Next Article in Journal
Development of New Antimycobacterial Sulfonyl Hydrazones and 4-Methyl-1,2,3-thiadiazole-Based Hydrazone Derivatives
Previous Article in Journal
Surgical Antimicrobial Prophylaxis in Patients of Neonatal and Pediatric Age Subjected to Eye Surgery: A RAND/UCLA Appropriateness Method Consensus Study
 
 
Article
Peer-Review Record

A Multicentre, Prospective, and Retrospective Registry to Characterize the Use, Effectiveness, and Safety of Dalbavancin in German Clinical Practice

Antibiotics 2022, 11(5), 563; https://doi.org/10.3390/antibiotics11050563
by Frank Hanses 1,2, Sebastian Dolff 3, Janina Trauth 4, Michael Seimetz 5 and Stefan Hagel 6,*
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Antibiotics 2022, 11(5), 563; https://doi.org/10.3390/antibiotics11050563
Submission received: 14 March 2022 / Revised: 19 April 2022 / Accepted: 20 April 2022 / Published: 22 April 2022

Round 1

Reviewer 1 Report

In this study, the authors report the data regarding the effectiveness and safety of the lipoglycopeptide dalbavancin in German clinical practice. The data described concern 9 patients, recruited in one year. Initially, the study was to be much larger including 150 patients recruited in two years. Hence, my biggest concern is the statistical validity of the reported data. However, the authors cope very well with this weakness, highlighting how this is only a preliminary study and comparing the results with others reported in the literature. Overall, the results are clear, the methods are accurately described, and the paper sounds good.

There are some aspects that can be improved such as:

1. Since this study includes a limited amount of data and is, therefore, preliminary, this aspect should be made explicit in the title and abstract.

2. The document seems to focus on the "Dalbavancin Utilization Registry in Germany". Please, add details on this Registry. What implications will this register have in the future?

3. Other statistical methods could be used to improve the study. I believe that a deeper analysis of the data at a statistical level could strengthen the study. For example: what is the probability that positive results are "false positives"?

4. Are all the dalbavancin studies shown in table 4? What criteria were used for the selection of the studies reported in this table?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear Authors

Good Day. 

Greetings. 

How do calculate sample size? 

What were the sampling methods? 

There is no control. Therefore no possibility of comparison. Please justify? 

So far my understanding is it will be difficult to generalize data for the whole community. Please mention with a separate heading study limitation.

Kindly remove the brand name of the antibiotic. 

Please add a future implication of this paper 

 

Good Luck with your paper. 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Back to TopTop