The effects of TGE and transoral endoscopy on blood pressure, heart rate, and SPO2
1. Introduction
Recent developments have made ultrathin endoscopes available for routine esophagogastroduodenoscopy and also for unsedated transnasal observation (Shaker, 1994). Transnasal endoscopy is known to be less of a burden for patients (more tolerable) than transoral endoscopy and has benefits including fewer effects on respiratory and cardiovascular status and reduced recovery time after the procedure (Campo et al., 1998; Dumortier et al., 1999; Mori et al., 2008). Transnasal ultrathin endoscopy has also been applied for biliary drainage, insertion of a nasoenteral feeding tube or a long intestinal tube, and percutaneous endoscopic gastrostomy (PEG) (Fang et al., 2005; Itoi et al., 2008; Sato et al., 2008; Vitale et al., 2005). PEG has become the primary access for long term enteral feeding since its introduction in 1980 (Gauderer et al., 1980; Ponsky & Gauderer, 1981), being a very easy and rapid method compared with the previous surgical technique to place a gastrostomy tube.
The endoscopic approach from the gastrocutaneous tract was first described by Chaurasia, et al. for the insertion of a jejunal feeding tube through the PEG (PEG-J) tract (Chaurasia & Chang, 1995). Although they used a pediatric bronchoscope at that time, an ultrathin endoscope later took its place for this purpose (Adler et al., 2002). We have been employing an ultrathin endoscope through the gastrostomy tract for observation, diagnosis, and treatment of various digestive diseases from April 2003. We also investigated the usefulness of such transgastrostomal endoscopy (TGE) for management of patients who had undergone percutaneous endoscopic gastrostomy (PEG).
2. Procedure of transgastrostomal endoscopy (TGE)
Patients with a mature gastrocutaneous tract are candidates for this procedure. Patients before the maturation of the fistula (2 to 4 weeks after the PEG) have higher risk of fistula disruption. If it is necessary to perform TGE in an earlier phase or at the PEG, tight affixation of the stomach to the abdominal wall using a gastropexy device (Rogers, 1989) allows TGE to be performed.
2.1. Equipment
The endoscopes we used were GIF-XP240, GIF-XP260 and GIF-XP260N (Olympus Optical Co., Ltd., Tokyo, Japan) (Fig. 1). The diameters of these endoscopes were 7.7, 6.5, and 5.2 mm, respectively. The diameter of the working channel of these endoscopes was 2.0 mm. GIF-XP240, GIF-XP260, and GIF-XP260N allow insertion through more than 24, 20 and 18 F fistula tracts, respectively.
2.2. Routine observation by TGE
The patient is usually infused with 500 ml of water instead of liquid feeds for several hours before the endoscopic examination to remove the gastric mucus. The patient is laid in the supine position and then the gastrostomy tube is removed. After applying lidocaine jelly around the stoma, an ultrathin endoscope is inserted through the fistural tract and proceeded into the gastric lumen (Fig. 2A). The endoscope is directed to the oral side and advanced to the esophagogastric junction (Fig. 2B). The junction can be observed in detail without disturbance, free from the shade of the endoscope. The endoscope is retrogradely inserted through the esophagus, and we can observe oral cavity as well as pharyngeal cavity if necessary. The endoscope is then withdrawn to the stomach and directed to the anal side. The endoscope is easily inserted up to the upper jejunum from the stoma (Fig. 2C). The endoscope is pulled back to the stomach and turned back to see around the stoma (Fig. 2D).
After the observation of the upper gastrointestinal tract, the endoscope is withdrawn from the stoma and a new gastrostomy tube is replaced via the skin incision.
3. Diagnostic application of TGE
Since most of patients who undergo PEG are of advanced age and also have comorbid diseases, they usually do not undergo esophagogastroduodenoscopy after PEG. As TGE is a less invasive procedure to perform even for critically ill patients, we have routinely conducted TGE at the periodical replacement of gastrostomy tubes. We have found variety of newly developed lesions in the upper gastrointestinal tracts and were able to treat them at an early stage without symptoms.
3.1. Cardiopulmonary tolerance during the procedure of TGE
TGE is a less invasive examination compared to transoral and transnasal endoscopy, because TGE does not pass through the pharynx. It requires neither local anesthesia of the pharynx or nasal tract nor conscious sedation before the procedure of TGE. We monitored blood pressure, heart rate, and percutaneous oxygen saturation concentration measurement of the arterial blood (SPO2) during TGE procedures in 20 consecutive patients. The values were compared to those for patients with the same clinical background during the examination of transoral endoscopy in routine esophagogastroduodenoscopy. Table 1 shows the comparison of cardiopulmonary effects during the procedures of TGE and transoral endoscopy. None of the TGE examinations required sedation, while the transoral examinations required 1 to 4 mg of intravenous administration of midazolam. There were no significant differences in the fluctuation of blood pressure and heart rate between TGE and transoral endoscopy. TGE displayed no effects on SPO2 during the procedure, while transoral endoscopy caused a decrease of SPO2 in 46% of the patients.
Transgastrostomal endoscopy (n=20) | Transoral endoscopy (n=13) | P value | |
Fluctuation of blood pressure (more than 20%) | 5% (1/20) | 23% (3/13) | P = 0.157 |
Fluctuation of heart rate (more than 20%) | 15% (3/20) | 15% (2/13) | P "/ 0.5 |
Decrease of SPO2
(more than 3%) |
0% (0/20) | 46% (6/13) | P < 0.01 |
3.2. Newly developed lesions in upper gastrointestinal tract after PEG
We have been conducting TGE for screening the upper gastrointestinal tract since April 2003. TGE was carried out 805 times in 245 PEG patients (Table 2) during routine replacement of gastrostomy tubes until November 2010. The demographic data are shown in Table 2. The mean age of the subjects was 80.8 years old, and the most frequent underlying disease was cerebrovascular disease. The endoscopic findings of TGE were compared to those at PEG placement. There were a variety of newly developed lesions after PEG.
Age (mean + SD) | 80.8 + 8.9 |
Gender (male : female) | 71 : 174 |
Underlying disease | |
Cerebrovascular disease Dementia Neuromuscular disease Pneumonia Malignant tumor Others |
143 |
42 23 20 6 11 |
Reflux esophagitis was found in 35 patients (14.2%), which was the most frequent lesion after PEG. Among the 35 patients, reflux esophagitis was absent at the PEG tube placement in the 30 patients. There were 5 severe cases with symptoms of hematemesis and this had subsequently led to stricture of the lower esophagus in 3 patients (Fig. 3A). Although in these 3 patients, we were unable to insert a transoral endoscope into the stomach, we were able to conduct endoscopic observation of the stomach by TGE (Fig. 3B).
The next most frequent lesion after PEG was polypoid lesion around the stoma. We classified polypoid lesions into 3 types according to their location e.g.; basal bumper type, peri-bumper type and opposite-bumper type. The basal-bumper type polyp was communicated with the fistural tract and the pathology was inflammatory granulation (Fig. 4A). The peri-bumper type polyp was located around the stoma and often observed in cases with decompression by a mushroom type gastrostomy tube (Fig. 4B). Pathological examination revealed hyperplasia of the gastric mucosa. The opposite-bumper type polyp was located on the contra lateral wall of the stoma, probably caused by the contact of the top of the bumper or tube (Fig. 4C). The pathology was hyperplasia. Most of the polypoid lesion was not significantly changed in the subsequent observation by TGE. One case with haemorrhage from the basal-bumper type granuloma was resected by polypectomy using TGE.
Gastric ulcer was observed in 6 patients. All of their lesions were accidentally found at TGE, because they did not complain of symptoms due to their comorbid diseases. Among the 6 patients, the lesions of 3 patients were supposed to have been caused by direct contact with the bumper. These ulcers were characterized by their location on the contralateral side of the bumper (Fig. 5A). The remaining 3 patients were sporadic cases (Fig. 5B). We have also experienced additional 2 cases of symptomatic gastric ulcer which were detected by transoral endoscopy. For their initial symptoms of hematemesis and tarry stool, emergent esophagogastroduodenoscopy was performed (Nishiwaki et al., 2010b). Hemostasis was obtained by endoscopic hemoclips and coagulation, respectively. TGE was not available for such therapy, because the working channel of the ultrathin endoscope was too narrow to insert the devices to achieve hemostasis.
Neoplastic lesions were also detected by TGE. We have experienced 3 cases of early gastric cancer and one case of carcinoid tumor that developed after PEG. One case was recurrence of early gastric cancer after endoscopic mucosal resection before the PEG. One case of early gastric cancer underwent coagulation therapy with argon plasma (Fig. 6A). The remaining two cancer (Fig. 6B) and one carcinoid tumor lesions were resected by endoscopic submucosal dissection assisted by TGE, the technique of which is described in section 4.2.1. The primary duodenal cancer was accidentally found by TGE at the first endoscopic examination 5 years after PEG.
Candida esophagitis was observed in 4 patients. All of the patients were administered proton pump inhibitor, which might have been involved in the development of the esophagitis.
Table 3 summarizes newly developed lesions after PEG. TGE was useful for the early detection and follow up of these lesions.
Pathological conditions | Number of patients |
Reflux esophagitis | 30 |
Polypoid lesion around the stoma | 28 |
Gastric ulcer | 6 |
Neoplastic lesions | |
Early gastric cancer | 3 |
Gastric carcinoid tumor | 1 |
Primary duodenal cancer | 1 |
Candida esophagitis | 4 |
Total | 73 |
4. Therapeutic application of TGE
TGE also offers several advantages for endoscopic treatments compared with a transoral endoscope. TGE can easily access up to the upper jejunum with less patient discomfort. TGE covers treatments for lesions from the upper esophagus to upper jejunum. Table 4 shows the list of TGE we have conducted for therapeutic purposes..
No. of trial | No. of success | |
Establishement of direct jejunal feeding access | ||
Placement of jejunal feeding thorough the PEG (PEG-J) | 27 | 27 |
Direct percutaneous endoscopic jejunostomy (DPEJ) | 36 | 33 |
Resection of upper gastrointestinal tumors | ||
TGE assisted endoscopic submucosal dissection (ESD) | 5 | 5 |
Endoscopic mucosal resection | 1 | 1 |
Polypectomy | 1 | 1 |
Palliative therapy for gastrointestinal obstruction | ||
Transgastrostomal gastroduodenal stenting | 5 | 4 |
Placement of long intestinal decompression tube | 2 | 2 |
Other usage for endoscopic treatments | ||
Recovery of disrupted gastrocutaneous tract | 5 | 5 |
Transgastrostomic biliary drainage | 1 | 1 |
4.1. Establishment of direct jejunal feeding access
Since the introduction of endoscopic placement of a gastrostomy tube, PEG has been the primary choice for long-term enteral access. However, there are some situations which postpyloric or jejunal feeding is preferable to gastric feeding. The indications for jejunal feeding include a high risk of aspiration, previous gastric resection, gastric pull-up, gastric outlet obstruction, an obstructed or nonfunctioning gastrojejunostomy, and gastric dysmotility (Shike & Latkany, 1998). The access for jejunal feeding was initially established as two methods, i.e.; transnasal insertion of a feeding tube and surgical placement of a jejunostomy tube. Alternative accesses methods for jejunal feeding have been developed; placement of a jejunal feeding tube through the PEG (PEG-J) and direct percutaneous endoscopic jejunostomy (DPEJ). We utilized TGE for the establishment of the two access methods.
4.1.1. Placement of a jejunal feeding tube through the PEG (PEG-J)
PEG-J was first described in 1984 (Ponsky & Aszodi, 1984). This technique was usually conducted for patients with high risk of aspiration due to gastroesophageal reflux of food after PEG (Nishiwaki et al., 2006). The tube is usually placed under fluoroscopic guidance. However, it is sometimes difficult to pass the feeding tube through the pyloric ring and duodenum using only fluoroscopy. A technique to assist the tube passage into the duodenum by transoral endoscopy has been reported (Duckworth et al., 1994; Leichus L et al., 1997). A transgastric technique using a pediatric bronchoscope or an ultrathin endoscope was described first in 1995 (Adler et al., 2002; Chaurasia & Chang, 1995).
We treated 27 patients using this technique with the following procedures: 1. Remove the gastrostomy tube. 2. Insert an endoscope through the gastrostomy and proceed up to the jejunum (Fig. 7A). 3. Insert a guidewire through the working channel of the endoscope. 4. Remove the endoscope retaining the guidewire (Fig. 7B). 5. Place a jejunal feeding tube over the guidewire through the PEG under fluoroscopic guidance (Figs. 7 C& D).
The jejunal feeding tubes (J-tube, Boston Scientific, Natick, MA, USA, or Jejunal Feeding Tube, Bard Access Systems, Inc., Salt Lake City, UT, USA) were successfully placed in all 27 patients. The duration of the procedure ranged from 4 to 20 minutes, which was remarkably quick compared with the procedure using only fluoroscopic guidance.
4.1.2. Direct percutaneous endoscopic jejunostomy (DPEJ)
Direct percutaneous placement of the jejunostomy tube was first described in 1987 (Shike et al., 1987). The first case series were for patients with previous gastric resection, and the technique was subsequently applied for patients with intact stomach. Becuase the length of the endoscope for the performance of DPEJ in the transoral route is 160 cm or longer, it requires the use of a long colonoscope or a small intestinal scope (Shike et al., 1991). However, the transgastrostomal approach facilitates access to the upper jejunum without difficulty. DPEJ using TGE was first described in 2002 (Baron, 2002). Although they successfully placed a jejunostomy tube through the mature gastrostomy tract, the gastrocutaneous tract was disrupted by passing through the bumper of the jejunostomy tube. We modified the procedure (Fig. 8) so that the jejunostomy tube does not pass the gastrostomy tract (Nishiwaki et al., 2009b).
After conscious sedation with intravenous administration of midazolam (1 to 2 mg) in most patients, an ultrathin endoscope was inserted through the gastrocutaneous tract immediately after the removal of the gastrostomy tube. The endoscope was inserted up to the jejunum and the site for placement was determined by transillumination to the abdominal wall and finger pressure against the jejunum (Fig. 8A). The abdominal skin surface of the placement area was cleansed with povidone-iodine, and local anesthesia was done administering 1% lidocaine. After test-puncturing with a 21-gauge needle, one or two point sutures were made using gastropexy devices (Create Medic Co., Ltd., Yokohama, Japan, or Easy Tie, Boston Scientific Japan K.K., Tokyo, Japan) to fix the jejunum against the abdominal wall. A 16- or 18-gauge Seldinger needle was inserted through the abdomen toward an open snare inserted through the TGE and a loop wire was passed through the outer sheath of the needle. The loop wire was grasped by the snare and pulled out through the gastrocutaneous tract with the endoscope (Fig. 8B). In most cases, the loop wire was then grasped in the stomach by biopsy forceps of an orally inserted endoscope (Fig. 8C) and pulled out via the mouth (Fig. 8D). In a few cases, the loop wire was directly brought out from the mouth by transgastrostomic endoscopy. Next, an jejunostomy tube was connected to the loop wire and the tube was placed in the jejunum by the pull-through technique (Fig. 8E). The systems to place the jejunostomy tube were One Step Button (Boston Scientific Co., Natick, MA, USA) or Bard Ponsky Pull PEG (Bard Access Systems, Inc., Salt Lake City, UT, USA)
Thirty six DPEJ procedures were attempted in 35 patients, with 33 (92%) successful placements. One unsuccessful placement was due to jejunum migration away from the abdominal wall during the puncture; this patient was maintained by feeding via the PEG-J. The other two failures were due to the lack of transillumination. The average + SD of duration of procedure, from insertion of the endoscope to the placement of the jejunostomy tube, was 29.6 + 13.1 (15 to 70) min.
Procedure-related complications consisted of 1 case each of pneumonia, colocutaneous fistula, and pneumoperitoneum. Of these, pneumonia and colocutaneous fistula were considered major complications. In the former case, placement of the jejunostomy tube took 50 min, and pneumonia occurred before administration of nutrients. The patient died 4 days after DPEJ due to respiratory failure. In the latter case, the fact that the jejunostomy tube had been placed penetrating through the colon was revealed at the second replacement of the jejunostomy tube. The colocutaneous fistula was naturally closed after removal of the replaced tube and DPEJ was successfully completed 28 days after the removal. Pneumoperitoneum, a minor complication, was observed in a severely obese patient.
4.2. Resection of upper gastrointestinal tumors
Endoscopic techniques to remove gastrointestinal tumors have been established as snare polypectomy and endoscopic mucosal resection (EMR). EMR allows the removal of flat lesions by injection with saline into the submucosal layer, but it is difficult to remove larger tumors. Recent development of the technique of endoscopic submucosal dissection (ESD) facilitates the removal of widely-spread neoplastic lesions compared to the EMR. We utilized TGE for these resection techniques of the upper gastrointestinal tumors.
4.2.1. TGE assisted endoscopic submucosal dissection (ESD)
ESD is a new technique for the resection of early gastric cancer (Ono et al., 2001). This technique allows en bloc resection even for larger lesions and allows for precise histological analyses. However, this method has a high incidence of complications such as bleeding and perforation, and the technique requires skill and a longer time for resection compared to conventional EMR (Gotoda 2006). One of the technical problems of ESD is the difficulty of maintaining a clear view of the submucosal layer of the gastric wall during submucosal dissection. We developed the technique of ESD assisted by TGE to dissolve the problem (Nishiwaki et al., 2009a).
Five patients previously established with PEG underwent TGE-assisted ESD (Table 5). ESD was carried out by a single-channel videoendoscope, GIF H260 (Olympus Optical Co., Ltd, Tokyo), attached with a short hood. The high-frequency electrosurgical unit was the Erbotom ICC 200 in Cases 1-3, and the VIO 300D in Cases 4 and 5 (ERBE Elektromedizin GmbH, Tubingen, Germany). After circumferential markings around the lesion, submucosal injection was carried out using a mixture of hyaluronic acid and glycerol containing indigocarmine, according to the method described by Fujishiro, et al (Fujishiro et al., 2006). Circumferential mucosal cutting and submucosal dissection were carried out using a needle knife, a Flex knife, an insulation-tipped (IT) knife, or a combination of the three. TGE was inserted through the mature gastrocutaneous tract and grasped the edge of the resecting specimen to achieve appropriate counter-traction during submucosal dissection (Figs. 9A-C). Hemostatic forceps were used when active bleeding took place during the procedure or to prevent bleeding from large vessels before cutting in soft coagulation mode. TGE also worked for additional submucosal injection, washing with water, aspiration of gastric fluid.
Case I.D. | 1 | 2 | 3 | 4 | 5 |
Age/Gender | 83/M | 82/F | 72/M | 79/F | 82/F |
Pathology | tub 1 | tub 1 | adenoma | tub 1 | carcinoid tumor |
Location | a.w. antrum | l.c. body | p.w. antrum | g.c. body | a.w. body |
Dissection size (mm) | 24 X 22 | 20 X 20 | 60 X 30 | 30 X 22 | 15 X 10 |
Operative duration (min) | 21 | 18 | 105 | 55 | 80 |
A laparoscopic port with a trocar is inserted into the gastric lumen percutaneously and assists in the ESD using an IT knife (Kondo et al., 2004). However these methods are thought to be limited for some tumor locations, as the assisting devices are not flexible. EMR assisted by a magnetic anchor has also been reported to create strong counter traction (Kobayashi et al., 2004). In our methods, an ultrathin endoscope is inserted through the gastrocutaneous tract after PEG, and assisted ESD is performed with an orally inserted videoendoscope. TGE can cover the whole stomach and esophagus and create appropriate counter traction for dissection of the lesion. Furthermore, TGE not only can provide traction control but can also support diverse procedures of ESD, such as marking, submucosal injection, washing out or aspiration of the intragastric contents. TGE facilitates ESD by this type of assistance, and the operation duration of TGE-assisted ESD has been found to be relatively short compared to the single-scope ESD in our hospital.
One of the problems of this method is light interference between the two endoscopes (Fig. 9D). When the two endoscopes faced each other, the position of the TGE had to be changed to eliminate the light interference. The other problem of this method is that it might be difficult to perform TGE before maturation of the gastrocutaneous tract. In the present case series, gastric tumors were found after PEG in three cases, but two tumors were found at the placement of PEG catheter. In general, it takes 3-4 weeks to complete the stoma. ESD was conducted in the two cases 3-4 weeks after PEG, after waiting for maturation of the gastrocutaneous tract. Pneumoperitoneum occurred after ESD in Case 3, although it recovered spontaneously. The use of gastropexy devices enabled us to fix the abdominal and gastric walls more tightly, such that pneumoperitoneum might be prevented even with prolonged endoscopic treatment. Narrowing of the ultrathin endoscope would be helpful for this method. Therefore, simultaneous PEG and TGE-assisted ESD using a gastropexy device and an ultrathin endoscope might be possible for large, difficult lesions.
4.2.2. Endoscopic mucosal resection (EMR) and polypectomy
The performance of ESD by TGE alone is quite difficult because the working channel of the ultrathin endoscope is too narrow to insert the devices for ESD. However, TGE itself can be used for polypectomy and EMR by using an appropriate injection needle and snare for the ultrathin endoscope. We carried out resection of two benign polyps by TGE. One case was a hyperplastic polyp on the prepylorus, and the other was inflammatory granuloma on the gastric stoma (Fig. 4A). Both polyps showed a tendency to develop during the periodical observation by TGE. Transgastric resections were successfully conducted without conscious sedation.
4.3. Palliative therapy for gastrointestinal obstruction using TGE
One of the indications of PEG is decompression of gastrointestinal content in cases of obstruction of gastrointestinal tract. On the other hand, some cases develop gastrointestinal obstruction after PEG. TGE also facilitates the palliative therapy in such cases.
4.3.1. Transgastrostomal stenting on gastroduodenal outlet obstruction
Endoscopic deployment of stents for malignant gastroduodenal outlet obstruction has been reported as a less invasive palliative therapy compared to a surgical bypass operation (Yim et al., 2001). It is more difficult to place the stent across the postpyloric lesion compared to the esophagus, because the delivery system often forms a loop in the stomach before passing through the obstruction. When the gastrostomy site is adjacent to the obstructed lesion, a transgastric approach for stenting is easier than a transoral approach. Transgastrostomic stenting for malignant duodenal obstruction was first described in 1993 (Keymling et al., 1993).
The schematic procedure of transgastrostomal stenting is shown in Fig. 10. An ultrathin endoscope is inserted through the mature gastrocutaneous tract followed by removal of the gastrostomy tube. A guidewire is advanced through the stenosis (Fig. 10A). The endoscope is once removed from the gastrostomy keeping the position of the guidewire. A balloon catheter is inserted over the guidewire and the stenosis is dilated with the balloon catheter (Fig. 10B). The ultrathin endoscope is inserted after the removal of the balloon catheter and advanced to the upper jejunum (Fig. 10C). A guidewire is inserted deeply in the jejunum through the endoscope and the endoscope is removed again, retaining the guidewire. A stent is finally deployed over the guidewire across the obstructed lesion using a delivery device (Figs. 10D& E).
We tried this procedure for 5 patients in whom we had failed to place a stent by the transoral approach (Table 6). In 4 out of 5 patients, a non-covered expandable metallic stent was successfully deployed (Ultraflex, Boston Scientific Co., Natick, MA, USA). The one case (#5) of failure was changed to the placement of a jejunal feeding tube through the obstruction. The duodenal obstruction of this case was too severe to pass the delivery device but a 12F jejunal feeding tube was successfully inserted through the obstruction. All of the patients could successfully feed again after the procedure.
Case I.D. | 1 | 2 | 3 | 4 | 5 |
Age/Gender | 88/F | 82/F | 84/M | 75/M | 72/F |
Tumors | Stomach | CBD | Stomach | CBD | Gall bladder |
Obstruction | Pylorus | Duodenum | Pylorus | Duodenum | Duodenum |
Operative duration (min) | 20 | 27 | 15 | 45 | |
Outcome | Success | Success | Success | Success | Failure |
4.3.2. Placement of long intestinal decompression tube
Placement of a long intestinal decompression tube was carried out for malignant lower intestinal obstructions. A series of radiographs of the procedure in the case with ascending colon cancer is shown in Fig. 11. The procedure was the same as for the placement of a PEG-J tube as described in 4.1.1. The benefit of this procedure is that it is more comfortable for the patient, especially in the prolonged period of decompression, compared to transnasal placement.
4.4. Other usage for endoscopic treatments
We utilized TGE to treat other pathological conditions. The applications for pancreaticobiliary and esophageal diseases have been reported. Alternative possibilities for the usage of TGE are discussed in this section.
4.4.1. Recovery of disrupted gastrocutaneous tract
Fistula disruption usually occurs upon early dislodgement of the gastrostomy tube or upon replacement of the tube. It might cause the serious complication of peritonitis due to leaking of gastric contents or feeding materials into the peritoneal cavity (Fang, 2007; McQuaid & Little, 1992; Romero et al., 1996). In order to place a new gastrostomy tube, a PEG procedure is often performed adjacent to the disrupted fistula (Galat et al., 1990). We employed TGE for recanalization of the disrupted fistula at the tube replacement (Nishiwaki et al., 2009 c). Schematic illustrations of the procedure are shown in Fig. 12.
We tried this technique for 5 patients with fistula disruption and in all of the patients, the disrupted fistula tract was successfully recovered. Before the introduction of this procedure, we inserted a guidewire to search for the gastric lumen from the skin incision under fluoroscopic guidance. However, in 2 out of 5 trials, we failed to reestablish a gastrocutaneous tract. The endoscopic view of TGE is helpful to locate the intragastric route from the skin incision.
4.4.2. Transgastrostomic biliary drainage
Endoscopic retrograde cholangiopancreatography from the gastrostomy have been reported (Baron & Vickers, 1998; Mori et al., 2007). Since all ultrathin endoscopes are forward-viewing, it is more difficult to insert a canula into a pancreatic or bile duct. Moreover, endoscopic treatment has limitations because of the limited availability of therapeutic devices. However, there is a benefit of direct access into the bile duct after sphincterotomy and the ability to conduct direct endoscopic observation of the bile tract (Larghi & Waxman, 2006).
We only experienced one case with transgastrostomal biliary drainage for obstructive jaundice in a patient with primary sclerosing cholangitis (Fig. 13). Placement of a transgastrostomal drain tube is more comfortable than that of a transnasal drain tube.
4.4.3. Transstomal endoscopy other than gastrostomy
Transstomal endoscopy is also available via other routes than the gastrostomy tract. We experienced two cases of endoscopic placement of a jejunal extension tube through direct endoscopic jejunostomy. The first case had undergone DPEJ and showed repeated aspiration due to jejunoesophageal reflux of the feeding materials. We inserted the top of a feeding tube to 40 cm anal direction from the jejunostomy site using a transjejunostomal endoscope (Fig. 14). Another case was jejunal obstruction caused by inflammatory granuloma of the jejunostomy site (Nishiwaki et al., 2007). We could successfully place a feeding tube on the anal side of the obstruction by transjejunostomal endoscope. Although a similar procedure has previously reported under fluoroscopic guidance (Luttman et al., 2005), it might be easier to place a feeding tube using the transjejunostomal endoscope.
We additionally experienced one case of transcecostomal endoscopy for stent placement across the ascending colon cancer (Nishiwaki et al., 2010b). The patient was too critically ill to undergo surgical treatment, and we were also unable to place a colonic stent by colonoscopy. After establishment of ultrasonography-guided percutaneous cecostomy, transcecostomal endoscopy facilitated the passing of a guidewire thorough the obstruction. The procedure of the stent deployment is shown in Fig. 15.
4.4.4. Other applications
TGE was reported to be useful for the dilatation of radiation-induced complete esophageal obstruction (Maple et al., 2006). The rendezvous technique for the dilatation of esophageal stricture was conducted by cooperative use of transoral endoscope and TGE. The use of TGE made the procedure more safe and effective.
Recent developments in super ultrathin endoscopes allow for insertion through the gastrostomy tube. The endoscope is then utilized for verification of a replaced gastrostomy tube into the stomach.
5. Conclusions
TGE offers a less invasive way to observe the upper gastrointestinal tract and allows for the detection of early stage gastrointestinal lesions. TGE also facilitates endoscopic treatments which might be more difficult by transoral endoscope. However, the narrow working channel of the ultrathin endoscope is a limitation of therapeutic TGE. Further improvements to the endoscope and applicable devices will be expected in the future.
6. Acknowledgements
We express our appreciation to Dr. Saito, Nishimino Kosei Hospital, for generous advice on these procedures. We also appreciate the members of the endoscopy unit of Nishimino Kosei Hospital (Dr. Maeda, Dr. Hayashi, Dr. Tagami, and Dr. Takada) for their assistance with the endoscopic procedures.
References
- 1.
Adler D. G. Gostout C. J. Baron T. H. 2002 Percutaneous transgastric placement of jejunal feeding tubes with an ultrathin endoscope.55 1 (Jan 2002),106 110 ,0016-5107 - 2.
Baron T. H. Vickers S. M. 1998 Surgical gastrostomy placement as access for diagnostic and therapeutic ERCP.48 6 (Dec 1998),640 641 ,0016-5107 - 3.
Baron T. H. 2002 Direct percutaneous endoscopic jejunostomy through a mature gastrostomy tract.56 6 (Dec 2002),946 947 ,0016-5107 - 4.
Campo R. Montserrat A. Brullet E. 1998 Transnasal gastroscopy compared to conventional gastroscopy: a randomized study of feasibility, safety, and tolerance.30 5 (Jun 1998),448 452 ,0001-3726 X - 5.
Chaurasia O. P. Chang K. J. 1995 A novel technique for percutaneous endoscopic gastrojejunostomy tube placement.42 2 (Aug 1995),165 168 ,0016-5107 - 6.
Duckworth P. F. Jr Kirby D. F. Mc Henry L. De Legge M. H. Foxx-Orenstein A. 1994 Percutaneous endoscopic gastrojejunostomy made easy: a new over-the-wire technique.40 3 (May-Jun 1994),350 353 ,0016-5107 - 7.
Dumortier J. Ponchon T. Scoazec J. Y. Moulinier B. Zarka F. Paliard P. Lambert R. 1999 Prospective evaluation of transnasal esophagogastroduodenoscopy: feasibility and study on performance and tolerance.49 3 (Mar 1999),285 291 ,0016-5107 - 8.
Fang J. C. Hilden K. Holubkov R. Di Sario J. A. 2005 Transnasal endoscopy vs. fluoroscopy for the placement of nasoenteric feeding tubes in critically ill patients.62 5 (Nov 2005),661 666 ,0016-5107 - 9.
Fang J. C. 2007 Minimizing endoscopic complications in enteral access.17 1 (Jan 2007),179 196 ,1052-5157 - 10.
Fujishiro M. Yahagi N. Nakamura M. Kakushima N. Kodashima S. Ono S. Kobayashi K. Hashimoto T. Yamamichi N. Tateishi A. Shimizu Y. Oka M. Ogura K. Kawabe T. Ichinose M. Omata M. 2006 Successful outcomes of a novel endoscopic treatment for GI tumors: endoscopic submucosal dissection with a mixture of high-molecular-weight hyaluronic acid, glycerin, and sugar.63 2 (Feb 2006),243 249 ,0016-5107 - 11.
Galat S. A. Gerig K. D. Porter J. A. Slezak F. A. 1990 Management of premature removal of the percutaneous gastrostomy.56 11 Nov 1990)733 736 ,0003-1348 - 12.
Gauderer M. W. Ponsky J. L. Izant R. J. Jr 1980 Gastrostomy without laparotomy: a percutaneous endoscopic technique.15 6 (Dec 1980)872 875 ,0022-3468 - 13.
Gotoda T. 2006 Endoscopic resection of early gastric cancer: the Japanese perspective.22 5 (Sep 2006)561 569 ,0267-1379 - 14.
Itoi T. Kawai T. Itokawa F. Sofuni A. Kurihara T. Tsuchiya T. Ishii K. Tsuji S. Ikeuchi N. Takeuchi M. Sakai Y. Moriyasu F. 2008 Initial experience of transnasal endoscopic biliary drainage without conscious sedation for the treatment of acute cholangitis (with video).67 2 (Feb 2008),328 332 ,0016-5107 - 15.
Keymling M. Wagner H. J. Vakil N. Knyrim K. 1993 Relief of malignant duodenal obstruction by percutaneous insertion of a metal stent.39 3 (May-Jun 1993),439 441 ,0016-5107 - 16.
Kobayashi T. Gotohda T. Tamakawa K. Ueda H. Kakizoe T. 2004 Magnetic anchor for more effective endoscopic mucosal resection.34 3 (Mar 2004)118 123 ,0368-2811 - 17.
Kondo H. Gotoda T. Ono H. Oda I. Kozu T. Fujishiro M. Saito D. Yoshida S. 2004 Percutaneous traction-assisted EMR by using an insulation-tipped electrosurgical knife for early stage gastric cancer.59 2 (Feb 2004),284 288 ,0016-5107 - 18.
Larghi A. Waxman I. 2006 Endoscopic direct cholangioscopy by using an ultra-slim upper endoscope: a feasibility study.63 6 (May 2006),853 857 ,0016-5107 - 19.
Leichus L. Patel R. Johlin F. 1997 Percutaneous endoscopic gastrostomy/jejunostomy (PEG/PEJ) tube placement: a novel approach.45 1 (Jan 1997),79 81 ,0016-5107 - 20.
Luttmann A. Deppe H. Wejda B. U. Huchzermeyer H. Dormann A. J. 2005 Placement of a jejunal enteral tube through a percutaneous endoscopic jejunostomy to prevent recurrent aspiration during intestinal feeding.61 3 (Mar 2005),492 493 ,0016-5107 - 21.
Maple J. T. Petersen B. T. Baron T. H. Kasperbauer J. L. Wong Kee. Song L. 2006 Endoscopic management of radiation-induced complete upper esophageal obstruction with an antegrade-retrograde rendezvous technique.64 5 (Nov 2006),822 828 ,0016-5107 - 22.
Mc Quaid K. R. Little T. E. 1992 Two fatal complications related to gastrostomy "button" placement.38 5 (Sep-Oct 1992),601 603 ,0016-5107 - 23.
Mori A. Ohashi N. Maruyama T. Tatebe H. Sakai K. Inoue H. Takegoshi S. Okuno M. 2007 Endoscopic retrograde cholangiopancreatography through a gastric stoma using an ultrathin endoscope: a novel approach.39 Suppl.1, (Feb 2007), E. 323,0001-3726 X - 24.
Mori A. Ohashi N. Maruyama T. Tatebe H. Sakai K. Shibuya T. Inoue H. Takegoshi S. Okuno M. 2008 Transnasal endoscopic retrograde cholangiopancreatography using an ultrathin endoscope: a prospective comparison with a routine oral procedure.14 10 (Mar 2008),1514 1520 ,1007-9327 - 25.
Nishiwaki S. Araki H. Goto N. Niwa Y. Kubota M. Iwashita M. Onogi N. Hatakeyama H. Hayashi T. Maeda T. Saitoh K. 2006 Clinical analysis of gastroesophageal reflux after PEG.64 6 (Dec 2006),890 896 ,0016-5107 - 26.
Nishiwaki S. Shirakami Y. Hayashi T. Saitoh K. 2007 Jejunal inflammatory granuloma: a complication of direct percutaneous endoscopic jejunostomy.39 Suppl.1, (Feb 2007), E. 116,0001-3726 X - 27.
Nishiwaki S. Araki H. Shirakami Y. Niwa Y. Iwashita M. Hatakeyama H. Saitoh K. 2009a Transgastrostomic endoscopy-assisted endoscopic submucosal dissection.41 Suppl.2, (Feb 2009), E. 13,0001-3726 X - 28.
Nishiwaki S. Araki H. Shirakami Y. Kawaguchi J. Asano T. Iwashita M. Tagami A. Hatakeyama H. Hayashi T. Maeda T. Naganawa S. Saitoh K. 2009b Direct percutaneous endoscopic jejunostomy using a transgastrostomic endoscope in patients with previous endoscopic gastrostomy.41 Suppl.2, (Mar 2009), E. 36-37,0001-3726 X - 29.
Nishiwaki S. Araki H. Takada J. Watanabe N. Asano T. Iwashita M. Tagami A. Hatakeyama H. Hayashi T. Maeda T. Saitoh K. 2009c Usefulness of an ultrathin endoscope in recanalization of a disrupted fistulous tract following percutaneous endoscopic gastrostomy or jejunostomy.41 Suppl.2, (Oct 2009), E. 273-274,0001-3726 X - 30.
Nishiwaki S. Hatakeyama H. Takada J. Watanabe N. Iwashiwa M. Araki H. Saito K. 2010a Transcecostomal colonic stent placement after US-guided percutaneous cecostomy.71 6 (May 2010),1097 1099 ,0016-5107 - 31.
Nishiwaki S. Araki H. Takada J. Watanabe N. Asano T. Iwashita M. Tagami A. Hatakeyama H. Hayashi T. Maeda T. Saito K. 2010b Clinical investigation of upper gastrointestinal hemorrhage after percutaneous endoscopic gastrostomy.22 3 (Jul 2010),180 185 ,0915-5635 - 32.
Ono H. Kondo H. Gotoda T. Shirao K. Yamaguchi H. Saito D. Hosokawa K. Shimoda T. Yoshida S. 2001 Endoscopic mucosal resection for treatment of early gastric cancer.48 2 (Feb 2001),225 229 ,0017-5749 - 33.
Ponsky J. L. Gauderer M. W. 1981 Percutaneous endoscopic gastrostomy: a nonoperative technique for feeding gastrostomy.27 1 (Feb 1981),9 11 ,0016-5107 - 34.
Ponsky J. L. Aszodi A. 1984 Percutaneous endoscopic jejunostomy.79 2 (Feb 1984),113 116 ,0002-9270 - 35.
Rogers B. H. Kaminski M. V. Jr All J. 1989 Stabilizing sutures for percutaneous endoscopic gastrostomy.35 3 (May-Jun 1989),241 243 ,0016-5107 - 36.
Romero R. Martinez F. L. Robinson S. Y. Sullivan K. M. Hart M. H. 1996 Complicated PEG-to-skin level gastrostomy conversions: analysis of risk factors for tract disruption.44 3 (Sep 1996),230 234 ,0016-5107 - 37.
Sato R. Watari J. Tanabe H. Fujiya M. Ueno N. Konno Y. Ishikawa C. Ito T. Moriichi K. Okamoto K. Maemoto A. Chisaka K. Kitano Y. Matsumoto K. Ashida T. Kono T. Kohgo Y. 2008 Transnasal ultrathin endoscopy for placement of a long intestinal tube in patients with intestinal obstruction.67 6 (May 2008),953 957 ,0016-5107 - 38.
Shaker R. 1994 Unsedated trans-nasal pharyngoesophagogastroduodenoscopy (T-EGD): technique.40 3 (May-Jun 1994),346 348 ,0016-5107 - 39.
Shike M. Schroy P. Ritchie M. A. Lightdale C. J. Morse R. 1987 Percutaneous endoscopic jejunostomy in cancer patients with previous gastric resection.33 5 (Oct 1987),372 374 ,0016-5107 - 40.
Shike M. Wallach C. Likier H. 37 37 1 (Jan-Feb 1991),62 65 ,0016-5107 - 41.
Shike M. Latkany L. 1998 8 3 (July 1998),569 580 ,1052-5157 - 42.
Vitale M. A. Villotti G. D’Alba L. De Cesare M. A. Frontespezi S. Iacopini G. 2005 Unsedated transnasal percutaneous endoscopic gastrostomy placement in selected patients.37 1 (Jan 2005),48 51 ,0001-3726 X - 43.
Yim H. B. Jacobson B. C. Saltzman J. R. Johannes R. S. Bounds B. C. Lee J. H. Shields S. J. Ruymann F. W. Van Dam J. Carr-Locke D. L. 2001 Clinical outcome of the use of enteral stents for palliation of patients with malignant upper GI obstruction.53 3 (Mar 2001),329 332 ,0016-5107