Nutritional support in patients undergoing oesophagectomy is of paramount importance in this usually malnourished patient group, but encountering significant clinical practice variation between units. Our aim was therefore to assess the strength of evidence behind nutritional support routes post-oesophagectomy. The Cochrane Library and Controlled Trials Registry, MEDLINE (Ovid) (1966–April 2009), PubMed, EMBASE (1966–April 2009), CINAHL, Web of knowledge and SCOPUS databases, were electronically searched for the highest level of evidence, with English language as a limit. Reference follow-up was also used. Studies were critically reviewed based on The NHS Public Health Resource Unit Critical Appraisal Skills Programme Tools. Five randomised control trials (RCTs) and one case-control trial, with 344 patients, were included in the review. There was a significant variation in the routes assessed (including intravenous fluid therapy, peripheral and central line nutrition, feeding jejunostomy, nasojejunal and nasoduodenal tubes) and the methodological quality of each study, with small patient numbers. No route was found to be superior over another in the RCTs. In the case-control trial, the combination of enteral parenteral nutrition led to shorter hospital stay compared with parenteral feeding alone. Nasojejunal and nasoduodenal tubes are associated with a significant rate of dislodgement. There is absence of strong direct evidence supporting a single feeding access route in oesophagectomy patients. Clinical decisions should be made based on available evidence from other types of gastrointestinal surgery, currently favouring enteral nutrition. If enteral feeding is chosen, feeding jejunostomy may be superior to nasojejunal or duodenal tubes.
The nutritional status of patients undergoing major surgery has been shown to impact significantly on postoperative recovery. Malnutrition weakens the immune response, impedes wound healing and promotes muscle wasting, leading to longer periods of hospitalisation and an increased risk of serious postoperative complications (Powell-Tuck, 2000; Bengmark et al., 2001; Saluja et al., 2002; Kirk and Heys, 2003; Ward, 2003).
The problem is further compounded in patients undergoing surgical resection for oesophageal cancer (Saito et al., 1990). In most cases, patients at presentation have had a prolonged period of reduced food intake as a result of the obstructive nature of the disease. Furthermore, due to the urgency of surgery in suitable patients and the effects of neo-adjuvant chemotherapy, this nutritional deficiency cannot be compensated for preoperatively. The operative magnitude of oesophagectomy and the physiological effect of reconstructing the upper gastrointestinal tract also mean that patients require nutritional support for weeks and often months following surgery (Sica et al., 2005; Ryan et al., 2006). These factors place this group of patients at a significant disadvantage and highlight the importance of optimising postoperative nutrition.
Several studies have suggested that following gastrointestinal surgery, enteral nutrition is superior to parenteral nutrition (PN), resulting in a reduced length of hospitalisation and serious postoperative complication rates (McCarter et al., 1996; Velez et al., 1997; Aiko et al., 2001; Gabor et al., 2005; Mazaki and Ebisawa, 2008). However, these benefits are not necessarily reproduced following oesophagectomy. The concern is that early enteral nutrition may be related to an increased incidence of aspiration pneumonia and may further complicate anastomotic leaks, resulting in mediastinitis, which carries a high mortality rate (Lassen and Revhaug, 2006; Athanassiadi, 2009). Although many oesophago-gastric surgeons attempt to bypass this problem by placing a surgical jejunostomy at operation, there is by no means consensus as to the optimum route of nutritional support in this patient group (Gerndt and Orringer, 1994; Wakefield et al., 1995; McCarter et al., 1996; Sica et al., 2005; Ryan et al., 2006). The aim of this study was therefore to examine the current body of evidence behind nutritional support routes in oesophagectomy patients by means of systematic review and meta-analysis.
Materials and methods
Data sources and search criteria
The Cochrane Library and Controlled Trials Registry, MEDLINE (Ovid) (1966–April 2009), PubMed, EMBASE (1966–April 2009), CINAHL, Web of knowledge and SCOPUS were searched. Medical subject headings from the National Library of Medicine used included ‘o/esophagectomy’, ‘enteral nutrition’, ‘parenteral nutrition’ and ‘jejunostomy’. Other terms used were ‘upper gastrointestinal surgery’ and ‘nutrition’. English language was the only limitation. References from full texts articles retrieved were interrogated to identify further studies. Two reviewers independently performed the above search.
This review initially included randomised control trials (RCTs) only in order analyse evidence with the highest level of methodology. Case-control studies were later introduced due to the small numbers of RCTs. Studies were eligible for inclusion if following oesophagectomy, access route(s) for nutrition were used, regardless of the type. Studies using patients undergoing oesophagectomy but also including other gastro-intestinal operations were excluded.
Primary outcomes being measured were procedure-related complication rates at 30 days and changes in nutritional status as defined biochemically. Secondary outcomes included length of hospital stay and overall complication rates at 30 days.
Two reviewers independently assessed all identified studies from abstracts against the inclusion criteria by use of a standard checklist and resolved differences by discussion. Similarly, studies deemed suitable were reviewed using their full texts by two independent reviewers. Pre-defined data extraction forms were used to extract required information from the selected studies. A methodological quality assessment was then performed by the two reviewers using the Critical Appraisal Skills Tools for RCTs and case-control trials, and risk of bias was assessed using established methods (Cochrane Reviewer's Handbook, 2009).
Our intention was to perform a meta-analysis on the collected data from RCTs, however, the small sample sizes and the methodological heterogeneity precluded this from taking place.
In total, 354 studies were identified. Five RCTs and one case-control study fulfilled the eligibility criteria (Figure 1). Tables 1 and 2 provide a summary and describe the outcomes of the studies, respectively. Table 3 describes the independent variables measured and tested for each study.
Intravenous fluids vs enteral or parenteral nutrition
Two RCTs compare postoperative nutrition (nasojejunal tube (Page et al., 2002) or via peripheral parenteral nutrition (PPN) (Cooper et al., 2006)) with intravenous (IV) fluid therapy only, followed by introduction of oral fluids on the fourth day and soft diet on the sixth or seventh postoperative day. Outcomes are summarised in Table 3. The results suggest that PPN or nasojejunal nutrition hold no significant benefit over IV fluid therapy alone.
In the study by Cooper et al. (2006), IV therapy was associated with an 18% 30 day mortality rate compared with 0% in the PPN group. However, at 1 year, this rose to 56.25 and 54.5%, respectively. This 30-day mortality was attributed by the authors to the higher median age of the group but with no statistical evidence supplied. PPN was also associated with a 12.5% incidence of phlebitis, and a 19% conversion rate to alternate route, with asymptomatic transient liver biochemistry abnormalities occurring in all patients. In the study by Page et al. (2002), the nasojejunal route was associated with a 37% incidence of accidental tube dislodgement around day 5.
Enteral and parenteral nutrition vs parenteral nutrition alone
Aiko et al. (2001) compare combined jejunostomy and central line feeding with central line feeding alone. Oral diet was introduced on the 5th postoperative day. Serum bilirubin and C-reactive protein levels were higher and serum lymphocyte count was lower in the combined group. There were no other statistically significant differences in complications or nutritional status at 7 days between the two groups. No jejunostomy related complications were reported. Gabor et al. (2005) similarly compared a combined regime (jejunostomy with central line feeding) with central line feeding alone in a case-control study. Intensive care and overall hospital stay was shorter in the combined group.
Enteral vs parenteral nutrition
Shiraishi et al. (2005) compared jejunostomy feeding with central line feeding with introduction of oral diet on the eighth postoperative day. Similarly there were no significant differences in outcomes between the two groups and no jejunostomy related complications were reported.
Enteral nutrition: nasoduodenal tube vs jejunostomy
Most recently, Han-Geurts et al. (2007) examined differences in outcomes between feeding jejunostomy and nasoduodenal feeding, with oral fluid intake introduced almost immediately postoperatively and soft diet at day 10. There were no significant differences in outcomes or route related complications between the two methods. However, jejunostomies were associated with insertion site infections (16%) and a 1% re-laparotomy rate. Postoperative patient nutritional status was not assessed.
All six studies compare different routes of nutritional support, although none of these use identical methods. At first glance, no single nutritional support route or combination of routes provide a significant benefit over another in terms of mortality, general complications, route related complications or nutritional status. However, there are several factors, which must be taken into account when considering these results.
Table 4 summarises the risk of bias assessment performed for each study in the review. This revealed that that the evidence is limited, of average quality due to methodological problems, and very heterogeneous among studies. No clear definitions are provided on the health status of selected patients within studies and between studies, allowing for a possible wide and unrepresentative patient sample variation of selected ‘medically fit’ patients for surgery. A variety of selection and performance bias affect both individual study rigour and external validity and the small number of patients included in the majority of studies allows for a significant risk of a type II statistical error (Neyman and Pearson, 1967; Cohen, 1988) where no difference between routes (null hypothesis) is accepted, although infact this is not the case and larger patient numbers are required. Performing a power calculation for their primary outcomes would help identify such potential errors but only one of the studies performed such a calculation (Cooper et al., 2006), suggesting that 40–60 patients and 100 patients per arm will be required to show a significant difference in mortality and 30 day complications, respectively.
Enteral vs parenteral nutrition
Shiraishi et al. (2005) and Aiko et al. (2001), attempt to examine the benefit of enteral feeding in patients undergoing oesophagectomy. As above, both studies carry a risk of bias with no reported randomisation technique on a small number of patients. Aiko et al. (2001) combine parenteral and enteral feeding and compare them to parenteral feeding alone. As it may be possible to attribute central line complications to the enteral feeding route in the combined group, it is difficult to assess the impact that enteral feeding has got. The authors report that parental feeding leads to raised bilirubin and C-reactive protein levels and a reduced lymphocyte count suggesting that, biochemically at least, enteral feeding holds some benefit. However, the clinical significance of this is uncertain, and again, in light of the potential bias discussed above, one must interpret such results with caution. Shiraishi et al. (2005) report no statistically significant difference in any outcomes between enteral and parenteral feeding groups, however, there were only 15 patients in this study.
Gabor et al. (2005) reported that patients in their combined group had a shorter stay on the intensive care unit and in the hospital overall, however, what is striking in this study is the extremely high anastomotic leak rate (48% for combined routes and 52% parenteral nutrition alone). The accepted anastomotic leak rate in oesophagectomy is less than 20,% (Briel et al. 2004; Ercan et al., 2005; Law et al., 2005) raising concerns regarding the external validity of this study.
Enteral feeding routes
When considering nutritional support routes, one must take into account route specific complications. Page et al. (2002) reported a high nasojejunal tube dislodgement rate of 37% on the fifth postoperative day. However, the follow-up period was very short (7 days), and may have missed subsequent complications. This is mirrored by Han-Geurts et al. (2007) who reported a nasoduodenal dislodgement rate of 23–35%. Neither study identifies the risks associated with re-insertion of nasojejunal or duodenal tubes.
Feeding jejunostomies on the other hand appear to dislodge less frequently, however, carry a greater risk of more severe complications such as intra-abdominal leak and a small but real risk of death. There are several large case series examining the use of feeding jejunostomy in oesophagectomy, some of which are summarised in Tables 5 and 6. These agree with the findings of Han-Geurts et al. (2007) in relation to complication rates. Although the quality of available evidence is limited, the suggestion is that if enteral feeding is to be used post-oesophagectomy, feeding jejunostomy should be preferred over nasojejunal or nasoduodenal tubes due to its lower dislodgement rates.
In view of the quality of available evidence, one must resort to examining postoperative nutrition routes in patients undergoing other gastrointestinal operations. This is suboptimal, but nonetheless necessary until more definitive results from higher quality studies become available.
Several studies including some meta-analyses examine this topic (Baigrie et al., 1996; Velez et al., 1997; Bozzetti et al., 2001; Mazaki and Ebisawa, 2008). In short, the evidence favours enteral over parenteral nutrition. Mazaki and Ebisawa's (2008) meta-analysed 29 trials with over 2500 patients and concluded that enteral nutrition results in fewer complications. However, the debate relating to the benefits of enteral nutrition over parenteral feeding in relation to bacterial gut translocation and gastro-intestinal function remains open.
In conclusion, evidence supporting an optimal route for nutritional support in post-oesophagectomy patients is weak. Given this, results from studies examining other types of gastro-intestinal surgery must be drawn upon to make clinical decisions, which is sub-optimal. If enteral feeding routes are to be used, jejunostomy may be superior to nasoduodenal or nasojejunal tubes.
Cochrane Reviewer's Handbook. (2009). Cochrane handbook for systematic reviews of interventions. Version 5.0.2 (updated September 2009). In: Higgins J and Green S (eds). The Cochrane Collaboration.
Aiko S, Yoshizumi Y, Sugiura Y, Matsuyama T, Naito Y, Matsizaki J et al. (2001). Beneficial effects of immediate enteral nutrition after oesophageal cancer surgery. Surg Today 31, 971–978.
Athanassiadi KA (2009). Infections of the mediastinum. Thorac Surg Clin 19, 37–45.
Baigrie RJ, Devitt PG, Watkin DS (1996). Enteral versus parenteral nutrition after oesophagogastric surgery: a prospective randomised comparison. Au N Z J Surg 66, 668–670.
Bengmark S, Anderson R, Mangiante G (2001). Uniterrupted peri-operative enteral nutrition. Gut 20, 11–19.
Bozzetti F, Braga M, Gianotti L, Gavazzi C, Mariani L (2001). Postoperative enteral versus parenteral nutrition in malnourished patients with gastrointestinal cancer: a randomised multicentre trial. Lancet 358, 1487–1492.
Briel JW, Tamhankar AP, Hagen JA, Demeester SR, Johansson J, Choustoulakis E et al. (2004). Prevalence and risk factors for ischaemia, leak, and stricture of esophageal anastomosis: gastric pull-up versus colon interposition. J Am Coll Surg 198, 541–542.
Cohen J (1988). Statistical Power Analysis for the Behavioural Sciences. 2nd edn. Lawrence Eribaum Associates Inc: New Jersey.
Cooper SC, Hulley CM, Grimley CE, Howden J, Mccluskey K, Norton RN et al. (2006). Perioperative peripheral parenteral nutrition for patients undergoing esophagectomy for cancer: a pilot study for safety, surgical and nutritional outcomes. Int Surg 91, 358–364.
Ercan S, Rice TW, Murthy SC, Rybicki LA, Blackstone EH (2005). Does esophagogastric anastomotic technique influence the outcome of patients with esophageal cancer? J Thora Cardio Surg 129, 623–631.
Gabor S, Renner H, Matzi V, Ratzenhofer B, Lindenmann J, Sankin O et al. (2005). Early enteral feeding compared with parenteral nutrition after oesophageal or oesophagogastric resection and reconstruction. Br J Nutr 93, 509–513.
Gerndt SJ, Orringer MB (1994). Tube jejunostomy as an adjunct to esophagectomy. Surgery 115, 164–169.
Han-Geurts IJM, Hop WC, Verhoef C, Tran KTC, Tilanus HW (2007). Randomised clinical trial comparing feeding jejunostomy with nasoduodenal tube placement in patients undergoing oesophagectomy. BJS 94, 31–35.
Han-Geurts IJM, Verhoef C, Tilanus HW (2004). Relaparotomy following complications of feeding jejunostomy in esophageal surgery. Digestive surgery 21, 192–196.
Kirk HJ, Heys SD (2003). Immunonutrition. BJS 90, 1459–1460.
Lassen K, Revhaug A (2006). Early oral nutrition after major upper gastrointestinal surgery: why not? Curr Opin Clin Nutr Metab Care 9, 613–617.
Law S, Swen DT, Wong KH, Kwok KF, Wong J (2005). A single-layer, continuous, hand-sawn method for oesophageal anastomosis: prospective evaluation in 218 patients. Arch Surg 140, 33–39.
Mazaki T, Ebisawa K (2008). Enteral versus parenteral nutrition after gastrointestinal surgery: a systematic review and meta-analysis of randomized controlled trials in the English literature. J Gastrointest Surg 12, 739–755.
Mccarter MD, Gomez ME, Daly JM (1996). Early post-operative enteral feeding following major upper gastrointestinal surgery. J Gastrointest Surg 1, 278–285.
Mercer CD, Mungara A (1996). Enteral feeding in oesophageal surgery. Nutrition 12, 200–201.
Neyman J, Pearson ES (1967). The testing of statistical hypotheses in relation to probabilities a priori. Cambridge University Press: Cambridge.
Page RD, Oo AY, Russel GN, Pennefather SH (2002). Intravenous hydration versus naso-jejunal enteral fedding after esophagectomy: a randomised study. Eur J Cardiothorac Surg 22, 666–672.
Powell-Tuck J (2000). Peri-operative nutritional support: does it reduce hospital complications or shorten convalescence? Gut 46, 749–750.
Ryan AM, Rowley SP, Healy LA, Flood PM, Ravi N, Reynolds JV (2006). Post-oesophagectomy early enteral nutrition via a needle catheter jejunostomy: 8-year experience at a specialist unit. Clin Nutr 25, 386–393.
Saito T, Zeze K, Kuwahara A, Miyahara M, Kobayashi M (1990). Correlations between preoperative malnutirtion and septic complications of oesophageal cancer surgery. Nutrition 6, 303–308.
Saluja SS, Kaur N, Shrivastava UK (2002). Enteral nutrition in surgical patients. Surg Today 32, 672–678.
Shiraishi T, Kawahara K, Yamamoto S, Maekawa T, Shirakusa T (2005). Postoperaative nutritional management after esophagectomy: is TPN the standard of nutritional care? Int Surg 90, 30–35.
Sica GS, Sujendaran V, Wheeler J, Soin B, Maynard N (2005). Needle catheter jejunostomy at esophagectomy for cancer. J Surg Onco 91, 276–279.
Velez JP, Lince LF, Restrepo JI (1997). Early enteral nutrition in gastrointestinal surgery: a pilot study. Nutrition 13, 442–445.
Wakefield SE, Mansell NJ, Baigrie RJ, Dowling BL (1995). Use of a feeding jejunostomy after oesophagogastric surgery. Br J Surg 82, 811–813.
Ward N (2003). Nutrition support to patients undergoing gastrointestinal surgery. Nutr J 1, 18.
The authors declare no conflict of interest.
Part of the study was conducted as an assignment for an MSc degree in Advanced Surgical Practise undertaken by G Markides at Cardiff University, Wales, UK.
About this article
Cite this article
Markides, G., Al-Khaffaf, B. & Vickers, J. Nutritional access routes following oesophagectomy—a systematic review. Eur J Clin Nutr 65, 565–573 (2011). https://doi.org/10.1038/ejcn.2010.292
- feeding routes
- critical review
Bedeutung der perioperativen Ernährungstherapie beim Risiko- und Komplikationsmanagement von Patienten mit Ösophaguskarzinom
Der Chirurg (2016)
Randomized Clinical Trial: Nasoenteric Tube or Jejunostomy as a Route for Nutrition After Major Upper Gastrointestinal Operations
World Journal of Surgery (2014)
Pre-therapy Laparoscopic Feeding Jejunostomy is Safe and Effective in Patients Undergoing Minimally Invasive Esophagectomy for Cancer
Journal of Gastrointestinal Surgery (2013)
Current Gastroenterology Reports (2012)