Optimizing nutritional access in oncology: A five-year experience with feeding jejunostomy

INTRODUCTION

Oncological patients undergoing various gastrointestinal (GI) as well as non-gastrointestinal procedures often require supplemental feeding due to an inability to consume adequate nutrition. There are two ways, namely, enteral and parenteral nutrition(PN). PN carries risks of electrolyte disturbances, hepatic dysfunction, central venous catheter-related infections and atrophy of gut mucosa. Whenever possible, enteral nutrition (EN) is preferred and can be provided through either oral or tube feeding.^[1] Tube feeding can be through a nasogastric or nasojejunal tube, gastrostomy or a jejunostomy tube.

Nasoenteric tubes are inconvenient for the patient cannot be used for long-term feeding. Percutaneous Jejunal tubes are the easiest to insert and maintain and are associated with the fewest complications. Most surgeons prefer to place a feeding jejunostomy (FJ) at the time of primary surgery whenever deemed necessary during major oncological cases. It can also be used as an alimentation method in head and neck cancers when the patient is unable to gain oral nutrition due to bulky tumoursor fistulas.

Jejunostomy tubes are not free of complications and various issues such as blockages, perijejunostomy leaks, intestinal rotation, re-exploration, and even mortality have been described in the literature.^[1] In the current era of enhanced recovery protocols, minimal tubes are propagated, and when required, supplementation is provided with PN. However, oncological cases differ from routine surgical procedures where the patients are marred by significant metabolic stress from disease and treatment and require long-term support. We aimed to evaluate the role, benefits and hazards of feeding jejunostomy in oncologic practice.

MATERIAL AND METHODS

Surgical technique

The procedure was carried out under appropriate anesthesia, either at the end of a major surgery or as a standalone procedure. A 14F Ryle’s tube is inserted through the left lateral abdominal wall at the level of umbilicus and tunnelled obliquely through the abdominal wall musculature. A seromuscular purse string suture using 3-0 polyglactin (vicryl) or polydioxanone (PDS) is placed on the antimesenteric wall of jejunum at least 15 cm distal to the lowest jejunal anastomosis or to the duodenojejunal flexure. The tube is inserted into the jejunal lumen and the purse string tightened around the opening. Saline is flushed through the tube to facilitate bowel dilatation and tunnel is created by placing serial seromuscular sutures ease of tube advancement [Figure 1]. A 4 cm long Witzel’s on the jejunal wall with 3-0 vicryl or PDS [Figure 2]. The jejunal wall is secured to the parietal abdominal wall at the distal most portion of the tunnel with 3-point fixation using the samesuture [Figure 3]. The tube is fixed to the skin of abdominal wall with 2-0 nylon and patency, and free flow of saline is confirmed [Figure 4]

Close

Figure 1:

The 14F Ryle’s tube secured into jejunum with a purse string suture

Export to PPT

Close

Figure 2:

The 4 cm long Witzel’s tunnel created with seromuscular sutures

Export to PPT

Close

Figure 3:

Tube fixed to the abdominal wall with 3-point fixation

Export to PPT

Close

Figure 4:

Tube secured to anterior abdominal wall

Export to PPT

RESULTS

A total of 613 patients who underwent an FJ placement during the 5-year study period were identified. Complete records were available for 534 patients, and these were included in the final analysis. The mean age of the patients was 52.8 ± 12.3 years. There were 282 males (52.8%) and 252 females (47.2%) in the population.

The most common indication for FJ was carcinoma oesophagus (39.3%), followed by periampullary carcinoma and gastric carcinoma (18%). 91.4% cases of oesophageal cancer had FJ insertion as part of surgery and rest for palliative feeding. 81.5 and 82.3 % cases of periampullary and gastric cancer had primary FJ during the surgery, respectively. Table 1 demonstrates the indications and procedures for a jejunostomy was placed.Thirty-six cases of advanced head and neck malignancies received palliative FJ. 12 of these were for laryngeal carcinomas, 17 for oropharyngeal and 7 for large oral cavity tumours with fungating masses [Table 1].

The most frequently performed procedure was VATS esophagectomy (32.5%) followed by Whipple’s procedure (25.8). We routinely perform VATS esophagectomy at our centre for carcinoma lower and mid oesophagus with cervical anastomosis. We also routinely place an FJ for all gastrectomy cases at our centre for early feeding. In 24 cases of gastrojejunostomy and 12 cases of triple bypass for advanced periampullary and pancreatic carcinomas, ancillary FJ was placed for early feeding. Fifty patients received a standalone tube procedure for various indications [Table 2].

FJ-related complications were seen in 97 (18.1%) patients. The most common complication was feed intolerance seen in 45 (8.4%) patients, which resulted in abdominal distension and pain. This was managed conservatively and resolved with the return of bowel motility. Other minor complications, such as tube dislodgement,were seen in 2.8% cases and managed with outpatient reinsertion of the tube. 2.2% had peritubal leak resulting in skin excoriation, which was managed with re-suturing of skin incision and barrier protective dressing. Tube occlusion was managed with flushing of the tube with saline and soda bicarbonate solution. Major complications were seen in 8 cases over the study period. 4 had a leak resulting in EC fistula, 2 cases of obstruction, 1 small bowel volvulus and 1 case of peritonitis [Table 3]. All The 4 cases of EC fistula resolved with tube removal and conservative management over varying periods of 6-12 weeks. 4 patients required re-exploration for obstruction, volvulus and peritonitis. There was no FJ-related mortality.

The highest overall complication rate was observed in patients undergoing Whipple’s procedure (29.7%), followed by total gastrectomy (27.8%) and D2 gastrectomy (16.7%). In contrast, procedures like hepatectomy, triple bypass, and distal gastrectomy were associated with no or minimal complications. Minor complications were most frequent in Whipple’s procedures (n=39), primarily comprising feed intolerance, tube dislodgement, or occlusion. Major complications, including peritubal leaks with EC fistula, intestinal obstruction, and peritonitis, were noted in 8 patients, most commonly after D2 gastrectomy and feeding jejunostomy tube placement [Table 4].

Mean pre op albumin was 3.2(±0.34), and mean postoperative albumin at day 15 was 2.9 (±0.28). FJ was retained for a mean duration of 17 (±0.7.3) days among the patients where the tube was placed with curative intent surgeries. In 104 patients where there were postoperative complications such as anastomotic leaks, sepsis or other surgical complications, the mean tube duration was 35 (±0.8.5) days. Among the cases undergoing tube placement for palliative care, the tube was left in situ indefinitely till deemed necessary by the treating team.

DISCUSSION

Malnutrition is a common problem among oncological patients and leads to worsening of performance status, poor quality of life and reduced survival. The cause of malnutrition can be multifactorial, including loss of appetite due to cancer-related cytokines, mechanical inability to feed, nausea and vomiting due to GI blockage from tumour or as a result of chemotherapy. The gut is a metabolically, immunologically and bacteriologically decisive organ during surgical stress response, and enteral nutrition has been proven to be superior beyond all doubt. Early initiation of enteral nutrition helps with faster recovery, shorter hospital stay and better healing.

However, routine placement of feeding tubes during major abdominal surgery has been called into question by many surgeons, as these are not without complications and associated morbidity. Prospective trials have shown that the reduction in postoperative infectious complications with tube use is offset by the increase in the number of tube-related and intestinal complications.^[2] The onus for deciding the benefit v/s risk for placement of FJ lies on the operating surgeon and must be guided by various factors, including the patient’s nutritional risk, need for prolonged supplementary alimentation, safety of tube placement, options for endoscopic tube insertion if required in the future and the option of parenteral nutritional support.

In our study, the most common indication for placement of FJ was in esophagectomy, followed by Whipple’s procedure. 80% of tubes were used as a part of a major oncological surgery, and in the remaining 20% cases, in cases of palliative treatment for feeding access, indicating this tube is useful for both temporary and permanent feeding solutions. Esophagectomy has a delicate anastomosis and requires a longer duration of healing as compared to other bowel anastomoses. Patients undergoing esophagectomy with gastro-oesophageal anastomosis usually have poor nutrition due to dysphagia preoperatively and are required to be nil per oral (NPO) for about a week postoperatively till the anastomosis heals. An FJ is an excellent method to provide nutrition during this phase to the patients.

Jejunostomy is useful in patients post-upper GI surgery who undergo anastomotic leaks and require prolonged nutritional support. Some authors have reported using needle catheter placement, percutaneous endoscopic Jejunostomy, but most prefer a Witzel’s tube at the time of surgery.^[1] A single centre reported the use of FJ routinely after upper GI surgery and most commonly in esophagectomy.^[3] In another prospective study evaluating long-term and post-discharge use of FJ, pancreatobiliary surgeries were the most common procedure where FJ was done, followed by esophagectomy and gastric resections.^[2]

Various techniques of constructing an FJ have been described, and each has its own merits and pitfalls. The commonly used methods include Witzel’s technique; use of a 16F red rubber catheter secured to the jejunum with a purse string and laparoscopic or robotic approach, where the bowel is hitched to the abdominal wall and accessed percutaneously with a needle and guidewire, subsequently being replaced by a tube and fixed with the Stamm technique.^[2,4] We use a modified Witzel’s technique using a 14F Ryle’s tube secured with a purse string to the jejunum and create a tunnel, but also tack the jejunal loop to the abdominal wall [Figures 1-4].

In a study from India, all patients undergoing esophagectomy received an FJ by Witzel’s technique using a 24F Foley catheter.^[1] An National Health Services, United Kingdom (NHS) study reported 70% cases received the FJ through the open technique, mostly using Witzel’s technique and 20% through minimally invasive methods, with the rest being percutaneously inserted.^[5] Various modifications of the technique exist, and each author performs it as per their convenience and practice. Minarich et al presented a simplified method where a red rubber tube is used and secured to the jejunum with a purse string suture without tunnelling or peritoneal fixation without any complications.^[2] Another study from India reported laparoscopic jejunostomy using a T tube with minimal complications and just 1 patient requiring resurgery.^[6] Varshney et.al reported lower complications with laparoscopic Witzel’s technique compared to open ( 0 vs 19%).^[7] Smaller-sizedtubes, such as 8F catheter and needle jejunostomies, have been used, but being quite thin and malleable, these tubes have a higher chance of knotting over themselves, leading to blockage, which requires tube replacement.^[3]

Complications associated with jejunostomy can be mechanical, infectious, GI or metabolic. Mechanical complications include coiling, kinking, malposition, retrograde flow or occlusion of the tube. Infectious problems can be due to peritubal leak and stoma site infection, whereas GI complications are small bowel obstruction, intussusception, distal narrowing, and extraluminal tracking. Metabolic complications can range from hyperglycemia, hypokalemia, hypophosphatemia, and hypocalcemia to water and sodium disturbances. The complication rate of 1.5 -2 % for a Witzel’s jejunostomy is reported in the literature.^[1] However, these only include mechanical complications. A more realistic overall complication rate in another study was 11% including occlusion in 4%, peritubal leak in 5% and accidental tube removal in 2%.^[2] Another NHS study reported a complication rate of 14% with 2% cases requiring laparotomy for obstruction and bezoar formation.^[8]

Catheter blockade is the most common complication, with an incidence of 7-11%.^[1,2]

Most of these can be managed by saline flushing, but some patients may require a resurgery in case of jejunal intussusception or other causes of tube block [Figure 5]. There are also documented cases of tube blockade by ascaris.^[1] We routinely use over-the-counter antacid fruit salts containing sodium carbonate and bicarbonate for tube flushing, which helps in cleaning the tube of food particles due to their effervescent action. Rarely, when a Foley’s catheter is used for a tube and the balloon is kept inflated for a prolonged duration, it may result in bowel intussusception due to invagination over the lead point, which may require resurgery.^[9-11]

Close

Figure 5:

Kinked Fj tube leading to block (black arrow)

Export to PPT

Peritubal leak may be present in 2-3% which can be managed with the use of balloon catheter for tamponade, purse string sutures, and emollient pastes. Skin excoriation may result in upto 5% patients due to peritubal leak of jejunal content, which is usually self-limiting once the wound heals and the stoma contracts. Tube displacement is also a rare complication which requires a simple reinsertion of the tube, as the tract is usually formed within a few days of surgery. The rate of tube dislodgement and leak in our study was around 2% each managed with tube reinsertion and suturing, respectively.

We report a complication rate of 18% overall, but the point to note is that feed intolerance was the problem in up to 8.5% of these cases, which is not routinely reported in studies. Temporary abdominal distention may be seen in some patients due to ileus and intolerance to feeds. This is almost always managed conservatively with temporary suspension of feeding followed by gradual reintroduction of low-volume boluses. Studies have also compared bolus feeding and continuous feeding, but failed to show any difference in the tolerance or nutritional benefit. However, bolus feeding is logistically simpler and aids patient mobility and is the preferred choice.^[12] Upto 10% patients may experience diarrhoea when started on jejunostomy feeds, which can be ameliorated with simple measures like the addition of probiotics, antimotility agents, and modifying the feed content, i.e., omission of milk.^[1] In our study, we noted that most complications were seen in patients undergoing

Whipple’s pancreaticoduodenectomy, which also aligns with the high rates of ileus and delayed gastric emptying known to happen after this procedure. Sluggish biliopancreatic flow leads to feed intolerance more commonly in these cases.

The major complication rate was only 1.5% with 4 cases of EC fistula, 2 cases of obstruction [Figure 6], 1 small bowel volvulus and 1 case of peritonitis [Table 3]. All 4 cases of EC fistula resolved with tube removal and conservative management over varying periods of 6-12 weeks. 4 patients required re-exploration for obstruction, volvulus and peritonitis with no FJ-related mortality. A rare but fatal complication is non-occlusive mesenteric ischemia (NOMI), which is seen shortly after initiation of feeding and likely due to high osmolarity formula, preoperative malnutrition and bowel dilation.^[4]

Close

Figure 6:

Coiling of the FJ tube (black arrow) resulting in small bowel obstruction

Export to PPT

Use of FJ helps the patients reach their nutritional needs early, and upto 82% patients accomplish adequate intake. We found that mean albumin levels reached preoperative levels within 15 days postoperatively. Tube feeding is short-lived in most cases, and reportedly, 77% of patients hardly require the tube beyond 30 days.^[4,5] A prospective study previously concluded that age, preoperative albumin levels, preoperative chemoradiation use or type of surgery were not related to prolonged FJ use but were mainly required in cases of stage IV malignancy and prolonged hospitalization.^[2] We found that the mean tube duration was 17 days, but prolonged to 35 days when patients had postoperative complications such as anastomotic leaks and sepsis.Removal of the FJ tube can be done on an outpatient basis, with the site healing over a few days with simple occlusive dressings. In patients with long-term tube use, epithelialization of the mucosa may result in delayed closure and may even require suture closure.

Irrespective of the controversial data, FJ placement in major abdominal procedures as a prophylactic measure is an acceptable practice with few complications and ease of procedure. The option of PN remains, but not without its complications, and the technique for laparoscopic FJ insertion in an emergency postoperative setting is not widely available.

CONCLUSION

Feeding jejunostomy is still a viable option for patients undergoing major oncological procedures to supplement nutrition postoperatively, which brings succour to the patient, aiding speedy recovery. It has minimal complications, which can be managed conservatively and can be safely removed within a month post-procedure. It is also a well-tolerated option for prolonged sustenance in cases of advanced malignancy.