| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular and Interventional Radiology |
1 From the Departments of Radiology (A.S.S., J.M., P.S.S.), Neurosciences (M.A.A.), and Dietetics (A.R.), King’s College Hospital, Denmark Hill, London SE5 9RS, England; and Department of Neurology, King’s MND Care & Research Centre, London, England (P.N.L.). Received September 15, 2003; revision requested November 28; revision received January 26, 2004; accepted February 17. Address correspondence to P.S.S. (e-mail: paul.sidhu@kingsch.nhs.uk).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Over a 12-month period (September 2002 through September 2003), 25 patients with ALS (mean age, 62.4 years; age range, 41–83 years; 15 men, 10 women) who had bulbar impairment and a body mass index of less than 20 kg/m2 or weight loss of greater than 10% were selected for placement of an enteral feeding tube. Patients with overnight oxygen desaturation or respiratory acidosis were referred for placement of the Entristar tube with radiologic guidance. This procedure was performed with local anesthesia and without sedation by using a modified percutaneous lateral fluoroscopic technique that aided tube insertion in patients with elevation of the hemidiaphragm and a "high" stomach position. Technical success and immediate and delayed procedure complications were recorded.
RESULTS: The Entristar tube was successfully inserted in all 25 patients. Pneumoperitoneum as an early complication was documented in one patient, and one patient developed a pelvic abscess that required drainage. Follow-up for a median of 112 days (range, 14–343 days) revealed superficial wound infections in four patients and weight gain in two patients; weight gain necessitated tube replacement in one patient. There were no procedure-related deaths.
CONCLUSION: Radiologically guided insertion of the Entristar skin-level gastrostomy tube is a safe procedure in patients with ALS that allows the creation of a permanent feeding gastrostomy without the need for sedation or endoscopy.
© RSNA, 2004
Index terms: Amyotrophic lateral sclerosis, 10.87, 30.87 • Gastrointestinal tract, radiography, 72.1267 • Gastrostomy, 72.1267 • Stomach, interventional procedures, 72.1267
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The insertion of a gastrostomy feeding tube may be performed with one of three methods: surgical, endoscopic, or radiologic. Surgical insertion of a gastrostomy is problematic in patients with ALS owing to the high risk associated with general anesthesia and has been replaced by a percutaneous endoscopic procedure known as percutaneous endoscopic gastrostomy (PEG) (8). PEG was first described in 1980 (9) and results in a reported mortality rate of 1%–2% and a morbidity rate of 3%–12% in all types of patients (10–12), although the requirement for patient sedation during PEG may preclude its use in patients with respiratory failure.
Radiologically guided insertion of a gastrostomy feeding tube, known as radiologic-inserted gastrostomy (RIG), was first described in 1983 (13,14). RIG is performed as a percutaneous procedure with fluoroscopic guidance and local anesthesia and has a low complication rate and high technical success rate (15). The reported disadvantages of an RIG include a small tube lumen size resulting in blockage, a tendency for the tube to dislodge, and the development of an intraperitoneal leak (16). Thornton et al (17) compared endoscopic and radiologically guided placement of gastrostomy tubes in patients with ALS and described a superior success rate with the latter procedure, although complication rates and overall survival did not differ significantly.
In our previous experience with primary radiologically guided placement of pigtail gastrostomy tubes in 35 patients with ALS over a 3-year period, no failure of tube insertion and no major complications were encountered (18). However, in our experience, these pigtail gastrostomy tubes have a number of shortcomings. The tube is secured with a "locked" pigtail that permits tube movement across the stomach and anterior abdominal wall, contributing to local inflammation and infection. The securing thread often dissolves, leading to tube slippage and position loss, and there continues to be a susceptibility to blockage.
The Entristar skin-level gastrostomy tube (Tyco Healthcare, Mansfield, Mass) (Fig 1) is a short 12-F-diameter tube (external diameter, 3 mm) with internal and external retention bolsters to secure its position and allow minimal movement. There is a feeding and/or decompression port fitted with an antireflux valve, which may be closed with a cap when not in use. As with all skin-level devices, the Entristar device is normally inserted as a replacement tube, when a "mature" tract has been established with a previous PEG procedure. The purpose of the present study was to retrospectively review our experience with a radiologic method of primary insertion of the Entristar tube in patients with ALS.
|
|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Patient Selection
Patients were assessed by a clinical team involved in the management of ALS that consisted of physicians (including P.N.L.), a clinical nurse specialist (M.A.A.), and a dietitian (A.R.) to reach a consensus view as to the need for enteral feeding. The following criteria were taken into consideration in assessing patients for enteral feeding: (a) bulbar impairment, which was defined by an ALS swallow severity score of 6/10 or lower (20), and (b) a body mass index of less than 20 kg/m2 and total body weight loss of more than 10% (21). Percentage weight loss was calculated on the basis of the body weight before illness. To determine the body mass index, which was calculated by dividing the weight in kilograms by the square of the height in meters, patient height and weight were measured with the Leicester height measuring device (Vogel & Halke, Hamburg, Germany) and Seca seated balance scales (Vogel & Halke).
Once the need for enteral feeding was agreed on, the patients were referred for a PEG or an RIG procedure on the basis of the criteria described immediately below.
Patients were admitted for noninvasive oximetry to record episodes of overnight oxygen desaturation, defined as the number of "dips" in saturation below baseline per hour. Arterial blood gas testing was performed shortly after the patient awoke while breathing room air to detect development of respiratory acidosis, defined as a PaCO2 of greater than 6.0 mm Hg (normal range, 3.5–5.0 mm Hg) and a bicarbonate level of greater than 26 mmol/L (normal range, 23–25 mmol/L). Patients with no episodes of oxygen desaturation and no respiratory acidosis were referred for a PEG procedure, which is the current standard departmental practice for enteral feeding tube placement.
Patients with normal arterial blood gases (ie, no respiratory acidosis) but in whom some overnight oxygen desaturation occurred (fewer than five episodes per hour) were referred for an RIG procedure to be performed without respiratory support. Patients with evidence of respiratory acidosis or oxygen desaturation (five or more episodes per hour) were also referred for an RIG procedure but received preprocedural and/or periprocedural noninvasive positive pressure ventilation. Measurement of vital capacity was not used as an indicator of respiratory failure (22).
The presence of any bulbar symptoms, irrespective of respiratory status, was an indication for enteral feeding through an RIG. Local experience suggests that patients with bulbar symptoms find insertion of the endoscope uncomfortable and traumatic to the pharyngeal structures.
Patients were also referred for an RIG procedure if there had been a primary failure to insert a feeding tube at PEG.
Over a 12-month period, from September 2002 to September 2003, 25 patients (mean age, 62.4 years; age range, 41–83 years; 15 men, 10 women) met our criteria for RIG placement.
Tube Insertion Technique and Recording of Data
The technique for inserting the Entristar skin-level gastrostomy tube primarily involves standard procedures for inserting an RIG tube that have been previously described (13,14,17); important differences are detailed below.
Patients were admitted to the hospital 1 day before the procedure, and informed consent for the placement of a gastrostomy tube was obtained. A nasogastric tube was positioned, and 200 mL of barium sulfate 100% (Baritop; Sakai Chemical Industry, Osaka, Japan), followed by 50 mL of water, was administered through the nasogastric tube to allow opacification of the transverse colon to occur before the RIG procedure. (Following a review of results in the first nine patients, the use of barium sulfate suspension was halted. The presence of barium sulfate within the large bowel was associated with constipation in patients, and colon opacification was not necessary when lateral fluoroscopy was used.) One hour before the procedure, 750 mg of an intravenous prophylactic broad-spectrum antibiotic (cefuroxime sodium, Britannia Pharmaceuticals, Redhill, England) was administered. Patient blood coagulation times (international normalized ratios) and platelet levels were confirmed to be in the normal range.
A fluoroscopy unit with a rotating C arm (Siemens Multistar; Siemens, Munich, Germany) was used. The patient was placed in a supine position on the fluoroscopy table, with continuous pulse oximetry and cardiac and blood pressure monitoring. A 20-mg dose of intravenous hyoscine bromide (Buscopan; Boehringer Ingelheim, Bracknell, England) was administered, and the stomach was inflated with 500–1000 mL of air through the nasogastric tube. The hyoscine bromide halts gastric peristalsis and reduces any discomfort arising from gastric insufflation.
During imaging in the anteroposterior plane, a marker (typically a pair of forceps) was placed on the skin surface over the body of the inflated stomach to the left of the midline (which normally should extend below the costal margins) so that a suitable site for RIG placement could be identified. However, if the inflated stomach lay above the costal margin, the marker was moved caudally until it was below the costal margin but directly inferior to the body of the stomach to the left of the midline (Fig 2a). The fluoroscopy unit was then rotated through 90°; lateral imaging ensured that no loops of bowel were interposed between the anterior abdominal wall and the stomach (Fig 2b). If interposed bowel loops were seen, the procedure was postponed for 24 hours.
|
|
|
|
|
Then, more local anesthetic was infiltrated to the center of the square formed by the four T-fasteners, and a transverse skin incision of approximately 4–6 mm in length was made. An 18-gauge needle was inserted through this incision, and a 0.035-inch Amplatz guidewire (Cook, Bjaeverskov, Denmark) was passed through the needle so that it coiled in the gastric cavity (Fig 4a). The needle was removed and replaced with a 6-F dilator (Cook); subsequently, serial dilators in 2-F increments (up to 20 F) were used to create a tract. The 20-F dilator was finally replaced with a 22-F peel-away sheath (Cook) and introducer (Fig 4b).
|
|
|
|
|
After the RIG procedure, each patient with ALS remained in the hospital for 2–3 days to learn the techniques of gastrostomy tube management. Patients were aggressively rehydrated through the gastrostomy tube to relieve any constipation (a well-recognized complication of ALS), and rectal and/or colonic enemas were administered as necessary (2). Hospital discharge was delayed until constipation had resolved. The T-fasteners were released at 5 days by cutting the external suture; this was usually performed by a community-based nurse. After the T-fasteners were removed, the Entristar tube was rotated through 360° each day (in accordance with the manufacturer’s recommendations). This aids the formation of a tract by preventing tissue attachment to the tube and also prevents epithelial overgrowth of the internal retention bolster. The tube requires rotation daily for as long as it is in situ.
Patient Follow-up
Complications were characterized as early (occurring <24 hours after the procedure) or late (occurring 
24 hours after the procedure) and were recorded. Those complications arising after the procedure were further defined as minor (abdominal pain, wound infection, fever, peristomal leakage, gastroparesis, simple dislocation, tract disruption, and catheter dislocation or fracture) or major (hemorrhage requiring blood transfusion, pneumoperitoneum, peritonitis, aspiration, and any complication of tube insertion requiring radiologic intervention or surgery) (23). Patient follow-up was performed by the specialist nurse (M.A.A.) in all cases 1 month after the procedure and every 3 months thereafter to document complications. The occurrence of death was recorded and assessed to determine whether it had any relationship to the gastrostomy procedure.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
There were no procedural failures with the tube insertion technique; all tubes were correctly positioned. No patient encountered any respiratory difficulties during the procedure. There was a single early complication in one patient (patient 4), who had been referred after primary failure of a PEG procedure owing to inadvertent puncture of the transverse colon. Pneumoperitoneum, which was asymptomatic and did not require further treatment, was demonstrated in this patient after the RIG procedure. A single patient (patient 14) developed a pelvic abscess, presumably because of a perigastrostomy leak (although this could not be demonstrated at fluoroscopy after administration of contrast medium). This abscess was detected 48 hours after the RIG procedure and was drained with CT guidance; the patient fully recovered. These episodes (n = 2) were classified as major complications.
Visual inspection after hospital discharge revealed local wound infection in four patients (patients 4, 5, 15, and 20); these infections responded to antibiotic treatment. No tube failure as a result of tube fracture, displacement, or dislodgment occurred. Weight gain in two patients after the procedure led to difficulties with the tube; in one patient (patient 3), the tube could only be rotated with the patient in certain positions, although the tube continued to function well otherwise; in the second patient (patient 5), weight gain necessitated changing the tube to a Kangaroo balloon gastrostomy tube (Tyco Healthcare) after 69 days. There were no difficulties encountered in replacing the tube at the bedside. All of these episodes (n = 6) were classified as minor complications. With the exception of patient 5, all patients had the Entristar tube in place at the time of death or the end of the follow-up period.
One patient (patient 2) died of a cerebrovascular accident 24 days after the procedure. One patient (patient 17) died of airway obstruction after impaction of a mucus plug 52 days after the procedure. Four patients (patients 1, 5, 7, and 11) died of the underlying ALS during the follow-up period. No tube-related death or lasting major morbidity occurred as a consequence of the RIG procedure.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Performing a PEG procedure in the patient with ALS entails a number of technical challenges. Sedation, which leads to further respiratory compromise and pharyngeal weakness, is required, and mechanical trauma from the gastroscope renders the procedure uncomfortable for the patient. Aspiration during a PEG procedure remains a considerable risk in comparison with the risk of aspiration at an RIG procedure (2.1% vs 0.6%) (27). The stomach often lies in a high position owing to weakness of the diaphragm and other respiratory muscles, so transillumination of the anterior abdominal wall with the gastroscope light is not possible.
Although there is evidence to support the primary placement of an enteral feeding tube with the RIG procedure in patients with ALS (17,18), complications associated with tube design remain a problem. Traditional RIG tubes have a Cope-loop internal fixation mechanism (28), and use of an additional skin fixation device to retain position and prevent migration is often required (29). Balloon-retaining devices are also available, but these too have limitations; weekly exchange of water in the balloon results in overdistention and rupture (17), and spontaneous balloon rupture is a major reason for tube replacement (30). The radiologically guided placement of pull-type gastrostomy tubes, in which a feeding device with a silicone "mushroom" is placed through the mouth after percutaneous stomach puncture (31–33), may overcome the tube-related problems of the RIG procedure, but the former procedure requires sedation and is not a suitable alternative for patients with ALS.
Traditional PEG tubes that employ the flange system have less tendency to dislodge but can result in the "buried-bumper syndrome," in which the gastric epithelium grows over the internal retention device (34). Nevertheless, PEG insertion with a mushroom-type device has advantages: The average longevity of the mushroom-type device exceeds that of a balloon-type device, with valve incompetence the usual reason for replacement (30). Therefore, transferring the advantages of a mushroom-type device to a procedure that involves percutaneous deployment without the need for sedation would be beneficial to patients with ALS. The development of the Entristar gastrostomy tube allows the primary placement with radiologic guidance of a well-designed tube in this population for the first time. The importance of a single-step procedure in patients with a terminal disease in whom quality of life needs to be optimized and hospital stay minimized must be emphasized.
The problems associated with a high-placed stomach in ALS are overcome with the radiologic method. Traditionally, the radiologic technique has involved imaging in a single plane, the anteroposterior direction (13,14). When the stomach lies below the costal margin, this method is adequate, but the ability to visualize the insufflated stomach in real time with both anteroposterior and lateral fluoroscopy allows accurate placement of the T-fasteners used to perform gastropexy and correct deployment of the feeding tube. The high-placed stomach can be approached safely through the subcostal route, avoiding an intercostal approach and without risking inadvertent pneumothorax (35). In the present study, with the addition of lateral fluoroscopy, the procedure time remained favorable in comparison with that estimated for conventional RIG (32.4 minutes ± 14.3) and PEG (35.7 minutes ± 13.0) procedures (27), and the radiation dose was acceptable. We would recommend that lateral fluoroscopy be used in all patients, even those with a subcostally located stomach. Furthermore, measurement of the length of the feeding tube that is required is performed in the lateral plane.
Patients quite often delay a decision to undergo a PEG procedure until they are in respiratory failure and are denied the benefits of enteral feeding. The American Academy of Neurology guidelines (26) advise that PEG should only be performed when vital capacity is more than 50% of the predicted value. However, vital capacity is a poor predictor of respiratory failure, which may be present even with a documented vital capacity of 70% of the predicted value (22). Exclusive use of the vital capacity guideline may expose patients who are in respiratory failure but have a vital capacity of more than 50% of the predicted value to increased risk during a PEG procedure.
We avoided the use of vital capacity guidelines in selecting the patients for enteral feeding tube insertion but chose to support the patients with ALS who were identified as having respiratory failure (according to methods detailed above) and who were thus unsuited for a PEG procedure with noninvasive positive pressure ventilation during the RIG procedure. Furthermore, the technique we describe avoids the need for sedation, which would further compromise respiratory reserve. Therefore, noninvasive positive pressure ventilation allows patients with poor respiratory reserve the opportunity to benefit from enteral feeding. If necessary, noninvasive ventilation may be continued throughout the RIG procedure, because the equipment used does not interfere with the operator or with the procedure.
The success of deployment of the Entristar gastrostomy tube surpassed expectations in this small group of patients with ALS. Dilation of the tract to 22 F exceeds the usual dilation to 14 or 18 F in other reported series (17,23,29), but, with the aid of lateral fluoroscopy, this was not problematic in our series. In two patients, postprocedural weight gain compromised the feeding tube, which was of insufficient length. As a result, insertion of longer tubes (4.5 or 5.0 cm) is recommended, with an allowance for weight gain even if the measured distance from the anterior abdominal wall to the stomach is less than 4.5 or 5.0 cm. Currently, 5.0 cm is the longest available tube of this design, although the manufacturer is reviewing the possibility of developing longer tubes. Furthermore, after a review of the initial cases, the use of the barium suspension was halted because it was thought to be noncontributory when lateral fluoroscopy was used and inevitably resulted in patient constipation, a common problem in ALS that is known to contribute to appetite loss (36,37). This enabled selected patients to be discharged within 2 days after RIG insertion.
The major limitation of this study was that it was carried out retrospectively, although the data were recorded contemporaneously. The relatively small number of patients included in this study reflects the prevalence of the disease, but, despite this, we have shown that RIG performed with the Entristar tube is a safe procedure in patients with ALS, even if they have respiratory failure. Furthermore, a survival benefit from gastrostomy in ALS has never been proved, but is often assumed. Indeed, such a probative study would be difficult to design and control; however, further research is required to evaluate the relationship between nutritional status, morbidity, and mortality in this population.
Following our initial success with this technique that has such advantages over endoscopic and surgical techniques, we are now evaluating longer-term outcomes. Moreover, given our success with the Entristar device, in conjunction with other recent reports (17), we are considering extending the use of an Entristar RIG procedure to all patients with ALS who require a gastrostomy. Further extending this policy to include any patient needing a gastrostomy and hence removing the need for endoscopic insertion of a feeding tube altogether is being considered. The major technical problem with extending the service to other patient groups remains the tube length, which may be a more substantial problem over the longer term.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, P.S.S.; study concepts, P.S.S., A.R., M.A.A.; study design, P.S.S.; literature research, A.S.S., A.R., M.A.A., P.S.S.; clinical studies, A.S.S., A.R., M.A.A., P.S.S., J.M.; data acquisition and analysis/interpretation, A.S.S., A.R., M.A.A., P.S.S.; manuscript preparation and final version approval, A.S.S., P.S.S.; manuscript definition of intellectual content, P.S.S., P.N.L.; manuscript editing, A.S.S., A.R., M.A.A., P.S.S.; manuscript revision/review, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
