| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 Departments of Interventional Radiology (T.d.B., R.C., V.K., P.C., J.P.D., A.R.)
2 Head and Neck Surgery (C.D., G.S.), Institut Gustave Roussy, 39 rue Camille Desmoulins, 94805 Villejuif, France.
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Five hundred cancer patients were referred for fluoroscopically guided gastrostomy, among whom percutaneous endoscopic gastrostomy was contraindicated or had been unsuccessful in approximately one-fourth. Five hundred eight fluoroscopically guided gastrostomies with T-fastener gastropexy were performed in 496 patients. The procedure was unsuccessful in four patients, and 12 patients needed a second gastrostomy.
RESULTS: Fluoroscopically guided gastrostomy was feasible in 99% of patients. During the first 30 postprocedure days, there were seven major complications (1.4%): cardiac failure (n = 1), hemorrhage (n = 1), and peritonitis (n = 5); one patient died of peritonitis. No major complications occurred after the 30th postprocedure day. There were 27 minor complications (5.4%) during the first 30 postprocedure days and 88 (17.6%) thereafter. Long-term minor complications mainly involved tube disturbances and nearly always resolved once the tube was exchanged. Such exchanges were easily performed under fluoroscopic guidance except in two patients, who required repeat fluoroscopically guided gastrostomy.
CONCLUSION: Fluoroscopically guided gastrostomy is highly feasible and safe and provides adequate feeding support, even when percutaneous endoscopic gastrostomy is impossible. Long-term complications, which are mainly tube disturbances, are easily treated.
Index terms: Gastrointestinal tract, interventional procedure, 72.1269 • Gastrostomy, 72.1269 • Stomach, interventional procedure, 72.1269
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
We performed a study to evaluate the feasibility, complications, adequacy of feeding support, and tolerability of fluoroscopically guided gastrostomy in cancer patients. The initial results and the results of extended follow-up were evaluated.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The initial 41 consecutive patients were referred for fluoroscopically guided gastrostomy after percutaneous endoscopic gastrostomy was unsuccessful due to a severe pharyngeal or esophageal stricture that obstructed passage of the endoscope, even though a pediatric endoscope was used. In the remaining patients, a severe malignant stricture (n = 107) was considered a contraindication to percutaneous endoscopic gastrostomy. Other reasons why percutaneous endoscopic gastrostomy was impossible were unmanageable trismus (n = 9); ulcers of the esophagus or base of the tongue (n = 3); recent pharyngolaryngeal surgery (n = 7), which included five cases of jejunal transplantation; large esophagotracheal fistula (n = 3); and inadequate transillumination (n = 1).
Of the 500 patients referred for fluoroscopically guided gastrostomy, 496 patients underwent 508 procedures. Fluoroscopically guided gastrostomy was unsuccessful in four patients. Twelve patients required a second procedure because of tumor recurrence 4–18 months after removal of the initial gastrostomy tube (n = 10) or because it was impossible to reposition the tube, which had been lost through the original tract (n = 2).
Technique
The inferior margins of the liver were determined with ultrasonography and drawn on the patient's skin. If a nasogastric tube was not already present, a 5-F straight catheter (Cook Europe, Bjaeverskov, Denmark) was placed in the stomach under fluoroscopic control; a 0.035-inch hydrophilic guide wire (Terumo, Tokyo, Japan) was used when needed. Placement of the straight catheter was not possible in 29 patients with tortuous stenoses due to cervical cancer. In 21 of these patients, a 6.5-F angled catheter (Cook Europe) was placed with use of various guide wires. In four of the remaining eight patients, the stomach was punctured with a 22-gauge Chiba needle under fluoroscopic guidance, with the small amount of air already in the stomach used as a target; the stomach was then directly inflated through the Chiba needle. In the remaining four patients, who had no air in the stomach, oral absorption of effervescent sodium bicarbonate powder (Gastrobul; Laboratoire Andre Guerbet, Aulnay-sous-Bois, France) allowed mild gastric ventilation and then direct puncture with the Chiba needle for additional insufflation.
Before inflation of the stomach, 1 mg of glucagon (Glucagen; Novo Nordisk, Bagsvaerd, Denmark) was administered intravenously to inhibit gastric motility and emptying. The skin was prepared with povidone-iodine and local anesthesia was achieved, then a puncture was made equidistant from the greater and lesser curvatures of the stomach at the junction of the upper two-thirds and lower one-third of the stomach and above the colon, which had not been opacified specifically for the procedure. Two or three T-fasteners (Brown/Mueller T-fastener; Medi-tech/Boston Scientific, Watertown, Mass) were inserted around the puncture site by using a specially modified 18-gauge needle with a 5-mm-long slot cut from the bevel. Before the T-fasteners were released, we confirmed that the needle tip was inside the stomach by aspiration of air bubbles or injection of contrast medium. The nylon filaments of the T-fasteners were pulled firmly but not excessively before the metallic clips were clinched to secure the T-fasteners.
The gastrostomy tract was created in the center of the T-fasteners with the 18-gauge needle. After serial dilation with a 0.035-inch Amplatz guide wire, a 16–18-F Malecot nephrostomy catheter (Medi-tech) was placed through a Peel-Away sheath (Cook Europe). The distal end of the gastrostomy catheter was placed in the fundus or antrum. The nasogastric tube was removed at the end of the procedure. Several physicians (two senior clinicians, two fellows, and 10 residents) participated in the procedures. Most of the procedures were performed by residents after a brief training period. Feeding was started within 2 days after the procedure, and the T-fasteners were removed 3 weeks later.
Local anesthesia alone was used in only 42 (14%) of the first 300 patients versus 144 (72%) of the last 200 patients. When administered alone, local anesthesia was achieved with 20–40 mL of lidocaine chlorhydrate, which largely infiltrated the puncture site. In the remaining patients, local anesthesia was combined with light sedation with fentanyl citrate and midazolam hydrochloride. Antibiotics were not administered routinely.
Follow-up
The patients were monitored strictly as inpatients during the first 2 or 3 days after the procedure and returned 3 weeks later for removal of the T-fasteners. Follow-up lasted until death or until removal of the tube (mean follow-up, 7.2 months; range, 1–56 months), except in the 24 patients referred from other hospitals. Follow-up was performed by the usual medical and surgical staff, with at least bimonthly evaluation of complications, tube disturbances, the adequacy of feeding support, and the fate of the tube. Complications or instances of dysfunction during follow-up were referred to the interventional radiology department. In case of problems with the tube, the patients were advised to contact the radiology department directly using a 24-hour telephone number, which was provided in the nursing care booklet given to each patient after the procedure.
Major complications were defined as procedure-related mortality, complications that required surgery or intensive care, or bleeding that required specific treatment. Minor complications were defined as self-limited wound infection, minimal leakage, or need for tube exchange because of occlusion, breakage, or dislodgment. Complications that occurred during the first 30 days were considered short-term; subsequent complications were considered long-term.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Thirty-one patients (6%) died within 30 days of the procedure. All except one of these deaths were disease related. Seven patients (1.4%) experienced short-term major complications; no long-term major complications occurred. One patient experienced cardiac failure with hypoxia after placement of the first T-fastener, and the procedure had to be postponed for 2 days after resuscitation. One patient developed gastric hemorrhage shortly after the procedure and required embolization of the left gastric artery. Five patients developed clinical symptoms of peritonitis during the 3 days after the procedure. Four of these patients recovered uneventfully after laparotomy, and one died. Major intraperitoneal leakage was found at laparotomy in two of the five patients, including the one who died. No internal leakage could be found in two patients, and minimal peristomal leakage was found in one. Among the latter three patients, one had anticancer drug–induced aplasia and two were receiving high-dose corticosteroid therapy.
Minor complications, which were mainly tube disturbances, occurred 115 times (23%), 27 (5.4%) in the short term and 88 (17.6%) in the long term (Table). Short-term minor complications consisted of peristomal leakage (n = 6), superficial skin infection (n = 5), tube dislodgment (n = 8), tube blockage (n = 6), and gastroparesis (n = 2). Long-term minor complications occurred 2–54 months after tube placement. These complications consisted of tube dislodgment (n = 51), which was associated with tube rupture in 17 cases; tube blockage due to kinking or residual food deposits (n = 30); peristomal leakage (n = 4); and superficial skin infection (n = 3).
|
Peristomal leakage of liquid food was treated by exchanging the catheter for a 2-F larger one and applying a cutaneous protection paste (Stomahesive; Laboratoire Convatec, Paris la Défense, France) to the skin around the tube. Leakage reappeared in four patients at 1 week to 2 months, whereupon another tube with a larger diameter was exchanged to resolve this complication. Superficial skin infection responded to local treatment associated with general antibiotic therapy. However, in one patient with recurrent infection associated with peristomal leakage, the gastrostomy was transformed into a gastrojejunostomy. Gastroparesis resolved spontaneously in two patients after discontinuation of morphinomimetic drugs, which were deemed responsible for the complication. We did not evaluate the frequency of exuberant granulation tissue, which did not appear to be related to other complications and which was easily removed by cauterization with silver nitrate.
The average duration of the gastrostomy was 7.2 months (range, 1–56 months); termination of the gastrostomy was due to death of the patient, removal of the tube, or exchange for a gastrostomy button. The tube was removed in 117 patients because dysphagia had resolved. Tube removal was painless and did not require anesthesia. A button was placed in 12 patients with persistent dysphagia who had a long life expectancy because their cancer had been cured or who had slowly progressing disease. At this writing, all patients who are alive have a functional tube and all patients who died had a functional tube at death.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The need for gastropexy is controversial in the literature because T-fasteners are considered capable of inducing considerable tension on the gastric wall. This tension in turn causes skin ischemia, which leads to parietal necrosis and finally enlargement of the gastrocutaneous tract. Fluid leakage then ensues around the tube (6). The risk of hemorrhage is theoretically increased, as is the number of punctures. The low rate of hemorrhage in our series (0.2%) and the potential benefit of tamponade induced by gastropexy (7), together with the low frequency of enlargement of the gastrocutaneous tract in our series (1.2%), prompted us to recommend the use of T-fasteners.
The advantages of gastropexy are that the gastrocutaneous tract matures rapidly, peritonitis and intraperitoneal tube replacement are avoided, and there is thus no need for a new fluoroscopically guided gastrostomy in case of tube dislodgment. Indeed, in our experience, all eight cases of short-term tube dislodgment and 49 of the 51 cases of long-term tube dislodgment were readily managed by means of tube replacement without complications. A new fluoroscopically guided gastrostomy was necessary in only two patients, who were referred to the hospital more than 5 days after tube dislodgment. In contrast, in the largest reported series of patients who underwent fluoroscopically guided gastrostomy without gastropexy, two cases of peritonitis occurred after early tube dislodgment, two cases of peritonitis occurred during late tube replacement, and a new gastrostomy was needed in 28 of 31 cases of tube dislodgment, even though patients were referred less than 24 hours after the dislodgment (8). Furthermore, use of T-fasteners permits initial placement of a large-bore (14–20-F) tube, whereas fluoroscopically guided gastrostomy without gastropexy most often results in placement of 10–11-F tubes (8), which nurses often find difficult to manage because feeding time is prolonged and administering crushed medication through the tube can be cumbersome.
The number of anchoring devices used varies from one device placed through the gastrostomy tract to four placed in the form of a square around the tract. Placing only one fastener through the tract does not seem adequate due to the risk of migration via the tract in the event of tube dislodgment. Moreover, there are additional difficulties related to nursing care of the tube, as well as a risk of balloon perforation when an expandable balloon gastrostomy tube is used. In our experience, two or three anchoring devices provide satisfactory results. However, although the difference was not statistically significant, enlargement of the tract was more frequent after tube placement by residents with limited experience with the technique; this trend was probably related to a tendency to overtighten the T-fasteners. We believe that three moderately tight T-fasteners instead of two very tight ones is probably the best option.
Fluoroscopically guided gastrostomy performed with T-fasteners appears to be safe, with a 1.4% rate of major complications in our series. This rate is comparable with the rates of adverse effects previously reported in smaller series (0%–2.8%). Peritonitis was the most frequent major complication in our series. Among the five cases, inner gastric leakage was abundant in two cases, mild in one case, and absent in two cases. This result emphasizes the extent to which peritonitis, which is not necessarily related to ample intraperitoneal leakage, can occur despite the use of T-fasteners but with minimal peritoneal contamination. In addition, peritonitis probably occurs more frequently in cancer patients because use of immunosuppressive medications, such as corticosteroid therapy or chemotherapy, is common in such patients. However, there were no late cases of peritonitis in our series, as has been reported in series in which gastropexy was not performed (8).
The overall rate of minor complications in our series (23%) appears unusually high compared with those in other series. However, most articles provide results from only the 1st month of follow-up (ie, short-term minor complications), and in this respect our results are comparable (5.4%). The 17.6% rate of long-term minor complications in our study reflects tube disturbances rather than real complications. Indeed, most of these adverse events consisted of tube dislodgment, obstruction, or kinking, which was well tolerated and did not produce unfavorable consequences. Furthermore, these adverse events were often easily managed by exchanging or repositioning the tube under fluoroscopic guidance. Unlike other investigators, we did not systematically exchange tubes after 5 or 6 months, even though doing so would probably have reduced the number of tube disturbances, because this problem is so easily managed that it does not compromise the long-term tolerability and functionality of fluoroscopically guided gastrostomy tubes. However, the importance of follow-up should be emphasized and patients should be well informed, preferably by the interventional radiology department. The eight cases of skin infection in our series occurred mainly at the beginning of our experience. The frequency of this complication has decreased considerably since we abandoned use of an occlusive dressing around the parietal tract; such a dressing is implicated in the occurrence of skin infection.
Wollman et al (9) compared the complication rates of percutaneous endoscopic gastrostomy and fluoroscopically guided gastrostomy using meta-analysis, which demonstrated that fluoroscopically guided gastrostomy produced significantly fewer major complications than did percutaneous endoscopic gastrostomy (5.9% vs 9.4%); no statistically significant difference was demonstrated for minor complications. The rate of tube-related complications was higher for percutaneous endoscopic gastrostomy. The theoretic advantages of percutaneous endoscopic gastrostomy are that it can be performed at the bedside and that endoscopic diagnosis is possible during the procedure. However, two operators are needed for percutaneous endoscopic gastrostomy, and seeding of the tract with cells from head and neck cancer, although unusual, has been reported (10). In our opinion, the major disadvantage of percutaneous endoscopic gastrostomy is that repeat endoscopy is required when tubes are exchanged, a frequent situation because tube disturbances are common.
In our experience, fluoroscopically guided gastrostomy is an effective means of feeding malnourished patients and is the technique of choice in patients with head and neck cancer because it is highly feasible and associated with a low risk. Use of T-fasteners helps reduce complications and facilitates tube exchange and replacement. The long-term tolerability and efficacy are good, and tubes can be readily exchanged when necessary.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, R.C.; study concepts, T.d.B.; definition of intellectual content, T.d.B.; literature research, R.C., T.d.B.; clinical studies, T.d.B., R.C., V.K., P.C., J.P.D., C.D., G.S., A.R.; data analysis, A.R.; manuscript preparation, P.C., R.C., T.d.B.; manuscript editing, T.d.B.; manuscript review, V.K., P.C.
Received February 4, 1998;
revision requested May 4, 1998; revision received July 13, 1998;
accepted October 16, 1998.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Wolch To the Editor: Nutr Clin Pract, December 1, 2007; 22(6): 689 - 690. [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Nosher, L. J. Bodner, W. S. Girgis, R. Brolin, R. L. Siegel, and C. Gribbin Percutaneous Gastrostomy for Treating Dilatation of the Bypassed Stomach After Bariatric Surgery for Morbid Obesity Am. J. Roentgenol., November 1, 2004; 183(5): 1431 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H-P Dinkel, K T Beer, P Zbaren, and J Triller Establishing radiological percutaneous gastrostomy with balloon-retained tubes as an alternative to endoscopic and surgical gastrostomy in patients with tumours of the head and neck or oesophagus Br. J. Radiol., April 1, 2002; 75(892): 371 - 377. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Funaki, R. Peirce, J. Lorenz, P. B. Menocci, J. D. Rosenblum, C. Straus, T. V. Ha, J. A. Leef, and G. X. Zaleski Comparison of Balloon- and Mushroom-Retained Large-Bore Gastrostomy Catheters Am. J. Roentgenol., August 1, 2001; 177(2): 359 - 362. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Funaki, G. X. Zaleski, J. Lorenz, P. B. Menocci, A. N. Funaki, J. D. Rosenblum, C. Straus, and J. A. Leef Radiologic Gastrostomy Placement: Pigtail- Versus Mushroom-Retained Catheters Am. J. Roentgenol., August 1, 2000; 175(2): 375 - 379. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Shingadia, R. M. Viani, R. Yogev, H. Binns, W. M. Dankner, S. A. Spector, and E. G. Chadwick Gastrostomy Tube Insertion for Improvement of Adherence to Highly Active Antiretroviral Therapy in Pediatric Patients With Human Immunodeficiency Virus Pediatrics, June 1, 2000; 105(6): 80e - 80. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. P. Brady, T. de Baere, R. Chapot, V. Kuoch, P. Chevallier, J. P. Delille, C. Domenge, G. Schwaab, and A. Roche Percutaneous Gastrostomy: US Guidance for Gastric Puncture • Dr de Baere and colleagues respond: Radiology, January 1, 2000; 214(1): 303 - 304. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
