Effect of Supplemental Parenteral Nutrition Versus Enteral Nutrition Alone on Clinical Outcomes in Critically Ill Adult Patients: A Systematic Review and Meta-Analysis of Randomized Controlled Trials
Round 1
Reviewer 1 Report
Dear Authors,
The paper needs English Editing. In addition:
- search string is not replicable, thus it should be better specified.
- figure 1 should be edited by PRISMA flow diagram. The number of studies at each stage is not visible.
- table 1: the number column is difficult to read in this way. A full list of abbreviations used in the table should be added.
- Paragraph "2.2.5. Duration of mechanical ventilation” and “2.3.1. Energy intake” have poor phrasing. Please rewrite the paragraph to better clarify the concept.
- In the analysis in which the random effects model was used, avoid putting in the figure also the results obtained with the fixed effects model, and vice-versa.
- “2.3.1. Energy intake" and "2.3.2. Protein intake" it would be useful to know how much calories and protein were administered with EN and PN respectively. Maybe the patients receiving EN+PN had only a better nutritional counseling ?
- The duration of the each intervention could be a key factor in influencing clinical outcomes i.e. PN for longer period has higher infectious risk. Did some study reported it? Also it is important to know if patients had already a CVC or if it had to be inserted in order to receive PN.
- is there in any paper given the percentage of patients who met the caloric goal? This may be useful to address in the discussion
Author Response
Reviewer 1
- The paper needs English Editing
Response: The manuscript was sent to English editing and was corrected
- In addition: - search string is not replicable; thus it should be better specified.
Response: The string has been added in lines (81-83) and referred to in Appendix 1
- figure 1 should be edited by PRISMA flow diagram. The number of studies at each stage is not visible.
Response: figure 1 is now edited by PRISMA flow diagram
- table 1: the number column is difficult to read in this way. A full list of abbreviations used in the table should be added.
Response: This has now been added to the footnote to Table 1
- Paragraph "2.2.5. Duration of mechanical ventilation” and “2.3.1. Energy intake” have poor phrasing. Please rewrite the paragraph to better clarify the concept.
Response: This has now been rephrased in lines 178 to 182 and lines 185 to 188.
- In the analysis in which the random effects model was used, avoid putting in the figure also the results obtained with the fixed effects model, and vice-versa.
Response: Figure 2 legend has been edited.
- “2.3.1. Energy intake" and "2.3.2. Protein intake" it would be useful to know how much calories and protein were administered with EN and PN respectively. Maybe the patients receiving EN+PN had only a better nutritional counseling?
Response: We added a column to Table 1 to report the energy and protein intakes and their sources i.e. EN or SPN.
- The duration of the each intervention could be a key factor in influencing clinical outcomes i.e. PN for longer period has higher infectious risk. Did some study reported it? Also it is important to know if patients had already a CVC or if it had to be inserted in order to receive PN.
Response: We added a column to Table 1 for reporting the duration of the intervention. The duration of the intervention was equal in both (SPE+EN and EN) groups. For trials starting with EN then adding SPN at day 4, the intervention duration was counted since the start of the SPN. The CVC was inserted for the indication of SPN in all RCTs.
- Is there in any paper given the percentage of patients who met the caloric goal? This may be useful to address in the discussion
Response: The caloric goal is 25-30 Kcal/kg/d according to the published guidelines. In the new column in Table 1, we added the percentage of achieving the caloric and protein goals between SPN+EN and EN groups. Regarding the percentage of patients who met the caloric goal, in 3 out of 5 RCTs (26, 27, 28), patients who reached 60% of the caloric goals (or 20 Kcal/kg/day) on the 4th day were assigned as SPN+EN (numbers are mentioned in Table 1), while in Fan et al (24) and Wischmeyer et al (25) the participants were randomized to one of the two groups from the start (numbers are mentioned in Table 1).
Author Response File: 
Author Response.docx
Reviewer 2 Report
Thank you for the opportunity to review this article that is a systematic review and meta-analysis of the effect of supplemental parental nutrition versus enteral nutrition alone on clinical outcomes in critically ill patients.
I have a major issue with the conduct of the search. The authors outline that they have followed the Cochrane Handbook but it doesn't appear that this is the case for search development. Traditionally for a systematic review +/- meta-analysis, a full version of the search is provided for at least one database. This is currently detailed on page 5 but not adequately.
Specifically (per the Cochrane methodology that the authors have reported they have followed)- how was the search developed? How was it tested? The detail provided in the search does not show which terms were searched as text words and/or subject headings, how they were linked together etc. There are many more possible terms to include in a search related to critical illness eg. intensive care unit, critical illness and other terms like mechanical ventilation. All of these terms should be tested in search development and then demonstrated how they have been used together and the full search for one data base (line by line) provided as an appendix to the article.
Further evidence of the inadequacy of the search is that small number of articles retrieved (163). A high quality review with a well conducted search should retrieve between 5000-10 000 or more articles before duplicate removal. One of the limitations mentioned by the authors is the small number of studies available; this may also be due to missed literature due to the search, introducing bias to your findings. I suggest that the authors review some search strategies for well conducted reviews in high quality journals.
I believe the search has to be redone.
Other comments:
- It is PRISMA not Prisma.
- OVID is the platform- what database did you use on this platform ie MEDLINE, EMBASE
- The date of registration on PROSPERO is after the search date and many of the study processes have been conducted. Usually this is done prior to anything being started (as recommended in PRISMA and Cochrane which the authors have stated they have followed).
- The authors should justify why they chose the Jadad scale for quality assessment over the Cochrane risk of bias.
Author Response
Reviewer 2:
- I have a major issue with the conduct of the search. The authors outline that they have followed the Cochrane Handbook but it doesn't appear that this is the case for search development. Traditionally for a systematic review +/- meta-analysis, a full version of the search is provided for at least one database. This is currently detailed on page 5 but not adequately.
Response: This has now been added to lines (81 to 83) and referred to in Appendix 1.
- Specifically (per the Cochrane methodology that the authors have reported they have followed)- how was the search developed? How was it tested? The detail provided in the search does not show which terms were searched as text words and/or subject headings, how they were linked together etc. There are many more possible terms to include in a search related to critical illness eg. intensive care unit, critical illness and other terms like mechanical ventilation. All of these terms should be tested in search development and then demonstrated how they have been used together and the full search for one data base (line by line) provided as an appendix to the article.
Response: This has now been added in detail to Appendix 1
- Further evidence of the inadequacy of the search is that small number of articles retrieved (163). A high quality review with a well conducted search should retrieve between 5000-10 000 or more articles before duplicate removal. One of the limitations mentioned by the authors is the small number of studies available; this may also be due to missed literature due to the search, introducing bias to your findings. I suggest that the authors review some search strategies for well conducted reviews in high quality journals.I believe the search has to be redone.
Response: We agree with the reviewer, however, the large number of articles were irrelevant to our research question because of the limited number of RCTs investigating SPN and its relatively recent clinical application of SPN.
- It is PRISMA not Prisma.
Response: This has been corrected in line (58)
- OVID is the platform- what database did you use on this platform ie MEDLINE, EMBASE
Response: This has been corrected in line (75)
- The date of registration on PROSPERO is after the search date and many of the study processes have been conducted. Usually this is done prior to anything being started (as recommended in PRISMA and Cochrane which the authors have stated they have followed).
Response: We agree with the reviewer and we confirm following the PRISMA and Cochrane recommendation, that was a typo and is now corrected. We searched until the preceding month to the date of PROSPERO registration i.e. January 2019. The search process was delayed until the start of the new academic year, so the correct date was September 2019 and January 2020
- The authors should justify why they chose the Jadad scale for quality assessment over the Cochrane risk of bias.
Response: We chose the Jadad scale as it is more frequently used and more familiar to us.
Author Response File: Author Response.docx
Reviewer 3 Report
The systematic review and meta-analysis written by Al-Sharif et al. discuss the effect of supplemental parenteral nutrition versus enteral nutrition alone on clinical outcomes in critically ill adult patients.
The significant limitations of this research is small number of studies (n=5), however, the authors have been reasonably explained that at the end of “discussion” section. Overall, the manuscript is well written.
Author Response
Reviewer 3
The systematic review and meta-analysis written by Al-Sharif et al. discuss the effect of supplemental parenteral nutrition versus enteral nutrition alone on clinical outcomes in critically ill adult patients.
The significant limitations of this research is small number of studies (n=5), however, the authors have been reasonably explained that at the end of “discussion” section. Overall, the manuscript is well written.
Response: Thanks for your valuable comment, we agree with the reviewer regarding the limited number of RCTs comparing SPN+EN vs EN from the start or following the guidelines i.e. after a while of EN.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
I have no further comments.
Author Response
- I have no further comments.
Response: Thanks a lot.
Author Response File: Author Response.docx
