Author’s response to reviews Title: Characteristics, changes and influence of body composition during a 4486km transcontinental ultramarathon. Results from the TransEurope FootRace mobile whole body MRI-Project. Authors:

Please define the abbreviations MRI, MR, T1w and TSE Lacking definitions are added now. In the Conclusion, you mention ‘unpreventable chronic negative energy balance’. However, you have not measured energy intake or energy expenditure, so you have not energy balance. The sentence was changed in: “Participants lost more than the half of their adipose soft tissue and even le a tissue volume decreased (mainly skeletal muscle tissue) and without any exception, ever y subject showed a significant loss of body volume. This indicates, that running an MSUM of nearly 4.500km without any day rest is linked with an unpreventable chronic negative ener gy balance due to the massive running burden.” The lack of energy balance intake measurement was mentioned in the last se ction “Limitations.” in the discussion. Introduction Page 3, first paragraph, line 5: the sentence should start with ‘However’ not wit h ‘But’ Corrected Materials and Methods In the section ‘Body composition analysis’ in the first line, change ‘fa’ to ‘ fat’. Corrected Discussion In the section ‘ultra endurance related changes in body composition’ or ‘mass loss’ I suggest inserting the reference Knechtle B, Duff B, Schulze I , Kohler G, A multi-stage ultra-endurance run over 1,200 km leads to a continuous accumulation of total body water, Journal of Sports Science and Medicine 7:357-364, 2008, where the authors showed a decrease in both fat mass and skeletal muscle mass in a multi-stage ultra-marathon. This reference was included in the sections ‘ultra endurance relate d changes in body composition’ and ‘mass loss’. As reviewer 2 mentioned to reduce the amount of r eference, this addition contradicts the request reviewer 2. To discuss the negative energy balance, I suggest the references Knecht le B, Knechtle P, Schück R, Andonie JL, Kohler G, Effects of a Deca Iron Triathlon on body composition: a case study, Int J Sports Med. 2008 Apr;29(4):343-51, and Enqvist JK, Mattsson CM, Johansson PH, Brink-Elfegoun T, Bakkman L, Ekblom BT, Energy turnover during 24 hours and 6 days of adventure racing, J Sports Sci. 2010 Jul;28(9):947-55. To follow the request of reviewer 2 in limiting the number of reference s, we decided not to add these 2 references because they are case reports and/or investigate other ultra endurance sports with a limited length and number of stages, compared to our collective. Seite 4 von 11 Error! Reference source not found. 4 Answers to Reviewer: Gregoire Millet, reviewer 2 Major Compulsory Revisions Provide data on the relative error and reproducibility for the quantification of the di fferent volumes and tissues. (Jürgen Machann) Data on reliability according to literature are implemented in ‘Material and M ethods’ Provide clear hypotheses regarding you study that appears as exclusively descriptive. Specific hypotheses are implemented in ‘Introduction’ now. Modify the discussion to discuss the most important results. The current discussion is very elusive. Beyond the exceptional setup of your data record, the nov lty of your results remains unclear. Complete revision of section ’Discussion’” is done. Most important resul t are emphasized. Elusive information is abandoned. Discussion of literature not specific t o ultramarathon is abandoned. Minor Essential Revisions P2 – Methods : MR – not defined, MRI – not defined, T1, TSE – not defined Lacking definitions are added now. P2 – Results: All abrev are not defined (TV, …) We do not understand this comment, there must be a misunderstanding: All abbreviations had been defined in the section ‘Materials and methods’ initially. P2 – conclusions: Energy balance was not measured The sentence “ the relative amount…” is not supported by data reported in the abstrac t. In the abstract the significance of difference between finisher and n on-finisher at start of TEFR09 is mentioned initially in section ‘results’. But for more prec ise reporting, we changed the sentence and added the absolute values of percentage visceral fat volume of t he two groups. “Before start the non-finisher group had significantly more percentage volu me of total visceral tissue compared to the finishers, due to more visceral fat (3,6vol% vs. 2,3vol%) .” In ‘conclusions’ the sentence was specified: “The ratio of adipose tissue contribution between the visceral and somatic com part ents had an influence on giving up during the first stages.” P3 – introduction: Ref 1,2 – too many self-citations with many in low-quality journals. I would recom mend you choose more wisely your references. We reduced the references regarding Knechtle B, because some of them we re not specific enough and reported results from studies on (ultra-)triathlon. References on mul tistage ultra-marathons are rare in literature. For discussion on this specific field of enduranc e sports, every reference is valuable. Quality of results is discussed critically in section ‘Disc u sion’. Though the reader should be able to decide about quality of our results compared to other data. Additional references on this topic from the last two years were implemented. Seite 5 von 11 Error! Reference source not found. 5 Knechtle B. is no author or Co-author and had no participation in any field of invest iga ion regarding this manuscript/article about the results of whole-body MRI-me asurements at TEFR09. Therefore, self-citation is not given from our point of view. Mean age.. please provide SD. SD were added. P3 – Material and Methods: Due to various reasons: please provide details. Body fa = body fat Corrected P4Image post-processing Please detail the error due to the absence of INF. How did you guarantee that t his absence of INF? The following data are provided now in section ‘image post-processing’: “At start of T EFR09, ABM and INF together account for 13.2% of total adipose tissue. Due to continuous los s of adipose body tissue, this ratio rises up to 28.2% till the end of race. For visceral adipose t ssu , the account for INF rises from 3% at start up to 65.4% at the end of TEFR09.” Report the CV in % for all volumes, tissues and compartments. To all investigated body compartments the absolute data are provide now in the new ‘Table 2’ for all measurement intervals. The table was placed at the beginning of chapter ‘re sults’ after the ‘casuistry’: “The absolute volume data of all investigated body tissue compartme n s and segments are shown in table 2.” (correction of table numbering was done). New figure 14 was added showing results of segmental data analysis. P5 Why is the mean difference in TAT so important 8.13% when compared to other tissues ? The difference between manual and automatic procedure rises with th e amount of volume measured, therefore our manual corrections are appropriate regarding the data of Wür slin et al.. For clarification of this the sentence in section ‘image post-processin g’ was corrected: … “procedure (2.07% for TV, 8.13% for TAT, 3.21% for VAT), the described additional man ual corrections regarding the small volumes of ABM and INF are appropriate.” P6 – Performance P-value is mentioned twice.