Researchers examined electronic health records data on 63,973 patients aged 18 and older who presented to primary care with unexpected weight loss from 2000 to 2012. Overall, a total of 908 patients (1.4%) had a cancer diagnosis within six months; nearly all of these cancer patients (97.1%) were over age 50.
Males who were current or former smokers and over 50, however, had a positive predictive value (PPV) of cancer above the 3% threshold that warrants urgent investigation under UK guidelines, the study found.
Several clinical features were also associated with PPV of cancer above the 3% threshold in both men and women: abdominal mass, abdominal pain, appetite loss, chest signs, iron deficiency, anemia, jaundice, and lymphadenopathy, according to the report in The BMJ.
"In the UK, national guidelines for suspected cancer in primary care recommend that symptoms with a PPV of 3% should be prioritized for investigation," said Monica Koo, a research fellow at University College London who wasn't involved in the study.
"While the PPV of unexpected weight loss among all adults is low - just under 1% in women and 2% in men - the results of this study show that the PPV may be higher if we combine information on other symptoms or clinical signs, patient characteristics, and test results," Koo said by email. "This has implications for GPs' clinical decision-making and could contribute to more timely diagnosis of cancer in individuals with weight loss."
In addition to clinical features that were present in both men and women with unexpected weight loss who developed cancer, some features were associated with cancer only in men: dysphagia, hemoptysis, and non-cardiac chest pain. In women only, back pain, venous thromboembolism, dyspepsia, and change in bowel habits were associated with cancer.
For men, positive likelihood ratios ranged from 1.86 for non-cardiac chest pain to 6.10 for an abdominal mass. Among women, positive likelihood ratios ranged from 1.62 for back pain to 20.9 for jaundice.
Certain abnormal blood tests associated with cancer included: low albumin levels, raised values for platelets, calcium, total white cell count, and C reactive protein. No single blood test in isolation could rule out cancer.
Lead author of the study Brian Nicholson of the University of Oxford didn't respond to emails seeking comment.
One limitation of the study is that if weight loss was recorded differentially for individuals with and without cancer, this may have influenced the estimation of PPVs, Koo said.
"Another word of caution rather than a limitation per se is that, while in the UK we have been using a PPV threshold of 3%, other countries may choose to use different thresholds," Koo said. "Therefore, the implications of the study will differ."
Even so, the findings could be used to update clinical guidelines to help GPs investigate suspected cancer in a more timely and appropriate manner, Koo said.
"There are many other causes of unexpected weight loss beyond cancer," Koo said. "This study found that unexpected weight loss alone is a poor predictor of cancer but combining this with other factors such as age, sex, smoking status, other symptoms, and test results can be helpful for deciding to investigate a patient for suspected cancer."
By Lisa Rapaport
SOURCE: https://bit.ly/35OB9cd The BMJ, online August 13, 2020.
Posted on
site created by:
© 2023 Medicom Medical Publishers. All rights reserved. Terms and Conditions | Privacy Policy