Abstract
Blanchard G, Amato C, André A, et al.
Owing to queries raised post-publication, the authors have been asked to provide additional clarification regarding the above study.
The authors would like to provide additional information regarding the study design and explanation for an apparent overlap with a published thesis in the public domain. 1 Indeed, long before the publication of their paper, 2 some of the data were used in a veterinary degree graduate thesis. 1 This work should therefore have been referenced in the published paper, and the results in the thesis presented as preliminary, since full data were not available at the time of its publication. Moreover, it should have been acknowledged in the published paper that Charline Baussart contributed to the data acquisition through her participation in carrying out the feeding trial.
The original aim of the study was first to evaluate the effects of moderate-sodium balanced home-prepared diets containing zucchini on urinary parameters of adult cats, and second, the effects of the consumption of zucchini on lowering the urine saturation in cats. These effects were compared with those of two commercial cat foods designed for the prevention of urolithiasis: a dry one using a high-sodium content strategy; and a wet one using a high-water content strategy. They were also compared with the effects of a standard commercial dry food for adult cat maintenance, given alone or with zucchini. Regarding the home-prepared diets, to try to discriminate between the effects of the diet water content from those of the sodium content on urinary parameters, two diets were formulated with different sodium contents. The first one had a low sodium content (HOME1, 0.09% sodium as fed), while sodium was added to the second one (HOME2, 0.15% sodium) at a level comparable to that of the wet diet intended for animals with a urinary crystal history. Therefore, a total of six trials (testing six diets) were run as a Latin square protocol: two veterinary diets (one dry, one wet) with a urinary focus; a standard commercial dry diet (one fed alone, one fed with zucchini); and two home-prepared diets.
These data were used in the published paper, excluding HOME1. As mentioned, the preliminary data had been gathered previously and used in the graduate thesis. 1 However, it was at a time when not all the data were available (notably because analyses, partly outsourced, had not been completed yet), since the student was working within significant time constraints. Therefore, it had not been possible to include the data regarding HOME2, and the standard diet with zucchini in the thesis.
At the time when the full results became available, we realised that there were concerns with the data published in the graduate thesis as a result of issues with the calculation of some digestibility coefficients, and that of the energy content of the diets. Moreover, the home-prepared diet, HOME1, had been poorly consumed (and the cats’ body weight decreased). When trying to understand the differences in consumption between the two home-prepared diets, it appeared that HOME1 was given to the cats all at once, while HOME2 was given in two meals, as planned. This was attributed to an unfortunate communication error that led the animal team to deviate from the original protocol. Being fed as one meal, HOME1 was poorly consumed by the test subjects, causing the results from HOME1 to not be biologically interpretable. It was not possible to consider the urine composition, which depends so much on the composition and the amount of food consumed, when the diet was only poorly consumed. This was critical regarding the data presented in the graduate work and for these reasons, it was decided not to include the HOME1 data in the published paper.
The graduate work presented preliminary data and results from only four diets. The published paper presented data from five diets, after the exclusion of one home-prepared diet (HOME1). Data from both home-prepared diets would have potentially allowed the effect of the sodium content to be highlighted. However, as the HOME2 diet had the same sodium content as the wet diet intended for animals with urinary crystal history, the omission of HOME1 did not contravene the aim of the study, nor affect its conclusions.
These differences should have been explained in the published paper, and it should have refenced the graduate work. The authors would like to apologise to readers of the journal for this omission.