
Each year, 2.3 million women worldwide are diagnosed with breast cancer, while there are 685,000 deaths. Two-thirds of these deaths occur in middle- and low-income countries where the cancers are often diagnosed at a late stage because there are no screening services available. Given that early detection and treatment can lead to a successful outcome, there is clearly an urgent need to find a way to introduce breast screening in low-income countries and reduce mortality. Dr Lydia Pace (seen above right, © Infinity Portrait Design), a primary care physician in Boston, has spent the past decade addressing this issue.
Pace and colleagues have just published a paper in the Journal of the American College of Radiology entitled “Clinical diagnoses and outcomes after diagnostic breast ultrasound by nurses and general practitioner physicians in rural Rwanda.” Collaborating with local healthcare workers in the central African country, the team found that local clinicians were able to evaluate lesions that were malignant with high sensitivity and acceptable positive predictive value (PPV). These encouraging findings support the breast ultrasound training and mentoring programme introduced by Pace.
There were 229 palpable findings (in 199 patients) that met the study’s eligibility criteria. The pie chart at left (© ACR) illustrates the pathologic diagnoses for 104 patients biopsied at their initial visit. The PPV of the trainees for biopsy recommendation was 35%, while their sensitivity was 93%. Of the 46 patients who did not receive biopsy and were told to return for imaging follow-up, 17 did not return. There is no mention of whether the 37 patients diagnosed with breast cancer received surgery or other therapy.
Pace told AuntMinnie.com their findings suggest “that diagnostic breast ultrasound capacity could be feasibly scaled up to district hospitals in Rwanda and similar settings, which could help decentralize a key aspect of breast cancer diagnosis and make services more accessible to patients.” However, given the time and effort required to train the local clinicians, and the time it takes to acquire hand-held ultrasound (HHUS) images, it seems reasonable to ask: Is there not a better way?
We believe there is. First, instead of HHUS, it would be preferable to use a novel automated breast ultrasound (ABUS) system, reducing image acquisition time and dependence on the operator’s skill. Second, instead of relying on inexperienced clinicians to make the diagnosis, the latest research by Zhuang et al. has demonstrated an accuracy of 94% when using an artificial intelligence algorithm on ABUS images (seen right, © CMIG). This approach would surely reduce breast cancer deaths in low-income countries.