Usefulness of abdominal symptoms in early diagnosis

In contrast to the predictable media coverage of Hamilton and colleagues’ claims about early detection of ovarian cancer,1 general practitioners are crying out for reliable information on the predictive value of symptoms to enable early diagnosis and select patients for further investigation.

One surprising omission from the study is a review of the community prevalence of the symptoms investigated. The prevalence of abdominal bloating in the control subjects was 2% (21 out of 1060 subjects). General practitioners active in clinical practice will find this surprisingly low, community surveys estimating it to be 16-30%.2 The positive predictive value for abdominal bloating that the authors calculate (0.3%) will be much lower if the prevalence of bloating is greater than the 2% in their controls.

They state: “Women with ovarian cancer usually have symptoms and report them to primary care, sometimes months before diagnosis.” But they found that women with ovarian cancer have the same variety of non-specific symptoms that many women experience, that some of these symptoms (abdominal pain, distension, and loss of appetite) are substantially more common in the two or three months before diagnosis, and that one symptom (abdominal bloating) is more common for longer than this, but only compared with the surprisingly low prevalence in their controls. All of this makes their final rather emotive claim—“ovarian cancer is not silent, rather its sound is going unheard”—and their comments to an unsceptical media rather surprising.