Applicability of Supraclavicular Oxygenated and Total Hemoglobin Evaluated by Near-Infrared Time-Resolved Spectroscopy as Indicators of Brown Adipose Tissue Density in Humans
, Sayuri Fuse 2,†, Shiho Amagasa 3, Toshiyuki Homma 4, Ryotaro Kime 2, Miyuki Kuroiwa 2, Tasuki Endo 2, Naoki Sakane 1, Mami Matsushita 5, Masayuki Saito 6, Yuko Kurosawa 2 and Takafumi Hamaoka 2,*
Round 1
Reviewer 1 Report
This study examined evidence that near-infrared time -resolved spectroscopy (NIRTRS) is capable of estimating human BAT density. The Authors determined seasonal (winter and summer) variations in BAT density and describe an association between BAT volume and CIT.
The authors concluded that NIRTRS is an useful tool for evaluating BAT density in a simple, non-invasive manner.Similar papers have already published, also from the same authors, and the results are not homogeneous.
The manuscript is potentially interesting but the data are not conclusive. However, as is the case whenever a novel idea is presented, the results need to be much more strong than if the study is adding to a well-established framework. Several points need to be clarified.
1) The authors present NIRTRS and CIT parameters between summer and winter seasons. The environmental characteristics are not described.
2) In winter, the (deoxy-Hb)DEL decreased; (oxy-Hb)SUP significantly increased but not (deoxy-Hb)SUP. Tha authors conclude by saying that "(oxy-Hb) but not (deoxy-Hb) has a similar accuracy as (total-Hb)". BAT oxidative metabolism increases after cold acclimation (winter) and an increase in deoxy-Hd would have been expected. Considering that this is a "applicability" study for evaluating BAT density, I would like to suggest to the authors to deepen this topic. Deoxy-Hb in supraclavicular region should be better investigated.
3) Supraclavicular region is a structurally heterogeneous region and it is not exclusively composed of brown adipose tissue. How do Authors measure the specific brown adipose tissue contribution to NIRS signal?
4) Why don't the Authors consider the oxy- and deoxy-myoglobin?
4) Have the authors data about other BAT regions?
Author Response
April 26, 2019
Editor, International Journal of Molecular Sciences
Dear Dr. Molly Lu
Thank you for your kind comments about our manuscript (ijms-491667). We have responded to the reviewers’ comments point-by-point below and have revised our manuscript accordingly. The corrections are highlighted in yellow.
Reviewer 1
This study examined evidence that near-infrared time -resolved spectroscopy (NIRTRS) is capable of estimating human BAT density. The Authors determined seasonal (winter and summer) variations in BAT density and describe an association between BAT volume and CIT.
The authors concluded that NIRTRS is an useful tool for evaluating BAT density in a simple, non-invasive manner. Similar papers have already published, also from the same authors, and the results are not homogeneous.
The manuscript is potentially interesting but the data are not conclusive. However, as is the case whenever a novel idea is presented, the results need to be much more strong than if the study is adding to a well-established framework. Several points need to be clarified.
Q1. The authors present NIRTRS and CIT parameters between summer and winter seasons. The environmental characteristics are not described.
A1. We have apologized for confusing that CIT was evaluated only in winter. We have edited that the CIT is evaluated only in winter (P6, line239 and P7, line 257).
Further, we have added minimum and maximum temperatures (P6, line 194-196).
Q2. In winter, the (deoxy-Hb)DEL decreased; (oxy-Hb)SUP significantly increased but not (deoxy-Hb)SUP. The authors conclude by saying that "(oxy-Hb) but not (deoxy-Hb) has a similar accuracy as (total-Hb)". BAT oxidative metabolism increases after cold acclimation (winter) and an increase in deoxy-Hd would have been expected. Considering that this is a "applicability" study for evaluating BAT density, I would like to suggest to the authors to deepen this topic. Deoxy-Hb in supraclavicular region should be better investigated.
A2. As, based on the previous study (Nirengi et al. Obesity 2015), NIRTRS indices did not change during an acute 2-hour cold exposure, NIRTRS indices could be interpreted as measurements of BAT density rather than BAT activity. Therefore, we believe that NIRTRS does not detect the oxidative metabolism of BAT in either winter or summer.
However, as pointed by the Reviewer 1, it was observed that the [deoxy-Hb]del in the muscle region decreased in winter while [deoxy-Hb]sup did not change. Therefore, we have added about the interpretation of [deoxy-Hb]sup in discussion (P4, line 109-111) and that NIRTRS indices could be parameters of BAT density rather than BAT activity in limitation (P5, line 185-187).
Q3. Supraclavicular region is a structurally heterogeneous region and it is not exclusively composed of brown adipose tissue. How do Authors measure the specific brown adipose tissue contribution to NIRS signal?
A3. As pointed out by the Reviewer 1, the supraclavicular region is heterogeneous and hence NIRTRS index results might be influenced by multiple tissue types other than BAT (e.g. white adipose tissue, muscle tissue, etc.). However, our previous work revealed a significant relationship between [total-Hb] and the 18FDG-PET/CT index in the supraclavicular region (r = 0.74), but not in deltoid muscle region. We added that NIRTRS indices were influenced by other tissue in limitation (P 5, line178-182).
Q4. Why don't the Authors consider the oxy- and deoxy-myoglobin?
A4. I appreciate for your comment. We have added sentences as below in method and limitation. I’d like to present as Hb for simplicity.
Method: For the sake of accuracy, it should be noted that NIRTRS cannot distinguish Hb from myoglobin (Mb). Hence, many studies have been reported in which results are presented as oxy-Hb/oxy-Mb, deoxy-Hb/deoxy-Mb, and total-Hb/total-Mb concentrations [17]. However, for simplicity, in this work we present the absorption measurements as [oxy-Hb], [deoxy-Hb], and [total-Hb]. (P6, line 221-225).
Limitation: Second, NIRTRS cannot differentiate Hb from myoglobin (Mb), and it has been reported that Mb is present in BAT [26]. In order to investigate the proportion of Hb in human BAT, 1proton (1H)-magnetic resonance spectroscopy measurements are needed. (P5, line182-185).
Q5. Have the authors data about other BAT regions?
A5. Thank you for greate suggestion. It is interesting whether seasonal changes have been observed in other BAT regions. However, we didn’t evaluate other BAT regions because the supraclavicular region is one of the representative locations for BAT deposits where NIRS can only be applied with a mean light depth of 20 mm. We have added limitation about other BAT regions (P5, line 187-190).
Author Response File: 
Author Response.docx
Reviewer 2 Report
Nirengi, Fuse et al use near-infrared imaging for assessing brown fat activity in humans in relation to season-dependent cold exposure. They find that oxygenated hemoglobin (among other parameters) can be used to detect seasonal changes in brown fat. I have a few minor comments the authors should consider.
A minor English language check is recommended - some wording is inappropriate and out of context.
In the end, confirming side by side with a conventional PET-CT that the new approach works would be ideal - this should be pointed out.
In the introduction, the authors should also mention short-wave infrared as a related brown fat imaging platform with contrast agents (Bartelt et al. Nature Medicine 2018).
The depiction of the data in bar graphs is misleading - I suggest plotting individual data points.
Author Response
April 26, 2019
Editor, International Journal of Molecular Sciences
Dear Dr. Molly Lu
Thank you for your kind comments about our manuscript (ijms-491667). We have responded to the reviewers’ comments point-by-point below and have revised our manuscript accordingly. The corrections are highlighted in yellow.
Reviewer 2
Nirengi, Fuse et al use near-infrared imaging for assessing brown fat activity in humans in relation to season-dependent cold exposure. They find that oxygenated hemoglobin (among other parameters) can be used to detect seasonal changes in brown fat. I have a few minor comments the authors should consider.
Q1. A minor English language check is recommended - some wording is inappropriate and out of context.
A1. We have re-checked English by Native English proofread service (Editage).
Q2. In the end, confirming side by side with a conventional PET-CT that the new approach works would be ideal - this should be pointed out.
A2. Thank you for recommended about NIRTRS is the new approach works would be ideal. We changed conclusion as below.
We conclude that NIRTRS is a new approach for evaluating BAT-d and could prove to be a useful tool for simple, non-invasive measurements in the clinic. However, a further validation study is definitely needed to confirm our results via side-by-side NIRTRS and conventional gold-standard 18FDG–PET/CT measurements. (P7, line 260-264)
Q3. In the introduction, the authors should also mention short-wave infrared as a related brown fat imaging platform with contrast agents (Bartelt et al. Nature Medicine 2018).
A3. Thank you for providing information. We mentioned a recent animal study using longer wavelength (1000-2000 nm) than ours (760-830 nm) with contrast agents, which are not suitable to human study. Further, light absorption by water limits photon penetration through biological tissue when wavelengths above 900 or 1000 nm are used for emission or detection [17], as was reported for the aforementioned BAT visualization experiment. To allow better penetration, almost all human NIRS studies, in common with our work, have utilized wavelengths in the region of 650-850 nm. We have added this topic in discussion (P5, line168-177).
Q4. The depiction of the data in bar graphs is misleading - I suggest plotting individual data points.
A4. Based on you comment, we have tried indicating individual data points (Please find the attachment). However, unfortunately, since number of samples are so many (n = 58) that we have worried for readers to confuse understanding. Therefore, we would like to present it as average bar graphs.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The authors have responded to all of my concerns. The new version is now approved to publish.
