- Email: [email protected]
Risks of Endoscopic Enteral Access
Risks of Endoscopic Enteral Access Mark H. DeLegge, MD, FASGE, FACG, AGAF, and Amy J. Berry, RD, CNSD Enteral nutrition therapy can improve outcomes in many disease states such as pancreatitis and inflammatory bowel disease, and in critically ill patients. Obtaining enteral access is fundamental in order to provide enteral nutrition. The endoscopist plays an important role in placing enteral access systems. The endoscopist should be familiar with the techniques to perform percutaneous endoscopic gastrojejunostomy (PEG/J), direct percutaneous jejunostomy (DPEJ) and bedside nasoenteric tube placement as well as the management of complications that can result from their placement. Tech Gastrointest Endosc 10:36-44 © 2008 Elsevier Inc. All rights reserved. KEYWORDS eneral access, gastrostomy, jejunostomy, nasoenteric tube, enteral nutrition, aspiration pneumonia
E
nteral access is necessary to provide enteral feeding in patients who cannot or will not eat. Enteral access techniques and commercialized kits have been designed to meet the specific needs of patients, such as into what site (gastric or jejunal) and for how long tube feedings should be given (Table 1). Enteral access techniques have been developed for use at the patient’s bedside or for use by the radiologist, the surgeon, or the endoscopist in a non-bedside location (Table 2). This article will focus on the complications associated with endoscopic enteral access.
Bedside Nasoenteric Access Bedside nasoenteric tube (NET) blind placement is the most common method to achieve enteral access. The final position of this enteral access tube is either the stomach (for gastric feedings) or the jejunum (for small bowel feedings). Obtaining blind nasogastric (NG) access is easier than obtaining nasojejunal (NJ) access. There are specific indications for NJ access. The use of small bowel feedings to prevent tube feeding aspiration events is a complicated and contentious issue. Some studies have shown a decrease in aspiration episodes when patients were fed directly into the small bowel compared with when feedings were instilled in the stomach.1,2 A recent, prospective trial by Neumann and coworkers compared the use of gastric and small bowel feedings in the intensive care unit, and showed no differences in aspiration episodes; it took longer to initiate small bowel feedings, however, because of the difficulty in obtaining adequate tube position.3 This deDigestive Disease Center, Medical University of South Carolina, Charleston, SC. Address reprint requests to Mark H. DeLegge, Medical University of South Carolina, Digestive Disease Center, 96 Jonathan Lucas Street, 210 Clinical Science Building, Charleston, SC 29425. E-mail: deleggem@ musc.edu
36
1096-2883/08/$-see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.tgie.2007.08.009
bate certainly will continue, but for patients with gastroparesis, patients intolerant to gastric feedings, or patients who have had a witnessed tube feeding aspiration event with gastric feedings, small bowel access and feedings should be initiated. Failure to pass a NJ tube blindly at the bedside requires the use of fluoroscopic or endoscopic methods to guide passage; the preference of technique is center dependent. In centers with available C-arm fluoroscopy and modified fluoroscopy beds, fluoroscopic passage of NJ tubes can be done at the patient’s bedside with a success rate that approaches 100%.4 In institutions without such bedside fluoroscopic capabilities, the transport of patients (especially critically ill patients) can be time consuming, expensive, and hazardous5; in these instances, endoscopic passage of NJ tubes is preferred.
Complications of Nasoenteric Tubes Complications of NET placement have been reported in up to 12% of patients6 (Table 3), and include pneumothorax, sinusitis, tube dislodgement, tube clogging, and aspiration pneumonia. Early identification of complications, such as tube dislodgement or clogging, is important in preventing interruptions in tube feeding regimens.7,8
Aspiration pneumonia Aspiration pneumonia is a leading cause of morbidity and mortality in the critically ill patient population,9 and a NET may increase the risk for this complication. Fay and coworkers looked at aspiration pneumonia occurrence in patients with NETs compared with percutaneous gastrostomy tubes (PEGs). This study found that, within 14 days, patients with NETs developed aspiration pneumonia significantly more than patients with PEGs (24% versus 6%).10 Defense mech-
Risks of endoscopic enteral access
37
Table 1 Endoscopic Enteral Access Methods Type of Access Surgical or percutaneous gastrostomy Surgical or percutaneous gastrojejunostomy Surgical or percutaneous jejunostomy Nasal/oral gastric tube Nasal/oral gastrojejunal tube
Nasal/oral jejunal tube
anisms present in healthy adults are altered in the critically ill, making this patient population more susceptible to aspiration pneumonia.9 Heyland and coworkers compared gastric and postpyloric feedings in medical and surgical ICU patients, and found that gastric feedings were associated with significantly more episodes of gastroesophageal regurgitation with a trend toward increased microaspiration.11 The North American Summit on aspiration in critically ill patients also noted the high risk of aspiration with NG tubes.9 They did not recommend decreasing tube diameter size or converting an NG tube to a PEG solely for prevention of aspiration pneumonia, but they did recommend elevation of the head of bed 30° to 45°, changing infusion from bolus-type tube feedings to continuous feedings, optimizing oral health, and utilizing small bowel feedings whenever possible.9 The decision to use a jejunal tube (J-tube) for feeding also should warrant some very specific instructions regarding its care. The lumen of these tubes is often much smaller than that of a gastric tube, and therefore, they are more prone to clogging. Jejunal feeding tubes should never be checked for residual content, as their residuas are a poor indicator of residual content of the small bowel. In addition, checking residuals through these small-bore tubes increases their probability of clogging.12 These tubes should be flushed after every tube feeding and medication instillation. Only liquid medications, or completely dissolved medications, should be
Function Gastric Gastric Gastric Jejunal Jejunal Gastric Gastric Gastric Gastric Jejunal Jejunal
feeding decompression decompression Feeding feeding feeding decompression feeding decompression feeding feeding
Length of Need >1 month 1–6 months >1 month <1 month <1 month
<1 month
placed through a J-tube to reduce the chances of tube occlusion. Care should be taken to stop tube feedings during infusion of medications such as theophylline or potassium chloride, products that are known to coagulate tube feedings and obstruct the J-tube.
Pneumothorax Inadvertent tube placement into the trachea causing pneumothorax occurs most often when nasoenteric tubes are placed blindly at the bedside. Schoten and coworkers reviewed 5 cases in which pneumothorax occurred from such malplacement, noting the significant morbidity and mortality that can occur with tube misplacement and the importance of appropriately evaluating the position of the tube after placement.13 Three of the 5 patients required chest tube thoracostomy; 1 patient was eventually discharged, 1 required 2 chest tubes (this patient died 2 days later), and the third developed chronic empyema requiring chest wall reconstruction. A cuffed endotracheal tube did not protect patients from feeding tube misplacement. Another study echoed the importance of confirmation of correct feeding tube placement by abdominal films and adequate training of personnel involved in placement of these tubes14; this study found 14 inadvertent tube misplacements out of 1100 attempts over a period of 18 months, including pneumothorax, hydrothorax, empyema, mediastinitis, pneumonia, and esophageal perforation.
Table 2 Methods of Obtaining Enteral Access
Sinusitis
Bedside-placed Tubes Nasal/oral gastric tube Nasal/oral jejunal tube Nasal/oral gastro/jejunal tube Radiologically-placed Tubes Percutaneous gastrostomy tube Percutaneous jejunostomy tube Percutaneous gastro/jejunostomy tube Nasoenteric tube Surgically-placed Tubes Gastrostomy tube Jejunostomy tube Gastro/jejunostomy tube Endoscopically-placed Tubes Percutaneous endoscopic gastrostomy tube Percutaneous endoscopic jejunostomy tube Percutaneous endoscopic gastro/jejunostomy tube Endoscopic nasoenteric tube
In a study by George and colleagues, nosocomial sinusitis in a medical intensive care unit was reported in 12 cases per 1000 patient days; 7.7% of patients with NJ tubes had sinusitis, and 19.8 cases were documented per 1000 NET days.15 Table 3 Complications of Nasogastric and Nasojejunal Tubes Aspiration pneumonia Nasal mucosal ulceration Naso-pharyngeal bleeding Otitis media Pharyngitis Pneumothorax Sinusitis Tracheoesophageal fistula Tube obstruction Tube migration
38 Table 4 Complications of Percutaneous Endoscopic Enteral Access Aspiration pneumonia Bleeding Buried bumper syndrome Clogging of tube Colocutaneous fistula Gastric outlet obstruction Hepatic injury Herniation of the stomach through the PEG site Ileus Inadvertant PEG/PEG-J removal Necrotizing fasciitis PEG site breakdown PEG site wound infection Peri-stomal leakage Peritonitits Pneumoperitoneum Replacement balloon tube failure Small bowel obstruction Tube deterioration Ulceration of stomach or small bowel
Van Zanten and coworkers found sinusitis as the sole source of fever in 16.2% of patients and as a contributing factor in 13.8% of critically ill patients with fever of unknown origin.16 Another study looked at patients with NG tubes placed after surgery and remaining in place longer than 48 hours.17 This study found that 75% of patients developed rhinitis, with no clinical symptoms, but with radiologic evidence of maxillary sinusitis. The authors concluded that local symptoms of sinusitis may not be present despite radiologic evidence of inflammation. The first line of treatment of sinusitis is removal of the NET tube; antibiotics are used as necessary.16,17 NETs are intended for short-term use only, and some facilities have switched to oroenteric tube placement because of complications such as sinusitis.16,18
M.H. DeLegge and A.J. Berry
Endoscopic Percutaneous Access PEG was developed by Ponsky and Gauderer in the early 1980s.23 The procedure involves the placement of a percutaneous gastrostomy tube after endoscopic trans-illumination of the stomach for an appropriate PEG access position. The use of prophylactic antibiotics before the procedure is important in the prevention of postprocedure infections.24 PEG tubes may be placed by either the Sachs-Vine (push) or Ponsky (pull) techniques, the decision being just a matter of physician preference.25 Prospective evaluations of PEG placement have found this procedure to be associated with known procedure-related complications (Table 4).26 In those patients in whom small bowel feedings are desired, endoscopic access to the bowel may be obtained by two methods. The first method, PEG/J, places a jejunal feeding tube through an existing PEG and into the small bowel using an over-the-guidewire method. DeLegge and coworkers reported a 100% success rate without major complications using this technique for PEG/J placement with a procedure time of approximately 26 minutes.27 This PEG/J system allowed for concurrent gastric decompression and small bowel feeding. The average longevity of this tube system was approximately 120 days when patients who died were excluded from the analysis. J-tube failures often were secondary to migration. Improper positioning of the J-tube through the PEG into the small bowel will result in J-tube migration back into the stomach, and any loop of J-tube left in the stomach after placement will ultimately result in J-tube migration. The Jtube needs to be seen exiting the PEG and directly passing through the pylorus into the small bowel with no evidence of gastric J-tube looping (Fig. 1). The second method, DPEJ, directly places a J-tube into the small bowel using an endoscope (Fig. 2). This procedure requires the use of an enteroscope or a pediatric colonoscope to reach a puncture position beyond the ligament of Treitz. A
Tube Dislocation and Clogging Tube dislocation and clogging are two commonly reported complications that cause tube feedings to be stopped. One study from a medical intensive care unit reported tube dislodgement in 48.5% of their patients and tube occlusion to occur 12.5% of the time.19 Another study looked at both NG and NJ tubes in the critically ill patient population and found that inadvertent removal of tubes occurred in one of every three patients, which the authors related to problems with patient sedation and securing of the tube.20 Patrick and coworkers studied NG and nasojejunal tubes in 54 critically ill patients and found 6% of patients had a clogged tube that did not require tube removal, whereas 3% had a clogged feeding tube that did require removal of the tube.21 They further discussed that clogging most often occurred because of infusion of inappropriate medications through the feeding tube or lack of flushing of the tube. They concluded that educating staff in the care and flushing of the tube may help prevent this complication. Irrigants, such as warm water, sodium bicarbonate, pancrealipase mixture, or a declogging device, can be used once clogging occurs.22
Figure 1 Abdominal plain film showing a jejunal tube exiting the PEG with no looping in the stomach.
Risks of endoscopic enteral access
39 those who did not receive antibiotic prophylaxis.31,32 Most PEG wound infections will respond to a first generation cephalosporin or a quinolone. Methicillin-resistant Staphylococcus aureus (MRSA) has emerged as an important cause of PEG site infections in some centers.33 One study found that nasopharyngeal decontamination of patients infected with MRSA (in addition to standard prophylactic antibiotics) significantly reduced the incidence of wound infections.34
Figure 2 Direct percutaneous jejunostomy; needle catheter in the small intestine. (Color version of figure is available at www.techgiendoscopy.com.)
standard 20- or 24-F PEG tube may be placed and used as the J-tube. Both Mellert and Shike have reported good success with this procedure, with only minor complications, including local DPEJ site infection, but no reported cases of peritonitis or bowel infarction.28 Other authors also have reported good success with DPEJ. A retrospective study by Fan and coworkers compared physician re-interventions for J-tube complications, such as J-tube occlusion or J-tube migration, in a group of patients who received PEG/J compared with another group of patients who received DPEJ. The DPEJ patients had significantly fewer endoscopic re-interventions.29 This same result was confirmed by DeLegge and co-workers in a prospective series of 44 patients randomized to PEG/J versus DPEJ enteral access.30
Complications of Percutaneous Endoscopic Enteral Access Complications from endoscopic percutaneous feeding tube placement generally involve either the PEG tube or the DPEJ tube. These are the tube systems that provide direct access to the stomach or the small bowel. There are some complications associated with the jejunal extension tube in the PEG/J system, although these are fewer in number. Complications from PEG tube placement can be classified as minor or major. Minor Complications Minor complications include wound infection, peristomal leakage, pneumoperitoneum, ileus, bleeding, gastric or small bowel ulceration, tube occlusion, tube degradation, balloon tube failure, and gastric or small bowel obstruction. Wound Infection. Wound infection is more likely to occur when a PEG or DPEJ has been placed through a contaminated procedure field, a contaminated oropharyngeal cavity, or with poor technique, especially in debilitated patients and
Peristomal Leakage. Peristomal leakage usually occurs within the first few days after PEG or DPEJ placement. It is more likely to occur in malnourished patients and those with diabetes who may have poor tissue healing and are prone to wound breakdown. In addition, placement of the external bolster of the PEG tube too tightly against the external abdominal wall will lead to poor tissue blood flow, wound breakdown, and peristomal leakage. Treatment should include correction of patient comorbidities, such as malnutrition and elevated blood sugars, loosening of the external bolster, and local measures to address skin breakdown (such as use of powdered absorbing agents or a skin protectant such as zinc oxide paste). Placement of a larger size gastrostomy or jejunostomy tube through the same feeding tube tract will not solve the problem. Once the PEG or DPEJ tube tract has started to leak, placing larger tubes through the same tract will serve only to further distend and distort the tract and will not promote tissue growth or healing. As a treatment method for peristomal leakage, the PEG or DPEJ tube can be removed for 24 to 48 hours, permitting the tract to close slightly; a replacement feeding tube can then be placed through the same, partially closed tract.35 This technique works well for patients whose PEG or DPEJ tube tract started to leak a month or more after initial insertion. It does not work as well for patients with early tract leakage because these patients usually have poor wound healing from other comorbid disease processes. In some patients with a mature PEG or DPEJ tube tract with persistent peristomal leakage, the feeding tube will need to be permanently removed, allowing the tract to close completely. Another PEG or DPEJ tube can then be placed at a different location on the abdominal wall. It is my experience that the new PEG or DPEJ tube can be placed when there is at least 50% closure of the old feeding tube tract, at which point initiation of feedings will not have a significant impact on leakage or inhibition of tissue healing through the old feeding tube tract. Pneumoperitoneum. Pneumoperitoneum is common after PEG or DPEJ placement.36 It is thought to be caused by insufflation of air associated with the endoscopic procedure and needle puncture of the gastric or small bowel wall. In the absence of peritonitis, it has no consequence and should not preclude feedings. Pneumoperitoneum may cause confusion for clinicians, however, in patients with clinical features that raise concern about a ruptured viscus. In these settings, a contrast radiology study should be obtained to confirm the position of the PEG or DPEJ tube within the stomach or small bowel and to exclude a leak. Subcutaneous air also has been described after PEG placement, and from air being introduced between the cutaneous
40 and subcutaneous tissues.37 In the absence of other findings, it is inconsequential and should not preclude feeding.37 Gastroparesis/Ileus. Some patients develop nausea and vomiting after PEG or DPEJ placement, which may be from transient gastroparesis. Rarely, an ileus develops, especially in patients with significant pneumoperitoneum.38 After a gastric, duodenal, or jejunal perforation has been excluded, patients who develop an ileus should be treated with bowel rest and, if necessary, gastric decompression. These patients can be identified by the presence of postprocedure abdominal distention, vomiting, and absence of bowel sounds. Feedings should be held until the ileus resolves. Bleeding. Hemorrhage following PEG or DPEJ tube placement is rare, but there have been reports of significant bleeding following PEG, including an aortic perforation and a retroperitoneal hemorrhage.39 Most PEG wound bleeding can be controlled by simple pressure. Appropriate measures should be taken to improve abnormal coagulation parameters. Gastric wall and rectus sheath hematomas also have been described, and in general are self-contained lesions that do not lead to significant surgical intervention.40 Bleeding occasionally develops in the PEG or DPEJ tube tract itself. In such cases, it is my experience that the external bumper of the feeding tube should be tightened against the abdominal wall, thereby pulling the internal bumper against the gastric mucosa and compressing the feeding tube tract. Compression should be released within 48 hours to avoid feeding tube tract wound breakdown. Only rarely will surgical intervention be necessary for PEG- or DPEJ-associated bleeding complications. It also is necessary to attempt improvement of significantly abnormal blood coagulation parameters before traction removal of PEG or DPEJ tubes to prevent tract hemorrhage. Gastrostomy or jejunostomy replacement tube devices with a balloon tip can be placed percutaneously at the bedside in patients with an established feeding tube tract who have abnormal blood coagulation parameters, unless it is anticipated that the gastrostomy or jejunostomy tube tract requires dilation before balloon tube insertion. Ulceration. In patients with longstanding PEG or DPEJ tubes, an ulcer may develop underneath the internal tube bolster or on the contralateral gastric or small bowel wall. Ulcers underneath the internal bolster often respond to loosening of the external bolster, which allows the PEG or DPEJ tube internal bolster to be released from the mucosa. Ulceration of the contralateral gastric or small bowel wall can occur with gastrostomy or jejunostomy balloon replacement tubes. In some balloon replacement tubes, the tip of the PEG or DPEJ may extend out from the inflated balloon and injure the contralateral mucosal wall. The balloon replacement tube should be removed and replaced with a nonballoon replacement tube or a replacement tube in which the tube tip is contained within the inflated balloon.41 Clogging. One of the most common problems is tube dysfunction secondary to clogging from medications or enteral formula. All medications should be dissolved in water or an appropriate liquid substance or delivered in liquid form, if such is available. A clinical pharmacist can be consulted for
M.H. DeLegge and A.J. Berry help in choosing the appropriate agent. Bulking agents such as psyllium and resins such as cholestyramine should never be placed through the PEG, PEG/J, or DPEJ tube. Patients and caregivers should be educated in the importance of flushing water through the feeding tube after all medications and enteral formula deliveries. In the event of feeding tube obstruction, flushing the tube with the contents of a 60-mL syringe is recommended. The best feeding tube irrigant for a tube lumen obstruction is warm water.42 Pancreatic enzymes (dissolved in a bicarbonate solution and left to dwell within the feeding tube before water flushing) also can be effective.43 If this technique fails, the gastrostomy or jejunostomy tube can be cleared with a specially designed tube-declogging brush. Tube Deterioration. A common cause of feeding tube dysfunction is deterioration of the tube material itself. Deterioration can be recognized by the presence of pitting, ballooning, and a characteristic smell. Although this presents no real risk to the patient, the tube can develop leaks and break, which makes infusion of tube feeding difficult or impossible. Microscopic examinations have demonstrated that tube deterioration is caused by yeast implantation into the wall of the tube.44 A randomized controlled trial suggested that deterioration leading to tube dysfunction was significantly more common with silicone than with polyurethane PEG tubes.45 No preventative measures have been established as being effective for preventing this problem. In my own practice, I recommend flushing the tube occasionally with 3 to 5 mL of ethanol in an attempt to “sterilize” the tube lumen. There is no absolute time period after which feeding tubes should be removed and exchanged to prevent tube dysfunction. The standard of care is to permit the tubes to remain in place until some tube dysfunction, such as clogging or deterioration, prevents adequate feedings or medication instillation. Replacement Balloon Tube Failure. The cause of replacement feeding tube failure is the early deflation of the balloon that serves as the internal bolster. Balloon dysfunction leads to the requirement for frequent gastrostomy or jejunostomy tube replacement. There are no prospective data comparing manufacturers’ feeding tube balloon tips, although it has been suggested that PEG tube balloons constructed of polyurethane may be more durable than balloons constructed of silicone.46 There are commercially available gastrostomy or jejunostomy replacement tube devices where a flexible bolster, rather than a balloon, serves as the internal bolster. The internal bolster is distended with a rigid stylet and passed through the PEG or DPEJ ostomy site, thus resolving the balloon deflation issue; the stylet can cause damage to the existing PEG or DPEJ tube tract on insertion, however, if not placed properly. Gastric Outlet Obstruction. PEG tubes can migrate forward into the duodenum and cause gastric outlet obstruction when the external bolster on the PEG tube migrates away from the abdominal wall, allowing the PEG tube to slide forward through the gastrostomy tract, its feeding tip migrating into the duodenum47; the internal bolster causes a gastric outlet obstruction. A similar problem has been reported with balloon gastrostomy tubes, where the inflated balloon migrates through the pylorus, resulting in obstruction.48 This
Risks of endoscopic enteral access complication can be avoided by ensuring that the external bolster remains at the same centimeter mark on the gastrostomy tube as determined after initial proper PEG tube positioning. Small Bowel Obstruction. If a replacement feeding tube has a balloon tip and is used in the small bowel, care must be taken to avoid over-inflation of the balloon as it could cause small bowel obstruction. The balloon on most gastrostomy replacement tubes should be filled with 10-15 mL of water, air, or saline. Jejunal balloon replacement tubes should have only 3-5 mL of air, water, or saline injected into their balloons. Major Complications Major complications of percutaneous endoscopic enteral access include necrotizing fasciitis, buried bumper syndrome, esophageal perforation, gastric perforation, small bowel perforation, colocutaneous fistula, and early, inadvertent feeding tube removal. Necrotizing Fasciitis. Necrotizing fasciitis is a rare complication of PEG or DPEJ tube placement.49,50 Patients with diabetes, wound infections, malnutrition, or a poor immune system are at increased risk. Traction and pressure on the PEG or DPEJ wound can also predispose to its development. One study demonstrated that patients who had their PEG tube external bolster set directly against the abdominal wall were more likely to develop wound infection, peristomal drainage, and fasciitis compared with patients whose external PEG bolster was left 3 cm from the abdominal wall.51 It was hypothesized that the distant placement of the external bumper prevented compression of the tissue in the PEG tube tract and wound breakdown. This hypothesis was confirmed in a study on dogs in which gastric mucosal histology showed significantly increased inflammation when the PEG tube external bolsters were placed directly against the abdominal wall compared with external bolsters that were left 4 cm from the abdominal wall.52 Prevention of necrotizing fasciitis is imperative, as treatment requires large surgical debridement, antibiotics, and extensive hospital support. It is important to allow the external bolster of the PEG or DPEJ tube to “free-float” 1 to 2 cm from the abdominal wall after feeding tube placement to prevent this complication. It is also important to make at least a 1-cm abdominal wall skin incision before PEG or DPEJ tube placement to avoid creating too tight a feeding tube wound once the PEG or DPEJ tube is pulled through the abdominal wall. Loose apposition of the stomach or small bowel to the abdominal wall does not result in peritoneal leakage because an early PEG or DPEJ tube tract forms as a result of tissue edema and associated tissue secretions. Wound care is important following PEG or DPEJ tube placement. As with any other surgical procedure, less manipulation is better than frequent tampering with a fresh wound. There are no prospective evaluations supporting the use of topical antibiotics as a preventative measure for wound infection after percutaneous feeding tube placement. It is my practice to simply clean the wound with full-strength hydrogen peroxide and to cover the wound with a gauze dressing. The gauze dressing is changed and the PEG or DPEJ wound is cleansed with hydrogen peroxide daily for 7 days. Following this regimen, the wound
41 can be cleaned with simple soap and water. The gauze dressing can be eliminated unless there is leakage around the PEG or DPEJ tube that is soiling a patient’s clothing. Gastric, esophageal, duodenal, and jejunal perforation with upper endoscopy is a known, but very rare complication of percutaneous feeding tube placement. Although this complication may be considered while placing PEG, PEG/J, or DPEJ tubes, there are no current reports of this problem in the literature. Buried Bumper Syndrome. Buried bumper syndrome is a long-term consequence of tight apposition of the external bolster of the PEG or DPEJ tube against the abdominal wall.53 The internal bolster of the feeding tube slowly erodes into the gastric and abdominal wall as tension is created on the feeding tube tract; ultimately this erosion causes pain and the inability to infuse tube feedings. The diagnosis can be confirmed on endoscopy, which will demonstrate the internal bolster buried within the gastric or small bowel mucosa. Often, just the distal tip of the feeding tube remains in the stomach or small intestine (Fig. 3). The treatment of buried bumper syndrome depends on the type of feeding tube.54 If the internal bolster is collapsible because it is on an externally removable PEG or DPEJ tube, the feeding tube can be removed by simple external traction. In a modification of this technique, the buried bumper PEG tube can be cut short and a guidewire passed through the shortened PEG tube into the gastric cavity.54 The guidewire then can be snared, pulled out of the oral cavity, and attached to a new PEG. As the guidewire at the abdominal surface is pulled, the new PEG is dragged into the gastric cavity, and the dilating portion of the new PEG engages the buried bumper on the old PEG. As the new PEG is pulled through the abdominal wall, the old PEG is pushed out of the abdominal wall and removed. The same procedure has been completed for DPEJ tubes.
Figure 3 Buried bumper syndrome; gastric mucosal view.
42 In contrast, if the internal bumper on the PEG or DPEJ tube is rigid, as it is on endoscopic-only removable feeding tubes, the feeding tube may have to be removed by PEG or DPEJ wound tract cut-down or the push–pull T-technique.55 The push–pull T-technique requires the PEG or DPEJ tube to be cut 3 cm from the abdominal wall. The patient is endoscoped and a snare is passed through the feeding tube opening in the gastric or small bowel wall to the outside of the patient through the feeding tube. An additional short piece of PEG or DPEJ tube is cut from the excess tubing. The snare is opened and this short piece of tubing is grasped and pulled back against the remaining PEG or DPEJ tube, creating a T-shape. A Kelly clamp is placed across the T-shape. The endoscopist slowly removes the endoscope, snare, and PEG or DPEJ tube orally as a second operator pushes the Kelly clamp and feeding tube into the gastric or small bowel lumen. This combined procedure frees the internal bumper from the gastric or small bowel wall. Once the feeding tube is removed, a new feeding tube can be placed back through the existing PEG or DPEJ tract using direct endoscopic visualization. A standard PEG or DPEJ tube placement technique should be used to permit the feeding tube dilator to re-expand the partially closed PEG or DPEJ tube tract. Prevention of the buried bumper syndrome requires good nursing care and patient instruction. As mentioned above, the external bolster of the PEG or DPEJ tube should be left 1 to 2 cm from the abdominal wall. Gauze pads should be placed over the external bolster, not underneath, which would create pressure on the PEG or DPEJ tube tract. In addition, the gastrostomy or jejunostomy tube itself should be pushed forward into the wound slightly and rotated during daily nursing care. This will ensure that the internal bumper does not become buried in the gastric or small bowel mucosa. After rotation, the feeding tube should be placed back in its original position. Colocutaneous Fistula. A colocutaneous fistula is a rare complication of PEG or DPEJ placement that occurs as a result of the interposition of colon between the anterior abdominal wall and the gastric or small bowel wall.56 The PEG or DPEJ tube is placed directly through the colon into the stomach or small intestine. Patients in whom this complication has occurred are often asymptomatic, except for transient fever or ileus, and the problem is usually discovered months after initial tube placement when the original tube is removed (because of “dysfunction”) and a replacement feeding tube is passed into position. As the replacement feeding tube is passed blindly at the bedside, it is pushed through the PEG or DPEJ tract opening and into the colon, but cannot find its way back into the stomach or small bowel. Once the tube feedings are restarted, the patient develops diarrhea from colonic tube feedings and dehydration from not receiving fluids or nutrition. This complication often can be treated by just removing the feeding tube to allow the fistula to close; however, surgery is sometimes necessary to correct the internal gastric– colon or small bowel– colon fistula.57 Prevention of this complication is related to the initial PEG or DPEJ tube procedure. Relying on the combination of transillumination and finger palpation of the abdominal wall in choosing an appropriate PEG or DPEJ tube site rather than
M.H. DeLegge and A.J. Berry using only one of these techniques alone will assure a safe feeding tube entrance site. In questionable situations, an 18or 22-gauge “probing” needle should be passed through the proposed PEG or DPEJ tube site before feeding tube placement. The needle should be advanced slowly with an attached syringe (filled with water or another fluid), creating back pressure. The presence of a sudden bolus of air within the syringe before the probing needle is endoscopically identified in the gastric or small bowel cavity suggests passage through the colon. However, this technique has not been subjected to a prospective evaluation. In questionable situations where a safe access site cannot be determined, ultrasound or CT scan guidance can assist in delineating a safe location. Inadvertent PEG or DPEJ Tube Removal. Inadvertent feeding tube removal is a common complication usually occurring in combative or confused patients who pull on the tube. Many PEG or DPEJ tubes today are designed to be removed by 10 to 14 pounds of external pull pressure. PEG or DPEJ tubes that are inadvertently removed within the first 4 weeks of feeding tube placement should not be replaced blindly at the bedside. Because the tube tract may not have matured adequately, the gastric or small bowel wall and the abdominal wall may have separated, leaving a rent in the stomach or small intestine. Blind replacement of the PEG or DPEJ tube may result in tube placement into the peritoneal cavity. If there is any concern about the possibility of a replacement feeding tube being positioned into the peritoneal cavity, a water-soluble contrast study through the feeding tube should be obtained to confirm proper tube tip position before initiation of tube feedings. If the replacement tube lies within the peritoneal cavity, it should be removed. The patient can be brought to the endoscopy suite for repeat PEG or DPEJ tube placement through the same feeding tube site.58 Patients in this situation should be treated with intravenous antibiotics and monitored for signs of peritonitis, which would require surgical intervention. Other Intraabdominal Complications. A variety of intraabdominal complications have been described in case reports. Although rare, they must be recognized by the interventional endoscopist. One complication is small bowel obstruction from a small bowel wall hematoma following PEG placement.59 In this particular case, the hematoma was on a loop of jejunum near the stomach. An operative procedure allowed evacuation of the hematoma and resolution of the small bowel obstruction. This situation could occur with DPEJ placement. Another complication is intrahepatic placement of a PEG tube.60 In this case, the originally inserted PEG tube malfunctioned and was replaced with a balloon gastrostomy tube 2.5 years after placement. The replacement tube was difficult to push back through the PEG fistula site. A contrast study showed that the balloon gastrostomy tube was inflated within the liver. Contrast from the tube entered the portal venous system. A fistula tract had developed between the liver and the stomach. Subsequent surgical exploration allowed the tube to be removed safely with resection of the gastrohepatic fistula tract; argon gas plasma coagulation of the liver bed was used to prevent bleeding, and the PEG was replaced with a
Risks of endoscopic enteral access Stamm gastrostomy. This complication could occur after DPEJ placement. Herniation of the stomach through a PEG tube fistula site was described in a patient noted to have a leaking PEG site 1 year following PEG tube insertion.61 A bulge was noted at the PEG tube site on the abdominal wall when the patient coughed. CT scanning demonstrated that a portion of the stomach had herniated through the PEG site. The PEG was removed, but the PEG fistula remained open. Surgical repair of the fistula was suggested; however, the patient died of aspiration pneumonia before definitive surgical therapy. This complication could occur after DPEJ placement. Abdominal wall pain can occur and persist after PEG or DPEJ placement. Workup should include a full examination to rule out infection of the abdominal wall. This may include a CT scan to rule out an abdominal wall abscess. In some cases, the pain will be consistent with neuropathic pain, in which case the remedy is often removal of the PEG or DPEJ and reinsertion at a different site. Abdominal wall injection with an anesthetic agent may also be helpful. Peritonitis. Leakage of contents from the stomach or the small intestine into the peritoneal cavity following PEG or DPEJ placement with the feeding tube in situ has been described, but is rare.62 Gastric or small bowel leakage can lead to peritonitis if not detected. In addition, infusion of tube feeding formula can lead to a combination of a chemical and bacterial peritonitis.63 It is hypothesized that if the introducer needle does not enter the stomach or small bowel directly through the abdominal wall during PEG or DPEJ placement, but instead approaches the stomach or small bowel tangentially, a long laceration may develop allowing the escape of gastric or small bowel contents. PEG or DPEJ Tract Tumor Seeding. Patients with proximal GI tract cancers, such as head and neck and esophageal cancers,64,65 are at risk of tumor seeding from the tumor site to the PEG or DPEJ tube tract by mechanical transfer. The feeding tube can transfer tumor cells as it is pushed or pulled by the tumor into its final position across the gastric or small bowel and abdominal wall. Some authors have suggested using an overtube during PEG placement to prevent this complication.
43
4.
5.
6.
7. 8.
9.
10.
11.
12.
13.
14. 15.
16.
17. 18. 19.
20.
21.
Conclusion Enteral access is necessary to provide enteral nutrition. The endoscopist plays an important role in this regard by placing nasoenteric tubes as well as gastrostomy and jejunostomy feeding tubes. With these enteral access techniques, there is the potential for both minor or major associated complications. The endoscopist should be aware of these complications, how to avoid them and how to treat them.
References 1. Burtch CD, Shatney CH: Feeding jejunostomy (versus gastrostomy) passes the test of time. Am Surg 53:54-57, 1987 2. Ho CS, Yee ACV, McPherson K: Complications of surgical and percutaneous non-endoscopic gastrostomy: review of 233 patients. Gastroenterology 95:206-210, 1988 3. Neumann DA, DeLegge MH: Gastric versus small bowel feeding in the
22.
23. 24.
25.
26.
27.
ICU: a prospective comparison of efficacy [abstract]. Gastroenterology 118:A774, 2000 Baskin WN, Johansen JF: An improved approach to the delivery of enteral nutrition in the intensive care unit. Gastrointest Endosc 42:161165, 1995 Fang JC, Hilden K, Holbukov R, et al: Transnasal endoscopy vs fluoroscopy for the placement of nasoenteric feeding tubes in critically ill patients. Gastrointest Endosc 62:661-666, 2005 Cataldi-Belcher EL, Selzer MH, Slocumb BA, et al: Complications during enteral nutrition therapy: a prospective study. JPEN 7:546-552, 1983 Leong SC, Mahanta V: Securing nasogastric tube in head and neck cancer patients. Laryngoscope 116:2089-2091, 2006 Whelan K, Hill L, Preedy VR, et al: Formula delivery in patients receiving enteral tube feeding on general hospital wards: the impact of nasogastric extubation and diarrhea. Nutrition 22:1025-1031, 2006 McClave SA, DeMeo MT, DeLegge MH, et al: North American Summit on Aspiration in the Critically Ill Patient: Consensus Statement. JPEN 26:S80-S85, 2002 Fay DE, Poplausky M, Gruber M, et al: Long term enteral feeding: a retrospective comparison of delivery via percutaneous endoscopic gastrostomy and nasoenteric tubes. Am J Gastroenterol 86:1604-1609, 1991 Heyland DK, Drover JW, MacDonald S, et al: Effect of postpyloric feeding on gastroesophageal regurgitation and pulmonary microaspiration: results of a randomized controlled trial. Crit Care Med 29:14951501, 2001 Powell KS, Marcuard SP, Farrior ES, et al: Aspirating gastric residuals causes occlusion of small bore feeding tubes. J Parenter Enteral Nutr 17:243-246, 1993 Scholten DJ, Wood TL, Thompson DR: Pneumothorax from nasoenteric feeding tube insertion. A report of 5 cases. Am Surg 52:381-385, 1986 McWey RE, Curry NS, Schabel SI, et al: Complications of nasoenteric feeding tubes. Am J Surg 155:253-257, 1988 George DL, Falk PS, Umberto Meduri G, et al: Nosocomial sinusitis in patients in the medical intensive care unit: a prospective epidemiological study. Clin Infect Dis 27:463-470, 1998 Van Zanten AR, Dixon JM, Nipshagen MD, et al: Hospital acquired sinusitis is a common cause of fever of unknown origin in orotracheally intubated critically ill patients. Crit Care 9:R583-R590, 2005 Desmond P, Raman R, Idikula J: Effect of nasogastric tubes on the nose and maxillary sinus. Crit Care Med 19:509-511, 1991 Baskin WN: Acute complications associated with bedside placement of feeding tubes. Nutr Clin Prac 21:40-55, 2006 Pancorbo-Hidalgo PL, Garcia-Fernandez FP, Ramirez-Perez C: Complications associated with enteral nutrition by nasogastric tube in an internal medicine unit. J Clin Nurs 10:482-490, 2001 Davies AR, Froomes PR, French CJ, et al: Randomized comparison of nasojejunal and nasogastric feeding in critically ill patients. Crit Care Med 30:586-590, 2002 Patrick PG, Marulendra S, Kirby DF, et al: Endoscopic nasogastricjejunal feeding tube placement in critically ill patients. Gastrointest Endosc 45:72-76, 1997 Russell M, Cromer M, Grant J: Complications of enteral nutrition therapy, in Gottschlich M (ed): The Science and Practice of Nutrition Support: A case based core curriculum. Dubuque, IA, American Society for Enteral and Parenteral Nutrition, 2001, pp 194-195 Gauderer MWL, Ponsky J, Izant RJ Jr: Gastrostomy without laparotomy: a percutaneous endoscopic approach. J Pediatr Surg 15:872-875, 1980 Jain NK, Larson DE, Schroeder KW, et al: Antibiotic prophylaxis for percutaneous, endoscopic gastrostomy: a prospective randomized, double blind clinical trial. Ann Intern Med 107:824-828, 1987 Hogan RB, DeMarco DC, Hamilton JK, et al: Percutaneous endoscopic gastrostomy-to push or to pull: a prospective, randomized trial. Gastrointest Endosc 32:253-258, 1986 Larson DE, Burton DD, Schroeder KW, et al: Percutaneous, endoscopic gastrostomy: Indications, success, complications and mortality in 314 consecutive patients. Gastroenterology 93:48-52, 1987 DeLegge MH, Patrick PG, Gibbs R: Percutaneous endoscopic gastrojejunostomy with a tapered tip, unweighted jejunal feeding tube: improved placement success. Am J Gastroenterol 91:1130-1134, 1996
M.H. DeLegge and A.J. Berry
44 28. Shike M, Berner YN, Gerdes H, et al: Percutaneous endoscopic gastrostomy and jejunostomy for long term feeding in patients with head and neck cancer. Otolaryngol Head Neck Surg 101:549-554, 1989 29. Fan AC, Baron TH, Rumalla A, et al: Comparison of direct percutaneous endoscopic jejunostomy and PEG with jejunal extension. Gastrointest Endosc 56:890-894, 2002 30. Delegge MH, Ginsberg G, McClave S, et al: Randomized prospective comparison of direct percutaneous endoscopic jejunostomy (DPEJ) vs. percutaneous endoscopic gastrostomy with jejunal extension (PEG-J) feeding tube placement for enteral feeding. Gastrointest Endosc 59: 158(A), 2004 31. Sharma VK, Howden CW: Meta-analysis of randomized, controlled trials of antibiotic prophylaxis before percutaneous endoscopic gastrostomy. Am J Gastroenterol 95:3133, 2000 32. Ahmad I, Mouncher A, Abdoolah A, et al: Antibiotic prophylaxis for percutaneous endoscopic gastrostomy: a prospective, randomized, double-blind trial. Aliment Pharmacol Ther 18:209, 2003 33. Chaudhary KIA, Smith OJ, Cuddy PG, et al: PEG site infections: the emergence of methicillin resistant Staphylococcus aureus as a major pathogen. Am J Gastroenterol 97:1713, 2002 34. Horiuchi A, Nakayama Y, Kajiyama M, et al: Nasopharyngeal decolonization of methicillin-resistant staphylococcus aureus can reduce PEG peristomal wound infection. Am J Gastroenterol 101:274, 2006 35. Tsang TK, Eaton D, Falconio MA: Percutaneous ostomy dilation: a technique for dilating the closed percutaneous endoscopic gastrostomy sites and inserting gastrostomies. Gastrointest Endosc 35:336-337, 1989 36. Gottfried EB, Plumser AB, Clair MR: Pneumoperitoneum following percutaneous endoscopic gastrostomy. Gastrointest Endosc 32:397, 1986 37. Stathopoulos G, Rudberg MA, Harig JM: Subcutaneous emphysema following PEG. Gastrointest Endosc 37:374, 1991 38. Dulabon GR, Abrams JE, Rutherford EJ: The incidence and significance of free air after percutaneous endoscopic gastrostomy. Am Surg 6:590593, 2002 39. Lau G, Siang HL: Fatal retroperitoneal haemorrhage: an unusual complication of percutaneous endoscopic gastrostomy. Forensic Sci Int 116:69-75, 2001 40. Ubogu EE, Zaidat OO: Rectus sheath hematoma complicating percutaneous endoscopic gastrostomy. Surg Laparosc Endosc Percutan Tech 12:430-432, 2002 41. Kazi S, Gunasekaran TS, Berman JH, et al: Gastric mucosal injuries in children from inflatable low-profile gastrostomy tubes. J Pediatr Gastroenterol Nutr 24:75, 1997 42. Metheny N, Eisenberg P, McSweeney M: Effect of feeding tube properties and three irrigants on clogging rates. Nurs Res 37:165, 1988 43. Sriram K, Jayanthi V, Lakshmi RG, et al: Prophylactic locking of enteral feeding tubes with pancreatic enzymes. JPEN J Parenter Enteral Nutr 21:353, 1997 44. Iber FL, Livak A, Patel M: Importance of fungus colonization in failure of silicone rubber percutaneous gastrostomy tubes (PEGs). Dig Dis Sci 41:226, 1996 45. Blacka J, Donoghue J, Sutherland M, et al: Dwell time and functional failure in percutaneous endoscopic gastrostomy tubes: a prospective randomized-controlled comparison between silicon polymer and poly-
46.
47.
48. 49.
50.
51. 52.
53.
54.
55.
56.
57. 58. 59. 60. 61. 62.
63.
64. 65.
urethane percutaneous endoscopic gastrostomy tubes. Aliment Pharmacol Ther 20:875, 2004 DeLegge RL, DeLegge MH: Percutaneous endoscopic gastrostomy evaluation of device materials: are we “failsafe?” Nutr Clin Pract 20:613617, 2005 Fischer LS, Bonello JC, Greenberg E: Gastrostomy tube migration and gastric outlet obstruction following percutaneous endoscopic gastrostomy. Gastrointest Endosc 33:381-382, 1987 Chong VH: Gastric outlet obstruction by gastrostomy tube balloon. Indian J Gastroenterol 23:90, 2004 Cave DR, Robinson WR, Brotsch EA: Necrotizing fasciitis following percutaneous endoscopic gastrostomy. Gastrointest Endosc 32:294, 1986 Martindale R, Witte M, Hodges G, et al: Necrotizing fasciitis as a complication of percutaneous endoscopic gastrostomy. JPEN J Parenter Enteral Nutr 11:583, 1987 Chung RS, Shertzer M: Pathogenesis of complications of percutaneous endoscopic gastrostomy. Am Surg 56:134, 1990 DeLegge MH, Lantz G, Kazacos R, et al: Effects of external bolster tension on PEG tube tract formation (abstract). Gastrointest Endosc 43:349, 1996 Klein S, Heare BR, Soloway RD: The “buried bumper syndrome”: a complication of percutaneous endoscopic gastrostomy. Am J Gastroenterol 85:448, 1990 Venu RP, Brown RD, Pastika BJ, et al: The buried bumper syndrome: a simple management approach in two patients. Gastrointest Endosc 56:582, 2002 Boyd JW, DeLegge MH, Shamburek RD, et al: The buried bumper syndrome: a new technique for safe, endoscopic PEG removal. Gastrointest Endosc 41:508, 1995 Saltzberg DM, Anand K, Juvan P, et al: Colocutaneous fistula: an unusual complication of percutaneous endoscopic gastrostomy. JPEN J Parenter Enteral Nutr 11:86, 1987 Berger SA, Zarling EJ: Colocutaneous fistula following migration of PEG tube. Gastrointest Endosc 37:86, 1991 Galat SA, Gerig KD, Porter JA, et al: Management of premature removal of percutaneous gastrostomy. Am Surg 56:733, 1990 Williams E, Sabol DA, Delegge M: Small bowel obstruction caused by bowel wall hematoma after PEG. Gastrointest Endosc 57:273, 2003 Chaer RA, Rekkas D, Trevino J, et al: Intrahepatic placement of a PEG tube. Gastrointest Endosc 57:763, 2003 Chuang CH, Chen CY: Gastric herniation through PEG site. Gastrointest Endosc 58:416, 2003 Haslam N, Hughes S, Harrison RF: Peritoneal leakage of gastric contents, a rare complication of percutaneous endoscopic gastrostomy. J Parenter Enteral Nutr 20:433-434, 1996 Larson DE, Burton DD, Schroeder KW, et al: Percutaneous endoscopic gastrostomy, indications, success, complications and mortality in 314 consecutive patients. Gastroenterology 93:48-52, 1987 Khurana V, Singh T: Percutaneous endoscopic gastrostomy site metastasis in esophageal cancer. Gastrointest Endosc 62:612, 2005 Mincheff TV: Metastatic spread to a percutaneous gastrostomy site from head and neck cancer: case report and literature review. JSLS 9:466471, 2005
E
nteral access is necessary to provide enteral feeding in patients who cannot or will not eat. Enteral access techniques and commercialized kits have been designed to meet the specific needs of patients, such as into what site (gastric or jejunal) and for how long tube feedings should be given (Table 1). Enteral access techniques have been developed for use at the patient’s bedside or for use by the radiologist, the surgeon, or the endoscopist in a non-bedside location (Table 2). This article will focus on the complications associated with endoscopic enteral access.
Bedside Nasoenteric Access Bedside nasoenteric tube (NET) blind placement is the most common method to achieve enteral access. The final position of this enteral access tube is either the stomach (for gastric feedings) or the jejunum (for small bowel feedings). Obtaining blind nasogastric (NG) access is easier than obtaining nasojejunal (NJ) access. There are specific indications for NJ access. The use of small bowel feedings to prevent tube feeding aspiration events is a complicated and contentious issue. Some studies have shown a decrease in aspiration episodes when patients were fed directly into the small bowel compared with when feedings were instilled in the stomach.1,2 A recent, prospective trial by Neumann and coworkers compared the use of gastric and small bowel feedings in the intensive care unit, and showed no differences in aspiration episodes; it took longer to initiate small bowel feedings, however, because of the difficulty in obtaining adequate tube position.3 This deDigestive Disease Center, Medical University of South Carolina, Charleston, SC. Address reprint requests to Mark H. DeLegge, Medical University of South Carolina, Digestive Disease Center, 96 Jonathan Lucas Street, 210 Clinical Science Building, Charleston, SC 29425. E-mail: deleggem@ musc.edu
36
1096-2883/08/$-see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.tgie.2007.08.009
bate certainly will continue, but for patients with gastroparesis, patients intolerant to gastric feedings, or patients who have had a witnessed tube feeding aspiration event with gastric feedings, small bowel access and feedings should be initiated. Failure to pass a NJ tube blindly at the bedside requires the use of fluoroscopic or endoscopic methods to guide passage; the preference of technique is center dependent. In centers with available C-arm fluoroscopy and modified fluoroscopy beds, fluoroscopic passage of NJ tubes can be done at the patient’s bedside with a success rate that approaches 100%.4 In institutions without such bedside fluoroscopic capabilities, the transport of patients (especially critically ill patients) can be time consuming, expensive, and hazardous5; in these instances, endoscopic passage of NJ tubes is preferred.
Complications of Nasoenteric Tubes Complications of NET placement have been reported in up to 12% of patients6 (Table 3), and include pneumothorax, sinusitis, tube dislodgement, tube clogging, and aspiration pneumonia. Early identification of complications, such as tube dislodgement or clogging, is important in preventing interruptions in tube feeding regimens.7,8
Aspiration pneumonia Aspiration pneumonia is a leading cause of morbidity and mortality in the critically ill patient population,9 and a NET may increase the risk for this complication. Fay and coworkers looked at aspiration pneumonia occurrence in patients with NETs compared with percutaneous gastrostomy tubes (PEGs). This study found that, within 14 days, patients with NETs developed aspiration pneumonia significantly more than patients with PEGs (24% versus 6%).10 Defense mech-
Risks of endoscopic enteral access
37
Table 1 Endoscopic Enteral Access Methods Type of Access Surgical or percutaneous gastrostomy Surgical or percutaneous gastrojejunostomy Surgical or percutaneous jejunostomy Nasal/oral gastric tube Nasal/oral gastrojejunal tube
Nasal/oral jejunal tube
anisms present in healthy adults are altered in the critically ill, making this patient population more susceptible to aspiration pneumonia.9 Heyland and coworkers compared gastric and postpyloric feedings in medical and surgical ICU patients, and found that gastric feedings were associated with significantly more episodes of gastroesophageal regurgitation with a trend toward increased microaspiration.11 The North American Summit on aspiration in critically ill patients also noted the high risk of aspiration with NG tubes.9 They did not recommend decreasing tube diameter size or converting an NG tube to a PEG solely for prevention of aspiration pneumonia, but they did recommend elevation of the head of bed 30° to 45°, changing infusion from bolus-type tube feedings to continuous feedings, optimizing oral health, and utilizing small bowel feedings whenever possible.9 The decision to use a jejunal tube (J-tube) for feeding also should warrant some very specific instructions regarding its care. The lumen of these tubes is often much smaller than that of a gastric tube, and therefore, they are more prone to clogging. Jejunal feeding tubes should never be checked for residual content, as their residuas are a poor indicator of residual content of the small bowel. In addition, checking residuals through these small-bore tubes increases their probability of clogging.12 These tubes should be flushed after every tube feeding and medication instillation. Only liquid medications, or completely dissolved medications, should be
Function Gastric Gastric Gastric Jejunal Jejunal Gastric Gastric Gastric Gastric Jejunal Jejunal
feeding decompression decompression Feeding feeding feeding decompression feeding decompression feeding feeding
Length of Need >1 month 1–6 months >1 month <1 month <1 month
<1 month
placed through a J-tube to reduce the chances of tube occlusion. Care should be taken to stop tube feedings during infusion of medications such as theophylline or potassium chloride, products that are known to coagulate tube feedings and obstruct the J-tube.
Pneumothorax Inadvertent tube placement into the trachea causing pneumothorax occurs most often when nasoenteric tubes are placed blindly at the bedside. Schoten and coworkers reviewed 5 cases in which pneumothorax occurred from such malplacement, noting the significant morbidity and mortality that can occur with tube misplacement and the importance of appropriately evaluating the position of the tube after placement.13 Three of the 5 patients required chest tube thoracostomy; 1 patient was eventually discharged, 1 required 2 chest tubes (this patient died 2 days later), and the third developed chronic empyema requiring chest wall reconstruction. A cuffed endotracheal tube did not protect patients from feeding tube misplacement. Another study echoed the importance of confirmation of correct feeding tube placement by abdominal films and adequate training of personnel involved in placement of these tubes14; this study found 14 inadvertent tube misplacements out of 1100 attempts over a period of 18 months, including pneumothorax, hydrothorax, empyema, mediastinitis, pneumonia, and esophageal perforation.
Table 2 Methods of Obtaining Enteral Access
Sinusitis
Bedside-placed Tubes Nasal/oral gastric tube Nasal/oral jejunal tube Nasal/oral gastro/jejunal tube Radiologically-placed Tubes Percutaneous gastrostomy tube Percutaneous jejunostomy tube Percutaneous gastro/jejunostomy tube Nasoenteric tube Surgically-placed Tubes Gastrostomy tube Jejunostomy tube Gastro/jejunostomy tube Endoscopically-placed Tubes Percutaneous endoscopic gastrostomy tube Percutaneous endoscopic jejunostomy tube Percutaneous endoscopic gastro/jejunostomy tube Endoscopic nasoenteric tube
In a study by George and colleagues, nosocomial sinusitis in a medical intensive care unit was reported in 12 cases per 1000 patient days; 7.7% of patients with NJ tubes had sinusitis, and 19.8 cases were documented per 1000 NET days.15 Table 3 Complications of Nasogastric and Nasojejunal Tubes Aspiration pneumonia Nasal mucosal ulceration Naso-pharyngeal bleeding Otitis media Pharyngitis Pneumothorax Sinusitis Tracheoesophageal fistula Tube obstruction Tube migration
38 Table 4 Complications of Percutaneous Endoscopic Enteral Access Aspiration pneumonia Bleeding Buried bumper syndrome Clogging of tube Colocutaneous fistula Gastric outlet obstruction Hepatic injury Herniation of the stomach through the PEG site Ileus Inadvertant PEG/PEG-J removal Necrotizing fasciitis PEG site breakdown PEG site wound infection Peri-stomal leakage Peritonitits Pneumoperitoneum Replacement balloon tube failure Small bowel obstruction Tube deterioration Ulceration of stomach or small bowel
Van Zanten and coworkers found sinusitis as the sole source of fever in 16.2% of patients and as a contributing factor in 13.8% of critically ill patients with fever of unknown origin.16 Another study looked at patients with NG tubes placed after surgery and remaining in place longer than 48 hours.17 This study found that 75% of patients developed rhinitis, with no clinical symptoms, but with radiologic evidence of maxillary sinusitis. The authors concluded that local symptoms of sinusitis may not be present despite radiologic evidence of inflammation. The first line of treatment of sinusitis is removal of the NET tube; antibiotics are used as necessary.16,17 NETs are intended for short-term use only, and some facilities have switched to oroenteric tube placement because of complications such as sinusitis.16,18
M.H. DeLegge and A.J. Berry
Endoscopic Percutaneous Access PEG was developed by Ponsky and Gauderer in the early 1980s.23 The procedure involves the placement of a percutaneous gastrostomy tube after endoscopic trans-illumination of the stomach for an appropriate PEG access position. The use of prophylactic antibiotics before the procedure is important in the prevention of postprocedure infections.24 PEG tubes may be placed by either the Sachs-Vine (push) or Ponsky (pull) techniques, the decision being just a matter of physician preference.25 Prospective evaluations of PEG placement have found this procedure to be associated with known procedure-related complications (Table 4).26 In those patients in whom small bowel feedings are desired, endoscopic access to the bowel may be obtained by two methods. The first method, PEG/J, places a jejunal feeding tube through an existing PEG and into the small bowel using an over-the-guidewire method. DeLegge and coworkers reported a 100% success rate without major complications using this technique for PEG/J placement with a procedure time of approximately 26 minutes.27 This PEG/J system allowed for concurrent gastric decompression and small bowel feeding. The average longevity of this tube system was approximately 120 days when patients who died were excluded from the analysis. J-tube failures often were secondary to migration. Improper positioning of the J-tube through the PEG into the small bowel will result in J-tube migration back into the stomach, and any loop of J-tube left in the stomach after placement will ultimately result in J-tube migration. The Jtube needs to be seen exiting the PEG and directly passing through the pylorus into the small bowel with no evidence of gastric J-tube looping (Fig. 1). The second method, DPEJ, directly places a J-tube into the small bowel using an endoscope (Fig. 2). This procedure requires the use of an enteroscope or a pediatric colonoscope to reach a puncture position beyond the ligament of Treitz. A
Tube Dislocation and Clogging Tube dislocation and clogging are two commonly reported complications that cause tube feedings to be stopped. One study from a medical intensive care unit reported tube dislodgement in 48.5% of their patients and tube occlusion to occur 12.5% of the time.19 Another study looked at both NG and NJ tubes in the critically ill patient population and found that inadvertent removal of tubes occurred in one of every three patients, which the authors related to problems with patient sedation and securing of the tube.20 Patrick and coworkers studied NG and nasojejunal tubes in 54 critically ill patients and found 6% of patients had a clogged tube that did not require tube removal, whereas 3% had a clogged feeding tube that did require removal of the tube.21 They further discussed that clogging most often occurred because of infusion of inappropriate medications through the feeding tube or lack of flushing of the tube. They concluded that educating staff in the care and flushing of the tube may help prevent this complication. Irrigants, such as warm water, sodium bicarbonate, pancrealipase mixture, or a declogging device, can be used once clogging occurs.22
Figure 1 Abdominal plain film showing a jejunal tube exiting the PEG with no looping in the stomach.
Risks of endoscopic enteral access
39 those who did not receive antibiotic prophylaxis.31,32 Most PEG wound infections will respond to a first generation cephalosporin or a quinolone. Methicillin-resistant Staphylococcus aureus (MRSA) has emerged as an important cause of PEG site infections in some centers.33 One study found that nasopharyngeal decontamination of patients infected with MRSA (in addition to standard prophylactic antibiotics) significantly reduced the incidence of wound infections.34
Figure 2 Direct percutaneous jejunostomy; needle catheter in the small intestine. (Color version of figure is available at www.techgiendoscopy.com.)
standard 20- or 24-F PEG tube may be placed and used as the J-tube. Both Mellert and Shike have reported good success with this procedure, with only minor complications, including local DPEJ site infection, but no reported cases of peritonitis or bowel infarction.28 Other authors also have reported good success with DPEJ. A retrospective study by Fan and coworkers compared physician re-interventions for J-tube complications, such as J-tube occlusion or J-tube migration, in a group of patients who received PEG/J compared with another group of patients who received DPEJ. The DPEJ patients had significantly fewer endoscopic re-interventions.29 This same result was confirmed by DeLegge and co-workers in a prospective series of 44 patients randomized to PEG/J versus DPEJ enteral access.30
Complications of Percutaneous Endoscopic Enteral Access Complications from endoscopic percutaneous feeding tube placement generally involve either the PEG tube or the DPEJ tube. These are the tube systems that provide direct access to the stomach or the small bowel. There are some complications associated with the jejunal extension tube in the PEG/J system, although these are fewer in number. Complications from PEG tube placement can be classified as minor or major. Minor Complications Minor complications include wound infection, peristomal leakage, pneumoperitoneum, ileus, bleeding, gastric or small bowel ulceration, tube occlusion, tube degradation, balloon tube failure, and gastric or small bowel obstruction. Wound Infection. Wound infection is more likely to occur when a PEG or DPEJ has been placed through a contaminated procedure field, a contaminated oropharyngeal cavity, or with poor technique, especially in debilitated patients and
Peristomal Leakage. Peristomal leakage usually occurs within the first few days after PEG or DPEJ placement. It is more likely to occur in malnourished patients and those with diabetes who may have poor tissue healing and are prone to wound breakdown. In addition, placement of the external bolster of the PEG tube too tightly against the external abdominal wall will lead to poor tissue blood flow, wound breakdown, and peristomal leakage. Treatment should include correction of patient comorbidities, such as malnutrition and elevated blood sugars, loosening of the external bolster, and local measures to address skin breakdown (such as use of powdered absorbing agents or a skin protectant such as zinc oxide paste). Placement of a larger size gastrostomy or jejunostomy tube through the same feeding tube tract will not solve the problem. Once the PEG or DPEJ tube tract has started to leak, placing larger tubes through the same tract will serve only to further distend and distort the tract and will not promote tissue growth or healing. As a treatment method for peristomal leakage, the PEG or DPEJ tube can be removed for 24 to 48 hours, permitting the tract to close slightly; a replacement feeding tube can then be placed through the same, partially closed tract.35 This technique works well for patients whose PEG or DPEJ tube tract started to leak a month or more after initial insertion. It does not work as well for patients with early tract leakage because these patients usually have poor wound healing from other comorbid disease processes. In some patients with a mature PEG or DPEJ tube tract with persistent peristomal leakage, the feeding tube will need to be permanently removed, allowing the tract to close completely. Another PEG or DPEJ tube can then be placed at a different location on the abdominal wall. It is my experience that the new PEG or DPEJ tube can be placed when there is at least 50% closure of the old feeding tube tract, at which point initiation of feedings will not have a significant impact on leakage or inhibition of tissue healing through the old feeding tube tract. Pneumoperitoneum. Pneumoperitoneum is common after PEG or DPEJ placement.36 It is thought to be caused by insufflation of air associated with the endoscopic procedure and needle puncture of the gastric or small bowel wall. In the absence of peritonitis, it has no consequence and should not preclude feedings. Pneumoperitoneum may cause confusion for clinicians, however, in patients with clinical features that raise concern about a ruptured viscus. In these settings, a contrast radiology study should be obtained to confirm the position of the PEG or DPEJ tube within the stomach or small bowel and to exclude a leak. Subcutaneous air also has been described after PEG placement, and from air being introduced between the cutaneous
40 and subcutaneous tissues.37 In the absence of other findings, it is inconsequential and should not preclude feeding.37 Gastroparesis/Ileus. Some patients develop nausea and vomiting after PEG or DPEJ placement, which may be from transient gastroparesis. Rarely, an ileus develops, especially in patients with significant pneumoperitoneum.38 After a gastric, duodenal, or jejunal perforation has been excluded, patients who develop an ileus should be treated with bowel rest and, if necessary, gastric decompression. These patients can be identified by the presence of postprocedure abdominal distention, vomiting, and absence of bowel sounds. Feedings should be held until the ileus resolves. Bleeding. Hemorrhage following PEG or DPEJ tube placement is rare, but there have been reports of significant bleeding following PEG, including an aortic perforation and a retroperitoneal hemorrhage.39 Most PEG wound bleeding can be controlled by simple pressure. Appropriate measures should be taken to improve abnormal coagulation parameters. Gastric wall and rectus sheath hematomas also have been described, and in general are self-contained lesions that do not lead to significant surgical intervention.40 Bleeding occasionally develops in the PEG or DPEJ tube tract itself. In such cases, it is my experience that the external bumper of the feeding tube should be tightened against the abdominal wall, thereby pulling the internal bumper against the gastric mucosa and compressing the feeding tube tract. Compression should be released within 48 hours to avoid feeding tube tract wound breakdown. Only rarely will surgical intervention be necessary for PEG- or DPEJ-associated bleeding complications. It also is necessary to attempt improvement of significantly abnormal blood coagulation parameters before traction removal of PEG or DPEJ tubes to prevent tract hemorrhage. Gastrostomy or jejunostomy replacement tube devices with a balloon tip can be placed percutaneously at the bedside in patients with an established feeding tube tract who have abnormal blood coagulation parameters, unless it is anticipated that the gastrostomy or jejunostomy tube tract requires dilation before balloon tube insertion. Ulceration. In patients with longstanding PEG or DPEJ tubes, an ulcer may develop underneath the internal tube bolster or on the contralateral gastric or small bowel wall. Ulcers underneath the internal bolster often respond to loosening of the external bolster, which allows the PEG or DPEJ tube internal bolster to be released from the mucosa. Ulceration of the contralateral gastric or small bowel wall can occur with gastrostomy or jejunostomy balloon replacement tubes. In some balloon replacement tubes, the tip of the PEG or DPEJ may extend out from the inflated balloon and injure the contralateral mucosal wall. The balloon replacement tube should be removed and replaced with a nonballoon replacement tube or a replacement tube in which the tube tip is contained within the inflated balloon.41 Clogging. One of the most common problems is tube dysfunction secondary to clogging from medications or enteral formula. All medications should be dissolved in water or an appropriate liquid substance or delivered in liquid form, if such is available. A clinical pharmacist can be consulted for
M.H. DeLegge and A.J. Berry help in choosing the appropriate agent. Bulking agents such as psyllium and resins such as cholestyramine should never be placed through the PEG, PEG/J, or DPEJ tube. Patients and caregivers should be educated in the importance of flushing water through the feeding tube after all medications and enteral formula deliveries. In the event of feeding tube obstruction, flushing the tube with the contents of a 60-mL syringe is recommended. The best feeding tube irrigant for a tube lumen obstruction is warm water.42 Pancreatic enzymes (dissolved in a bicarbonate solution and left to dwell within the feeding tube before water flushing) also can be effective.43 If this technique fails, the gastrostomy or jejunostomy tube can be cleared with a specially designed tube-declogging brush. Tube Deterioration. A common cause of feeding tube dysfunction is deterioration of the tube material itself. Deterioration can be recognized by the presence of pitting, ballooning, and a characteristic smell. Although this presents no real risk to the patient, the tube can develop leaks and break, which makes infusion of tube feeding difficult or impossible. Microscopic examinations have demonstrated that tube deterioration is caused by yeast implantation into the wall of the tube.44 A randomized controlled trial suggested that deterioration leading to tube dysfunction was significantly more common with silicone than with polyurethane PEG tubes.45 No preventative measures have been established as being effective for preventing this problem. In my own practice, I recommend flushing the tube occasionally with 3 to 5 mL of ethanol in an attempt to “sterilize” the tube lumen. There is no absolute time period after which feeding tubes should be removed and exchanged to prevent tube dysfunction. The standard of care is to permit the tubes to remain in place until some tube dysfunction, such as clogging or deterioration, prevents adequate feedings or medication instillation. Replacement Balloon Tube Failure. The cause of replacement feeding tube failure is the early deflation of the balloon that serves as the internal bolster. Balloon dysfunction leads to the requirement for frequent gastrostomy or jejunostomy tube replacement. There are no prospective data comparing manufacturers’ feeding tube balloon tips, although it has been suggested that PEG tube balloons constructed of polyurethane may be more durable than balloons constructed of silicone.46 There are commercially available gastrostomy or jejunostomy replacement tube devices where a flexible bolster, rather than a balloon, serves as the internal bolster. The internal bolster is distended with a rigid stylet and passed through the PEG or DPEJ ostomy site, thus resolving the balloon deflation issue; the stylet can cause damage to the existing PEG or DPEJ tube tract on insertion, however, if not placed properly. Gastric Outlet Obstruction. PEG tubes can migrate forward into the duodenum and cause gastric outlet obstruction when the external bolster on the PEG tube migrates away from the abdominal wall, allowing the PEG tube to slide forward through the gastrostomy tract, its feeding tip migrating into the duodenum47; the internal bolster causes a gastric outlet obstruction. A similar problem has been reported with balloon gastrostomy tubes, where the inflated balloon migrates through the pylorus, resulting in obstruction.48 This
Risks of endoscopic enteral access complication can be avoided by ensuring that the external bolster remains at the same centimeter mark on the gastrostomy tube as determined after initial proper PEG tube positioning. Small Bowel Obstruction. If a replacement feeding tube has a balloon tip and is used in the small bowel, care must be taken to avoid over-inflation of the balloon as it could cause small bowel obstruction. The balloon on most gastrostomy replacement tubes should be filled with 10-15 mL of water, air, or saline. Jejunal balloon replacement tubes should have only 3-5 mL of air, water, or saline injected into their balloons. Major Complications Major complications of percutaneous endoscopic enteral access include necrotizing fasciitis, buried bumper syndrome, esophageal perforation, gastric perforation, small bowel perforation, colocutaneous fistula, and early, inadvertent feeding tube removal. Necrotizing Fasciitis. Necrotizing fasciitis is a rare complication of PEG or DPEJ tube placement.49,50 Patients with diabetes, wound infections, malnutrition, or a poor immune system are at increased risk. Traction and pressure on the PEG or DPEJ wound can also predispose to its development. One study demonstrated that patients who had their PEG tube external bolster set directly against the abdominal wall were more likely to develop wound infection, peristomal drainage, and fasciitis compared with patients whose external PEG bolster was left 3 cm from the abdominal wall.51 It was hypothesized that the distant placement of the external bumper prevented compression of the tissue in the PEG tube tract and wound breakdown. This hypothesis was confirmed in a study on dogs in which gastric mucosal histology showed significantly increased inflammation when the PEG tube external bolsters were placed directly against the abdominal wall compared with external bolsters that were left 4 cm from the abdominal wall.52 Prevention of necrotizing fasciitis is imperative, as treatment requires large surgical debridement, antibiotics, and extensive hospital support. It is important to allow the external bolster of the PEG or DPEJ tube to “free-float” 1 to 2 cm from the abdominal wall after feeding tube placement to prevent this complication. It is also important to make at least a 1-cm abdominal wall skin incision before PEG or DPEJ tube placement to avoid creating too tight a feeding tube wound once the PEG or DPEJ tube is pulled through the abdominal wall. Loose apposition of the stomach or small bowel to the abdominal wall does not result in peritoneal leakage because an early PEG or DPEJ tube tract forms as a result of tissue edema and associated tissue secretions. Wound care is important following PEG or DPEJ tube placement. As with any other surgical procedure, less manipulation is better than frequent tampering with a fresh wound. There are no prospective evaluations supporting the use of topical antibiotics as a preventative measure for wound infection after percutaneous feeding tube placement. It is my practice to simply clean the wound with full-strength hydrogen peroxide and to cover the wound with a gauze dressing. The gauze dressing is changed and the PEG or DPEJ wound is cleansed with hydrogen peroxide daily for 7 days. Following this regimen, the wound
41 can be cleaned with simple soap and water. The gauze dressing can be eliminated unless there is leakage around the PEG or DPEJ tube that is soiling a patient’s clothing. Gastric, esophageal, duodenal, and jejunal perforation with upper endoscopy is a known, but very rare complication of percutaneous feeding tube placement. Although this complication may be considered while placing PEG, PEG/J, or DPEJ tubes, there are no current reports of this problem in the literature. Buried Bumper Syndrome. Buried bumper syndrome is a long-term consequence of tight apposition of the external bolster of the PEG or DPEJ tube against the abdominal wall.53 The internal bolster of the feeding tube slowly erodes into the gastric and abdominal wall as tension is created on the feeding tube tract; ultimately this erosion causes pain and the inability to infuse tube feedings. The diagnosis can be confirmed on endoscopy, which will demonstrate the internal bolster buried within the gastric or small bowel mucosa. Often, just the distal tip of the feeding tube remains in the stomach or small intestine (Fig. 3). The treatment of buried bumper syndrome depends on the type of feeding tube.54 If the internal bolster is collapsible because it is on an externally removable PEG or DPEJ tube, the feeding tube can be removed by simple external traction. In a modification of this technique, the buried bumper PEG tube can be cut short and a guidewire passed through the shortened PEG tube into the gastric cavity.54 The guidewire then can be snared, pulled out of the oral cavity, and attached to a new PEG. As the guidewire at the abdominal surface is pulled, the new PEG is dragged into the gastric cavity, and the dilating portion of the new PEG engages the buried bumper on the old PEG. As the new PEG is pulled through the abdominal wall, the old PEG is pushed out of the abdominal wall and removed. The same procedure has been completed for DPEJ tubes.
Figure 3 Buried bumper syndrome; gastric mucosal view.
42 In contrast, if the internal bumper on the PEG or DPEJ tube is rigid, as it is on endoscopic-only removable feeding tubes, the feeding tube may have to be removed by PEG or DPEJ wound tract cut-down or the push–pull T-technique.55 The push–pull T-technique requires the PEG or DPEJ tube to be cut 3 cm from the abdominal wall. The patient is endoscoped and a snare is passed through the feeding tube opening in the gastric or small bowel wall to the outside of the patient through the feeding tube. An additional short piece of PEG or DPEJ tube is cut from the excess tubing. The snare is opened and this short piece of tubing is grasped and pulled back against the remaining PEG or DPEJ tube, creating a T-shape. A Kelly clamp is placed across the T-shape. The endoscopist slowly removes the endoscope, snare, and PEG or DPEJ tube orally as a second operator pushes the Kelly clamp and feeding tube into the gastric or small bowel lumen. This combined procedure frees the internal bumper from the gastric or small bowel wall. Once the feeding tube is removed, a new feeding tube can be placed back through the existing PEG or DPEJ tract using direct endoscopic visualization. A standard PEG or DPEJ tube placement technique should be used to permit the feeding tube dilator to re-expand the partially closed PEG or DPEJ tube tract. Prevention of the buried bumper syndrome requires good nursing care and patient instruction. As mentioned above, the external bolster of the PEG or DPEJ tube should be left 1 to 2 cm from the abdominal wall. Gauze pads should be placed over the external bolster, not underneath, which would create pressure on the PEG or DPEJ tube tract. In addition, the gastrostomy or jejunostomy tube itself should be pushed forward into the wound slightly and rotated during daily nursing care. This will ensure that the internal bumper does not become buried in the gastric or small bowel mucosa. After rotation, the feeding tube should be placed back in its original position. Colocutaneous Fistula. A colocutaneous fistula is a rare complication of PEG or DPEJ placement that occurs as a result of the interposition of colon between the anterior abdominal wall and the gastric or small bowel wall.56 The PEG or DPEJ tube is placed directly through the colon into the stomach or small intestine. Patients in whom this complication has occurred are often asymptomatic, except for transient fever or ileus, and the problem is usually discovered months after initial tube placement when the original tube is removed (because of “dysfunction”) and a replacement feeding tube is passed into position. As the replacement feeding tube is passed blindly at the bedside, it is pushed through the PEG or DPEJ tract opening and into the colon, but cannot find its way back into the stomach or small bowel. Once the tube feedings are restarted, the patient develops diarrhea from colonic tube feedings and dehydration from not receiving fluids or nutrition. This complication often can be treated by just removing the feeding tube to allow the fistula to close; however, surgery is sometimes necessary to correct the internal gastric– colon or small bowel– colon fistula.57 Prevention of this complication is related to the initial PEG or DPEJ tube procedure. Relying on the combination of transillumination and finger palpation of the abdominal wall in choosing an appropriate PEG or DPEJ tube site rather than
M.H. DeLegge and A.J. Berry using only one of these techniques alone will assure a safe feeding tube entrance site. In questionable situations, an 18or 22-gauge “probing” needle should be passed through the proposed PEG or DPEJ tube site before feeding tube placement. The needle should be advanced slowly with an attached syringe (filled with water or another fluid), creating back pressure. The presence of a sudden bolus of air within the syringe before the probing needle is endoscopically identified in the gastric or small bowel cavity suggests passage through the colon. However, this technique has not been subjected to a prospective evaluation. In questionable situations where a safe access site cannot be determined, ultrasound or CT scan guidance can assist in delineating a safe location. Inadvertent PEG or DPEJ Tube Removal. Inadvertent feeding tube removal is a common complication usually occurring in combative or confused patients who pull on the tube. Many PEG or DPEJ tubes today are designed to be removed by 10 to 14 pounds of external pull pressure. PEG or DPEJ tubes that are inadvertently removed within the first 4 weeks of feeding tube placement should not be replaced blindly at the bedside. Because the tube tract may not have matured adequately, the gastric or small bowel wall and the abdominal wall may have separated, leaving a rent in the stomach or small intestine. Blind replacement of the PEG or DPEJ tube may result in tube placement into the peritoneal cavity. If there is any concern about the possibility of a replacement feeding tube being positioned into the peritoneal cavity, a water-soluble contrast study through the feeding tube should be obtained to confirm proper tube tip position before initiation of tube feedings. If the replacement tube lies within the peritoneal cavity, it should be removed. The patient can be brought to the endoscopy suite for repeat PEG or DPEJ tube placement through the same feeding tube site.58 Patients in this situation should be treated with intravenous antibiotics and monitored for signs of peritonitis, which would require surgical intervention. Other Intraabdominal Complications. A variety of intraabdominal complications have been described in case reports. Although rare, they must be recognized by the interventional endoscopist. One complication is small bowel obstruction from a small bowel wall hematoma following PEG placement.59 In this particular case, the hematoma was on a loop of jejunum near the stomach. An operative procedure allowed evacuation of the hematoma and resolution of the small bowel obstruction. This situation could occur with DPEJ placement. Another complication is intrahepatic placement of a PEG tube.60 In this case, the originally inserted PEG tube malfunctioned and was replaced with a balloon gastrostomy tube 2.5 years after placement. The replacement tube was difficult to push back through the PEG fistula site. A contrast study showed that the balloon gastrostomy tube was inflated within the liver. Contrast from the tube entered the portal venous system. A fistula tract had developed between the liver and the stomach. Subsequent surgical exploration allowed the tube to be removed safely with resection of the gastrohepatic fistula tract; argon gas plasma coagulation of the liver bed was used to prevent bleeding, and the PEG was replaced with a
Risks of endoscopic enteral access Stamm gastrostomy. This complication could occur after DPEJ placement. Herniation of the stomach through a PEG tube fistula site was described in a patient noted to have a leaking PEG site 1 year following PEG tube insertion.61 A bulge was noted at the PEG tube site on the abdominal wall when the patient coughed. CT scanning demonstrated that a portion of the stomach had herniated through the PEG site. The PEG was removed, but the PEG fistula remained open. Surgical repair of the fistula was suggested; however, the patient died of aspiration pneumonia before definitive surgical therapy. This complication could occur after DPEJ placement. Abdominal wall pain can occur and persist after PEG or DPEJ placement. Workup should include a full examination to rule out infection of the abdominal wall. This may include a CT scan to rule out an abdominal wall abscess. In some cases, the pain will be consistent with neuropathic pain, in which case the remedy is often removal of the PEG or DPEJ and reinsertion at a different site. Abdominal wall injection with an anesthetic agent may also be helpful. Peritonitis. Leakage of contents from the stomach or the small intestine into the peritoneal cavity following PEG or DPEJ placement with the feeding tube in situ has been described, but is rare.62 Gastric or small bowel leakage can lead to peritonitis if not detected. In addition, infusion of tube feeding formula can lead to a combination of a chemical and bacterial peritonitis.63 It is hypothesized that if the introducer needle does not enter the stomach or small bowel directly through the abdominal wall during PEG or DPEJ placement, but instead approaches the stomach or small bowel tangentially, a long laceration may develop allowing the escape of gastric or small bowel contents. PEG or DPEJ Tract Tumor Seeding. Patients with proximal GI tract cancers, such as head and neck and esophageal cancers,64,65 are at risk of tumor seeding from the tumor site to the PEG or DPEJ tube tract by mechanical transfer. The feeding tube can transfer tumor cells as it is pushed or pulled by the tumor into its final position across the gastric or small bowel and abdominal wall. Some authors have suggested using an overtube during PEG placement to prevent this complication.
43
4.
5.
6.
7. 8.
9.
10.
11.
12.
13.
14. 15.
16.
17. 18. 19.
20.
21.
Conclusion Enteral access is necessary to provide enteral nutrition. The endoscopist plays an important role in this regard by placing nasoenteric tubes as well as gastrostomy and jejunostomy feeding tubes. With these enteral access techniques, there is the potential for both minor or major associated complications. The endoscopist should be aware of these complications, how to avoid them and how to treat them.
References 1. Burtch CD, Shatney CH: Feeding jejunostomy (versus gastrostomy) passes the test of time. Am Surg 53:54-57, 1987 2. Ho CS, Yee ACV, McPherson K: Complications of surgical and percutaneous non-endoscopic gastrostomy: review of 233 patients. Gastroenterology 95:206-210, 1988 3. Neumann DA, DeLegge MH: Gastric versus small bowel feeding in the
22.
23. 24.
25.
26.
27.
ICU: a prospective comparison of efficacy [abstract]. Gastroenterology 118:A774, 2000 Baskin WN, Johansen JF: An improved approach to the delivery of enteral nutrition in the intensive care unit. Gastrointest Endosc 42:161165, 1995 Fang JC, Hilden K, Holbukov R, et al: Transnasal endoscopy vs fluoroscopy for the placement of nasoenteric feeding tubes in critically ill patients. Gastrointest Endosc 62:661-666, 2005 Cataldi-Belcher EL, Selzer MH, Slocumb BA, et al: Complications during enteral nutrition therapy: a prospective study. JPEN 7:546-552, 1983 Leong SC, Mahanta V: Securing nasogastric tube in head and neck cancer patients. Laryngoscope 116:2089-2091, 2006 Whelan K, Hill L, Preedy VR, et al: Formula delivery in patients receiving enteral tube feeding on general hospital wards: the impact of nasogastric extubation and diarrhea. Nutrition 22:1025-1031, 2006 McClave SA, DeMeo MT, DeLegge MH, et al: North American Summit on Aspiration in the Critically Ill Patient: Consensus Statement. JPEN 26:S80-S85, 2002 Fay DE, Poplausky M, Gruber M, et al: Long term enteral feeding: a retrospective comparison of delivery via percutaneous endoscopic gastrostomy and nasoenteric tubes. Am J Gastroenterol 86:1604-1609, 1991 Heyland DK, Drover JW, MacDonald S, et al: Effect of postpyloric feeding on gastroesophageal regurgitation and pulmonary microaspiration: results of a randomized controlled trial. Crit Care Med 29:14951501, 2001 Powell KS, Marcuard SP, Farrior ES, et al: Aspirating gastric residuals causes occlusion of small bore feeding tubes. J Parenter Enteral Nutr 17:243-246, 1993 Scholten DJ, Wood TL, Thompson DR: Pneumothorax from nasoenteric feeding tube insertion. A report of 5 cases. Am Surg 52:381-385, 1986 McWey RE, Curry NS, Schabel SI, et al: Complications of nasoenteric feeding tubes. Am J Surg 155:253-257, 1988 George DL, Falk PS, Umberto Meduri G, et al: Nosocomial sinusitis in patients in the medical intensive care unit: a prospective epidemiological study. Clin Infect Dis 27:463-470, 1998 Van Zanten AR, Dixon JM, Nipshagen MD, et al: Hospital acquired sinusitis is a common cause of fever of unknown origin in orotracheally intubated critically ill patients. Crit Care 9:R583-R590, 2005 Desmond P, Raman R, Idikula J: Effect of nasogastric tubes on the nose and maxillary sinus. Crit Care Med 19:509-511, 1991 Baskin WN: Acute complications associated with bedside placement of feeding tubes. Nutr Clin Prac 21:40-55, 2006 Pancorbo-Hidalgo PL, Garcia-Fernandez FP, Ramirez-Perez C: Complications associated with enteral nutrition by nasogastric tube in an internal medicine unit. J Clin Nurs 10:482-490, 2001 Davies AR, Froomes PR, French CJ, et al: Randomized comparison of nasojejunal and nasogastric feeding in critically ill patients. Crit Care Med 30:586-590, 2002 Patrick PG, Marulendra S, Kirby DF, et al: Endoscopic nasogastricjejunal feeding tube placement in critically ill patients. Gastrointest Endosc 45:72-76, 1997 Russell M, Cromer M, Grant J: Complications of enteral nutrition therapy, in Gottschlich M (ed): The Science and Practice of Nutrition Support: A case based core curriculum. Dubuque, IA, American Society for Enteral and Parenteral Nutrition, 2001, pp 194-195 Gauderer MWL, Ponsky J, Izant RJ Jr: Gastrostomy without laparotomy: a percutaneous endoscopic approach. J Pediatr Surg 15:872-875, 1980 Jain NK, Larson DE, Schroeder KW, et al: Antibiotic prophylaxis for percutaneous, endoscopic gastrostomy: a prospective randomized, double blind clinical trial. Ann Intern Med 107:824-828, 1987 Hogan RB, DeMarco DC, Hamilton JK, et al: Percutaneous endoscopic gastrostomy-to push or to pull: a prospective, randomized trial. Gastrointest Endosc 32:253-258, 1986 Larson DE, Burton DD, Schroeder KW, et al: Percutaneous, endoscopic gastrostomy: Indications, success, complications and mortality in 314 consecutive patients. Gastroenterology 93:48-52, 1987 DeLegge MH, Patrick PG, Gibbs R: Percutaneous endoscopic gastrojejunostomy with a tapered tip, unweighted jejunal feeding tube: improved placement success. Am J Gastroenterol 91:1130-1134, 1996
M.H. DeLegge and A.J. Berry
44 28. Shike M, Berner YN, Gerdes H, et al: Percutaneous endoscopic gastrostomy and jejunostomy for long term feeding in patients with head and neck cancer. Otolaryngol Head Neck Surg 101:549-554, 1989 29. Fan AC, Baron TH, Rumalla A, et al: Comparison of direct percutaneous endoscopic jejunostomy and PEG with jejunal extension. Gastrointest Endosc 56:890-894, 2002 30. Delegge MH, Ginsberg G, McClave S, et al: Randomized prospective comparison of direct percutaneous endoscopic jejunostomy (DPEJ) vs. percutaneous endoscopic gastrostomy with jejunal extension (PEG-J) feeding tube placement for enteral feeding. Gastrointest Endosc 59: 158(A), 2004 31. Sharma VK, Howden CW: Meta-analysis of randomized, controlled trials of antibiotic prophylaxis before percutaneous endoscopic gastrostomy. Am J Gastroenterol 95:3133, 2000 32. Ahmad I, Mouncher A, Abdoolah A, et al: Antibiotic prophylaxis for percutaneous endoscopic gastrostomy: a prospective, randomized, double-blind trial. Aliment Pharmacol Ther 18:209, 2003 33. Chaudhary KIA, Smith OJ, Cuddy PG, et al: PEG site infections: the emergence of methicillin resistant Staphylococcus aureus as a major pathogen. Am J Gastroenterol 97:1713, 2002 34. Horiuchi A, Nakayama Y, Kajiyama M, et al: Nasopharyngeal decolonization of methicillin-resistant staphylococcus aureus can reduce PEG peristomal wound infection. Am J Gastroenterol 101:274, 2006 35. Tsang TK, Eaton D, Falconio MA: Percutaneous ostomy dilation: a technique for dilating the closed percutaneous endoscopic gastrostomy sites and inserting gastrostomies. Gastrointest Endosc 35:336-337, 1989 36. Gottfried EB, Plumser AB, Clair MR: Pneumoperitoneum following percutaneous endoscopic gastrostomy. Gastrointest Endosc 32:397, 1986 37. Stathopoulos G, Rudberg MA, Harig JM: Subcutaneous emphysema following PEG. Gastrointest Endosc 37:374, 1991 38. Dulabon GR, Abrams JE, Rutherford EJ: The incidence and significance of free air after percutaneous endoscopic gastrostomy. Am Surg 6:590593, 2002 39. Lau G, Siang HL: Fatal retroperitoneal haemorrhage: an unusual complication of percutaneous endoscopic gastrostomy. Forensic Sci Int 116:69-75, 2001 40. Ubogu EE, Zaidat OO: Rectus sheath hematoma complicating percutaneous endoscopic gastrostomy. Surg Laparosc Endosc Percutan Tech 12:430-432, 2002 41. Kazi S, Gunasekaran TS, Berman JH, et al: Gastric mucosal injuries in children from inflatable low-profile gastrostomy tubes. J Pediatr Gastroenterol Nutr 24:75, 1997 42. Metheny N, Eisenberg P, McSweeney M: Effect of feeding tube properties and three irrigants on clogging rates. Nurs Res 37:165, 1988 43. Sriram K, Jayanthi V, Lakshmi RG, et al: Prophylactic locking of enteral feeding tubes with pancreatic enzymes. JPEN J Parenter Enteral Nutr 21:353, 1997 44. Iber FL, Livak A, Patel M: Importance of fungus colonization in failure of silicone rubber percutaneous gastrostomy tubes (PEGs). Dig Dis Sci 41:226, 1996 45. Blacka J, Donoghue J, Sutherland M, et al: Dwell time and functional failure in percutaneous endoscopic gastrostomy tubes: a prospective randomized-controlled comparison between silicon polymer and poly-
46.
47.
48. 49.
50.
51. 52.
53.
54.
55.
56.
57. 58. 59. 60. 61. 62.
63.
64. 65.
urethane percutaneous endoscopic gastrostomy tubes. Aliment Pharmacol Ther 20:875, 2004 DeLegge RL, DeLegge MH: Percutaneous endoscopic gastrostomy evaluation of device materials: are we “failsafe?” Nutr Clin Pract 20:613617, 2005 Fischer LS, Bonello JC, Greenberg E: Gastrostomy tube migration and gastric outlet obstruction following percutaneous endoscopic gastrostomy. Gastrointest Endosc 33:381-382, 1987 Chong VH: Gastric outlet obstruction by gastrostomy tube balloon. Indian J Gastroenterol 23:90, 2004 Cave DR, Robinson WR, Brotsch EA: Necrotizing fasciitis following percutaneous endoscopic gastrostomy. Gastrointest Endosc 32:294, 1986 Martindale R, Witte M, Hodges G, et al: Necrotizing fasciitis as a complication of percutaneous endoscopic gastrostomy. JPEN J Parenter Enteral Nutr 11:583, 1987 Chung RS, Shertzer M: Pathogenesis of complications of percutaneous endoscopic gastrostomy. Am Surg 56:134, 1990 DeLegge MH, Lantz G, Kazacos R, et al: Effects of external bolster tension on PEG tube tract formation (abstract). Gastrointest Endosc 43:349, 1996 Klein S, Heare BR, Soloway RD: The “buried bumper syndrome”: a complication of percutaneous endoscopic gastrostomy. Am J Gastroenterol 85:448, 1990 Venu RP, Brown RD, Pastika BJ, et al: The buried bumper syndrome: a simple management approach in two patients. Gastrointest Endosc 56:582, 2002 Boyd JW, DeLegge MH, Shamburek RD, et al: The buried bumper syndrome: a new technique for safe, endoscopic PEG removal. Gastrointest Endosc 41:508, 1995 Saltzberg DM, Anand K, Juvan P, et al: Colocutaneous fistula: an unusual complication of percutaneous endoscopic gastrostomy. JPEN J Parenter Enteral Nutr 11:86, 1987 Berger SA, Zarling EJ: Colocutaneous fistula following migration of PEG tube. Gastrointest Endosc 37:86, 1991 Galat SA, Gerig KD, Porter JA, et al: Management of premature removal of percutaneous gastrostomy. Am Surg 56:733, 1990 Williams E, Sabol DA, Delegge M: Small bowel obstruction caused by bowel wall hematoma after PEG. Gastrointest Endosc 57:273, 2003 Chaer RA, Rekkas D, Trevino J, et al: Intrahepatic placement of a PEG tube. Gastrointest Endosc 57:763, 2003 Chuang CH, Chen CY: Gastric herniation through PEG site. Gastrointest Endosc 58:416, 2003 Haslam N, Hughes S, Harrison RF: Peritoneal leakage of gastric contents, a rare complication of percutaneous endoscopic gastrostomy. J Parenter Enteral Nutr 20:433-434, 1996 Larson DE, Burton DD, Schroeder KW, et al: Percutaneous endoscopic gastrostomy, indications, success, complications and mortality in 314 consecutive patients. Gastroenterology 93:48-52, 1987 Khurana V, Singh T: Percutaneous endoscopic gastrostomy site metastasis in esophageal cancer. Gastrointest Endosc 62:612, 2005 Mincheff TV: Metastatic spread to a percutaneous gastrostomy site from head and neck cancer: case report and literature review. JSLS 9:466471, 2005