Biomakers in Chronic Chagas Cardiomyopathy
Round 1
Reviewer 1 Report
The article entitled "BIOMAGENS IN CHAGASIC HEART DISEASE", submitted to the journal Microorganism by Angela Braga Rodrigues et al, is an interesting study in which they show a correlation between a series of biochemical and immunological markers present in the serum of patients affected by Chagasic heart disease and the prognosis of these patients. Although some of the markers of inflammation, such as C-reactive protein, uric acid or Ca-125, there does not seem to be a correlation between circulating serum values and the severity of the disease. Elevated levels of Th1 inflammatory interleukins (IL-12 and IL-8) show a correlation between the levels of these interleukins and the severity of cardiac ventricular lesions. Interestingly, elevated IL-8 correlates with IL-6 in patients with right ventricular dilatation. They also found that levels of BNP natriuretic peptide, a vasodilator peptide and antagonist of the renin-angiotensin-aldosterone system, may be an effective marker of cardiac damage and severity of damage caused by Chagas disease.
In my opinion, the authors should have determined, in addition to the interleukins studied, the levels of IFN g (Th1) and the Th17 response by determining the values of IL-17 and IL-23 as a Th17 inflammatory response and IL-23 as an interleukin that maintains this Th17 response. Experimental studies have observed an increase of IL-17 in infected individuals and perhaps it would have been interesting to verify the relationship between these interleukins and BNP levels.
The work may be publishable in the journal, although the authors could have tried to publish it in a more clinically oriented journal with a wider dissemination in cardiology.
Author Response
#
Thank you very much for your valuable comment. Determination of Th17 response is an excellent suggestion. It would be great if further studies could address the IL-17 and IL 23 role in CCC. It is a limitation of our investigation.
In the new version, we have now added a paragraph under the discussion section, commenting on the possible important role of IL-17 in the immunomodulatory response in CCC ( Sousa GR,2017). We suggest that further studies should examine the Th17 response in CCC and determine IL-17 and Il-23 circulating levels and analyze associatively with BNP.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper addresses one of the most important topics of research in Chagas disease, such as the search for biomarkers of progression and prognoses.
Although the patients included in the study have significant cardiac involvement, the results obtained may have a relevant clinical value.
The work is well written and the data is well presented. Anyway I have some doubts about the way in which they have been analyzed
After the multivariate analysis, the variable that has a greater association with an unfavorable clinical event is IL-1β. What strikes me is that when comparing the means of that cytokine in patients who have cardiac events or who die (tables 2 and 3) that variable appears practically the same in the groups (in fact in table 3 it has practically the same mean value, and has a p of 0.08) even with a p in table 2 of 0.3, so a priori it would not have to be entered into the model. Surprisingly, it is the one that appears with the greatest relationship with death. This analysis has not been clear to me.
Likewise, when the three levels of analysis are done in the multivariate, I do not understand how they have been chosen. In material and methods it is literally commented that:
“variables were divided in three different blocks accordingly to clinical, echocardiography or laboratorial categories. The final model was set with the most significant variables for predicting death”
But in the way they are shown, the echocardiographic and laboratory variables are mixed.
Continuing with Table 5 of multivariate, it seems that RV has a protective effect against death, risk of 0.8 while in Table 4 it clearly appears as a risk of more than 6 times. That's right?
I believe that these analyses would have to be reassessed in order to have greater clarity on the conclusions of the study.
Author Response
This paper addresses one of the most important topics of research in Chagas disease, such as the search for biomarkers of progression and prognoses.
Although the patients included in the study have significant cardiac involvement, the results obtained may have a relevant clinical value.
The work is well written and the data is well presented. Anyway I have some doubts about the way in which they have been analyzed
After the multivariate analysis, the variable that has a greater association with an unfavorable clinical event is IL-1β. What strikes me is that when comparing the means of that cytokine in patients who have cardiac events or who die (tables 2 and 3) that variable appears practically the same in the groups (in fact in table 3 it has practically the same mean value, and has a p of 0.08) even with a p in table 2 of 0.3, so a priori it would not have to be entered into the model. Surprisingly, it is the one that appears with the greatest relationship with death. This analysis has not been clear to me.
#
Thank you so much for giving us the opportunity to improve our work. There were incorrect values in table 3 due to editing mistake. In the new version, we have now corrected those values( page 19 and 20). Thank you for alerting us.
In the univariate analysis, IL 1β and IL-12 have shown significance (p=0.006 and p=0.009), thus they have taken part in the multivariate model, even though when comparing their median values between those patients that survived and those that did not, there was no significance.
Likewise, when the three levels of analysis are done in the multivariate, I do not understand how they have been chosen. In material and methods it is literally commented that:
“variables were divided in three different blocks accordingly to clinical, echocardiography or laboratorial categories. The final model was set with the most significant variables for predicting death”
But in the way they are shown, the echocardiographic and laboratory variables are mixed.
Continuing with Table 5 of multivariate, it seems that RV has a protective effect against death, risk of 0.8 while in Table 4 it clearly appears as a risk of more than 6 times. That's right?
I believe that these analyses would have to be reassessed in order to have greater clarity on the conclusions of the study.
#
Thank you for mentioning that and helping us to ameliorate our work.
We have now corrected table 5( page 22) In the new version, table 5 now shows only the most important variable from each category. For the clinical block: pulmonary rales; for the echocardiogram, RV dilation and for laboratory variables, IL1 β.
To sum up, in the multivariate analyzes, three blocks of variables were examined: clinical, echocardiography and laboratory variables. We have then selected the most important clinical (pulmonary rales), echocardiography (RV dilation) and laboratory (IL-1β) form each block. In the final step, these three variables where then analyzed together to determine the most important indicator of death.
Regarding, RV dilation analyzes, it interesting that in the univariate analysis it was a powerful event, with a 6 times higher risk of death. When analyzed in the presence of IL 1 β, it lost its power and significance. If we further examine, RV dilation, without IL-1β, it reaches significance, in the presence of pulmonary rales (4 times risk of death) and is then the most important predictor of death ( page 20-21)
|
Variables |
Significance |
Estimated Risk |
CI |
|
Pulmonary rales |
0.159 |
1.878 |
0.782 – 4.512 |
|
RV dilation |
0.129 |
4.105 |
0.661 – 25.479 |
|
|
|
|
|
|
RV dilation |
0.037 |
6.081 |
1.112 – 33.259 |
Author Response File: Author Response.pdf
Reviewer 3 Report
The article written by Rodrigues et al. involves important topic concerning Chagas disease, which is still a parasitological problem in the world. The diagnostics and treatment are poorly developed, therefore new research is significant. However, paper written by Rodrigues et al. is poorly written and has many dark sides.
English correction is obligatory.
The title of the work indicates the biomarkers that can be used to determine CCC. And the aim of the study is to determine biomarkers that may indicate a severe course of CCC.
Introduction is too short and a reader is not properly introduced to the subject.
Why didn’t you choose to determine NT-proBNP?
Why did you choose to examine CRP? You cite Yin et al. [2003] who examined hs-CRP not CRP. CRP levels rise in inflammation, infections, and a heart attack, surgery, and trauma. CRP is often referred to as an acute phase protein. There are two different tests for the determination of CRP: the standard CRP test and CRP determined by the highly sensitive method, the so-called highly sensitive CRP (hs-CRP). Each of them allows the determination of different ranges of CRP in the blood. The hs-CRP test is a more accurate (and more sensitive) determination of lower protein levels than the standard CRP test and is used to assess an individual risk of coronary artery disease.
Material and methods section is shortly written. A reader doesn’t know how many patients were involved in the study. Were they divided into groups? Please describe NHYA classes. What tests did you choose to examine all studied markers?
Line 95. So how many patients were classifies as NYHA – 22 or 2?
Table 1: What data is shown in the brackets? Did you involve 50 patients who were divided into LVEF>35% group and LVEF<35% group? If in the brackets are min-max, the range of the two subgroups should be total patients (n = 50). For example, if the BNP range of patients (n =50) is 154-445, then how come the sudden range starts with 40 in LVEF patients>35%? This happened in each studied marker. Moreover, death- 5 patients out of 50. But then 3 out of 28 (in LVEF>35%), and 3 out of 22 (LVEF<35%), which gives 6 out of 50- so how many died? I’m confused.
In table 1, I suggest to put reference values of the studied markers in patients with cardiomiopathy.
Line 105. “Right ventricular dilation (RV dilation) has been shown to be an important prognostic factor not only in patients with CCC but also in other etiologies of HF.” Did you check it in your study? If not, why is this in the result section?
Figure 1. It doesn’t show a relationship. You didn’t examine correlation between two paramethers. You simply showed concentration of BNP, Ca-125, uric acid and CRP in patients with and without RV dilation.
You should have control group of patients (only with cardiomiopathy). Because you dont know how these bioamarkers change in the serum of patients only with cardiomiopathy. Only then, you can say then a parameter may be a biomarker of specific disease.
Author Response
The article written by Rodrigues et al. involves important topic concerning Chagas disease, which is still a parasitological problem in the world. The diagnostics and treatment are poorly developed, therefore new research is significant. However, paper written by Rodrigues et al. is poorly written and has many dark sides.
English correction is obligatory.
#
Thank you for alerting us and helping us to improve the quality of our work. We have now sent it to a professional English review and will soon make the necessary correction and adjustments.
The title of the work indicates the biomarkers that can be used to determine CCC. And the aim of the study is to determine biomarkers that may indicate a severe course of CCC.
Introduction is too short and a reader is not properly introduced to the subject.
#
Thank you for helping us to improve our study. We have extensively reviewed the introduction and made the necessary adjustments in the new version (page 4 and 5)
Why didn’t you choose to determine NT-proBNP?
#
Thank you for your question.
Unfortunately, our financial resources for research are scarce and we have to work on a limited budgeted. At the time of our investigation it was available to us a BNP triage kit test.
Why did you choose to examine CRP? You cite Yin et al. [2003] who examined hs-CRP not CRP. CRP levels rise in inflammation, infections, and a heart attack, surgery, and trauma. CRP is often referred to as an acute phase protein. There are two different tests for the determination of CRP: the standard CRP test and CRP determined by the highly sensitive method, the so-called highly sensitive CRP (hs-CRP). Each of them allows the determination of different ranges of CRP in the blood. The hs-CRP test is a more accurate (and more sensitive) determination of lower protein levels than the standard CRP test and is used to assess an individual risk of coronary artery disease.
#
We sought to determine standard CPR due to its greater availability and lower cost. Our aim was to investigate simple parameters that could be measured in most places.
Thank you for mentioning Yin et al (2003), they have indeed studied Hs-CPR. Therefore, in the new version we have changed our reference for one that is more appropriate ( page 28).
In the new version we now cite Anand ( 2005 ).The authors performed a retrospective analysis of the predictive values for C-reactive protein for long term outcomes and concluded that higher C-reactive levels were associated with morbidity and mortality ( page 28).
Material and methods section is shortly written. A reader doesn’t know how many patients were involved in the study. Were they divided into groups? Please describe NHYA classes. What tests did you choose to examine all studied markers?
#
Thank you for giving us the opportunity to improve the quality of our work. We have revised the material and methods section and made the necessary adjustments in the new version.
50 patients with CCC were selected. NYHA classification is: NYHA I: 22/50 patients, NYHA II 15/50 patients, NYHA III 11/50 patients and NYHA IV 2/50 patients (page 6).
Patients were subdivided in groups to compare the markers accordingly to hemodynamic parameters and to investigate if there were significant differences between them. So we have verified if patients with LVEF less than 35% had higher levels of BNP, ca-125, uric acid and C-reactive protein comparing to patients with LVEF of 36-55%. The subgroup of patients with LVEF less than 35% showed higher BNP, ca-125 and uric acid serum levels. No difference was found for C-reactive protein
The same investigation was performed regarding the presence of RV dilation on echocardiogram. Patients enrolled in the study were subdivided according to the presence or the absence of RV dilation on echo scans. We have verified if BNP, ca-125, uric acid and C-reactive levels were significantly different between the two subgroups. Patients that showed the presence of RV dilation on echo scans had higher levels ca-125,uric acid and C-reactive serum levels.
Line 95. So how many patients were classifies as NYHA – 22 or 2?
#
Thank you for giving us the opportunity to clarify our work. This is now clearer in the new version and under the materials and methods section( page 6) The classification is:
NYHA I -> 22 patients (out of 50 patients)
NHYA II ->15 patients (out of 50 patients)
NYHA III -> 11 patients (out of 50 patients)
NYHA IV -> 2 patients (out of 50 patients)
Table 1: What data is shown in the brackets? Did you involve 50 patients who were divided into LVEF>35% group and LVEF<35% group? If in the brackets are min-max, the range of the two subgroups should be total patients (n = 50). For example, if the BNP range of patients (n =50) is 154-445, then how come the sudden range starts with 40 in LVEF patients>35%? This happened in each studied marker. Moreover, death- 5 patients out of 50. But then 3 out of 28 (in LVEF>35%), and 3 out of 22 (LVEF<35%), which gives 6 out of 50- so how many died? I’m confused.
#
Thank you for helping us to improve our work. We have studied 50 patients with CCC. Those patients were further grouped according to LVEF of 35% to compare the markers serum levels between the two groups (page 7-8)
In table 1, we have reported patients’ features in addition to the results concerning the serum markers levels and circulating cytokines accordingly to LEVEF of 35%. This had led to some confusion. So in the new version, this is corrected, now we have one table for reporting patients’ features and one figure to show the markers differences in patients grouped according to LVEF of 35% ( page 10 and 12).
In the brackets the data shown regards 25th and 75 th percentiles, since we have reported median levels for variables that were not normally distributes. In the new version this is now clear and we indicate this in the table legend.
There is indeed an editing mistake. There were 6 patients who died. This is corrected in the new version. Thank you for noting this.
In table 1, I suggest to put reference values of the studied markers in patients with cardiomiopathy.
#
Thank you for mentioning this. I am not sure what the reference values for these markers in patients with cardiomyopathy. To the best of my knowledge, there is not a clear definition on those levels.
Concerning uric acid, a traditionally marker for the diagnosis of gout, the threshold for the risk of lifelong hyperuricemia is 6mg/dl. In the study by Hamaguchi (2011), hyperuricemia was defined as levels higher than 7 mg/dl. In the study by Stefan de Anker (2003), in the ROC analysis, the best uric acid levels for predicting survival in patients with chronic HF was 9.5 mg/dl.
Regarding ca-125 levels, a well established tumor marker, the usual serum values are 35 U/ml. In the study by Nunez (2016), in patients hospitalized for acute HF, the cut off value for guiding therapy was also 35 U/ml. According to Hung (2013) the range of value and potential clinical implication of ca-125 remains unknown, with variability between investigations involving acute or chronic HF and even healthy volunteers.
The Val-HeFt study demonstrated that patients with CPR above 3.23mg/l had lower LVEF were in more advanced functional class and had an increased risk of an adverse event or death. In patients with acute HF there is a positive correlation with the risk of death with CPR levels above 12 mg/dl(Huynh, 2015; Anand,2005).
Plasma BNP values are set at a value of 100 ng/l. To the best of my knowledge there are no definite normal values of cytokines in patients with cardiomyopathy.
Line 105. “Right ventricular dilation (RV dilation) has been shown to be an important prognostic factor not only in patients with CCC but also in other etiologies of HF.” Did you check it in your study? If not, why is this in the result section?
#
Thank you for helping us to improve the quality of our work. Now we mention that RV dilation is an important prognostic factor under the methods section ( page 7) Our aim was to compare the markers in patients with and without RV dilation and investigate if there were significant differences.
Figure 1. It doesn’t show a relationship. You didn’t examine correlation between two paramethers. You simply showed concentration of BNP, Ca-125, uric acid and CRP in patients with and without RV dilation.
#
Thank you for mentioning this. In the new version, we have changes the figure title to: RV dilation and BNP, ca-125, uric acid and CPR ( page 13).
You should have control group of patients (only with cardiomiopathy). Because you dont know how these bioamarkers change in the serum of patients only with cardiomiopathy. Only then, you can say then a parameter may be a biomarker of specific disease.
#
Thank you for pointing that out. This is a limitation of our study; we did not have a control group of patients with other etiologies of cardiomyopathy.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Thank you for considering my suggestions
Author Response
Thank you very much for your suggestions and in helping us improve the quality of our work.
Reviewer 3 Report
Manuscript has been changed. Thank you for putting most of my suggestion into text. However, I still have some minor remarks.
According to the authors, the manuscript was checked by English native speaker. However, English correction is still obligatory (i.e. line 51 where is “tree” instead of “three”).
Line 23: I don’t understand why authors wrote how may male were in the study. I would suggest to delete this information (“32 male”). It’s confusing because a few lines higher is the information that there were 50 patients.
Table 1: Still don’t know how may patients died. In the response to my review, authors wrote that 6, in the paper is 5, but when we calculate the %, it occurred to be 6.
Line 176: The values for patients with LVEF less than 35% are not shown in table 1
Line 186: “NYHA” instead of “HYHA”
In the reviewer response, the Authors wrote about the reference values for studied markers in patients with cardiomyopathy. I really regret that authors did not put this information in the discussion section.
Author Response
Manuscript has been changed. Thank you for putting most of my suggestion into text. However, I still have some minor remarks.
#
Thank you very much for your suggestions and giving us the opportunity to improve the quality of our work.
According to the authors, the manuscript was checked by English native speaker. However, English correction is still obligatory (i.e. line 51 where is “tree” instead of “three”).
#
We will send the manuscript to professional English reviewers, but they ask for a few days to get it done. Thank you.
Line 23: I don’t understand why authors wrote how may male were in the study. I would suggest to delete this information (“32 male”). It’s confusing because a few lines higher is the information
that there were 50 patients.
## Thank you for mentioning this. In the new version, we have now deleted the information about 32 male.
Table 1: Still don’t know how patients died may. In the response to my review, authors wrote that 6, in the paper is 5, but when we calculate the %, it occurred to be 6.
#
I apologize for this persistent mistake. In fact, 6 patients died. Thank you very much for helping us to improve the quality of our work.
Line 176: The values for patients with LVEF less than 35% are not shown in table 1
#
Thank you for noticing it. We have now added these values in table 1.
Line 186: “NYHA” instead of “HYHA”
#
We have now corrected this in the new version. Thank you.
In the reviewer response, the Authors wrote about the reference values for studied markers in patients with cardiomyopathy. I really regret that authors did not put this information in the discussion section.
#
Thank you for the interesting suggestion. We have now added in the new version, the discussion about reference values.
Author Response File: 
Author Response.docx
