I read this article [1] with interest; the authors’ objective was to report the recurrence and progression rates of 69 patients with pT1G3 TCC of the bladder who were treated by intravesical BCG after TURBT. The authors then added a report of results of the same outcome measures in patients with the same disease stage and grade who were allocated, at least initially, two other methods of treatment, i.e. TURBT only or cystectomy (22 and 20 patients, respectively). They also reported on the survival of the three groups.

There are major problems in objectively assessing the results of this study. First, the authors make no mention of the reason(s) for allocating patients to any of the three treatment groups. Second, there is no attempt at any statistical comparison of the outcome measures among the three groups; this is left to the subjective impression of the reader.

In view of these reasons it will be difficult to justify the authors’ conclusion of an alleged benefit of intravesical BCG in facilitating bladder preservation, reducing recurrence and delaying progression in many patients. This type of conclusion can only be drawn from properly conducted randomized controlled trials that are analysed statistically, and not from an observational study of such a cohort of patients.

If there is a significant difference between the outcome measures in the three groups, then to attribute this difference to a particular treatment requires assurance that the groups had comparable prognostic characteristics at the outset. If such data are missing, the results cannot be assessed.