Radiology guided antegrade GASTROSTOMY deployment of mushroom (pull type) catheters with classical and modified methods in patients with oropharyngeal, laryngeal carcinoma, and anesthesia risk

Abstract

Objective:

The aim of study is to evaluate the results of deployment of Percutaneous Radiological Gastrostomy (PRG), which is a good alternative to Surgical Gastrostomy (SG), with transoral approach in cases where Percutaneous Endoscopic Gastrostomy (PEG) is contraindicated, difficult or unsuccessful, in patients with high risk of American Society of Anesthesiologists with four scores. In addition, we aimed to demonstrate the advantages of mushroom pull type catheters over push type gastrostomy catheters.

Methods:

This retrospective study included a total of 40 patients (18 females and 22 males) aged 21–92 years who underwent PRG with the antegrade transoral approach. PRG was performed by retrograde passing through the esophagus or snaring the guidewire from the stomach and taking out of the anterior abdominal wall. Patients’ demographic data, indications for PRG, procedural outcomes and complications were screened and recorded.

Results:

PRG was performed in 39 of 40 patients included in the study. Technical success rate was 97.5%. Procedure-dependent major complications such as death, aspiration, colon perforation, and deep abscess were not observed. Aspiration occurred in the first patient during the first feeding on the day after the procedure. Major complication rate was 2.5%. The total minor complication rate was 17.5% in 7 patients; parastomal leakage in 2 patients (5%), skin rash and infection in 3 (7.5%) patients, minor bleeding in 2 (5%) patients with oropharynx cancer, minimal bleeding from the gastrostomy catheter 1 week after the procedure in 1 (2.5%) patient. None of the cases had buried buffer. Tube functionality was preserved in all patients without any damage.

Conclusion:

Mushroom tip (pull type) gastrostomy catheter is a safe treatment method for patients requiring prolonged feeding because of wide diameter, endurance, long staying opening duration, less excessive dilatation and parastomal leakage, and no need for gastropexy. Lower cost and easier access are advantageous for mushroom tip pull type catheters compared to push type gastrostomy catheters in our country. The less invasive PRG is an alternative option in patients who are difficult to administer PEG, are at high anesthesia risk and cannot be sedated.

Advances in knowledge:

This article is valuable in terms of its contribution to develop an alternative radiological method for the deployment of gastrostomy tubes in medical difficult patients. This method has shortened the duration of the procedure and increased the success rate in patients with difficulty in transition from the stomach to the esophagus or with difficulty in the upper gastrointestinal tract. Mushroom tip catheters can be placed successfully by radiological methods.

Introduction

A gastrostomy is an option for enteral feeding in patients who cannot be fed orally and need long-term nutritional support.^1,2 The first percutaneous approaches emerged in the 1980s and have been preferred to open surgical procedures (SG) until today.¹

Endoscopic and radiologic methods can be used in the deployment of percutaneous gastrostomy catheters.³ Percutaneous endoscopic gastrostomy (PEG) is frequently used and has been the first accepted method.^3–8 Percutaneous radiological gastrostomy (PRG) is performed with antegrade (transoral or transabdominal) pathways and is a good non-surgical alternative to PEG for gastrostomy.

PRG is a good alternative for patients who have stenosis or obstruction that prevent the endoscopic passage or patients who cannot tolerate the endoscopic procedure due to high risk of anesthesia. Compared to PEG, it has high procedural success rates and low complication rates. Different techniques have been developed since it was defined in 1981. PRG is mainly applied by antegrade (transoral) or retrograde (transabdominal) routes. It has been described in many modified methods.

The retrograde (transabdominal) approach is performed with a gastric puncture, and then serial dilatations followed by insertion of a catheter with a balloon or pig-tail fixation mechanism using the Seldinger technique. This approach is mainly preferred in patients with oropharyngeal, hypopharyngeal and esophageal malignancy.^9–20

Antegrade (transoral) approach is performed with a gastric puncture and then snaring of the guidewire which is advanced to the stomach by passing through the esophagus or from the gastric puncture with a snare catheter passed from the mouth into the stomach. After this step, the procedure is the same as PEG, and mushroom tipped catheters are used. A limited number of articles have been published in the literature for this alternative approach.^21–31

The objective of this study was to evaluate transoral deployment of PRG, which is a good alternative of Surgical Gastrostomy (SG) where PEG is contraindicated, difficult or failed in high-risk patients with American Society of Anesthesiologists 4 (ASA 4) score, and to reveal the advantages of mushroom pull type catheters over push type gastrostomy catheters.

Methods and materials

Patients

After approval of the local ethics committee, PRG procedures performed with the antegrade (transoral) approach under the guidance of fluoroscopy and ultrasonography between January 2016 and December 2018 were retrospectively examined.

Seven patients who had failed PEG trials with oropharyngeal and laryngeal malignant diseases were included in the study and PRG was performed. 22 patients with ASA 4 risk who were followed in the intensive care unit due to neurological diseases and 12 patients with swallowing difficulties due to malignancy and chronic diseases were included in the study. Only one patient had previous gastric surgery. PEG was tried first in all patients. No patients were included in the study for gastric decompression, PEG was applied to ASA 2 and 3 patient groups without upper gastrointestinal obstructive lesions, and they were excluded from the study. In patients with ASA 4 risk who can tolerate endoscopy, PEG was first tried.

The mean age was 70.75 (range 21–92) years. PRG was performed in 18 female (mean age: 71.61, range 37–92) and 22 male (mean 70.04, range 22–91) patients. Patients’ demographic data, indications for PRG, procedural outcomes and complications were screened and recorded. The complications were defined as major and minor according to the guidelines of the Interventional Radiology and Gastroenterology Society.^31,32 Indications of PRG are listed in Table 1.

Table 1.

Indications of PRG in 40 patients with the antegrade transoral approach

PRG indications	Female (number-age)	Male (number-age)	Total (number-age)
Neurological	13	8	21
Stroke	6 (60-83/77,83)	6 (64-81/70,66)	12 (60-83/74,25)
Amyotrophic lateral sclerosis	1 (61)	-	1 (61)
Alzheimer’s – Parkinson's – Dementia	4 (84-92/87,25)	1 (80)	5 (80-92/85,8)
Glioblastoma	1 (64)	-	1 (64)
Subarachnoid hemorrhage – Epidural hematoma	1 (37)	1 (21)	2 (21-37/29)
Head-neck cancers	2	5	7
Oral cavity malignancies	1 (72)	1 (82)	2 (72-82/77)
Hypopharyngeal carcinoma	1 (37)	1 (68)	2 (37-68/52,5)
Nasopharyngeal carcinoma	-	1 (83)	1 (83)
Laryngeal carcinoma	-	2 (73-75/74)	2(73-75/74)
Other (swallowing difficulty)	3	9	12
Trauma	-	2 (59-64/61,5)	2 (59-64/61,5)
Colon carcinoma	1 (44)	-	1 (44)
Endometrial carcinoma	1 (76)	-	1 (76)
Lung carcinoma	-	4 (59-80/68,5)	4 (59-80/68,5)
Endocrine pancreatic carcinoma	-	1 (59)	1 (59)
Chronic renal failure	1 (82)	-	1 (82)
Gastrointestinal decompression	-	-	-
Total	18 (37-92/71,61)	22 (21-90/70,04)	40 (21-92/70,75)

PRG, Percutaneous Radiological Gastrostomy.

Technique

All procedures were performed in the Interventional Radiology unit by a single Interventional Radiologist who had 13 years’ experience. Routine coagulopathy values were checked before the procedure. The operation was performed in patients with an INR <1.5 and a platelet count >50.000×10³/LL (50.000 × 10⁹/L). All patients were evaluated by anesthesiologists and ASA scoring was performed before the procedure.

The nasogastric tube was not fed for at least 8 h before the procedure. All patients were followed up under intensive care conditions and received appropriate antibiotic therapy.

After informed consent was obtained, the patient received moderate sedation by intravenous midazolam (Dormicum 50 mg / 10 ml) and fentanyl (Talinate 0,5 mg / 10 ml) by an anesthesiologist. Povidone-iodine (Betadine 30,100) was used to perform abdominal skin antisepsis with standard brush protocol.

In all patients, the stomach was accessed with 5 Fr angiography catheter (Radifocus^® Optitorque^™ Angiographic catheter) and 0.035-inch 180 cm length hydrophilic guidewire (GLIDEWIRE^® Hydrophilic Coated Guidewire – Terumo) under fluoroscopy by oral route and inflated from the catheter with room air. Ultrasonography was routinely performed in all patients in order to identify the left lobe margin of the liver and rule out the presence of a colon in front of the stomach. After the access site was determined by ultrasonography, the middle part of the stomach corpus-antrum was targeted by fluoroscopy. Local anesthesia to the access site was applied with prilocaine hydrochloride (Priloc %2) and then a deep incision was performed 1 cm wide. Ultrasonography and fluoroscopy puncture was performed in the stomach with a 17 G needle. After exhausting the air of the stomach previously inflated through the needle with room air, five ccs opaque agent (ULTRAVIST^® 370 mg/mL- Bayer) was delivered inside the stomach and was confirmed with fluoroscopy. A hard-hydrophilic guidewire was advanced to the stomach via the access needle. 8 Fr sheath (Avanti sheath introducer-cordis) was placed into the stomach via the guidewire. In order to keep the stomach adjacent to the abdominal wall, no T connection was needed in any case. After this stage, two methods were applied for the insertion of guidewire between the mouth and the anterior abdominal wall.

In the first method; 180 cm long soft hydrophilic guidewire, folded in half in the middle, was pushed through the 8 Fr intraducer from the fold and delivered into the stomach lumen. Free ends of the guidewire remained outside. Thus, many wide rings facing in different directions were formed in the stomach with the guidewire. In addition to the 5 Fr angiographic catheter sent orally to the stomach, another guidewire was directed to the formed rings in the lumen of the stomach and passed through it. The wire folded through the anterior abdominal wall was pulled from the free ends and the guidewire sent from the mouth was caught like a snare catheter, and then pulled together with the 8F sheath and taken out from the anterior abdominal wall. Thus, a guidewire was placed between the mouth and the anterior abdominal wall.

Figure 1.

Fluoroscopic images show that, after retrograde access to the stomach, the guidewire sent through the mouth with an angiography catheter is captured in the stomach and removed from the anterior abdominal wall and finally a mushroom tip (pull-type) gastrostomy catheter is inserted. (A) The corpus or antrum of the stomach, which is inflated with room air via a nasogastric catheter or a 5 Fr angiographic catheter inserted through the mouth, is punctured through the middle section with a needle. The hydrophilic guidewire is advanced through the needle into the stomach. After needle removal, the 8 Fr introducer is placed in the stomach. (B) The folded hydrophilic guidewire is advanced through the introducer into the stomach and large rings are formed in different directions. (C) The hydrophilic guidewire is advanced through the mouth through an angiography catheter to the rings in the stomach and is taken out of the anterior abdominal wall with the introducer. (D) Once one end of the guidewire is removed from the anterior abdominal wall, the other end on the mouth side is attached to the loop at the end of the 20 Fr mushroom tip gastrostomy catheter; then the gastrostomy catheter is pulled into the stomach.

In the second method; after the 8F sheath was guided to the esophagogastric junction, the esophagus was entered with a 5 Fr angiographic catheter and guidewire. The guidewire was then pulled out from the mouth with the support of the catheter. Thus, a guidewire was provided between the mouth and anterior abdominal wall.

Figure 2.

Fluoroscopic images show retrograde access to the stomach through the esophagus, removal of the guidewire from the mouth and subsequent insertion of a mushroom tip (pull type) gastrostomy catheter. (A) The corpus or antrum of the stomach, which is inflated with room air via a nasogastric catheter or a 5 Fr angiographic catheter inserted through the mouth, is punctured through the middle section with a needle. The hydrophilic guidewire is advanced through the needle into the stomach. After needle removal, the 8 Fr introducer is placed in the stomach. The esophagogastric junction is then passed through an angiographic catheter and guidewire. (B) The guidewire is advanced from the esophagus and removed from the mouth. (C) After the guidewire is removed from the patient’s mouth, the guidewire is attached to the ring at the end of the 20 Fr mushroom tip gastrostomy catheter and the gastrostomy catheter is pulled into the stomach.

The procedure after providing a guidewire between the mouth and anterior abdominal wire with both techniques was the same with PEG. Mushroom tip (pull type) catheter was attached to the guidewire from the ring at the end. The catheter was then supported from the back side, pulled out from the anterior abdominal wall together with the guidewire and deployed into the stomach with the transoral antegrade approach. The string and dilator at the end of the catheter were cut and a feeding port was fixed to the catheter. Then, the fixing disk was lowered to the skin and dressed. The procedure was completed by checking under contrast-enhanced fluoroscopy. The PRG was considered successful when the gastrostomy catheter was properly placed in the stomach and functioning properly. In the absence of any emergency complications, catheter feeding was initiated 24 h after the procedure.

Statistical analysis

While the average and standard deviations were given as descriptive statistics of numerical variables, the numbers and percentages were given as descriptive statistics of categorical variables. SAS University Edition 9.4 software was used to obtain the analyzes.

Results

PRG was performed in 39 of 40 patients. Of all patients, 18 were female (mean age: 71.61) and 22 were male (mean age 70.04). Our technical success rate for PRG was 97.5%. In only one case where the procedure failed, there was a previous Billroth one type gastric surgery. The procedure was discontinued in the patient who did not have gastric distension, could not be approached to the anterior abdominal wall, and who had colon transposition. In 20 (50%) of the procedures, the guidewire was caught in stomach lumen by the first method and removed out from the anterior abdominal wall, while in the rest 19 (47.5%) procedures, the guidewire was pulled to the esophagus lumen and then was taken out from the mouth to complete the placement of gastrostomy tube. No procedure-related major complications such as mortality, aspiration, colon perforation and deep abscess were observed. However, in one patient, aspiration occurred during the initial feeding on the day after the procedure. The major complication rate was 2.5%. No procedure-related mortality occurred, but at 30 day follow-up, the rate of documented all-cause mortality was 27.5%. The cause of death in 34 of the patients who were followed in the intensive care unit were comorbidities. The total minor complication rate was 17.5% in 7 patients. In all patients, the tube patency was preserved and there was no fracture in the tubes. Because only 2 (5%) patients had parastomal leakage, the amount of fluid delivered from the gastrostomy catheter was reduced and leakage was lost in these patients. Due to erythema and infection, 3 (7.5%) patients were treated with intravenous antibiotics in addition to frequent catheter care and topical antibiotics. During the procedure, 2 (5%) patients with oropharyngeal cancer had minor bleeding and limited without treatment. In one patient (2.5%) who initially could not tolerate the PEG procedure and underwent PRG, although there was no bleeding during the procedure, a small amount of bleeding occurred from the gastrostomy catheter 1 week after the procedure. Bleeding was found to be due to gastritis in diagnostic endoscopy. None of the cases had a buried bumper.

Discussion

Since the 1980s, PEG and PRG have been implemented with high success rates. In this retrospective study, which included 40 patients with comorbidities, all of whom were followed up under intensive care conditions, except 7 patients with head and neck cancer, we evaluated the results of PRG procedures by the antegrade (transoral) route and found that this was an appropriate and effective alternative method for enteral nutrition.

Gastrostomy is performed for nutritional support or decompression. There are various methods for performing gastrostomies such as SG, PEG and PRG.^1,2

SG is an invasive procedure that requires general anesthesia and is performed with a high rate of complication (19,9–57%).^3,7,33 However, SG has an exact indication in patients with abdominal surgery and in patients with surgically altered gastrointestinal anatomy. Today, PEG and PRG offer superior access compared to surgical methods. So, SG can only be used when PEG and PRG fail.³³

PEG is the first choice in most patients and requires varying degrees of anesthesia support in practice. However, it has limitations in oropharyngeal, esophageal cancers, patients in whom abdominal wall transillumination cannot be performed and obese patients.^3–8 This has led to recommendations for PRG in high-risk patients with comorbidities, and those with technically difficult and contraindicated endoscopy.^12,17,18,20 Our study group consisted of patients in whom PEG was unsuccessful and patients with ASA 4 risk with co-morbidities that prevented endoscopic procedures.

PRG is performed by fluoroscopic control, rarely with ultrasonography or CT-guided and with local anesthesia or mild sedoanalgesia. Compared with PEG, PRG has higher success rates.^7,26 PRG can be performed with retrograde (transabdominal) or antegrade (transoral) approaches. Mostly the retrograde (transabdominal) approach is preferred and push-type catheters with pigtail or balloon fixation mechanisms are used.^9–20 After serial dilatations and insertion of removable sheaths for insertion of push-type catheters with pig-tail and balloon fixation mechanisms, over expansion of the track can lead to gastric contents penetrating through the gastrostomy site, peritonitis, skin irritation, and infection. These catheters are prone to obstruction, break-off and spontaneous dislocation when compared with mushroom tip pull type gastrostomy catheters.^28,29 On the other hand, retrograde insertion of push type catheter has some advantages; there is no risk of tumor seeding into the gastrostomy tract and there is no contamination of oropharyngeal flora with this technique.^16–18 The antegrade transoral approach is a less widely used method and there are limited publications in the literature. In the present study, all gastrostomy catheters were deployed with the antegrade transoral approach. Technical success of mushroom tip pull type catheters ranges from 98 to 100%.^21–31 Transverse colon and liver left lobe between the stomach and abdominal anterior wall or previous gastric surgery to disrupt the anatomy are the reasons for the failure of the procedure. In a single case with a transverse colon between the stomach and anterior abdominal wall with the failed procedure, the stomach could not be distended and approached to the anterior abdominal wall due to previous surgery (Billroth one gastric surgery). Our technical success rate is 97.5% and is consistent with the literature. Malignancies of oropharynx and esophagus are difficult procedures of PRG by the antegrade (transoral) route. In our study, there were seven patients with oropharyngeal carcinoma and laryngeal carcinoma in which PEG failed, and one patient who developed cardiac arrest during PEG insertion had difficulty in anesthesia and stenosis that prevented the passage of the endoscope. In these patients, after a mild sedoanalgesia and oropharynx spray lidocaine anesthesia, angiography catheter with a finer calibration than the endoscope and a guidewire could be passed through the stenosis and a mushroom tip pull type gastrostomy catheter was inserted. Bleeding due to the passage of the mushroom tip pull type catheter was self-limited and did not require intervention (5%). There was no massive bleeding or rupture.

Complication rates of PRG range from 0 to 16.7%.^{9–11,13,14,21–32,34} Funaki et al reported that pigtail and balloon-locked catheters (36 and 2%, respectively) had higher complication rates than mushroom tip pull type catheters, while other major and minor complication rates were higher in pigtail and balloon-locked catheters (3 and 8%, respectively) than mushroom tip pull type catheters (0 and 2%, respectively).^10,29

In our study, major and minor complication rates were 2.5 and 17.5%, respectively. In the antegrade (transoral) method, wound infection rates due to oropharyngeal contamination are expected to be higher than the retrograde (transabdominal) method. However, superficial cellulite rates have been reported to be higher in push type catheters than in pull type catheters.^10,29 In our study, the rate of parastomal superficial cellulitis was 7.5%. Catheter-related complications such as a buried bumper, catheter breakdown, catheter obstruction, and accidental catheter dislocation may reach up to 36% with pull type catheter.^9–20,28,29 Mushroom tip pull type catheters have longlasting and adequate patency rates.^10,29 Our results were consistent with previous studies in the meaning of catheter patency.

Usually, in antegrade (transoral) approach, the esophagus is passed in a retrograde route and the wire is pulled out from the mouth.^21–31 We used the same method in patients with easy catheterization of the esophagus. In the study performed by Piton et al,³⁵ the guidewire sent to the stomach after gastric puncture was captured by a snare catheter sent from the mouth and pulled out from the mouth. In our study, instead of using a snare catheter, we created a very wide loops through the 8 Fr sheath that we placed in the stomach and folded in two directions and advanced the guidewire we sent through the mouth with the help of angiographic catheter, and snared the guidewire in 8 Fr sheath by pulling from its two ends. In the literature, we did not find any pulling out of the guidewire from the anterior abdominal wall by this method. In patients with difficult catheterization of the esophagus or in patients with the difficult or retrograde passage of oropharynx or esophageal stenosis, we think that this method provides convenience, shortens the procedure time and reduces the exposed radiation dose.

In PRG, T-fixation elements are used in order to fix the stomach to the anterior abdominal wall.^10–16 Funaki et al used one T-fixation and release it just before deployment of mushroom tip pull type catheter.¹⁰ In our study, we did not use T-fixation. The fixation of the stomach by the T-fixation provides ease of entry and safety, but the rigid hydrophilic guidewire sufficiently charged to the inflated stomach with room air provided the 8 Fr sheath to be inserted into the stomach without problems. In addition, T-fixation elements may cause an increase in complications such as cellulitis.¹⁰

In our country, push type catheters can be obtained with high costs and have significant cost disadvantages compared to pull type catheters. Cost values far above the health practice declaration applied in our country make push-type catheter procurement difficult. Lower cost and easier access are advantageous for mushroom tip pull type catheters. In our study, a push-type catheter could not be provided due to high costs.

Limitations of the present study are its retrospective nature, short life expectancy in patient selection, the inclusion of patients with morbidity that may affect the success of the procedure, absence of long-term follow-up, lack of body weight and serum albumin levels, and no comparison with an alternative technique.

Conclusion

Although it has the disadvantage of passing the esophagus and oropharynx during placement, mushroom tip pull type gastrostomy catheters provide a safe treatment method in patients requiring long-term nutrition, because of wide diameter, endurance, prolonged luminal patency, no excessive dilatation, less parastomal leakage and no need for gastropexy. When inserting the catheter, instead of capturing the guidewire between the mouth and anterior abdominal wall with a snare catheter delivered through the mouth, the conversion of a guidewire with very wide and different directions of rings folded in half with an 8 Fr sheath already inserted into the stomach reduces our procedure time and patient radiation exposure. Low cost and easy accessibility compared to push type gastrostomy catheters are advantageous for our country. Endoscopic placement is the first choice for the placement of mushroom tip pull type catheters in patients with no obstructive lesion in the upper GIS and without a high risk of anesthesia. However, in patients with failed anterior abdominal wall transillumination, upper GIS occlusion but stenosis that does not allow passage of the endoscope and cannot be sedated because of the high risk of anesthesia, mushroom tip pull type catheter deployment using the less invasive radiological method is the alternative to endoscopy-guided catheter placement.