Hot and cold biopsy: implications of study design on outcomes

To the Editors:

We read with interest the article in the European Respiratory Journal by Tremblay et al. 1, wherein the authors conclude that the use of hot biopsy forceps for endobronchial biopsy does not appear to have a negative impact on the pathological samples, and that there was a statistically significant (albeit clinically insignificant) reduction in bleeding score with hot biopsy forceps. However, many of the conclusions of the study have limitations because of the study design of alternate hot and cold biopsies.

The authors state that the quantification of bleeding was carried out and recorded by the bronchoscopists between each biopsy on a four-point scale. However, the interval between the two biopsies was not stated. Moreover, it is a common observation that the ooze from an endobronchial lesion continues after the biopsy has been taken and thus it becomes difficult to ascertain whether it was the first or the second “bite” that contributed to the bleeding. This is a potential confounding factor in the analysis. Furthermore, bleeding can continue for 48 h, and even longer, after biopsy, making it difficult to decide which technique has contributed to the bleeding. Therefore, it becomes important to use a protocol in which the hot and cold biopsies are performed in alternate patients rather than alternate biopsies in the same patient. If we follow this design, a semi-quantitative assessment can be made by quantifying the amount of saline instilled and the return amount and comparing between the two. This would also complement the qualitative assessment made by the bronchoscopist.

In addition, the authors have stated minimal damage even with cold biopsies. Whether this is an effect of the previous hot biopsy is also unclear, as multiple biopsies have been taken from the same lesion. Due to the small sample size, and given the fact that previous studies have shown significant pathological changes in the tissues after endobronchial electrocoagulation 2, 3, further investigation is required to confirm the findings of Tremblay et al. 1.