Key Publication

Reviewer: Martin Daniel Rosenthal, MD Assistant Professor of Trauma, Acute Care Surgery, and Critical Care; Chair of UF Nutrition Committee, University of Florida, Gainesville, FL

References: Martinez et al. Effect of Preoperative Administration of Oral Arginine and Glutamine in Patients with Enterocutaneous Fistula Submitted to Definitive Surgery: a Prospective Randomized Trial. Journal of Gastrointestinal Surgery. 2020 Feb;24(2):426-434. doi: 10.1007/s11605-018-04099-4. Epub 2019 Feb 1. PMID: 30710211

What is the study question (in PICO format)? In Patients with enterocutaneous fistulas, does Intervention with immunonutrition, as Compared to standard treatment, improve the clinical Outcomes of fistula closure, inflammatory biomarkers, and reduce infectious complications?

Why is this study important? Treatment of enterocutaneous fistulas (ECF) and entero-atmospheric fistulas (EAF) remains complex as often times a ‘one size fits all’ approach does not apply to management strategies. Outside of the acute phase of fistula management where most experts agree on the classic principles of resuscitation, correction of electrolyte imbalances, source control of sepsis, skin protection, and metabolic support there is a diverse approach to ongoing treatment algorithms. Imbedded in the foundation of fistula management is nutritional support, but not all agree upon the how, what, why, and when of nutritional support. Countless discussion of parenteral nutrition versus enteral nutrition support have been had, but there is little to no recent randomized control trials guiding evidentiary based nutritional support. Since the late Stanely Dudrick gifted us parenteral nutrition (when we first saw a dramatic improvement in mortality) the literature has remained stagnant aside from large case series, most of them single center. This article  presented here, published in 2020, is noted for efforts in pushing ECF management forward, as well as, possibly birthing an ERAS model for those ECFs undergoing surgery.

Summary: From Jan 2011-2013 the authors performed a randomized single blinded study using a closed envelop randomization scheme to allocate 40 patients with ECF to 2 equal cohorts 1 week before undergoing surgical reversal.  The control cohort continued one with current nutritional support that they were previously on and the experimental arm received conventional treatment and additional 4.5g arginine coupled with 10g glutamine orally for seven days prior to surgery. The researchers determined that the patients needed at least 1 meter of functional small bowel prior to the fistula for absorption. Type and timing of the surgery was determined by the attending surgeon. Primary endpoint was ECF recurrence and secondary endpoints pertained to inflammatory biomarkers (IL-6, CRP) or infectious complications (pneumonia, UTI, central line infection, bacteremia).  

Of the 40 patients entered, 27 were male, median age was 54, 26 came were referred from outside hospitals, the origin was postoperative complications in all.  Interestingly, in this study colostomy takedown had the highest risk associated with fistula creation. Fistula was located in the jejunum (n=20), ileum (n=16), and colon (n=4). 30 patients had enteroatmospheric fistulas, which in 22 was a result of open abdomens from damage control operations. 25 patient were determined to be malnourished, while all patients could tolerate enteral nutrition to some varying degree as 29 patients still required some parenteral support for a median time of 139 days.  Surgery consisted of fistula resection and anastomosis (30 hand sewn and 10 stapled). Complete abdominal closure was feasible in 14 patients.  None of the demographic characteristics of the cohorts were statistically different, but trends towards significance was seen as the control group had more patients with ASA >3 and patients with multiple fistulas.  ECF recurrence occurred in 2 patients in the experimental group and nine patients admits the control (p <0.04).  Median time to recurrence was 6 days.  IL-6 and CRP had statistically significant lower serum concentration preoperative and post-operative days 1 (for IL-6), but not days 3 and 7 post-operative. Infectious morbidity was higher in the control group and not coincidentally all these patients had fistula recurrence.  

Commentary: The unfortunate aspect of fistula management is that there is scant contemporary level one evidence that drives clinical practice. This is impart secondary to ECFs relative low prevalence in society, but remains difficult generate high powered studies quickly. A well-executed prospective randomized control trial would take years at a single center, but perhaps a multicenter consortium of protocoloized fistula management could answer best practice questions for ECF management. Several significant questions that remain are not based on surgical technique but rather how to manage these patients in the time period between fistula emersion and operative reversal. Best practice for nutritional support remains highly debated even among ECF experts.    

In this manuscript Martinez and colleagues attempt to add amino acids as an oral supplement to enhance outcomes. Though not powered for specific clinical outcomes this trial has a flavor of Enhanced Recovery After Surgery (ERAS). Immunonutrition 1 week prior to surgery coupled with a 50g glucose load the morning of has been demonstrated to improve infectious and non-infectious complication, length of stay, insulin resistance, and even anastomotic leak rate.[1-6]  Most immune-nutrient formulas include arginine and omega 3 fatty acids, while some include leucine, glutamine, and nucleotides.[7-11]  The authors should be commended in their attempt to drive literature forward in determining best practice for ECF management, but I do have a few critiques. 

  1. No power analysis: it is essential to demonstrate a power analysis to strengthen clinical outcomes. Though there are little studies to derive such a power analysis on there exist well designed trials discussing clinical outcomes such as anastomotic leak in using immunonutrition preoperatively. Which at the end of the day what is a fistula reversal if not a GI anastomosis once the fistula is excised.
  2. Cohort mismatch: though none of the demographics were statistically different (p <0.05) there was a trend to the control group having multiple fistulas and were sicker based off ASA classification >3. The lack of statistical significance is likely related to low number of patients included in the trial.  Increased number of fistulas in the control group appears to naturally drive an increased number of anastomoses in the surgical outcomes that was required to reverse the fistulas.  Any time there is more anastomoses there is an inherent greater chance for leak.  More anastomoses coupled with half (n=10) the control group having an ASA >3 could skew primary outcomes of fistula recurrence before any intervention is even studied such as immunonutrition.
  3. Amino acid profile: in determining clinical relevance of an oral supplement I have always wondered if absorption factors into a potential confounder.  The authors rightfully enrolled only fistula patients with >1meter of bowel prior to the GI opening.  This should be plenty of length to absorb amino acids, but something as simple as an amino acid profile (inexpensive, quick test) could demonstrate that arginine and glutamine oral supplementation actually makes serum concentration improve.  Additionally, citrulline levels on the amino acid profile could serve as a rudimentary surrogate for GI absorption capacity. Lastly, most immunonutrition supplements for preoperative ERAS protocols also contain omega 3 fatty acids, which has proven to be beneficial in reducing inflammation. More as a question I am curious why fish oil wasn’t added to enhance recovery? 
  4. Criteria for parenteral: the authors make mention that “all patients were able to tolerate enteral diet while 29 patients (73%) received parenteral support. For a median of 139 days”. It would be interesting to see the percentage of support for these patients.  Was the patient on total parenteral nutrition despite being able to tolerate enteral diet? Was parenteral support used as supplemental parenteral nutrition? Was it added if there were signs of malabsorption and how did they determine that? To say that all patients tolerated enteral nutrition to me means they would be eating not ¾ of the study be on parenteral support. I do recognize supplemental parenteral support for those who cannot achieve caloric needs either through GI intolerance or malabsorption, but this was poorly discussed in the article.  
  5. Infectious morbidity: the authors begin their discussion with “the results of this study show that the preoperative use of oral arginine and glutamine in patients with ECF have positive effects in regard to fistula recurrence, postoperative inflammatory markers, and infectious complications”.  Additionally, the authors correctly mention that all infectious complications were in patients that had recurrence of their fistulas.  To make conclusions that the immunonutrients, arginine and glutamine, decreased post-operative infectious morbidity because all the infections occurred in the control group would be erroneous. I would submit that the infectious morbidity, and increased length of stay, in the control group was driven solely by the fistula recurrence requiring prolonged hospitalization and interventions. Again this could be in part reflected for the increased risk of anastomotic leak in the control group who had more fistulas and were sicker (ASA>3) at baseline.

Again the authors should be commended for studying a very difficult patient population, as the authors point out to us in their conclusions the tenets of fistula management remains the same for decades: “source control, appropriate management of sepsis, and nutritional support are key steps in the preoperative phase, and they need to be fulfilled before planning the operation itself”. I would add to their statement that correcting electrolyte imbalances and skin care are the complete foundation of fistula management, but there remains a nebulous middle ground that needs to be further explored.  After the acute phase (>2months) when the fistula is likely not going to spontaneously close there is a vast black hole of literature providing guidance for nutritional support. Most experts would agree that the non-caloric benefits of oral/enteral nutrition (even if it is oral food for comfort or trophic feeds) far outweigh parenteral nutrition[12, 13], but still fistula patients remain on parenteral support throughout their course awaiting surgical intervention without transitioning to or challenging the GI tract. The authors mentioned that all patients could tolerate diets and had bowel length >1m prior to the fistula, but the rationale for parenteral support was lacking (perhaps the patients couldn’t tolerate >60% of their goal calories). Nevertheless, at every turn in managing ECF we ought to be trying to get more and more enteral stimulation all the while checking for GI intolerance, malabsorption, dehydration, and other potential complications of intra-luminal nutrition. Nutritional support for fistulas should be dynamic and every attempt to de-escalate parenteral support should be made throughout the ‘middle ground’ management between fistula appearance and reversal.  Prior to reversal using studies like Martinez et al and Wang et al, there is no reason ECF surgical intervention shouldn’t follow most ERAS protocols using a week of immunonutrition to improve outcomes and shorten lengths of stay.  


  1. Marimuthu, K., et al., A meta-analysis of the effect of combinations of immune modulating nutrients on outcome in patients undergoing major open gastrointestinal surgery. Ann Surg, 2012. 255(6): p. 1060-8.
  2. Osland, E., et al., Effect of timing of pharmaconutrition (immunonutrition) administration on outcomes of elective surgery for gastrointestinal malignancies: a systematic review and meta-analysis. JPEN J Parenter Enteral Nutr, 2014. 38(1): p. 53-69.
  3. Mazaki, T., Y. Ishii, and I. Murai, Immunoenhancing enteral and parenteral nutrition for gastrointestinal surgery: a multiple-treatments meta-analysis. Ann Surg, 2015. 261(4): p. 662-9.
  4. Amer, M.A., et al., Network meta-analysis of the effect of preoperative carbohydrate loading on recovery after elective surgery. Br J Surg, 2017. 104(3): p. 187-197.
  5. Bilku, D.K., et al., Role of preoperative carbohydrate loading: a systematic review. Ann R Coll Surg Engl, 2014. 96(1): p. 15-22.
  6. Smith, M.D., et al., Preoperative carbohydrate treatment for enhancing recovery after elective surgery. Cochrane Database Syst Rev, 2014(8): p. CD009161.
  7. Braga, M. and L. Gianotti, Preoperative immunonutrition: cost-benefit analysis. JPEN J Parenter Enteral Nutr, 2005. 29(1 Suppl): p. S57-61.
  8. Braga, M., et al., Preoperative oral arginine and n-3 fatty acid supplementation improves the immunometabolic host response and outcome after colorectal resection for cancer. Surgery, 2002. 132(5): p. 805-14.
  9. Buzquurz, F., et al., Impact of oral preoperative and perioperative immunonutrition on postoperative infection and mortality in patients undergoing cancer surgery: systematic review and meta-analysis with trial sequential analysis. BJS Open, 2020.
  10. Cerantola, Y., et al., Immunonutrition in gastrointestinal surgery. Br J Surg, 2011. 98(1): p. 37-48.
  11. Gade, J., et al., The Effect of Preoperative Oral Immunonutrition on Complications and Length of Hospital Stay After Elective Surgery for Pancreatic Cancer--A Randomized Controlled Trial. Nutr Cancer, 2016. 68(2): p. 225-33.
  12. Kudsk, K.A., Effect of route and type of nutrition on intestine-derived inflammatory responses. Am J Surg, 2003. 185(1): p. 16-21.
  13. Kudsk, K.A., et al., Enteral feeding of a chemically defined diet preserves pulmonary immunity but not intestinal immunity: the role of lymphotoxin beta receptor. JPEN J Parenter Enteral Nutr, 2007. 31(6): p. 477-81.

Read more key publication reviews

Physician Community homepage