Improved Microbiological Diagnosis of Bone and Joint Infections Using Mechanical Bead-Milling Extraction of Bone Specimens with the Ultra-Turrax^® System

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript compared the performance of the Ultra-Turrax® bead grinding system with the standard vortex homogenization method in the microbial diagnosis of bone and joint infections through a prospective cohort study. At the same time, it combined traditional statistics with the Bayesian subcategory model analysis to make up for the lack of a gold standard. The results showed that the Ultra-Turrax® system significantly improved the detection sensitivity, especially in bone sample and hip joint prosthesis infections, without increasing the risk of contamination, and had important clinical and laboratory application value. However, some questions in this manuscript should be addressed.

1. The sample size of the knee joint prosthesis infection subgroup was only n=25, which led to the sensitivity improvement not reaching statistical significance and might limit the generalization of the conclusion of this subgroup. Suggestingto clearly state the limitations in the discussion to avoid over-interpretation.
2. 16S rRNAgene PCR was used for the detection of cultured negative samples, but key parameters such as primer sequences, amplification conditions and product validation were not described in detail, which affected the repeatability of the experiment. Recommending to supplement the complete process and verification data of the PCR experiment to ensure that other laboratories can reproduce it.
3. Only evaluated the diagnostic performance such as sensitivity, specificity and detection rate, and did not explore the impact of this method on the clinical outcomes of patients. Suggesting to supplementclinical outcome data or propose future research directions in the discussion to further verify the clinical value of this diagnostic method.
4. The current discussions only mention the operational advantages of Ultra-Turrax® over ultrasonic treatment, but lack a direct comparison of diagnostic performance with other mechanical homogenization techniques such as ultrasonic methods.
5. It is poorly formattedin table 2, such as ambiguous correspondence between methods and indicators, and disordered numerical arrangement, which affects readability. Suggesting to reorganize the table structure and unify the data presentation format.
6. The terms "bead-beating homogenization" and "mechanical bead-milling" are alternately used in the text. Recommendingto unify the terms for enhancing the rigor of the thesis.

Author Response

Reviewer 1

-Comments 1 : The sample size of the knee joint prosthesis infection subgroup was only n=25, which led to the sensitivity improvement not reaching statistical significance and might limit the generalization of the conclusion of this subgroup. Suggesting to clearly state the limitations in the discussion to avoid over-interpretation.

Response 1 : We fully agree with the reviewer. This limitation was acknowledged in the revised discussion section, where we explicitly state that the small sample size in the knee prosthesis subgroup may restrict the generalizability of our findings. This change can be found : page 9, paragraph Discussion, lines 918-920.

 

-Comments 2 : 16S rRNAgene PCR was used for the detection of cultured negative samples, but key parameters such as primer sequences, amplification conditions and product validation were not described in detail, which affected the repeatability of the experiment. Recommending to supplement the complete process and verification data of the PCR experiment to ensure that other laboratories can reproduce it.

Response 2 : Thank you for pointing this out. All details of the protocol are clearly described in Reference 17 (page 3, paragraph 2.4, line 176 and  References section).

Specifically, the method includes proteinase K treatment (2 g/L for 3 h at 65°C), automated DNA extration using the MagNA Pure system (Roche), and real-time PCR using SYBR Green targetting the 5′ region of the 16S rRNA gene (forward primer 27F, 5′-AGA GTT TGA TCM TGG CTC AG-3′; reverse primer 685R3, 5′-TCT RCG CAT TYC ACC GCT AC-3′; expected amplicon size : 658 b). Amplicons were sequenced bidirectionnally and compared with sequences in the BIBI and BLAST databases for species-level identification (≥98% similarity). All of these procedures follow the published protocol in Reference 17 without any modification.

 

-Comments 3 : Only evaluated the diagnostic performance such as sensitivity, specificity and detection rate, and did not explore the impact of this method on the clinical outcomes of patients. Suggesting to supplementclinical outcome data or propose future research directions in the discussion to further verify the clinical value of this diagnostic method.

Response 3 : We agree with the reviewer that assessing clinical outcomes would provide additional value. However, this was beyond the scope of the present study, which focused on microbiological performance. We have added a statement in the discussion highlighting the need for future prospective studies to evaluate whether improved diagnostic accuracy translates into better clinical outcomes. This change can be found : page 9, paragraph Discussion, lines 925-926.

 

-Comments 4 : The current discussions only mention the operational advantages of Ultra-Turrax® over ultrasonic treatment, but lack a direct comparison of diagnostic performance with other mechanical homogenization techniques such as ultrasonic methods.

Response 4 : We appreciate this suggestion. We have added a paragraph in the discussion comparing our results with published data on ultrasonic homogenization. For example, Liu et al. (Eur J Clin Microbiol Infect Dis, 2017 – Ref 29) reported a sensitivity of 79% and specificity of 95% after sonication, while Hoekstra et al. (J Bone Joint Infect, 2020 – Ref 30) reported sensitivity of 80.5% and specificity of 97.8%. These comparisons are now included to contextualize the performance of bead-milling. This change can be found : page 8, paragraph Discussion, lines 871-875.

 

-Comments 5 : It is poorly formatted in table 2, such as ambiguous correspondence between methods and indicators, and disordered numerical arrangement, which affects readability. Suggesting to reorganize the table structure and unify the data presentation format.

Response 5 : We thank the reviewer for pointing this out. Table 2 has been completely reformatted for clarity, ensuring consistent alignment of indicators and numerical values. This change can be found : page 5, paragraph Table 2, lines 265-266.

 

-Comments 6 : The terms "bead-beating homogenization" and "mechanical bead-milling" are alternately used in the text. Recommending to unify the terms for enhancing the rigor of the thesis.

Response 6 : We agree and have standardized terminology througout the manuscript, exclusively using the term « bead-milling ».

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for letting me review this excellent paper. It has scientific merit and I believe it will be of interest to a range of readers.

Line 81: please define the term “recent surgery.”

Line 96: replace “peripherical” with “peripheral.”

Line 102: specify the type and brand of blood culture bottles/vials used – for example, anaerobic and aerobic. Were fungal blood culture vials used?

Table 1: some spelling mistakes are present, i.e. neutrophiles (instead of neutrophils) and materiel (instead of material).

Line 191: rephrase “more positive cultures that…” as “more positive cultures than…”

Line 236: the authors claim that no fungi were isolated. Do explain how this was ascertained when 1) the incubation duration of agar plates was rather short (i.e. 7 days only vs. the usual 2-4 weeks for fungal cultures) and 2) the use of fungal selective media such as SDA was not mentioned.

The discussion was well-written and the conclusion was sound.

Author Response

Reviewer 2

-Thank you for letting me review this excellent paper. It has scientific merit and I believe it will be of interest to a range of readers.

We thank the reviewer for this encouraging feedback.

-Line 81: please define the term “recent surgery.”

We clarified that « recent surgery » refers to procedures perfomed within the last 3 months. This change can be found : page 2, paragraph 2.2, line 122.

-Line 96: replace “peripherical” with “peripheral.”

We modified accordingly. This change can be found : page 2, paragraph 2.4, lines 150.

-Line 102: specify the type and brand of blood culture bottles/vials used – for example, anaerobic and aerobic. Were fungal blood culture vials used?

We added details: synovial fluid samples were inoculated into BD BACTEC™ Plus Aerobic and BD BACTEC™ Lytic Anaerobic bottles. This change can be found : page 3, paragraph 2.4, lines 156-157

Fungal-specific bottles were not used, as they are reserved for candidemia investigations in our institution. However, the bottles used can detect yeasts if present (French Referentiel of Microbiology REMIC, French Society of Microbiology and Manual of Clinical Microbiology, ASM Press).

-Table 1: some spelling mistakes are present, i.e. neutrophiles (instead of neutrophils) and materiel (instead of material).

We corrected the spelling mistakes accordingly. This change can be found : page 4, paragraph Table 1, line 239

-Line 191: rephrase “more positive cultures that…” as “more positive cultures than…”

We corrected the typo accordingly. This change can be found : page 5, paragraph 3.3, line 271.

-Line 236: the authors claim that no fungi were isolated. Do explain how this was ascertained when 1) the incubation duration of agar plates was rather short (i.e. 7 days only vs. the usual 2-4 weeks for fungal cultures) and 2) the use of fungal selective media such as SDA was not mentioned.

As described in the  section ‘2.5 Microbiological processing’ (P. 3, line 170), samples were incubated for 14 days, which is sufficient for yeast recovery according to French (REMIC, SFM) and American (Manual of Clinical Microbiology, ASM Press) guidelines. These references state that Candida species grow well on blood agar and can even grow under anaerobic conditions on enriched media. Although selective fungal media were not used, the extended incubation and enriched media strongly support the absence of Candida in our collection.

-The discussion was well-written and the conclusion was sound.

We thank the reviewer for this positive assessment.