This study examined mammogram reports of women with Down syndrome (DS) treated in the largest medical facility specifically serving adults with DS in the United States. Records of 684 women and results of 993 mammograms were reviewed, including 902 screening and 93 diagnostic mammograms. Only 2 (0.7%) women had a diagnosis of breast cancer. This study found a low rate of breast cancer among women with DS, a finding which is consistent with previous studies. The financial cost per finding was high, the benefit of the mammograms questionable, and the potential for harm greater. With less than 1% of the women with DS in this study developing breast cancer, further discussion is recommended to assess whether mammography is a beneficial, cost-effective part of health maintenance for women with DS.
Life expectancy for people with DS has increased from 25 in 1983 to 60 in 2012 (National Down Syndrome Society, NDSS, 2012). People with DS are reaching the age where they can and should be considered for participation in adult health screening recommendations for the general population. However, with greater emphasis being placed on screening and individualized care, efforts should be taken to establish which screening tests should be recommended for this population. Our clinical experience suggests that the screening recommendations for tests such as mammograms should be reevaluated for people with DS. This is particularly important since people with DS are at lower risk for solid tumors such as breast cancer (NDSS, 2012).
Although breast cancer is the most common cancer (second only to the wide grouping of skin cancers) in women without DS, multiple studies have shown that women with DS have a much lower incidence of breast cancer. In the Danish Cancer Registry, Hasle (2001) noted 60 cases of cancer in people with DS compared to 49.8 expected; a disproportionately low amount, 24 compared to 47.8, were found to have solid tumors. Hasle, Clemmensen, and Mikkelsen (2000) also observed in an earlier study about half as many solid tumors in people with DS compared to the general population, and noted that this decrease applied across all age groups. A Japanese study reviewed 1,514 cases from the Annual of Pathological Autopsy Cases of Japan between 1974 and 2000. Eight patients with 10 benign tumors and 104 patients with malignant tumors were recorded. Of the 104 cases of malignant tumors, 17 (16.3%) cases had solid tumors (Ehhara, Kousaku, & Hisao, 2011). The records of 1,298 people with DS identified from the Disability Services Commission of Western Australia and the State Cancer Registry (725 males, 573 females) showed eight cases of solid tumors compared to 18.3 expected (Sullivan, Hussain, Glasson, & Bittles, 2007). Patja, Eero, and Iivanainen (2001) noted no solid tumors in 102 women with DS in Finland, whereas Patja, Eero, Sund, Iivanainen, and Kaski (2006) found that although the overall risk of cancer was similar in patients with DS compared to the general population, there were fewer solid tumors in the former group.
The mortality of women with DS from breast cancer is also reported to be lower than women in the general population. Using the INSERM national mortality statistics in France that compared deaths from female breast cancer in the general French population with the cancer deaths in women with DS over a 24 year period, only five deaths in women with DS were identified (Satge, Sasco, Pujol, & Rethore, 2001). In a cohort study of 1,425 British and Scottish women with DS, only one case of breast cancer was recorded as cause of death (Hermon, Alberman, Veral, & Swerdlow, 2001). Although 7.3 cases were expected based on data from the general population, no cases of breast cancer were identified from the 2,814 Danish individuals with DS (Hasle et al., 2000). Similar findings have been reported with Australian (Sullivan et al., 2007), Japanese (Ehhara et al., 2011), and Finnish women with DS (Patja et al., 2006).
Despite these findings, women with DS continue to fall under the breast cancer screening recommendations for the general population. Although current recommendations vary between organizations, they all recommend screening mammograms as part of a breast cancer screening protocol. The American Cancer Society (ACS; 2014) and the American College of Obstetricians and Gynecologists (2011) recommend yearly mammographic screening for every woman over the age of 40 after a discussion of the risks and benefits of screening mammography. They also recommend yearly clinical breast exams after the age of 40 with breast exams every 3 years between the ages of 20 and 39, as well as yearly Magnetic Resonance Imaging (MRI) for patients who are at high risk for breast cancer.
As greater emphasis is placed on preventive screening and individualized care, efforts should be taken to establish which screening tests should be recommended for women with DS. However, no studies, to date, have looked at the incidence rate of breast cancer among women with DS in the United States. The primary purpose of this study was to evaluate the risk of breast cancer in a population of American women with DS served in our practice by identifying the number of patients with breast cancer, the frequency of mammograms, and the number and type of abnormalities found in mammograms. The study also aimed to record the types of additional testing recommended based on mammogram results and to calculate the approximate cost of breast cancer screening per abnormality.
The study population consisted of a convenience sample of 684 female patients with DS ages 35 and older when evaluated at a center serving adolescents and adults with DS from January 1996 to May 2012. Approval and a waiver of authorization for access to patient private health information were obtained from the facility's Institutional Review Board.
Data was collected through a retrospective chart review of information related to patients' histories of breast cancer screening and breast health. Data recorded included the number and dates of screening, diagnostic and follow-up mammograms received and their results, as well as the number and dates of additional tests or procedures performed and their results. The types of additional tests or procedures were ultrasounds, breast MRIs, presurgery X-rays, needle localizations, biopsies, and lumpectomies. Data were collected through a standardized chart review protocol. A randomly selected subset of records (approximately 10%) was reviewed by a second researcher to confirm accuracy and completion of data collection.
Analysis and Results
Data are presented descriptively as mean, standard deviation or number and (%). Descriptive analyses were performed using SPSS for Windows, version 22.0 (SPSS Inc., Chicago, IL). The study included 684 female patients who met the inclusion criteria. Their mean age was 48.3 (± 6.4) years old. Of these patients, 422 (61.7%) received a mammogram. However, only 290 (68.7%) of the records contained a report from the performing radiologist or facility (Figure 1). The reports from these patients were used for further data analysis. The other 132 women had mammograms ordered by other providers at other locations and the reports were not available; however, there were no reports of breast cancer in these women during this time period.
For the 290 patients for whom reports were available, the average age at the time of each mammogram was 48.4 years, the oldest was 71.1 years and the youngest was 31.2 years. The distribution by age of each of the individuals at the time of their most recent mammogram (in the timeframe of this study) is described in Table 1.
For these 290, there were 993 mammograms on record; 902 of these were routine mammograms, and 91 were diagnostic mammograms. Additional testing was recommended following the results of 74 of the routine mammograms. The results of 34 diagnostic mammograms indicated the need for additional testing (Figure 2). The additional testing consisted of ultrasounds, MRIs, needle localizations, biopsies, one X-ray, and one lumpectomy. Thirty four patients underwent additional testing, totaling 51 tests. Twelve of these additional tests were abnormal and further testing was recommended. However, only two (0.7%) patients had a diagnosis of breast cancer (Figure 3). The age of the two women was 57 years and 46 years at diagnosis, respectively (see Table 2).
The low rate of breast cancer among women with DS included in our study is consistent with low rates of breast cancer reported in previous studies (e.g., Ehhara et al., 2011; Hasle et al., 2000; Hermon et al., 2001; Satge et al., 2001; Sullivan et al., 2007). We found only two cancers; one being a noninvasive ductal adenocarcinoma in situ and one a phyllodes tumor with borderline malignant potential. No invasive cancers were found. Only one of the two cancers was found via a truly screening mammogram. The phyllodes tumor was initially found on physical exam and confirmed by a mammogram. Coupled with the very low rate of breast cancer in this study is that the cases of breast cancer that we found were unlikely to be life-threatening. Neither patient opted to continue with the recommended full course of treatment and neither had a recurrence. In light of the development of Alzheimer disease in these two patients, the noninvasive nature of the tumors, and the nonrecurrence (despite receiving less than the recommended follow-up treatment), it is not clear how much, if any, the mammograms extended the life of the two women.
Despite the limited benefit of mammogram screening in our practice, the cost of the mammogram screening was substantial. Thus, these findings have significant implications for financial, time, and psychosocial costs of mammograms and additional tests as well as increased risk from radiation exposure and the use of anesthesia.
The estimated cost of a digital bilateral screening mammogram is $280, according to The Healthcare Blue Book (2014), an online resource that provides the typical fee that providers accept as payment from insurance companies. This estimate was based on the zip code for our center. Thus, the estimated cost of the mammograms in this study was $278,040 (993 × $280 per mammogram), which is over $139,000 per cancer found. The Healthcare Bluebook's estimate for the cost of a bilateral breast ultrasound is $185. Applying this cost to our study, an estimated $7,215 was spent on ultrasounds (39 × $185 per ultrasound). The other additional tests added significant costs as well; an estimated $63 for X-rays (1 × $63 per X-ray), $2,518 for MRIs (2 × $1,259 per MRI), $890 for needle localizations (2 × $445), $6,992 for breast biopsies with device (4 × $1,748), $6,096 for open incision biopsies (2 × $3,048), and $13,767 for lumpectomies (1 × $13,767 per lumpectomy). In total, the estimated cost of breast cancer screening and treatment in this study was $315,581, or just under $158,000 per cancer found. Since it is not clear that any additional years of life were added, cost per year of life added cannot be calculated, however, it seems to be greater than for women without DS. The cost of screening in the female population without DS has been determined to be $18,999 per additional life-year for biennial film mammography for women aged 50 to 64 years and $106,428 per additional life-year for annual film mammography starting at age 40 (Melnikow et al., 2013).
The amount of time required to receive mammograms also factors into the overall cost of breast cancer screening. Screening mammogram appointments usually are scheduled to last 20 minutes but can last up to 1 hour depending on the speed of the sign-in process, the number of other patients, and the degree to which patients cooperate with testing. Since very few of our patients are able to take themselves to the appointments, the time commitment is shared by family members or service providing staff members (e.g., staff of a residential facility) of the women with DS. Many of the patients find the mammogram challenging and additional time commitment is required.
Although psychological costs are more difficult to measure than financial and time costs, they should also be taken into account. At least 51 women in our study could not cooperate with mammograms. Women with DS often find a mammogram quite challenging. Radiology departments have requested that the our providers not refer some women with DS for subsequent mammograms because the experience was too frightening and overwhelming for the patients, making the procedure “dangerous” for both the patients and the radiology technicians. Patients were described as becoming aggressive, pulling away from the machine in the middle of the procedure, and creating an uncomfortable environment for other women undergoing mammograms.
Radiation exposure from mammograms and anesthesia exposure from subsequent testing also pose a risk to patients. Ma, Hill, Bernstein, and Ursin, (2008) found that women in the general population who had seven or more mammograms had a higher risk of breast cancer (odds ratio, 1.80) than women who never had a mammogram. According to the Health Physics Society (2012), the average American receives approximately 3.0 mSv of exposure from natural background radiation each year. The effective radiation dose of a mammogram (4 views) was calculated as 0.7 mSv. A single bilateral two-view digital or screen-film mammogram is associated with a lifetime risk of inducing fatal breast cancer due to radiation exposure of 1.3–1.7 cases in 100,000 women aged 40 years at exposure (Hendrick, 2010). This radiation exposure may put patients with DS at greater risk than the general population, given that patients with DS are more susceptible to effects of ionizing radiation. Hannan, Waghray, Sigut, and Ozand (1992) found that the lymphoblastoid cell lines (LCL) of a patient with DS were more radiosensitive than the LCL of the control. According to Takeshita et al. (1992), chromosomal changes from radiation were approximately 1.3 times more frequent in patients with DS as compared to the control patients.
Although some studies have concluded that the benefits of mammogram screening tests outweigh the risks posed by ionizing radiation for women without DS (Kong et al., 2012; Yaffe & Mainprize, 2010), no studies could be found that analyzed the risk-benefit ratio specifically for adults with DS. Nevertheless, the potentially increased risk from radiation exposure among people with DS should be considered in discussions of breast cancer screening recommendations for women with DS.
The risk of additional testing also needs to be considered. Additional testing is recommended if abnormalities are detected during a mammogram screening and it may include another mammogram, a breast MRI exam, a ductogram, an ultrasound, and/or a biopsy. These tests enable physicians to determine if a detected abnormality is benign or malignant. Biopsies are often performed under local anesthesia (ACS, 2012). However, people with DS often require general anesthesia in order to be able to cooperate with procedures such as a biopsy. A higher risk of complications from anesthesia (e.g., bradycardia, post-intubation stridor, natural airway obstruction, difficult intubation, bronchospasm, etc.) is noted in patients with DS compared to the general population (Borland, Colligan, & Brandom, 2004; Meitzner & Skurnowicz, 2005). In fact, difficult airway anatomy, altered respiratory mechanics, gastric reflux, cardiovascular disorder, and neuromuscular problems should be considered when administering general anesthesia to patients with DS. Desensitization and relieving anxiety are also important steps to perform before administering sedation or anesthesia to a patient with an intellectual or developmental disability (Butler, Hayes, Hathaway, & Begleiter, 2000).
This study has several limitations that should be carefully considered. We relied on a convenience sample of women with DS seen at our center. Therefore, the results of this study are not necessarily representative for other women 35 years and older with DS. The age of our sample is a particular source of bias when comparing our patients to the population without DS. Due to a lower life expectancy of people with DS, it is expected that our study would not include a similar number of older women as a study of women without DS. However, at the time of the most recent mammogram, 49% of the women in this study were 50 years of age or older. Thus, a group of women potentially at higher risk for breast cancer was represented, as seen in Table 1.
Another limitation is the number of mammograms recorded is an underestimate because some of the patients also received care from primary care physicians and gynecologists who are unaffiliated with our health system. If reports from mammograms ordered by other healthcare providers were not sent to our center, then those reports would not have been included in the data for this study. However, regardless of these limitations, this is the first study looking at the breast cancer rate in women with DS in the United States.
For women with DS, there are considerable financial, time, and psychosocial costs related to mammograms. There is also an increased risk posed by radiation exposure and the use of anesthesia. With limited, questionable, or potentially no significant benefit of mammograms in this population, the cost-benefit ratio becomes very high. Of even more concern is the potential that little or no benefit is being realized and there is potential or actual harm. The potential for mammograms contributing to the onset of some breast cancers in women without DS has been confirmed. Since the tissue of people with DS is more sensitive to the effect of ionizing radiation, the potential for harm from repeated mammograms may be greater. Additionally, the higher rate of need for anesthesia for follow-up testing and the greater risk for anesthesia complications raises the potential for increased harm from testing algorithms. The low rate of breast cancer found in this study suggests that the cost-benefit ratio of mammogram screening for women with DS should be re-evaluated. Also, further discussion to assess whether mammography is a beneficial, cost-effective part of health maintenance for women with DS is warranted.
We would like to thank Maria Rozo, BS, for her help with manuscript review and Aaron Perlow, Leora Perlow, Elena Perlow, and Anne Rojas for their contribution to the data collection process. Portions of this project have been presented as a poster at the 2014 Advocate Research & Innovation Forum at Advocate Lutheran General Hospital, Park Ridge, IL. This project did not receive any funding.
Brian Chicoine, Advocate Medical Group Adult Down Syndrome Center & Advocate Lutheran General Hospital Family Medicine Residency, Park Ridge, IL USA; Melody Roth, Advocate Lutheran General Hospital Family Medicine Residency, Park Ridge, IL USA; Laura Chicoine, Advocate Medical Group Adult Down Syndrome Center, Park Ridge, IL USA; and Suela Sulo, James R. & Helen D. Russell Institute for Research & Innovation, Advocate Lutheran General Hospital, Park Ridge, IL USA.