Inflammatory breast cancer (IBC) is a rare and aggressive form of breast cancer characterized by diffuse erythema and edema (peau d’orange) of the skin of the breast. It does not include neglected locally advanced cancer of the breast presenting late in the course of the disease. IBC is primarily a clinical diagnosis, but also requires pathologic confirmation of breast cancer. The purpose of this study was to examine the reliability of medical records for diagnosing IBC in patients at two centers in Tanta, Egypt (Gharbiah Cancer Society and Tanta Cancer Center) which participated in an epidemiologic case-case study of IBC and non-IBC in 2009-2010. IBC patients were recruited to the case-case study at diagnosis based on clinical examination; detailed information on the extent and duration of signs used for the diagnosis of IBC was recorded. Information on signs of IBC similar to those collected in the case-case study was abstracted from the medical records of the hospitals where case-case study patients were recruited during the same time period, then linked and compared to the signs observed from clinical examination for the case-case study.
A total of 40 IBC cases were recruited into the case-case study during 2009-2010. Medical record information was available for 31 of these cases. Twenty-eight of these 31 cases were reported to have erythema and 28 edema/peau d’orange based on the clinical examination for the case-case study. Of those with erythema, 16 (57.1%) also had erythema/inflammation listed in the medical record; of those with edema/peau d’orange, 23 (82.1%) had this mentioned in the medical record. Erythema was noted on the medical record but not recorded in the case-case study for 2 cases; there were no cases in which edema/peau d’orange was present in the medical record and not in the case-case study. Our analysis indicates that medical records at these cancer centers often have missing information regarding the presence of signs that are necessary to the diagnosis of IBC. Overall, this study suggests that medical record review is not sufficient to identify IBC cases for an epidemiologic study.