We read with interest the article by Nadalin et al1 in which they, using their White test, reported a 74% bile leak rate, which allowed identification and control of biliary radicals, reducing the leak rate to 5.1%. We concur with McFadden in his accompanying critique that it is a pity that additional details were not provided as to the number of patients who underwent resection. In addition, the inclusion of the indication for surgery (primary or metastatic hepatic carcinoma or cholangiocarcinoma), extent of resection, and morbidity as a result of bile leaks would also have added value to the article. We are also surprised that, even after instillation of the fat emulsion, the clinical leak rate was 5.1%. This represents only a 2.9% reduction in leak rate compared with their own series referenced in the article. Unfortunately, there is no reference to this data, and no statistical comparison was performed. Furthermore, if 74% of patients had a leak identified on the White test, the authors' limited data would suggest that the vast majority of these were not clinically significant, otherwise the rate in their unpublished series would be expected to have been much higher than 8%. This would suggest that the White test has a significant false-positive rate, at least in terms of significant leaks, and there is also still a significant false-negative rate, as there was a 5.1% leak rate following the introduction of the test.