Chapter 98: Endoscopic Placement of Feeding Tubes

Nutritional support is an important component of care for the critically ill. There are numerous modalities to provide specialized nutrition support including oral dietary therapy, enteral nutrition, and parenteral nutrition. For patients with a functioning gastrointestinal (GI) tract, enteral nutrition is preferred to parenteral nutrition. It is safer, more physiologic, and economical. In the intensive care unit (ICU) setting, enteral nutrition is associated with a decreased likelihood of developing infections when compared to parenteral nutrition.¹ In patients with acute pancreatitis, use of enteral nutrition is also associated with a reduction in hospital length of stay and a trend toward reduced organ failure when compared to parenteral nutrition.² Patients who have a functioning GI tract but are unable to safely ingest oral intake are fed via enteral access. In patients in whom long-term enteral access is required, endoscopically placed enteral access is recommended. Percutaneous endoscopically placed gastrostomy (PEG) and jejunostomy (PEJ) tubes are utilized for long-term enteral nutrition.

A majority of patients have enteral feeding tubes (EFTs) inserted through the mouth (orogastric) or nares (nasogastric) at the bedside that terminate in the gastric antrum. EFTs allow noninvasive access to the intestinal tract in patients who are unable to eat or drink. Examples of patients requiring EFT include sedated or unconscious patients such as those who are mechanically ventilated or have head injury. Severe oropharyngeal dysfunction that occurs in stroke patients or patients unwilling to have oral intake such as severe depression or anorexia are also candidates for EFT. Use of EFT is short term, typically less than 30 days. If enteral access is required beyond 30 days, a PEG or PEJ tube should be inserted. Contraindications for the placement of EFT include bowel ischemia, intestinal obstruction, and ileus.

Endoscopy is not required for the insertion of most EFTs. Indications for endoscopy include the presence of an esophageal obstruction or to facilitate postpyloric EFT placement. For esophageal obstruction the gastroscope is passed to the stricture, the feeding tube guided through the stricture under direct vision and advanced to the desired depth. Alternatives include placement of a guidewire through the stricture using the gastroscope biopsy channel, exchanging the gastroscope and guidewire to leave the wire in place and then placing an EFT over the guidewire. Also, the stricture can be dilated endoscopically prior to bedside placement.

Postpyloric feeding should be considered in patients with delayed gastric emptying, those with a history of regurgitation^3,4 as well as patients with acute severe pancreatitis. Jejunal feeds minimize pancreatic exocrine secretions by bypassing the upper GI tract and is a core aspect in treating acute severe pancreatitis. Comparisons between gastric and jejunal feedings have shown mixed results and a large prospective study is currently underway.^5,6,7 Patients with acute severe pancreatitis should receive jejunal feeds until more evidence is available. There are 4 endoscopic methods of postpyloric feeding tube placement:

1. Drag and pull. A suture is placed at the distal tip of the feeding tube and then passed via nares to the stomach followed by the gastroscope. Biopsy forceps through the gastroscope grasp the suture and advance together into the small bowel. The feeding tube is then left in position by grasp, advancement, release, and regrasp of the feeding tube while the gastroscope is retracted. Friction of the feeding tube against the gastroscope may lead to retraction of the EFT. Helpful maneuvers include decompression of the stomach to reduce looping within the stomach, liberal lubrication applied to the EFT and gastroscope, and using a feeding tube with an internal guidewire to provide stiffening (one 0.052-in or two 0.035-in guidewires placed through the EFT tube without exiting the tip of the tube). Alternatively, rather than using biopsy forceps an endoscopic clip may be used to grasp the attached suture and advance the EFT as desired. At the desired depth of insertion the hemoclip can then be used to affix the EFT suture to the small bowel mucosa. This approach has been demonstrated to increase the initial success rate of EFT placement and decrease retrograde tube migration with subsequent need for repeat endoscopy.⁸

2. Over the guidewire. Advance a gastroscope into the small bowel and exchange with a guidewire down the biopsy channel leaving the guidewire in situ. The guidewire exits orally and is changed to achieve nasal exit. A nasopharyngeal transfer tube is placed via the nares into the pharynx. The distal tip of the tube is grabbed using forceps to allow the distal end to exit orally. The guidewire is fed through this catheter to exit at the nose. Next the catheter is retracted nasally to leave guidewire in situ (Figure 98–1). The EFT is fed over the guidewire to the small bowel.

3. Through scope. A small-diameter feeding tube can be fed through the biopsy channel of a therapeutic scope. A 240-cm 8- or 10-Fr gauge feeding tube is placed through the endoscope biopsy channel. The scope is exchanged with the EFT once the distal end of the feeding tube is in a suitable position. The extended length of the feeding tube allows it to remain in position as the scope is exchanged and removed over the feeding tube. The feeding tube is then cut to the desired length and oronasal transfer performed as presented earlier.

4. Ultraslim gastroscope. An ultrathin gastroscope is passed through the nares down the upper GI tract into the small bowel. A soft-tipped guidewire can be inserted in the channel and then exchanged with the gastroscope, leaving the guidewire in place. The EFT is then advanced over the guidewire and guidewire is removed once a suitable depth on EFT insertion is achieved.

Endoscopically placed gastrostomy tubes were first described in 1980 as an alternative to open surgical gastrostomy placement.⁹ Since then numerous techniques have been developed and described. The Ponsky-Pull method is the most commonly practiced method illustrated here.

Indications for PEG

Suitable candidates for insertion of PEG tubes for nutritional supplementation include the following:

• Inability to perform motor mechanisms of oral intake

• Functioning GI tract

• Anticipated requirement of nutritional support for over 30 days

Common indications for the placement of PEG tubes for nutritional supplementation include patients with stroke, head and neck cancer, as well as brain injury or neurologic disorders such as amyotrophic lateral sclerosis. Conversely, feeding tubes in patients with dementia have not shown improved clinical outcomes and are generally ineffective in prolonging life, improving function, or reducing risk of pressure sores or infections.¹⁰

Contraindications to PEG Placement

Absolute contraindications include a nonfunctional GI tract, active GI bleeding, uncorrected coagulopathy, hemodynamic instability, peritonitis, and abdominal wall infection over insertion site. The presence of ascites is a relative contraindication to PEG placement and depends on the amount, location, and type of ascites.

Complications

PEG tube insertion is a well-tolerated procedure with low incidence of complications; rates up to 4% have been reported. Complications may relate to either the insertion of the PEG tube or complications thereafter. These include stomal site infection, abdominal wall ulceration, necrotizing fasciitis, peristomal leakage, bleeding, transient gastroparesis, bowel perforation, gastric outlet obstruction from the internal bumper, buried bumper syndrome, colocutaneous fistula, peritonitis, and liver puncture. Pneumoperitoneum may be identified on radiologic imaging post-PEG insertion. This is an expected finding and is not worrisome in the absence of clinical signs of peritonitis.

Anesthesia

Insertion of endoscopically placed PEG or PEJ tubes is done with a combination of local anesthesia and sedation. A short-acting benzodiazepine (ie, midazolam) in combination with an opioid such as fentanyl is a common practice. In some instances patients may receive deeper sedation via propofol, or in some cases general anesthesia, under the guidance of a nurse anesthetist or anesthesiologist.

Equipment Required

See Figure 98–2 for typical PEG kit components.

Personnel

PEG tube insertion generally requires 2 individuals to perform the procedure. One individual functions as the endoscopist and the other as the surgical assistant.

Preparation and Antibiotic Prophylaxis

The patient is kept NPO the night before the procedure. There is creation of a surgical wound during the creation of the gastrostomy and antibiotic prophylaxis is given to prevent abdominal wall infection. A single dose of a first-generation cephalosporin is given prior to PEG tube insertion to reduce the risk of infection. Patients in the ICU often receive significant antimicrobial coverage and as long as there is preexisting gram-positive organism coverage, additional antibiotics are not required. The patient is supine with the head turned to the side and suction available for significant oral secretions. A routine upper GI examination is performed prior to PEG insertion.

Landmarking

The epigastrium is the typical location for PEG placement but can vary. The final site is dependent on appropriate landmarking. The final location for the gastrostomy tube should be at least an inch away from the costal margins and clear of the xiphoid process.

The stomach is inflated to maximally appose the gastric wall with the abdominal wall. This facilitates the 2 maneuvers to identify a suitable PEG insertion site, the absence of any underlying vessel or interposing tissue such as bowel or liver edge. First is diaphanoscopy, adequate transillumination of the endoscopy light through the abdominal wall. The endoscopy suite lights are dimmed and transilluminated light from the endoscope should be clearly visible on the anterior abdominal wall (Figure 98–3). The second is 1:1 transmission of abdominal wall pressure. The endoscopic view is directed at the gastric wall corresponding to the point of external point pressure. When point pressure is applied to the abdominal wall, the endoscopic image shows 1:1 transmission of movement (Figure 98–4). Mark the site PEG placement with surgical pen marking.

Site Preparation

PEG placement requires sterile technique with skin disinfection, sterile gloves, and draping. The landmark is washed with povidone/iodine or chlorhexidine wash, 3 times in concentric swirls starting from the landmark site outwards. After the surgical drape is applied local anesthesia with a total of 3 to 5 mL of 1% lidocaine should be injected subcutaneously, followed by injection toward the gastric lumen. Direct the syringe perpendicular to the abdominal wall and apply local anesthesia with periodic suction to inspect for blood return and possible puncture of a blood vessel. Significant heme return should prompt a different site for PEG insertion. The syringe will puncture into the gastric lumen, seen endoscopically to confirm an acceptable PEG location. The depth and direction of the syringe serves as a guide for insertion of the trocar/catheter apparatus.

Inflate the stomach to facilitate the trocar penetration through the muscular gastric wall. The trocar/catheter apparatus is applied to the abdominal wall in a firm and steady fashion. At all times the endoscope visualizes the point of trocar insertion to monitor the site for location confirmation and applied force of the trocar toward the gastric lumen. This avoids trauma of the posterior gastric wall and underlying structures by overzealous force. Remove the trocar from the trocar/catheter apparatus after it has entered the gastric lumen. A guidewire is inserted through the catheter. Advance a snare through the channel of the gastroscope and grasp the guidewire. One end of the guidewire and gastroscope are pulled out through the mouth. The surgical assistant must ensures that at all times a portion of the guidewire remains external to the catheter on the abdominal wall. Once out of the mouth, the guidewire is released from the snare and affixed to the gastrostomy tube. The loop of the guidewire is inserted through the loop of the gastrostomy tube and then opened and passed over top of the gastrostomy bumper. The 2 pieces of equipment are then pulled, interlocking the loops of the guidewire and gastrostomy tube. The abdominal wall is prepared to allow penetration of the large caliber PEG tube through the abdominal wall. Cut an approximately 1-cm epidermal incision along the trocar puncture site. A deeper incision is not required and will only worsen postprocedural pain and possible bleeding. Apply a small amount of lubricant onto the gastrostomy tube. Significant amounts may result in pooling of lubricant in the hypopharynx increasing possible aspiration risk and airway compromise. The surgical assistant then pulls the guidewire with the attached gastrostomy tube away from the catheter/abdominal wall. The PEG tube slides down through the mouth and through the abdominal wall. The catheter remains in place during this motion to avoid damage to the surrounding tissue. While pulling on the guidewire resistance is met once the transition from the guidewire to the gastrostomy tube is encountered by the gastric wall. Brace the abdominal wall with taut pressure on the guidewire until the tip of the PEG tube passes through the abdominal wall. Gently continue pulling the PEG tube until resistance is again met, when the internal bumper rests up against the gastric wall (Figure 98–5). An alternative to tactile feedback to indicate appropriate PEG positioning is to insert the gastroscope into the esophagus with direct visualization of the gastrostomy tube while it is pulled into position. In either case the endoscope should visualize the internal gastrostomy site to ensure adequate hemostasis. Cut the guidewire from the PEG tube and slide the external bumper over the tube and down toward the abdominal wall to affix the tube in place. At an appropriate tautness, the external bumper should rest comfortably just above abdominal wall without evidence of significant tension on the skin surrounding the gastrostomy site. There should be no more than 1 cm of PEG tube between the external bumper and abdominal wall when the PEG tube is lifted taut. Decide an appropriate length for the gastrostomy tube, cut and discard the remainder, before placing the clamp and gastrostomy cap onto the PEG tube. The gastrostomy tube can be used for feeding and medication following insertion.

Patients with a PEG tube can have an extension placed through the PEG tube into the small bowel. For these cases a large-bore PEG (usually 24 or 28 Fr) is used and the jejunal extensions are 9 to 12 Fr. For PEG-J extensions there are 3 methods described.

1. Pull method. A jejunostomy tube (JT) is advanced through the preexisting PEG tube and grasped by biopsy forceps through the gastroscope channel. The gastroscope is advanced into the small bowel and the biopsy forceps advance the JT distally. After a suitable placement is achieved, as the gastroscope is retracted the biopsy forceps continue to advance the JT until the gastroscope reaches the gastric body and is removed.

2. Guidewire. A gastroscope is advanced from the mouth into the stomach and a guidewire is passed through the PEG tube into the stomach. The wire is then grasped by biopsy forceps and pulled as far as possible into the small bowel using the gastroscope. The gastroscope and forceps are removed and a JT is fed over the guidewire into position with subsequent guidewire removal.

3. Ultrathin gastroscope: An ultrathin gastroscope is advanced directly through the PEG. Via the biopsy channel a guidewire is placed. The gastroscope is removed via exchange with the guidewire to leave the guidewire in place. The jejunal extension is passed over the guidewire into the small bowel and the guidewire is subsequently removed.

The placement of direct percutaneous endoscopic jejunostomy (DPEJ) tubes has become more common in recent years. Indications for DPEJ insertion rather than PEG include (1) high risk for aspiration, (2) status postgastric resection, (3) status postesophagectomy, (4) gastric outlet obstruction, (5) obstructed or nonfunctioning gastrojejunostomy, and (6) gastric dysmotility. In comparison to PEG-J and nasojejunal (NJ) tubes, DPEJ has been found to decrease the risk of aspiration in patients.¹¹ The lack of effectiveness in preventing aspiration in NJ and PEG-J tubes may relate to their smaller caliber with a greater propensity for blockage and tube migration (DPEJ tubes are 18-20 Fr, larger than PEG-J tubes 9-12 Fr). Contraindications to DPEJ insertion are the same as for PEG tubes. Potential complications include bowel perforation, jejunal volvuli, bleeding, aspiration, chronic enterocutaneous fistulae, site infections, and persistent site pain.¹²

Procedure

Many of the maneuvers to insert a DPEJ are similar to PEG tube insertion. However, there are several notable aspects that differ between DPEJ and PEG placement. Employment of a pediatric colonoscope may be required to reach the jejunum; a gastroscope may be used if there has been a prior gastrectomy. For the choice of jejunostomy tube, balloon-type inflatable mechanisms should be avoided as they are prone to obstruct the bowel lumen. Bumper-style jejunostomy tubes are preferred.

Landmarking

In terms of location for DPEJ placement, the site may vary throughout the abdomen. It is not unusual for the DPEJ to be placed in the suprapubic or flank regions. In patients with prior surgery the DPEJ operators should be aware of the altered anatomy and be cognizant for potential limitations or preferences for the site of DPEJ placement (ie, afferent vs efferent jejunal limb). Successful site landmarking requires a greater degree of precision in DPEJ in comparison to PEG. The diaphanoscopy may be falsely reassuring due to the existence of a “periscope effect” from the tortuous course of the small intestine. Transillumination on the abdomen may be the result of the endoscope light source reflecting throughout different points of the bowel wall rather than representing the true position of the endoscope tip. For this reason finger indentation must be particularly precise, in combination with diaphanoscopy, to locate an appropriate DPEJ site. Due to the mobile nature of the small bowel it may take several passes through the jejunum until a suitable location is identified. Once diaphanoscopy and finger indentation prospect a site the negative needle aspiration test can be performed. Following sterile preparation and during local anesthesia, aspiration of the needle is intermittently applied. The operator should look to the syringe now not only for blood that would indicate an underlying blood vessel, but stool as well that would indicate needle insertion into the colon. If this occurs, the needle should be withdrawn and a different location for DPEJ insertion sought. All landmarking maneuvers must be done swiftly due to the mobile nature of the small bowel.

Following successful landmarking and injection of local anesthesia, once the needle has been identified endoscopically it remains held in place in the jejunum. This differs from PEG insertion where the needle is removed. The trocar is then applied beside the needle as close as possible to imitate its course through the abdomen. This reduces the likelihood of inadvertently penetrating abdominal vessels or structures. Once the trocar/catheter apparatus is successfully inserted into the small bowel the needle can be removed and the remainder of DPEJ insertion is similar to PEG insertion via the “pull” technique as presented earlier.

Nutrition support is a vital component of care for critically ill patients. There are multiple modalities at the disposal of the medical team to ensure adequate nutrient delivery to the patient. The choice of intervention is dictated by the clinical scenario and technical expertise available.