Fractures in CKD Patients—Risk Analysis in RRT Lombardy Patients

Round 1

Reviewer 1 Report

This retrospective study depict the fracture situation in CKD patients in a particular region with a lot of data and findings that are more or less in line with the other reports from various region, anyway worth doing it.

Comments:

Figure 1 is too simple and questionable for presentation as such, while 2 may be enlarged for better overview although maybe the least interesting for the readers.

Discussion section is too long although with a lot of useful data, and a bit difficult to follow. It may be subdivided into various paragraphs with some subheadings as well as presenting data in a table for an improved overview.

Although difficult to draw any real recommendation from a retrospective study, still conclusion should be more focused with a couple of points instead of narative repetition of the previous results and discussion.   

Author Response

This retrospective study depicts the fracture situation in CKD patients in a particular region with a lot of data and findings that are more or less in line with the other reports from various region, anyway worth doing it.

Thank you for your kind comment.

Comments:

Figure 1 is too simple and questionable for presentation as such, while 2 may be enlarged for better overview although maybe the least interesting for the readers.

Thank you for the comment. We agree with the observation and decide to delate Fig 1, and transfer the information in the text. Fig 2 may be ameliorated.

Discussion section is too long although with a lot of useful data, and a bit difficult to follow. It may be subdivided into various paragraphs with some subheadings as well as presenting data in a table for an improved overview.

The observation is correct and the authors gladly accepted the suggestion to divide the text into subsections so as to simplify reading

Although difficult to draw any real recommendation from a retrospective study, still conclusion should be more focused with a couple of points instead of narrative repetition of the previous results and discussion. 

The authors accepted the observation and modified the conclusion

Reviewer 2 Report

In this MS the authors report the results of a large retrospective study on fractures in a large cohort of Italian  RRT patients. The study is analytically described, and  the results are important.  I have a few concerns, which are released to the authors as follows:

 

Majot

 

-I would suggest adding a table (Table 1)  with demographics data (age, gender  BMI at least) in the different age groups.

 

It would be of major interest  understanding how increased is the risk  of fracture risk in CKD as compared to age-matched control subjects in Lombardy, 

 

In the discussion,the authors need adding a paragraph enlighteninig strenghts and limitations of the study

 

 Conclusions: “On multivariate analysis, 31 age >65 years, female gender, PPI therapy, and cerebrovascular disease were found to be strongly 32 associated with fractures in dialysis patients, whereas undergoing renal transplantation presented a reduced risk”

 

Comment: it is unclear to this reviewer if a younger age (if so) in KTRs may explain the low risk of fracture.

 

Author Response

 In this MS the authors report the results of a large retrospective study on fractures in a large cohort of Italian RRT patients. The study is analytically described, and the results are important.  I have a few concerns, which are released to the authors as follows:

-I would suggest adding a table (Table 1) with demographics data (age, gender BMI at least) in the different age groups.

 

Thank you for your comment, but as we report in in the Materials and Methods chapter the study was made using data from the “Administrative database of Lombardy Health Service” which unfortunately does not include clinical or anthropometric findings. So, the authors could only consider the age and gender of the patients, as reported in the text.

 

It would be of major interest understanding how increased is the risk of fracture risk in CKD as compared to age-matched control subjects in Lombardy

 

Your comment is very important, but unfortunately, we are unable to extract this result because only subjects with CKD on RRT are available in the database.  The data in the literature regarding the general population in Italy on a regional basis are scanty, and from the only paper published in Archives of Osteoporosis (2019) 14:81, we can find an incidence of hip fractures in Lombardy in the year 2011 of 6.82 per 1000 subjects, a value that is about 1/3 of what was obtained in our study. However, it is important to emphasize that this result refers to the total number of hospital admissions for fracture, with no indication of whether the admission is unique or whether multiple admissions are counted for the same patient, unlike in the present paper.  However, we mentioned this result in the discussion.

 

In the discussion, the authors need adding a paragraph enlightening strengths and limitations of the study

Thank for your comment: Strengths and weaknesses of the study are already reported in the text, but willingly the authors will devote a paragraph to this.

 

Conclusions: “On multivariate analysis, 31 age >65 years, female gender, PPI therapy, and cerebrovascular disease were found to be strongly 32 associated with fractures in dialysis patients, whereas undergoing renal transplantation presented a reduced risk” 

Comment: it is unclear to this reviewer if a younger age (if so) in KTRs may explain the low risk of fracture.

 

Your comment is correct: really patients who underwent transplantation have a significantly lower mean age than non-transplant recipients, which together with the reduced presence of comorbidities gives reason for the reduced fracture risk. The results were added in the text The variable age was also analysed as a continuous variable (year), without losing its significance as a risk factor for fractures. The use of a cut-off of 65 years is justified by the distribution of age in the groups. mean age of KTx patients 50.14 years vs. 68.40 in the nontransplant recipients.

	1st Qu.	Median	Mean	3rd Qu
No KTx (7571)	61,156	71,372	68,397	78,292
KTx (538)	41,537	51,131	50,144	610,899

Reviewer 3 Report

This manuscript is an analysis of a large database with a large number of samples. Some suggestions are as follows:

1.      Abbreviations should be included in the full text for the first time.

2.      Some variables should be included in the overall analysis, such as the history of CKD, the history of osteoporosis, and the cause of fracture, otherwise it is easy to cause bias when interpreting the results.

3.      The fracture rate of CKD patients compared with the general population should be included in the discussion.

 

Author Response

This manuscript is an analysis of a large database with a large number of samples. Some suggestions are as follows:

1. Abbreviations should be included in the full text for the first time.

The authors have reviewed the text for possible shortcomings.

2. Some variables should be included in the overall analysis, such as the history of CKD, the history of osteoporosis, and the cause of fracture, otherwise it is easy to cause bias when interpreting the results.

The authors are aware of the importance of this information, but unfortunately, as reported in the materials and methods chapter, this kind of information was not present in the administrative database of the Lombardy region health service. Moreover, a valid interpretation of these findings can be more properly performed in a prospective study than in a retrospective one.

3. The fracture rate of CKD patients compared with the general population should be included in the discussion.

The authors accept this suggestion, but unfortunately, we are unable to extract this result because only subjects with CKD on RRT are available in the database.  The data in the literature regarding the general population in Italy on a regional basis are scanty, and from the only paper published in Archives of Osteoporosis (2019) 14:81, we can find an incidence of hip fractures in Lombardy in the year 2011 of 6.82 per 1000 subjects, a value that is about 1/3 of what was obtained in our study. However, it is important to emphasize that this result refers to the total number of hospital admissions for fracture, with no indication of whether the admission is unique or whether multiple admissions are counted for the same patient, unlike in the present paper.  However, we mentioned this result in the discussion.

 

Round 2

Reviewer 3 Report

The author has responded and explained the comments. Agree to publish.