Article Content

Invisible blood in urine may be an early warning sign of bladder cancer, according to research by a team at the University of Exeter Medical School. The study, published in the September 1 issue of British Journal of General Practice (2014;64:e584-e589) found that one in 60 people over the age 60 who had invisible blood in their urine (identified by their general practitioner testing their urine) went on to be diagnosed with bladder cancer.

Figure. No caption a... - Click to enlarge in new windowFigure. No caption available.

That percentage was about half that of those found to have visible blood in their urine, the best known indicator of bladder cancer, but was still higher than that for other potential symptoms of bladder cancer that warrant further investigation.


The lead author, Sarah Price, a PhD student, said in a news release, "It is well known that if you see blood in your urine you should contact your GP, who is likely to refer you for tests. But there is no clear guidance for GPs on what to do if they detect blood that is not visible during routine tests. We are hopeful that our findings will now lead to robust guidance that it warrants further investigation."


The study examined approximately 26,000 people whose anonymized data contributed to the Clinical Practice Research Datalink, a large research database used by the Exeter team in several other cancer studies. The team found that the risk of bladder cancer was 1.6 percent in people over 60 who had invisible blood in their urine.


The researchers-senior author was William T. Hamilton, MD, Professor of Primary Care Diagnostics and clinical lead for the U.K.'s ongoing revision of the National Institute for Health and Care Excellence (NICE) guidance on investigation of suspected cancer-note that the results of the study also have implications for those conducted in large electronic datasets that analyze coded data only:


"The preferential use of a coded format for visible hematuria is important; if this differential coding is also used for other 'red-flag' symptoms of other cancers, then positive predictive value estimates may be artificially high," they said.


"This finding should be tested in other cancers and their key symptoms... Researchers should be aware that they risk missing important information if they omit uncoded data in studies using CPRD data, potentially introducing errors that may inflate or reduce estimates."