Origin and Clinical Impact of Early Multidrug-Resistant (MDR) Contamination in Patients Undergoing Pancreaticoduodenectomy

Round 1

Reviewer 1 Report

I congratulate the authors on their study on the impact of MDR contamination on the clinical outcomes of patients undergoing pancreaticoduodenectomies.

Overall, the topic of this study is interesting, relevant and worthy of publication. The manuscript can benefit from some English language editing as there were several spelling errors.

Specific comments:

Methodology: 

* Why were MPC assessed in a 90-day window rather than in a 30-day window (like the SSI)?

* Were the patients on any type of antibiotic prior to the operation? what is the routine prophylaxis given for stent placement?

Results:

* What are the characteristics of patients who underwent stenting in this cohort? What are the characteristics of Bile MDR+ vs. Bile MDR- patients?  Is it possible that MDR contamination is just a surrogate marker to malignant lesions, or more advanced stage requiring stenting and neoadjuvant therapy?

 * Can you please provide a more detailed account of the complications included in MPC? what was their timing in relation with the diagnosis of MDR ?

* What does 're-intervention' mean? is that re-operation?

* The CR-POPF and PPH rates are fairly higher than usually reported, what was the case mix in terms of complexity, and how did it relate to MDR status?

 * What was the length of stay in both groups, and overall? 

 Discussion:

* "However, the rate of MDR bacteria after pancreaticoduodenectomy, their presence at the time of the procedure in the bile, and their finding on the fifth post- operative day in the surgical drain, indicating the surgical site, clearly identify the pre- operative and early postoperative period as the timeframe in which MDR bacteria were selected and began to colonize patients undergoing pancreaticoduodenectomy."  - Since there was no assessment of the native bile pre-stenting and post-stenting, it is difficult to conclude exactly when colonization began. 

Some review of the language is advised: 

for example: 'DREN'  in table #5.

Author Response

Dear Editor/Revierw,

we want to thank you for the support and opportunity to further improve our work. As requested, we provide in this document our point-by-point answer to the reviewers’ comments.

Reviewer 1

I congratulate the authors on their study on the impact of MDR contamination on the clinical outcomes of patients undergoing pancreaticoduodenectomies.

Overall, the topic of this study is interesting, relevant and worthy of publication. The manuscript can benefit from some English language editing as there were several spelling errors.

Thank you for carefully reviewing the manuscript. We asked our language service core to review the spelling and the grammar of the manuscript.

Specific comments:

Methodology: 

* Why were MPC assessed in a 90-day window rather than in a 30-day window (like the SSI)?

Thank you for pointing out this aspect. As you point-out 30-day mobility is commonly used in general surgery to asses postoperative outcomes. Despite that, we rather use a 90-day window for postoperative complications since this is the time frame commonly used to asses post- operative complications in pancreatic surgery, especially after pancreaticoduodenectomy. Here some references: J Gastrointest Surg. 2017 Mar;21(3):506-515. doi: 10.1007/s11605-016-3346-1. Epub 2017 Jan 5.The Impact of Increasing Hospital Volume on 90-Day Postoperative Outcomes Following Pancreaticoduodenectomy; Surg Oncol. 2021 Jun:37:101319. doi: 10.1016/j.suronc.2020.01.002. Epub 2020 Jan 27.Pancreatic resections: 30 and 90-day outcomes in octogenariansShoshana T Levi; J Gastrointest Surg. 2016 Dec;20(12):1975-1985. doi: 10.1007/s11605-016-3286-9. Epub 2016 Oct 11.Preoperative Chemoradiation for Pancreatic denocarcinoma Does Not Increase 90-Day Postoperative Morbidity or Mortality Jason W Denbo; etc….

* Were the patients on any type of antibiotic prior to the operation? what is the routine prophylaxis given for stent placement?

Thank you for your questions. The preoperative prophylaxis is described in the methods section for patients with and without the pancreatic stent (see line 104). Regarding prophylaxis administration before the stent placement, at the time of the procure no antibiotic therapy is routinely administered following the ESGE guideline (Dumonceau Jean-Marc et al. Endoscopic biliary stenting: indications, choice of stents, and results: European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline – Updated October 2017. Endoscopy 2018; 50: 910–930). These guidelines are routinely adopted in our center and in centers that refer patients to us. 

Results:

* What are the characteristics of patients who underwent stenting in this cohort? What are the characteristics of Bile MDR+ vs. Bile MDR- patients?  Is it possible that MDR contamination is just a surrogate marker to malignant lesions, or more advanced stage requiring stenting and neoadjuvant therapy?

Thank you for pointing out this aspect. Considering bile MDR + and – patients we didn’t find statistically relevant differences in age (68.9 vs 65.8 years), BMI (23.3 vs 24.1), ASA score ≥ 3 (26.0% vs 21.3%), the rate of neoadjuvant therapy (22.7% vs 18.1) or malignant disease (52.6% vs 52.1%). We added this information in the text (line 150).

 * Can you please provide a more detailed account of the complications included in MPC? what was their timing in relation with the diagnosis of MDR ?

Thank you. We included in MPC all the postoperative complication grade III of higher considering the Clavien-Dindo classification (methods, line 123). So, any complication that required interventional procedure (radiological, endoscopic or surgical), admission to ICU for organ failure or death has been taken in consideration. Considering the timing of the MPC, all of them happen after the intraoperative sampling of the bile by definition and so after the first proof of a contamination by a MDR bacteria. Regarding the drain, only in 11 cases the MPC occurred before POD 5 (the day of the routine cultural exam of the drain material). Considering that, in almost all cases MDR bacteria were present in bile or at the surgical site before the major event and so they could potentially play a role in MPC developing. We added this concept to the discussion (line 288).

* What does 're-intervention' mean? is that re-operation?

Thank you for pointing out this misleading term. We change re-intervention with re-operation in the text (line 292).

* The CR-POPF and PPH rates are fairly higher than usually reported, what was the case mix in terms of complexity, and how did it relate to MDR status?

We thank you for you observation that allow us to make our work clearer to the readers. Due to the timeframe of the observations, we used both the old and new classification of the ISGPS. In the table is reported the old overall fistula rate, as reported in the methods, we define as CR-POPF the new classification that exclude biochemical leak, the overall rate of CR-POPF in our population is 21.2% that is comparable to reported previous studies (J Hepatobiliary Pancreat Sci (2011) 18:601–608DOI 10.1007/s00534-011-0373-x Predictive risk factors for clinically relevant pancreatic fistulaanalyzed in 1,239 patients with pancreaticoduodenectomy:multicenter data collection as a project study of pancreaticsurgery by the Japanese Society of Hepato-Biliary-PancreaticSurgery) especially considering the low percentage of patients affect by an adenocarcinoma in our series. We modify the text and table to clarify this point. Regarding the PPH, the overall rate is coherent with previous published report and less than the one reported by the ISGPS  (Saudi J Gastroenterol. 2020 Aug 18;26(6):337-343. doi: 10.4103/sjg.SJG_145_20. Post pancreaticoduodenectomy hemorrhage: A retrospective analysis of incidence, risk factors and outcomeangenbecks Arch Surg. 2011 Aug;396(6):783-91. doi: 10.1007/s00423-011-0811-x. Epub 2011 May 25.Critical appraisal of the International Study Group of Pancreatic Surgery (ISGPS) consensus definition of postoperative hemorrhage after pncreatoduodenectomy). Also, we considered all the PPH, the rate of severe hemorrages is 8.3% in our cohort. We further stress the correlation between postoperative complications and MDR infections present in the paper.

 * What was the length of stay in both groups, and overall? 

Thank you for your request. The median length of stay of the whole population is 11 days with an interquartile range of 8 days (mean 14.13 standard deviation 8.9). Dividing for stented and non-stented patient’s median, interquartile rage and mean and standard deviation were respectively 11 [7] vs 12 [9] days and 14.13 (8.9) vs 14.33 (9.6) p= 0.773, we added it in the text (table1). An analysis of the length of stay in the different MDR + groups is reported at line 159 and 181.

 Discussion:

* "However, the rate of MDR bacteria after pancreaticoduodenectomy, their presence at the time of the procedure in the bile, and their finding on the fifth post- operative day in the surgical drain, indicating the surgical site, clearly identify the pre- operative and early postoperative period as the timeframe in which MDR bacteria were selected and began to colonize patients undergoing pancreaticoduodenectomy."  - Since there was no assessment of the native bile pre-stenting and post-stenting, it is difficult to conclude exactly when colonization began.

Thank you for your comment, we changed the sentence accordingly. As reported in literature, by definition, and the biliary tree is sterile and so is the native bile. Pre- operative biliary stenting creates a communication between the biliary system and the gastrointestinal tract, with potential modification of the bile bacterial flora, probably leading also to the development/selection of MDR bacteria.  (Gut. 1971 Jun;12(6):487-92. doi: 10.1136/gut.12.6.487. Bacteria and disease of the biliary tract A J Scott; BMC Gastroenterol. 2016 Mar 31:16:43. doi: 10.1186/s12876-016-0460-1. Role of preoperative biliary stents, bile contamination and antibiotic prophylaxis in surgical site infections after pancreaticoduodenectomy Francesca Gavazzi).

Reviewer 2 Report

This study would describe the impact of the early multidrug resistant contamination in patient undergoing pancreatoduodenectomy.

Major issues: unfortunately, there is no huge novelty in the paper, comparing with what already is published in the literature. Therefore, the authors should dress the paper with a new light.

Second major issue: abstract is quite long and many results can be cut short and better explained in the paper. Also, it could be important to find the date or timing of placement of the biliary drainage and understand if the long or short period of drainage (before the operations) might influence the contamination. 

The quality of English is relatively good. 

Author Response

Dear Editor/Revierw,

we want to thank you for the support and opportunity to further improve our work. As requested, we provide in this document our point-by-point answer to the reviewers’ comments.

This study would describe the impact of the early multidrug resistant contamination in patient undergoing pancreatoduodenectomy.

Major issues: unfortunately, there is no huge novelty in the paper, comparing with what already is published in the literature. Therefore, the authors should dress the paper with a new light.

We thank the reviewer for this costructive comment. We believe that the novelty of our work resides in the early, even intraoperative, evaluation of MDR contamination in patients that underwent PD. This could open a new filed for the interventions aimed to mitigate the negative impact that these contaminations had on early postoperative outcome, focusing not only on the postoperative period but also on the pre and intraoperative. We also assess the impact of MDR contamination in our work as had been done in previous studies to confirm the trend also in our population and stress the relevance of this contamination. We clarify this aspect in the discussion (line 276).

Second major issue: abstract is quite long and many results can be cut short and better explained in the paper. Also, it could be important to find the date or timing of placement of the biliary drainage and understand if the long or short period of drainage (before the operations) might influence the contamination.

Thank you for your advice. Considering the time frame between the pre-operative biliary stenting and  the operation, we didn’t find any correlation with the MDR contamination of the bile. MDR + bile patients had a median stenting period of 66 [interquartile range 45.5] vs 46.5 [98.5] days p= 0.475. Also, the percentage of patients that underwent neoadjuvant chemotherapy among MDR + Bilicolture is 22.7 % vs 18.1% of patients that didn’t undergo neoadjuvant chemotherapy p = 0.817. We add this concept to the text and modified it accordingly. We are planning to better understand this finding evaluating the rate of cholangitis and needing/rate of antibiotic therapy in the preoperative period (line 356).

Round 2

Reviewer 2 Report

Thanks for reviewing but, considered the data and what has Benn already published, the study can be improved in terms of aim, analysis and comments. 

Author Response

Dear reviewer,

we thank you for you constructive opinion and the time you have spent in reviewing/improuving our work. Unfortunately the editor  required us to upload the revised manuscript in 72 hours maxium. In light of this, we would be more than happy to made further edits upon more specific requests in order to accomodate your suggestions the best way possibile.

Best regards.