = 381). Patients who received TBDP were characteristically those patients who presented to the PLC with multiple atypical nevi, with or without a personal history of melanoma. Not all patients presenting to the PLC received photography, however. This latter group consisted of patients who had melanomas in the past, but did not have multiple atypical nevi. There were also patients who opted to not undergo photography, mostly because of cost, even if they had multiple atypical nevi. To select potentially eligible patients from our database, we randomly chose and obtained charts from each of the two groups. Obtained charts were then reviewed individually and accepted for the study if the following eligibility criteria were met: being an actual patient of the PLC, having multiple atypical moles as indicated by a physician’s note, and having at least 1 year of follow-up after the initial visit to the PLC. Sample size and power calculation Charts were obtained until we had at least 59 patients in each group. This number was determined by a sample size calculation that ensured 80% power to detect a difference of 0.7 biopsies in the first year of care between each group assuming an acceptable type I error of 0.05. To our knowledge, there is no current literature available that establishes a clinically significant biopsy difference between two diagnostic modalities for melanoma detection. Therefore, we determined our 0.7 effect size from a preliminary analysis of 50 charts from each group and based our sample size calculation on these preliminary results. Data abstracted Data abstracted from the charts included: age at the second visit to the PLC, sex, melanoma history, family history of melanoma, number of biopsies before entering the PLC, number and severity of biopsies consistent with dysplastic nevi before entering the PLC, frequency of visits to the PLC in the first year of visiting the PLC, number of providers seen, provider experience during the first year of visits to the PLC, number of biopsies in the first year of attending the PLC, and the pathology reports of those biopsies. For our study, specimens were reviewed by board-certified dermatopathologists who defined a dysplastic nevus to include the following criteria: (1) bridging of melanocytes/melanocytic nests along the rete ridges; (2) fibrosis; and (3)
430 Risser et al
J AM ACAD DERMATOL SEPTEMBER 2007
melanocytic atypia. Furthermore, mild, moderate, and severe dysplasia was differentiated by the cytology of the individual melanocytes and the architecture of the nevus. Data were collected on the number of biopsies before entering the PLC from out institution and outside pathology reports. Each patient’s number and severity of dysplastic nevi before entering the PLC was determined using the same criteria. Total biopsy number in the first year of care included only those biopsies performed at our PLC. For patients who received only a TBSE, biopsy number during the first year of care was calculated from the second visit to the PLC. For those patients who received both TBSE and TBDP, biopsy number during the first year of care was calculated from the first visit during which the patient had established photographs. This was typically the office visit after the initial visit given that most patients had photographs taken at their initial visits to the PLC. Therefore, our outcome variables of interest, including number of biopsies, dysplastic nevi, and melanomas diagnosed in 1 year, were measured from the same starting point in each group, the second visit. Patients who received TBDP had regional photographs taken of their body by one of two experienced technicians and close-up photographs were not routinely taken. Of note, although dermoscopy (epiluminescence microscopy) is currently used at the PLC, it was not used by any providers before 2003 and, thus, is not applicable to the time period examined in our study. Statistical analyses Descriptive statistics to describe the patient population were determined. To compare the number of biopsies in the first year of care between those with TBDP and those without, the nonparametric Wilcoxon rank sum test was performed. Multiple linear regression was used to determine the relationship of the total number of biopsies in the first year of visits at the PLC with presence of TBDP, adjusting for the variables outlined in the previous paragraph. Using a backward stepwise regression approach, with appropriate interactions terms, only those variables that achieved a type I error equal to or less than .05 were considered statistically significant. The regression model was repeated with the number of dysplastic nevi (not including melanoma) diagnosed in the first year as the outcome variable. Finally, to establish whether the severity of dysplastic nevi diagnosed in the first year of visits to the PLC differed based on whether or not patients had received TBDP, we performed a chi-square test of proportions. To increase the clinical relevance of this result, we also performed a chi-square test of
proportions to determine whether there was a difference in the proportion of people given the diagnosis of severe lesions between the two groups. For the purposes of our study, those patients given a diagnosis of mild, mild-moderate, and moderate dysplastic nevi were categorized as having nonsevere lesions. Those patients with moderate-severe and severe dysplastic nevi were categorized as having severe lesions. Those patients who were given the diagnosis of melanoma were also identified and a chi-square test was performed to determine whether there was a difference in the proportion of people given the diagnosis of melanomas in each group.
RESULTS Table I provides basic demographic information about our patient population and a summary of melanoma risk factors and melanoma outcome events. The mean age for patients in our population was approximately 36 years, with 55.5% of patients being female. There were no statistically significant differences in age, sex, personal history of melanoma, or family history of melanoma between our patients who received TBDP and those who did not. There were no statistically significant differences found between those patients with and without TBDP in the mean number of severe dysplastic nevi diagnosed before their initial visit to the PLC (1.4 compared with 1.1) or the mean number of biopsies performed before their initial visit to the PLC (3.5 compared with 3.5). Because the numbers of dysplastic nevi and biopsies in the first year of care were not normally distributed, we analyzed these differences using nonparamentric testing. We found no statistically significant differences between the two groups for the median number of severe dysplastic nevi diagnosed (P = .19) and the median number of biopsies performed (P = .47) before the patient’s visit to the PLC. The mean provider experience and mean number of visits in the first year of care at the PLC in those who did receive TBDP was equal to those patients who did not receive TBDP (Table II). In addition, 98% of patients were seen by the same provider at each visit and there were a total of 3 providers throughout the time period of this study. The mean number of biopsies performed on patients in their first year of care at the PLC in those who did not receive TBDP was equal the mean number of biopsies performed on patients who did receive TBDP (0.82 and 0.8, respectively). Nonparametric testing between the two groups revealed no statistically significant difference in the median number of
Risser et al 431
J AM ACAD DERMATOL VOLUME 57, NUMBER 3
Table I. Patient characteristics
Patient characteristics
Demographics Age, y, mean; median Female (%) No. of biopsies before visit to PLC, mean; median Melanoma risk factors Personal history of MM (%) Family history of MM (%) No. of severe dysplastic nevi before visit to PLC, mean; median Events in first year Provider experience, y No. of visits to PLC, mean; median No. of biopsies, mean; median No. of dysplastic nevi, mean; median
Overall population n = 128
Patients with TBDP n = 64
Patients without TBDP n = 64
35.7; 33.6 71 (55.5) 3.5: 2.0
35.5; 33.7 35 (54.7) 3.5: 2.0
35.9; 33.6 36 (56.2) 3.5: 2.0
.36 .89 .47
52 (40.6) 24 (18.8) 1.3; 0.0
22 (34.4) 16 (25) 1.4; 1.0
30 (46.9) 8 (12.5) 1.1; 0.0
.15 .07 .19
5.5; 5.0 2.38; 2.0 .81; 0.0 .39; 0.0
5.6; 3.0 2.4; 2.0 0.82; 0.0 .39; 0.0
5.5; 6.0 2.3; 2.0 0.80; 0.0 .39; 0.0
.89 .7483 .43 .50
P value*
PLC, Pigmented lesion clinic; MM, malignant melanoma; TBDP, total body digital photography. *Denotes P value result from nonparametric comparison testing.
Table II. Final linear regression model: Number of biopsies Variable
Regression coefficient
Intercept Total body digital photography Provider experience Personal history of melanoma X history of severe dysplastic nevi
SE
P value
1.835 0.007
0.353 \.0001 0.201 .971
0.041 0.903
0.031 0.312
.191 .005
X, Interaction between the variables.
Table III. Final linear regression model: Number of dysplastic nevi Variable
Intercept Total body digital photography No. of biopsies before visit to PLC Provider experience Personal history of melanoma X history of severe dysplastic nevi
Regression coefficient
SE
P value
1.789 0.054
0.259 0.134
\.0001 .690
0.028
0.014
.050
0.047 1.209
0.021 0.218
.029 \.0001
PLC, Pigmented lesion clinic; X, interaction between the variables.
biopsies performed (P = .43). Overall, there were 51 biopsies performed in 29 patients (1.76 biopsies/ patient of those biopsied) without photographs and 53 biopsies performed in 28 patients (1.89 biopsies/ patient of those biopsied) with photographs. The mean number of dysplastic nevi diagnosed in patients with TBDP in the first year of care was also the same in patients who did not receive TBDP (0.39 and 0.39, respectively). Nonparametric testing between the two groups revealed no statistically significant difference in the median number of dysplastic nevi diagnosed in the first year of care (P = .5). Of the dysplastic nevi diagnosed in our population, 26 dysplastic nevi (3 being severe) and 3 melanomas were diagnosed in 19 patients (1.53 lesions/patient) without photographs, whereas 25 dysplastic nevi (6 being severe) and no melanomas were diagnosed in 16 patients (1.56 lesions/patient) with photographs. Of the 3 patients given the
diagnosis of melanoma, the Breslow thicknesses of their melanomas were 0.3, 0.44, and 1.1 mm. Table II details our regression model of the predictors of the number of biopsies in the first year of care at the PLC. The interaction term between a lack of personal history of melanoma and severe dysplastic nevi (e0.930, P = .005) has a significant protective effect on the number of biopsies. TBDP was not a statistically significant predictor of the number of biopsies (0.007, P = .791) in our population. Table III details our regression model of predictors of the number of dysplastic nevi diagnosed in the first year of care at the PLC. The interaction term between a lack of personal history of melanoma and severe dysplastic nevi (e1.209, P\.0001), increasing years of provider experience (e0.047, P = .029), and increasing number of biopsies before the initial PLC (e0.028, P = .050) have a statistically significant
432 Risser et al
J AM ACAD DERMATOL SEPTEMBER 2007
protective effect on the number of dysplastic nevi diagnosed in the first year of PLC. TBDP did not have a statistically significant effect on the number of dysplastic nevi diagnosed in the first year of PLC visits (0.054, P = .690). We next compared the proportion of severe lesions diagnosed in the two groups to determine whether a clinically significant difference between the groups existed. We first investigated the proportion of the detected dysplastic lesions that were severe lesions and found no statistically significant difference between the TBDP and the non-TBDP groups (chi-square = 1.362, P = .243). We then investigated the proportion of people given the diagnosis of severe lesions and found no statistically significant difference between the two groups (chisquare = 1.076, P = .300). There was no statistically significant difference in the proportion of people given the diagnosis of melanoma between the two groups (chi-square = 3.072, P = .08). The very small number of melanomas diagnosed precluded further analysis of this group.
DISCUSSION To our knowledge, this study is one of the first to attempt to objectively evaluate the effect of TBDP used by dermatologists on biopsy rates in patients at increased risk for melanoma. This is an important question that needs to be addressed because, if TBDP is found to be a useful adjunctive screening measure in this patient population, then it has the potential to decrease morbidity, mortality, the cost of care, or a combination of these. In our study, we were not able to demonstrate that the use of TBDP in patients at high risk for melanoma has an impact during a 1-year study period on the total number of biopsies or the number of dysplastic nevi diagnosed in patients. The most significant factor predicting biopsy number included the interaction of having a personal history of melanoma and a personal history of severe dysplastic nevi. In our study, the negative value of this interaction term illustrates that not having a history of melanoma and dysplastic nevi means one is less likely to be biopsied. This could also be interpreted to mean that those who have both a personal history of melanoma and of dysplastic nevi are more likely to be biopsied, which intuitively makes sense. We also found that those patients without both a personal history of melanoma and dysplastic nevi, those seen by providers with greater years of experience, and those with increasing numbers of biopsies before coming to the PLC are less likely to have dysplastic nevi diagnosed in the first year of care. These findings also have face validity. One would
expect that patients with fewer risk factors (personal history of melanoma and dysplastic nevi) would have fewer dysplastic nevi. One would also expect providers with greater years of experience to not biopsy mildly dysplastic nevi, but only those with severe dysplasia. Although our model for number of biopsies did not significantly identify provider experience as a predictor, it did show a trend toward predicting biopsy number (e.047, P = .19). Lastly, the more dysplastic nevi that were removed before attending the PLC, the fewer left behind to biopsy in the PLC. Our results are similar to those found by Hanrahan et al,13 who conducted a randomized prospective trial evaluating the impact of photography on biopsy rates and pathology of biopsied lesions by primary care physicians in older men from Australia. The researchers found that fewer numbers of benign pigmented lesions (typically seborrheic keratoses) were removed in patients with photographs, although this was not true for dysplastic or malignant lesions where there was no difference in their removal between those with photographs and those without, in concordance with our findings. This study also identified that photographs increased the specificity for detecting nonmelanoma skin cancers, but could not comment on whether photographs allowed for more specificity in diagnosing melanoma, secondary to low numbers of melanoma diagnosed.13 Our results challenge a chart review by Feit et al,12 which concluded that photographic follow-up was valuable in avoiding unnecessary biopsy in suggestive, but stable lesions. This same study further demonstrated that photographically assisted followup helped detect new and subtly changing melanomas that did not meet the classic clinical features of melanoma.12 Our study differs in that we have a comparison group that allowed us to demonstrate a lack of statistically significant difference between the number of biopsies performed between those with digital photographs and those who did not have TBDP performed. Interestingly, in our study 3 melanomas were diagnosed in the cohort followed up with TBSE alone and none in the group with photographs. Although the finding between the two groups is provocative, increased numbers of melanoma need to have been diagnosed to be able to draw any conclusions about any differences observed. Although at first glance the results of this study appear to reveal a disappointing trend, it is likely that our study does not capture the time frame in which TBDP is most useful to prevent unnecessary biopsies. Our study population reflects both patients who have never seen a dermatologist and those who had
J AM ACAD DERMATOL
Risser et al 433
VOLUME 57, NUMBER 3
already been followed up by a dermatologist, and referred when TBDP became available. The latter group may have already had the majority of their dysplastic nevi removed; in our data analyses, we discovered that there are usually a large number of biopsies of atypical nevi that are taken at a person’s first visit to the PLC. Although we could have included the first visit in our follow-up period, we thought that because the first visit involved taking the photographs, rather than using the photographs, in the TBDP group, we needed to begin the follow-up period with the second visit. It is also possible that our study underestimated the full potential of TBDP to decrease the number of biopsies because the duration of the retrospective review was for only 1 year. One year may not be enough time for the TBDP to detect morphologic change of atypical nevi, given that many melanomas are slow growing. Our patient cohort was also relatively young and, thus, may not have had sufficient melanomas to detect in our 1-year period. Other potential limitations to our study include its design as a retrospective chart review. With this design, it was not possible to discern the exact reason a biopsy was performed. Biopsies are performed based both by the clinician’s suspicion and to allay patient fears and this could not be accounted for in our study. In addition, there is a potential selection bias in that a few patients with multiple atypical nevi opted to not have photography; these patients cited cost as the primary reason to not pursue photography. Lastly, this study reviews the experience of a relatively small number of providers and patients at a single academic institution, thus, the ability to generalize the results are limited. Despite these limitations, until a randomized controlled trial can be performed, retrospective studies provide our best evidence and generate hypotheses for future studies. Our methods used rigid eligibility criteria for the inclusion of patients in our study. In addition, our cohorts are matched and can be expected in terms of patient characteristics (demographics, personal and family history of melanoma, number of biopsies and severe dysplastic nevi at the time of presentation to PLC) and provider characteristics (mean provider experience and total number of visits in the first year of care at the PLC). Thus, given our findings, one could argue that TBDP may be most beneficial to those patients with atypical mole syndrome who have not already had multiple atypical nevi removed. TBDP may also potentially be useful during an extended period of time, beyond the short 1-year period that we studied. Lastly, TBDP may provide relief from cancer worry for the at-risk population. Preliminary data indicate
that patients are highly satisfied with the digital photographs and are less worried about changes in pigmented lesions going unnoticed.15 Although a primary purpose of TBDP is to detect melanomas earlier and to avoid unnecessary biopsies, photography may change the criteria by which dermatologists manage suggestive lesions. One possibility may be that physicians become more complacent with photographs and not biopsy those lesions with very subtle changes. However, the reverse may be true too. Physicians may increase their tendency to biopsy because previously undetectable changes in nevi are now apparent with baseline photographs. It is important that future prospective studies analyzing TBDP as an adjunct screening measure explore this issue. The practice of following up patients with TBDP continues to grow, without the knowledge of how and when TBDP benefits patients. Future studies should be directed at prospectively evaluating whether TBDP is both more sensitive and specific than TBSE alone for detecting dysplastic nevi and, more importantly, melanoma. Based on our study results and our observations, these studies may need to be conducted during a period of time longer than 1 year in duration and also include patients who have not already undergone extensive biopsies. Future studies should also evaluate the merit of dermoscopy being used in conjunction with total body examinations and total body photographs. Although during the period of the study, the dermatologists involved did not use dermoscopy, studies have subsequently shown that dermoscopy is a useful adjunct to the screening of melanoma, particularly during a followup period of between 3 to 21 months.16-19 Lastly, cost-effectiveness and psychologic impact should be included as study outcomes in future efforts.
REFERENCES 1. Holly E, Kelly J, Shpall S, Chiu S-H. Number of melanocytic nevi as a major risk factor for malignant melanoma. J Am Acad Dermatol 1987;17:459-68. 2. Rhodes A, Weinstock M, Fitzpatrick T, Mihm M, Sober A. Risk factors for cutaneous melanoma: a practical method of recognizing predisposed individuals. JAMA 1987;258: 3146-54. 3. Tucker M, Halpern A, Holly E, Hartge P, Elder D, Sagebiel R, et al. Clinically recognized dysplastic nevi: a central risk factor for cutaneous melanoma. JAMA 1997;277:1439-44. 4. Masri G, Clark W, Guerry D, Halpern A, Thompson C, Elder D. Screening and surveillance of patients at high risk for malignant melanoma result in earlier detection of disease. J Am Acad Dermatol 1990;22:1042-8. 5. Marghoob A, Slade J, Kopf A, Salopek T, Rigel D, Bart R. Risk of developing multiple primary cutaneous melanomas in patients with the classic atypical-mole syndrome: a case-control study. Br J Dermatol 1996;135:704-11.
434 Risser et al
J AM ACAD DERMATOL SEPTEMBER 2007
6. Tierstan A, Grin C, Gottlieb G, Bart R, Rigel D, Friedman R, et al. Prospective follow-up for malignant melanoma in patients with atypical-mole (dysplastic-nevus) syndrome. J Dermatol Surg Oncol 1991;17:44-8. 7. Freedberg K, Geller A, Miller D, Lew R, Koh H. Screening for malignant melanoma: a cost-effective analysis. J Am Acad Dermatol 1999;41:738-45. 8. Tripp J, Kopf A, Marghoob A, Bart R. Management of dysplastic nevi: a survey of fellows of the American Academy of Dermatology. J Am Acad Dermatol 2002;46:674-82. 9. Nehal K, Oliveria S, Marghoob A, Christos P, Dusza S, Tromberg J, et al. Use of and beliefs about baseline photography in the management of patients with pigmented lesions: a survey of dermatology residency programs in the United States. Melanoma Res 2002;12:161-7. 10. Banky J, Kelly J, English D, Yeatman J, Dowling J. Incidence of new and changed nevi and melanomas detected using baseline images and dermoscopy in patients at high risk for melanoma. Arch Dermatol 2005;141:998-1006. 11. Kelly J, Yeatman J, Regalia C, Mason G, Henham A. A high incidence of melanoma found in patients with multiple dysplastic naevi by photographic surveillance. Med J Aust 1997;167:191-4. 12. Feit N, Dusza S, Marghoob A. Melanomas detected with the aid of total cutaneous photography. Br J Dermatol 2004;150: 706-14.
13. Hanrahan P, D’Este C, Menzies S, Plummer T, Hersey P. A randomized trial of skin photography as an aid to screening skin lesions in older males. J Med Screen 2002;9:128-32. 14. Halpern A. Total body skin imaging as an aid to melanoma detection. Semin Cutan Med Surg 2003;22:2-8. 15. Pressley Z, Delong L, Risser J, Curiel-Lewandrowski C, Chen S. Quality of life and cancer anxiety impact of total body digital photography in patients with atypical mole syndrome [Abstract]. J Am Acad Dermatol 2007;56:2203. 16. Carli P, De Giorgi V, Crocetti E, Mannone F, Massi D, Chiarugi A, Giannotti B. Improvement of malignant/benign ratio in excised melanocytic lesions in the ‘dermoscopy era’: a retrospective study 1197-2001. Br J Dermatol 2004;150: 687-92. 17. Carli P, De Giorgi V, Chiarugi A, Nardini P, Weinstock M, Crocetti E, et al. Addition of dermoscopy to conventional naked-eye examination in melanoma screening: a randomized study. J Am Acad Dermatol 2004;50:683-9. 18. Carli P, Ghigliotti G, Ghone M, Chiarugi A, Crocetti E, Astorino S, et al. Baseline factors influencing decisions on digital followup of melanocytic lesions in daily practice: an Italian multicenter survey. J Am Acad Dermatol 2006;55:256-62. 19. Menzies S, Gutenev A, Avramidis M, Batrac A, McCarthy W. Short-term digital surface microscopic monitoring of atypical or changing melanocytic lesions. Arch Dermatol 2001;137: 1583-9.
T
he incidence of melanoma continues to increase at an alarming rate in this country. Early diagnosis and treatment of melanoma
remain the best means of decreasing the number of deaths attributable to this potentially fatal malignancy. Several populations have been deemed to be
From the Department of Dermatologya and Center for Outcomes Research,b Emory University; and Department of Health Services Research and Development,c Atlanta Department of Veterans Affairs Medical Center. Drs Chen and Veledar are supported in part by a Mentored Patient Oriented Career Development Award (No. K23AR02185-01A1). Conflicts of interest: None declared. An abstract of this work was presented at the American Academy of Dermatology Annual Meeting, San Francisco, Calif, on March 4, 2006 and published in the poster abstract edition.
Accepted for publication February 25, 2007. Reprint requests: Suephy C. Chen, MD, MS, Department of Dermatology, Emory University, 101 Woodruff Circle, Atlanta, GA 30322. E-mail: [email protected]. Published online July 12, 2007. 0190-9622/$32.00 ª 2007 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2007.02.036
428
J AM ACAD DERMATOL
Risser et al 429
VOLUME 57, NUMBER 3
at increased risk for developing melanoma including patients with multiple nevi, especially those with multiple atypical nevi.1-3 Patients with multiple nevi have a relative risk of 5 to 12 of developing melanoma whereas those with multiple atypical nevi have a relative risk of 7 to 27.2 Given the elevated risk of developing melanoma in persons with multiple atypical nevi, increased regular skin cancer screening for this population is recommended.3-7 Total body skin examination (TBSE) by physicians has been the standard of care for performing routine skin cancer screening for such patients at high risk for melanoma. It is increasingly common practice to augment TBSE with photographic surveillance to monitor patients at high risk by increasing the objectivity of the TBSE.8,9 With the aid of baseline photographs, physicians are able to detect subtle changes in nevi that may have otherwise gone unnoticed and detect new pigmented lesions more easily. The use of total body photography in patients with multiple atypical nevi has been suggested by several studies6,10-12 and it has been suggested that photography may increase the sensitivity and specificity of melanoma diagnoses.12-14 Using total body photographs in conjunction with traditional TBSE may also increase the threshold at which physicians excise suggestive lesions in patients with multiple clinically atypical nevi, leading to decreased numbers of perhaps unnecessary biopsies in patients with multiple atypical nevi.10,12 However, limited data exist testing such a hypothesis and the converse may also be true: the use of total body photographs may result in increased numbers of biopsies in patients. This study seeks to determine the effect of total body digital photography (TBDP) on biopsy rates in patients with multiple atypical nevi presenting to our pigmented lesion clinic (PLC). Given that TBDP was introduced at our PLC in December 2000, this study compares biopsy rates in the first year of care between patients who received TBSE alone before December 2000, and those who received both TBSE and TBDP after December 2000.
METHODS Study population A chart review was performed of patients who attended our PLC between the years 1998 to 2003. To identify patients who would be eligible for the study, a database of the patients who had attended the PLC between 1998 and 2003 was generated (n = 3632). Patients were considered for the study if they had attended the PLC at least 3 times (n = 946). Patients who met these initial criteria were then stratified into a pre-TBDP group composed of patients who
presented to the PLC before December 2000 (n = 565) and a posttotal digital photography group composed of those patients who presented to the PLC after December 2000 (n = 381). Patients who received TBDP were characteristically those patients who presented to the PLC with multiple atypical nevi, with or without a personal history of melanoma. Not all patients presenting to the PLC received photography, however. This latter group consisted of patients who had melanomas in the past, but did not have multiple atypical nevi. There were also patients who opted to not undergo photography, mostly because of cost, even if they had multiple atypical nevi. To select potentially eligible patients from our database, we randomly chose and obtained charts from each of the two groups. Obtained charts were then reviewed individually and accepted for the study if the following eligibility criteria were met: being an actual patient of the PLC, having multiple atypical moles as indicated by a physician’s note, and having at least 1 year of follow-up after the initial visit to the PLC. Sample size and power calculation Charts were obtained until we had at least 59 patients in each group. This number was determined by a sample size calculation that ensured 80% power to detect a difference of 0.7 biopsies in the first year of care between each group assuming an acceptable type I error of 0.05. To our knowledge, there is no current literature available that establishes a clinically significant biopsy difference between two diagnostic modalities for melanoma detection. Therefore, we determined our 0.7 effect size from a preliminary analysis of 50 charts from each group and based our sample size calculation on these preliminary results. Data abstracted Data abstracted from the charts included: age at the second visit to the PLC, sex, melanoma history, family history of melanoma, number of biopsies before entering the PLC, number and severity of biopsies consistent with dysplastic nevi before entering the PLC, frequency of visits to the PLC in the first year of visiting the PLC, number of providers seen, provider experience during the first year of visits to the PLC, number of biopsies in the first year of attending the PLC, and the pathology reports of those biopsies. For our study, specimens were reviewed by board-certified dermatopathologists who defined a dysplastic nevus to include the following criteria: (1) bridging of melanocytes/melanocytic nests along the rete ridges; (2) fibrosis; and (3)
430 Risser et al
J AM ACAD DERMATOL SEPTEMBER 2007
melanocytic atypia. Furthermore, mild, moderate, and severe dysplasia was differentiated by the cytology of the individual melanocytes and the architecture of the nevus. Data were collected on the number of biopsies before entering the PLC from out institution and outside pathology reports. Each patient’s number and severity of dysplastic nevi before entering the PLC was determined using the same criteria. Total biopsy number in the first year of care included only those biopsies performed at our PLC. For patients who received only a TBSE, biopsy number during the first year of care was calculated from the second visit to the PLC. For those patients who received both TBSE and TBDP, biopsy number during the first year of care was calculated from the first visit during which the patient had established photographs. This was typically the office visit after the initial visit given that most patients had photographs taken at their initial visits to the PLC. Therefore, our outcome variables of interest, including number of biopsies, dysplastic nevi, and melanomas diagnosed in 1 year, were measured from the same starting point in each group, the second visit. Patients who received TBDP had regional photographs taken of their body by one of two experienced technicians and close-up photographs were not routinely taken. Of note, although dermoscopy (epiluminescence microscopy) is currently used at the PLC, it was not used by any providers before 2003 and, thus, is not applicable to the time period examined in our study. Statistical analyses Descriptive statistics to describe the patient population were determined. To compare the number of biopsies in the first year of care between those with TBDP and those without, the nonparametric Wilcoxon rank sum test was performed. Multiple linear regression was used to determine the relationship of the total number of biopsies in the first year of visits at the PLC with presence of TBDP, adjusting for the variables outlined in the previous paragraph. Using a backward stepwise regression approach, with appropriate interactions terms, only those variables that achieved a type I error equal to or less than .05 were considered statistically significant. The regression model was repeated with the number of dysplastic nevi (not including melanoma) diagnosed in the first year as the outcome variable. Finally, to establish whether the severity of dysplastic nevi diagnosed in the first year of visits to the PLC differed based on whether or not patients had received TBDP, we performed a chi-square test of proportions. To increase the clinical relevance of this result, we also performed a chi-square test of
proportions to determine whether there was a difference in the proportion of people given the diagnosis of severe lesions between the two groups. For the purposes of our study, those patients given a diagnosis of mild, mild-moderate, and moderate dysplastic nevi were categorized as having nonsevere lesions. Those patients with moderate-severe and severe dysplastic nevi were categorized as having severe lesions. Those patients who were given the diagnosis of melanoma were also identified and a chi-square test was performed to determine whether there was a difference in the proportion of people given the diagnosis of melanomas in each group.
RESULTS Table I provides basic demographic information about our patient population and a summary of melanoma risk factors and melanoma outcome events. The mean age for patients in our population was approximately 36 years, with 55.5% of patients being female. There were no statistically significant differences in age, sex, personal history of melanoma, or family history of melanoma between our patients who received TBDP and those who did not. There were no statistically significant differences found between those patients with and without TBDP in the mean number of severe dysplastic nevi diagnosed before their initial visit to the PLC (1.4 compared with 1.1) or the mean number of biopsies performed before their initial visit to the PLC (3.5 compared with 3.5). Because the numbers of dysplastic nevi and biopsies in the first year of care were not normally distributed, we analyzed these differences using nonparamentric testing. We found no statistically significant differences between the two groups for the median number of severe dysplastic nevi diagnosed (P = .19) and the median number of biopsies performed (P = .47) before the patient’s visit to the PLC. The mean provider experience and mean number of visits in the first year of care at the PLC in those who did receive TBDP was equal to those patients who did not receive TBDP (Table II). In addition, 98% of patients were seen by the same provider at each visit and there were a total of 3 providers throughout the time period of this study. The mean number of biopsies performed on patients in their first year of care at the PLC in those who did not receive TBDP was equal the mean number of biopsies performed on patients who did receive TBDP (0.82 and 0.8, respectively). Nonparametric testing between the two groups revealed no statistically significant difference in the median number of
Risser et al 431
J AM ACAD DERMATOL VOLUME 57, NUMBER 3
Table I. Patient characteristics
Patient characteristics
Demographics Age, y, mean; median Female (%) No. of biopsies before visit to PLC, mean; median Melanoma risk factors Personal history of MM (%) Family history of MM (%) No. of severe dysplastic nevi before visit to PLC, mean; median Events in first year Provider experience, y No. of visits to PLC, mean; median No. of biopsies, mean; median No. of dysplastic nevi, mean; median
Overall population n = 128
Patients with TBDP n = 64
Patients without TBDP n = 64
35.7; 33.6 71 (55.5) 3.5: 2.0
35.5; 33.7 35 (54.7) 3.5: 2.0
35.9; 33.6 36 (56.2) 3.5: 2.0
.36 .89 .47
52 (40.6) 24 (18.8) 1.3; 0.0
22 (34.4) 16 (25) 1.4; 1.0
30 (46.9) 8 (12.5) 1.1; 0.0
.15 .07 .19
5.5; 5.0 2.38; 2.0 .81; 0.0 .39; 0.0
5.6; 3.0 2.4; 2.0 0.82; 0.0 .39; 0.0
5.5; 6.0 2.3; 2.0 0.80; 0.0 .39; 0.0
.89 .7483 .43 .50
P value*
PLC, Pigmented lesion clinic; MM, malignant melanoma; TBDP, total body digital photography. *Denotes P value result from nonparametric comparison testing.
Table II. Final linear regression model: Number of biopsies Variable
Regression coefficient
Intercept Total body digital photography Provider experience Personal history of melanoma X history of severe dysplastic nevi
SE
P value
1.835 0.007
0.353 \.0001 0.201 .971
0.041 0.903
0.031 0.312
.191 .005
X, Interaction between the variables.
Table III. Final linear regression model: Number of dysplastic nevi Variable
Intercept Total body digital photography No. of biopsies before visit to PLC Provider experience Personal history of melanoma X history of severe dysplastic nevi
Regression coefficient
SE
P value
1.789 0.054
0.259 0.134
\.0001 .690
0.028
0.014
.050
0.047 1.209
0.021 0.218
.029 \.0001
PLC, Pigmented lesion clinic; X, interaction between the variables.
biopsies performed (P = .43). Overall, there were 51 biopsies performed in 29 patients (1.76 biopsies/ patient of those biopsied) without photographs and 53 biopsies performed in 28 patients (1.89 biopsies/ patient of those biopsied) with photographs. The mean number of dysplastic nevi diagnosed in patients with TBDP in the first year of care was also the same in patients who did not receive TBDP (0.39 and 0.39, respectively). Nonparametric testing between the two groups revealed no statistically significant difference in the median number of dysplastic nevi diagnosed in the first year of care (P = .5). Of the dysplastic nevi diagnosed in our population, 26 dysplastic nevi (3 being severe) and 3 melanomas were diagnosed in 19 patients (1.53 lesions/patient) without photographs, whereas 25 dysplastic nevi (6 being severe) and no melanomas were diagnosed in 16 patients (1.56 lesions/patient) with photographs. Of the 3 patients given the
diagnosis of melanoma, the Breslow thicknesses of their melanomas were 0.3, 0.44, and 1.1 mm. Table II details our regression model of the predictors of the number of biopsies in the first year of care at the PLC. The interaction term between a lack of personal history of melanoma and severe dysplastic nevi (e0.930, P = .005) has a significant protective effect on the number of biopsies. TBDP was not a statistically significant predictor of the number of biopsies (0.007, P = .791) in our population. Table III details our regression model of predictors of the number of dysplastic nevi diagnosed in the first year of care at the PLC. The interaction term between a lack of personal history of melanoma and severe dysplastic nevi (e1.209, P\.0001), increasing years of provider experience (e0.047, P = .029), and increasing number of biopsies before the initial PLC (e0.028, P = .050) have a statistically significant
432 Risser et al
J AM ACAD DERMATOL SEPTEMBER 2007
protective effect on the number of dysplastic nevi diagnosed in the first year of PLC. TBDP did not have a statistically significant effect on the number of dysplastic nevi diagnosed in the first year of PLC visits (0.054, P = .690). We next compared the proportion of severe lesions diagnosed in the two groups to determine whether a clinically significant difference between the groups existed. We first investigated the proportion of the detected dysplastic lesions that were severe lesions and found no statistically significant difference between the TBDP and the non-TBDP groups (chi-square = 1.362, P = .243). We then investigated the proportion of people given the diagnosis of severe lesions and found no statistically significant difference between the two groups (chisquare = 1.076, P = .300). There was no statistically significant difference in the proportion of people given the diagnosis of melanoma between the two groups (chi-square = 3.072, P = .08). The very small number of melanomas diagnosed precluded further analysis of this group.
DISCUSSION To our knowledge, this study is one of the first to attempt to objectively evaluate the effect of TBDP used by dermatologists on biopsy rates in patients at increased risk for melanoma. This is an important question that needs to be addressed because, if TBDP is found to be a useful adjunctive screening measure in this patient population, then it has the potential to decrease morbidity, mortality, the cost of care, or a combination of these. In our study, we were not able to demonstrate that the use of TBDP in patients at high risk for melanoma has an impact during a 1-year study period on the total number of biopsies or the number of dysplastic nevi diagnosed in patients. The most significant factor predicting biopsy number included the interaction of having a personal history of melanoma and a personal history of severe dysplastic nevi. In our study, the negative value of this interaction term illustrates that not having a history of melanoma and dysplastic nevi means one is less likely to be biopsied. This could also be interpreted to mean that those who have both a personal history of melanoma and of dysplastic nevi are more likely to be biopsied, which intuitively makes sense. We also found that those patients without both a personal history of melanoma and dysplastic nevi, those seen by providers with greater years of experience, and those with increasing numbers of biopsies before coming to the PLC are less likely to have dysplastic nevi diagnosed in the first year of care. These findings also have face validity. One would
expect that patients with fewer risk factors (personal history of melanoma and dysplastic nevi) would have fewer dysplastic nevi. One would also expect providers with greater years of experience to not biopsy mildly dysplastic nevi, but only those with severe dysplasia. Although our model for number of biopsies did not significantly identify provider experience as a predictor, it did show a trend toward predicting biopsy number (e.047, P = .19). Lastly, the more dysplastic nevi that were removed before attending the PLC, the fewer left behind to biopsy in the PLC. Our results are similar to those found by Hanrahan et al,13 who conducted a randomized prospective trial evaluating the impact of photography on biopsy rates and pathology of biopsied lesions by primary care physicians in older men from Australia. The researchers found that fewer numbers of benign pigmented lesions (typically seborrheic keratoses) were removed in patients with photographs, although this was not true for dysplastic or malignant lesions where there was no difference in their removal between those with photographs and those without, in concordance with our findings. This study also identified that photographs increased the specificity for detecting nonmelanoma skin cancers, but could not comment on whether photographs allowed for more specificity in diagnosing melanoma, secondary to low numbers of melanoma diagnosed.13 Our results challenge a chart review by Feit et al,12 which concluded that photographic follow-up was valuable in avoiding unnecessary biopsy in suggestive, but stable lesions. This same study further demonstrated that photographically assisted followup helped detect new and subtly changing melanomas that did not meet the classic clinical features of melanoma.12 Our study differs in that we have a comparison group that allowed us to demonstrate a lack of statistically significant difference between the number of biopsies performed between those with digital photographs and those who did not have TBDP performed. Interestingly, in our study 3 melanomas were diagnosed in the cohort followed up with TBSE alone and none in the group with photographs. Although the finding between the two groups is provocative, increased numbers of melanoma need to have been diagnosed to be able to draw any conclusions about any differences observed. Although at first glance the results of this study appear to reveal a disappointing trend, it is likely that our study does not capture the time frame in which TBDP is most useful to prevent unnecessary biopsies. Our study population reflects both patients who have never seen a dermatologist and those who had
J AM ACAD DERMATOL
Risser et al 433
VOLUME 57, NUMBER 3
already been followed up by a dermatologist, and referred when TBDP became available. The latter group may have already had the majority of their dysplastic nevi removed; in our data analyses, we discovered that there are usually a large number of biopsies of atypical nevi that are taken at a person’s first visit to the PLC. Although we could have included the first visit in our follow-up period, we thought that because the first visit involved taking the photographs, rather than using the photographs, in the TBDP group, we needed to begin the follow-up period with the second visit. It is also possible that our study underestimated the full potential of TBDP to decrease the number of biopsies because the duration of the retrospective review was for only 1 year. One year may not be enough time for the TBDP to detect morphologic change of atypical nevi, given that many melanomas are slow growing. Our patient cohort was also relatively young and, thus, may not have had sufficient melanomas to detect in our 1-year period. Other potential limitations to our study include its design as a retrospective chart review. With this design, it was not possible to discern the exact reason a biopsy was performed. Biopsies are performed based both by the clinician’s suspicion and to allay patient fears and this could not be accounted for in our study. In addition, there is a potential selection bias in that a few patients with multiple atypical nevi opted to not have photography; these patients cited cost as the primary reason to not pursue photography. Lastly, this study reviews the experience of a relatively small number of providers and patients at a single academic institution, thus, the ability to generalize the results are limited. Despite these limitations, until a randomized controlled trial can be performed, retrospective studies provide our best evidence and generate hypotheses for future studies. Our methods used rigid eligibility criteria for the inclusion of patients in our study. In addition, our cohorts are matched and can be expected in terms of patient characteristics (demographics, personal and family history of melanoma, number of biopsies and severe dysplastic nevi at the time of presentation to PLC) and provider characteristics (mean provider experience and total number of visits in the first year of care at the PLC). Thus, given our findings, one could argue that TBDP may be most beneficial to those patients with atypical mole syndrome who have not already had multiple atypical nevi removed. TBDP may also potentially be useful during an extended period of time, beyond the short 1-year period that we studied. Lastly, TBDP may provide relief from cancer worry for the at-risk population. Preliminary data indicate
that patients are highly satisfied with the digital photographs and are less worried about changes in pigmented lesions going unnoticed.15 Although a primary purpose of TBDP is to detect melanomas earlier and to avoid unnecessary biopsies, photography may change the criteria by which dermatologists manage suggestive lesions. One possibility may be that physicians become more complacent with photographs and not biopsy those lesions with very subtle changes. However, the reverse may be true too. Physicians may increase their tendency to biopsy because previously undetectable changes in nevi are now apparent with baseline photographs. It is important that future prospective studies analyzing TBDP as an adjunct screening measure explore this issue. The practice of following up patients with TBDP continues to grow, without the knowledge of how and when TBDP benefits patients. Future studies should be directed at prospectively evaluating whether TBDP is both more sensitive and specific than TBSE alone for detecting dysplastic nevi and, more importantly, melanoma. Based on our study results and our observations, these studies may need to be conducted during a period of time longer than 1 year in duration and also include patients who have not already undergone extensive biopsies. Future studies should also evaluate the merit of dermoscopy being used in conjunction with total body examinations and total body photographs. Although during the period of the study, the dermatologists involved did not use dermoscopy, studies have subsequently shown that dermoscopy is a useful adjunct to the screening of melanoma, particularly during a followup period of between 3 to 21 months.16-19 Lastly, cost-effectiveness and psychologic impact should be included as study outcomes in future efforts.
REFERENCES 1. Holly E, Kelly J, Shpall S, Chiu S-H. Number of melanocytic nevi as a major risk factor for malignant melanoma. J Am Acad Dermatol 1987;17:459-68. 2. Rhodes A, Weinstock M, Fitzpatrick T, Mihm M, Sober A. Risk factors for cutaneous melanoma: a practical method of recognizing predisposed individuals. JAMA 1987;258: 3146-54. 3. Tucker M, Halpern A, Holly E, Hartge P, Elder D, Sagebiel R, et al. Clinically recognized dysplastic nevi: a central risk factor for cutaneous melanoma. JAMA 1997;277:1439-44. 4. Masri G, Clark W, Guerry D, Halpern A, Thompson C, Elder D. Screening and surveillance of patients at high risk for malignant melanoma result in earlier detection of disease. J Am Acad Dermatol 1990;22:1042-8. 5. Marghoob A, Slade J, Kopf A, Salopek T, Rigel D, Bart R. Risk of developing multiple primary cutaneous melanomas in patients with the classic atypical-mole syndrome: a case-control study. Br J Dermatol 1996;135:704-11.
434 Risser et al
J AM ACAD DERMATOL SEPTEMBER 2007
6. Tierstan A, Grin C, Gottlieb G, Bart R, Rigel D, Friedman R, et al. Prospective follow-up for malignant melanoma in patients with atypical-mole (dysplastic-nevus) syndrome. J Dermatol Surg Oncol 1991;17:44-8. 7. Freedberg K, Geller A, Miller D, Lew R, Koh H. Screening for malignant melanoma: a cost-effective analysis. J Am Acad Dermatol 1999;41:738-45. 8. Tripp J, Kopf A, Marghoob A, Bart R. Management of dysplastic nevi: a survey of fellows of the American Academy of Dermatology. J Am Acad Dermatol 2002;46:674-82. 9. Nehal K, Oliveria S, Marghoob A, Christos P, Dusza S, Tromberg J, et al. Use of and beliefs about baseline photography in the management of patients with pigmented lesions: a survey of dermatology residency programs in the United States. Melanoma Res 2002;12:161-7. 10. Banky J, Kelly J, English D, Yeatman J, Dowling J. Incidence of new and changed nevi and melanomas detected using baseline images and dermoscopy in patients at high risk for melanoma. Arch Dermatol 2005;141:998-1006. 11. Kelly J, Yeatman J, Regalia C, Mason G, Henham A. A high incidence of melanoma found in patients with multiple dysplastic naevi by photographic surveillance. Med J Aust 1997;167:191-4. 12. Feit N, Dusza S, Marghoob A. Melanomas detected with the aid of total cutaneous photography. Br J Dermatol 2004;150: 706-14.
13. Hanrahan P, D’Este C, Menzies S, Plummer T, Hersey P. A randomized trial of skin photography as an aid to screening skin lesions in older males. J Med Screen 2002;9:128-32. 14. Halpern A. Total body skin imaging as an aid to melanoma detection. Semin Cutan Med Surg 2003;22:2-8. 15. Pressley Z, Delong L, Risser J, Curiel-Lewandrowski C, Chen S. Quality of life and cancer anxiety impact of total body digital photography in patients with atypical mole syndrome [Abstract]. J Am Acad Dermatol 2007;56:2203. 16. Carli P, De Giorgi V, Crocetti E, Mannone F, Massi D, Chiarugi A, Giannotti B. Improvement of malignant/benign ratio in excised melanocytic lesions in the ‘dermoscopy era’: a retrospective study 1197-2001. Br J Dermatol 2004;150: 687-92. 17. Carli P, De Giorgi V, Chiarugi A, Nardini P, Weinstock M, Crocetti E, et al. Addition of dermoscopy to conventional naked-eye examination in melanoma screening: a randomized study. J Am Acad Dermatol 2004;50:683-9. 18. Carli P, Ghigliotti G, Ghone M, Chiarugi A, Crocetti E, Astorino S, et al. Baseline factors influencing decisions on digital followup of melanocytic lesions in daily practice: an Italian multicenter survey. J Am Acad Dermatol 2006;55:256-62. 19. Menzies S, Gutenev A, Avramidis M, Batrac A, McCarthy W. Short-term digital surface microscopic monitoring of atypical or changing melanocytic lesions. Arch Dermatol 2001;137: 1583-9.