- Email: [email protected]
Randomized Controlled Trial Comparing Radiologic Pigtail-Retained Gastrostomy and Radiologic Mushroom-Retained Gastrostomy
CLINICAL STUDY
Randomized Controlled Trial Comparing Radiologic Pigtail-Retained Gastrostomy and Radiologic Mushroom-Retained Gastrostomy Kichang Han, MD, Man-Deuk Kim, MD, Joon Ho Kwon, MD, Yong Seek Kim, MD, Gyoung Min Kim, MD, Junhyung Lee, MD, Woosun Choi, MD, Jong Yun Won, MD, and Do Yun Lee, MD ABSTRACT Purpose: To prospectively compare 2 different types of percutaneous fluoroscopic gastrostomy procedures (pigtail-retained gastrostomy [PG] vs mushroom-retained gastrostomy [MG]). Materials and Methods: Between March 2014 and February 2016, 100 patients were randomly assigned to receive 14-F PG or 20-F MG. Block randomization (block size 4) was performed, and sample size was calculated to assess the difference in minor complications. One patient withdrew from the study after allocation. Baseline characteristics were not significantly different between groups (P > .05). Technical success, defined as successful placement of gastrostomy tube, and procedural complications were evaluated. Procedural complications were divided into major and minor complications according to the Society of Interventional Radiology criteria. Results: Technical success rate was 100%. In the PG group, the major complication rate was 2% (1 of 50); 1 patient had a misplaced PG in the peritoneal cavity between the gastric and abdominal walls and developed peritonitis that had to be surgically treated. The minor complication rate was 34% (17 of 50) in the PG group. In the MG group, the major complication rate was 0%, and the minor complication rate was 12.2% (6 of 49). The most common minor complication was tube occlusion. Minor complication rate was significantly higher in the PG group (P ¼ .016). Mean fluoroscopy time was significantly longer in the MG group (P ¼ .013). Conclusions: Both PG and MG demonstrated high technical success rates in all indications. MG had lower complication rates than PG at the cost of an increase in fluoroscopy times.
ABBREVIATIONS MG ¼ mushroom-retained gastrostomy, PG ¼ pigtail-retained gastrostomy, RIG ¼ radiologically inserted gastrostomy
Percutaneous fluoroscopically guided gastrostomy has been widely used as a safe and effective technique for feeding, medication, or gastric decompression (1–7). Radiologically inserted gastrostomy (RIG) is performed by placing a
From the Department of Radiology, Severance Hospital, Research Institute of Radiological Science, Yonsei University College of Medicine, 50 Yonsei-ro Seodaemun-gu, Seoul 120-752, Korea. Received May 12, 2017; final revision received June 21, 2017; accepted June 23, 2017. Address correspondence to M.-D.K.; E-mail: [email protected] None of the authors have identified a conflict of interest. From the SIR 2017 Annual Scientific Meeting. © SIR, 2017 J Vasc Interv Radiol 2017; ▪:1–6 http://dx.doi.org/10.1016/j.jvir.2017.06.031
pigtail-retained gastrostomy (PG) or balloon-retained gastrostomy catheter through the Seldinger technique with serial dilation of the insertion site. A decade after the introduction of RIG, peroral placement of a mushroomretained gastrostomy (MG) tube was reported by some investigators (8–13), in which the esophagus was catheterized after gastric puncture in a retrograde fashion, and then the gastrostomy was pulled from the mouth into the stomach. In previous studies comparing the 2 techniques, the technical success rates were uniformly high; however, different results were reported concerning complications. Although Kuo et al (14) demonstrated that there was no significant difference in the complication rate between the 2 techniques, Yang et al (13) reported that more complications occurred in PG than in MG. It is desirable that the differences between the 2 gastrostomy types be confirmed with randomized data. At the present time, only 1 randomized
2 ▪ Pigtail-Retained Gastrostomy vs Mushroom-Retained Gastrostomy
EDITORS’ RESEARCH HIGHLIGHTS Prospective randomized controlled trial compared 14-F pigtail-retained gastrostomy (PG; 50 patients) and 20-F mushroom-retained gastrostomy (MG; 49 patients) (level of evidence 1b). Technical success rate was 100% for both groups. Minor complication rate was higher in the PG group (34%) than in the MG group (12.2%; P ¼ .016) owing to higher rates of tube occlusion (16% vs 0%) or dislodgment (4% vs 0%) and peristomal leakage (6% vs 2%). Major complication rate was 2% in the PG group and 0% in the MG group (P > .05).
controlled trial comparing PG and MG exists; however, the study included only patients with head and neck cancer, and there was no significant difference in complication rates, technical success, and procedure times between the 2 groups (15). Therefore, complication rates and catheter performance need to be documented in a larger patient cohort with broader indications for gastrostomy. The purpose of this study was to prospectively compare 2 different types of gastrostomy (PG and MG) in terms of technical success, procedure-related and catheter-related complications, and mean fluoroscopy time.
MATERIALS AND METHODS Patients This was a parallel-group randomized trial approved by the institutional review board. Between March 2014 and February 2016, 127 patients were assessed for eligibility. Recruitment took place in a radiology outpatient clinic where radiologists interviewed the patients and/or their caregivers. The inclusion criteria were as follows: (i) informed consent; (ii) age > 18 years; and (iii) need for a gastrostomy for enteral feeding, medication or gastric decompression. The exclusion criteria included (i) patient refusal; (ii) previous gastrectomy; (iii) no treatment for head and neck cancer; (iv) radiologic evidence of residual or recurrent tumor after anticancer treatment; (v) ventriculoperitoneal shunt; and (vi) uncorrectable coagulopathy (Table 1). Of 127 patients, 100 patients who met the inclusion criteria for undergoing percutaneous RIG were enrolled in this study. Each patient provided a signed written informed consent form before study participation. One patient withdrew from the study after allocation to MG because of relocation to a city far from the study institution.
Sample Size and Randomization The sample size was based on the assumption that the minor complication rate would be 13% in the MG group (8) and 31% in the PG group (15). Using PASS (NCSS, LLC, Kaysville, Utah) statistical software, it was calculated that 50 patients including a dropout rate of 5% would be required in
Han et al ▪ JVIR
Table 1. Eligibility Criteria for Participants Inclusion Criteria
Exclusion Criteria
Signed informed consent
Patient refusal
Age > 18 y
Previous gastrectomy
Gastrostomy needed for enteral feeding, medication, or gastric decompression
Residual or recurrent tumor after anticancer treatment in patients with head and neck or esophageal cancer Ventriculoperitoneal shunt Uncorrectable coagulopathy No treatment for head and neck or esophageal cancer
each group to achieve a power of .80 and a significance level of .05 for assessing the difference between the 2 groups. Randomization was performed through block randomization (block size 4) to balance the gastrostomy assignment. The patients and interventionists could not be blinded to the gastrostomy type. The Consolidated Standards of Reporting Trials flow diagram shows the randomization and flow of patient evaluations throughout the trial (Fig).
Baseline Demographics Baseline patient characteristics are presented in Table 2. The mean age of all patients was 60.1 y ± 14.9, and 20.2% were women (P ¼ .653 between groups). The indications for gastrostomy were equally distributed between the 2 groups: head and neck cancer (57.6% of patients), cerebrovascular accident (19.2% of patients), neuromuscular disease (9.1% of patients), and other conditions (14.1% of patients) (P ¼ .726 between groups).
Procedure Details Gastrostomy procedures were performed by 2 boardcertified interventional radiologists with > 10 years of clinical experience. The puncture site was chosen under fluoroscopy taking into account the following 3 anatomic relationships: gastroesophageal junction, costal margin, and colon. All patients undergoing gastrostomy received antibiotic prophylaxis with intravenous administration of 1 g of cefazolin (Ildong Pharmaceutical, Seoul, Korea).
Pigtail-Retained Gastrostomy The stomach was inflated by using either a nasogastric tube or a 5-F end-hole catheter under fluoroscopic guidance. After subcutaneous administration of lidocaine, the mid to lower gastric body of the stomach was punctured with a 17-gauge needle. Contrast agent was injected to confirm the appropriate position of the needle in the stomach. Two gastropexy T-fasteners (Cope Gastrointestinal Suture Anchor Set; Cook, Inc, Bloomington, Indiana) were routinely placed around the intended catheter insertion site. A 5-mm incision was made between the 2 T-fastener sites. The stomach was accessed through the incision, and the tract was serially dilated to accept a 14-F PG catheter (Wills-Oglesby gastrostomy
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
3
Figure. Consolidated Standards of Reporting Trials flow diagram.
Mushroom-Retained Gastrostomy
Table 2. Baseline Patient Characteristics Characteristic
PG (n ¼ 50) MG (n ¼ 49) P Value
Sex Men
.653 39
40
Women
11
9
Mean age, y
60.9 ± 16.6
59.8 ± 15.9
Indications
.718 .726
Head and neck cancer
26
31
Cerebrovascular accident
10
9
Neuromuscular disease
6
3
Other malignancy
4
2
Others
4
4
MG ¼ mushroom-retained gastrostomy; PG ¼ pigtail-retained gastrostomy.
catheter; Cook, Inc). After gastrostomy placement, contrast agent was injected to confirm the correct position. The gastrostomy tubes were directed toward the gastric fundus to avoid the risk of bowel perforation. The catheters were secured to the patient with sutures, and feeding or medication was initiated 24 hours from the time of placement in the absence of any immediate complication. T-fasteners were cut 2 weeks after tube placement.
The stomach was insufflated by using either a nasogastric tube or a 5-F end-hole catheter under fluoroscopic guidance. After administration of local anesthesia, a 5- to 10-mm incision was made over the intended gastric puncture site. The stomach was punctured through the incision with a 17gauge needle and 180-cm, 0.035-inch guide wire (Terumo Corporation, Tokyo, Japan). Contrast medium was injected to confirm the appropriate position of the needle in the stomach. For MG, gastropexy was not used. Instead, an 8-F vascular sheath was placed over the guide wire, and a 5-F end-hole catheter was introduced to facilitate the selection of the esophagus in a retrograde manner. The interventionist then advanced the wire out of the mouth of the patient under fluoroscopy. The 5-F catheter was exchanged for a 7-F endhole catheter, and the thread included in the kit was advanced through the catheter from the abdominal wall to the mouth. The snare attached to the 20-F gastrostomy catheter (Ponsky Pull PEG Kit; Bard Access Systems, Inc, Salt Lake City, Utah) was tied to the thread, and the tube was pulled in anterograde fashion into the stomach. The catheter was secured to the patient with sutures, and the dilator on the end of the catheter was cut. After gastrostomy placement, contrast agent was injected to confirm the correct position of the gastrostomy tube. Feeding or medication was
4 ▪ Pigtail-Retained Gastrostomy vs Mushroom-Retained Gastrostomy
initiated 24 hours from the time of placement in the absence of any immediate complication. If the guide wire failed to enter the esophagus retrogradely, the snare was introduced into the stomach via the sheath, and the guide wire was advanced from the mouth to the stomach by using a 5-F end-hole catheter. The guide wire was captured by the snare in the stomach and pulled out of the stomach through the abdominal wall. The tube was then placed in exactly the same manner as described earlier.
Han et al ▪ JVIR
complication rate was significantly higher in the PG group (P ¼ .016) (Table 3). The 30-day mortality rates were 2% (1 of 50) and 4% (2 of 49) for the PG and MG groups, respectively (P ¼ .617). No patient deaths were related to gastrostomy placement. The mean fluoroscopy time was significantly longer in the MG group than in the PG group (2.7 min for PG vs 6 min for MG; P ¼ .013). In patients who received PG catheters, the major complication rate was 2% (1 of 50), and the minor complication rate was 34% (17 of 50). The major complication was peritonitis that developed 1 day after gastrostomy placement. Computed tomography revealed that the locking pigtail of the catheter was located between the abdominal and gastric walls, and the gastric contents were spilled into the peritoneum. This patient needed surgical repair and aggressive antibiotic treatment. Minor complications included tube occlusion (n ¼ 7), peristomal leakage (n ¼ 3), superficial cellulitis (n ¼ 3), tube dislodgment (n ¼ 2), transient bleeding (n ¼ 1), and pain in the insertion site (n ¼ 1). In 7 patients, occlusion was resistant to saline flushing and was treated by removing the catheter and inserting a new catheter via the existing gastrostomy tract. In 3 patients with pericatheter leakage, the PG catheter was exchanged for an 18-F balloon-retained gastrostomy catheter via the existing tract. Dislodgment was managed by placing a new catheter via the existing tract in 2 patients. Pain subsided with administration of the nonnarcotic analgesic ketorolac tromethamine (Keromin; Hana Pharmaceutical Co, Ltd, Seoul, Korea) 30 mg intravenously, and superficial cellulitis was managed with cefazolin. In patients who received MG catheters, the major complication rate was 0%, and the minor complication rate was 12.2% (6 of 49). Minor complications included pain and/or discomfort in the catheter site (n ¼ 3), superficial cellulitis (n ¼ 1), peristomal leakage (n ¼ 1), and partial tube fracture (n ¼ 1). Pain subsided with administration of ketorolac tromethamine 30 mg intravenously, and superficial cellulitis was managed with cefazolin. Pericatheter leakage and fractured catheters were managed by placing a 24-F balloon-retained gastrostomy catheter via the existing tract.
Follow-up The primary endpoint was procedural complications. Technical success was defined as successful placement of the gastrostomy tube. Procedural complications were divided into major and minor complications according to the Society of Interventional Radiology (SIR) classification system (16). Follow-up was conducted on a monthly basis for 6 months after catheter placement for gastrostomy-related complications or dysfunction or until gastrostomy removal if earlier than 6 months. For patients who were discharged from the hospital, phone calls were made every month to ask if they experienced any complications. The patients were not scheduled for routine catheter exchange; rather, tubes were changed only when they malfunctioned.
Statistical Analysis Comparisons between groups were performed using Student t test for continuous data and c2 or Fisher exact test for categorical data. All analyses were by intention to treat and performed with IBM SPSS Statistics for Windows Version 23 (IBM Corp, Armonk, New York). P values < .05 were considered statistically significant.
RESULTS After randomization, 50 patients underwent PG creation, and 49 patients underwent MG creation. All procedures (100%) were technically successful. The minor
Table 3. Complication Rates in Both Groups Parameters
PG (n ¼ 50)
Interval (d)
MG (n ¼ 49)
Interval (d)
1 (2%)
1
0
—
Catheter misplacement
7 (16%)
24–126
0
— 2
Minor complications Tube occlusion
.016
Skin infection
3 (8%)
1–9
1 (5%)
Peristomal leakage
3 (6%)
1–3
1 (2%)
1
Tube dislodgment Catheter site pain
2 (4%) 1 (2%)
36–92 1
0 3 (6%)
— 1–4
0
—
1 (2%)
48
1 (2%)
1
0
—
Tube fracture Transient bleeding
P Value > .999
Major complications
MG ¼ mushroom-retained gastrostomy; PG ¼ pigtail-retained gastrostomy.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
DISCUSSION Since RIG was first described in 1983, knowledge about the efficacy and safety of percutaneous gastrostomy has been increasing. Percutaneous gastrostomy has largely replaced surgical gastrostomy owing to its lower complication rates and less invasiveness (7,17–19). For the varied percutaneous access techniques for PG and MG, a wide range of efficacy and complication rates have been reported in retrospective and prospective studies (3,4,8,20,21). Comparative studies between PG and MG have been published; however, these studies reported conflicting results and had nonnegligible selection bias (13–15). In a recent randomized trial comparing the 2 techniques, the measured outcomes, including pain scores, technical success, and complication rates, were comparable in the 2 groups (15). However, the study included only patients with head and neck cancer who were undergoing prophylactic gastrostomy creation before radiation therapy. The present study demonstrated that the minor complication rates were significantly higher in the PG group, and the mean total fluoroscopy times were significantly longer in the MG group. In prior studies, PG and MG had equally high technical success rates (98%–100%) (3,4,15,20). In this study, the technical success rate was 100% in both groups. Concerning procedure time, MG had a significantly longer fluoroscopy time than PG, which was in accordance with a previous report (15). In some patients, retrograde catheterization of the guide wire into the esophagus or capture of the guide wire by using a snare within the insufflated stomach substantially prolonged the fluoroscopy time. The most common indication for gastrostomy was to provide nutrition for patients with head and neck cancer, esophageal cancer, cerebrovascular accident, or neuromuscular disease. This finding was also in agreement with findings of most prior studies (3,4,13). Other indications for gastrostomy included corrosive esophagitis, deep neck infection, and gastric decompression. Concerning complications, in 1 patient, the locking pigtail was located between the abdominal and gastric walls, and the gastric juice spilled into the peritoneal cavity, causing peritonitis. Although 2 gastropexy T-fasteners were routinely placed to prevent the gastric wall from retracting from the abdominal wall, misplacement of the locking pigtail still occurred. This could have been prevented if the patient had received MG because the catheter would have been deployed from the mouth to the stomach and thus would have provided better apposition between the 2 walls. Skin infection was reported to occur in 1.6%–4.4% of patients who underwent RIG (1,2,5,6). Compared with PG, the infection rates were higher in MG because of the oropharyngeal route of tube placement (14,22,23). However, in the present study, superficial cellulitis was more common in patients with PG than in patients with MG (4 in PG group vs 1 in MG group), and no major infection occurred. This was likely due to antibiotic prophylaxis and peristomal leakage.
5
Peristomal leakage was more common in the PG group, and it might have caused skin ulceration, increasing the risk for subsequent infection. Pain in the catheter site was more commonly seen in MG, and this could be attributed to the larger diameter of the MG tube. In general, catheter performance is important for patient and caregiver satisfaction, and catheter complications can adversely affect referrals for gastrostomy. In the present study, the catheter complication rate was significantly higher in the PG group. Funaki et al (4) and Laasch et al (20) also reported a significantly higher catheter complication rate in the PG group in their comparative studies. However, in the study by Bernstein et al (15), the catheter complication rate was not significantly different between the 2 techniques, and the discrepancy might be attributed to the selective patient cohort having head and neck cancer. The tube occlusion rate was the most distinctive difference between the 2 groups (7 tube occlusions in the PG group vs no tube occlusions in the MG group). Tube patency is determined as a function of diameter and length. It is plausible that MG is less likely to be occluded than PG because MG has a larger diameter and a shorter length than PG. Tube dislodgment was more common in the PG group (2 in PG group vs 0 in MG group). The stability of a gastrostomy catheter is determined by its retaining mechanism. In PG, the locking pigtail is retained only by a thread, and the thread is vulnerable to the acidic gastric juice. In MG, the internal and external retention plates provide more stability than PG, making MG more difficult to dislodge (4). Peristomal leakage was also more common in the PG group (3 in PG group vs 1 in MG group). The possible reason for this was that serial dilation of the gastrostomy tract may lead to overdilatation, making it easier for the gastric contents to leak through the catheter site (4). In MG, however, the dilator attached to the tube exactly matches the diameter of the gastrostomy catheter; thus, overdilatation rarely occurs. Seeding metastasis to the gastrostomy tract is a rare but catastrophic complication of MG with a reported incidence of 0.7%–1% (13,24,25). In a study by Cappell (25), squamous cell carcinoma, large tumor size, advanced tumor stage, tumor histology less well differentiated, and pharyngoesophageal cancer were associated with malignant seeding. In the present study, malignant tumor seeding did not occur. That may be because sterile patients with head and neck cancer or esophageal cancer were included, which means that only patients who underwent anticancer treatment and had no evidence of residual or viable tumor on cross-sectional imaging were enrolled in this study. In a study by Yang et al (13), malignant seeding developed 243 days after MG placement; thus, the relatively short-term follow-up period in the present study might have missed such a complication. However, the superiority of MG over PG in terms of overall complications seems to cancel out the low possibility of tumor seeding. This study has some limitations. First, aspiration pneumonia that was included in prior studies as a major complication was not evaluated owing to pre-existing risks
6 ▪ Pigtail-Retained Gastrostomy vs Mushroom-Retained Gastrostomy
of aspiration in this patient cohort. Second, this randomized study could not be conducted using gastrostomy catheters of the same diameter because all the available MG tubes in the country of this study were 20 F in diameter. As the tube diameter affects the likelihood of tube occlusion, excluding 7 tube occlusions leaves 10 complications in the PG group (20%) and 6 in the MG group (12%). Third, the follow-up period was 6 months, and there was a possibility of missing some complications that occurred > 6 months after the procedure. Fourth, radiation exposure in each group was not measured and compared in this study. In conclusion, both PG and MG demonstrated high technical success rates in all indications. However, this randomized controlled study demonstrates the advantages of MG over PG in terms of the overall complication rate.
REFERENCES 1. Bell SD, Carmody EA, Yeung EY, Thurston WA, Simons ME, Ho CS. Percutaneous gastrostomy and gastrojejunostomy: additional experience in 519 procedures. Radiology 1995; 194:817–820. 2. de Baere T, Chapot R, Kuoch V, et al. Percutaneous gastrostomy with fluoroscopic guidance: single-center experience in 500 consecutive cancer patients. Radiology 1999; 210:651–654. 3. Funaki B, Peirce R, Lorenz J, et al. Comparison of balloon- and mushroomretained large-bore gastrostomy catheters. AJR Am J Roentgenol 2001; 177:359–362. 4. Funaki B, Zaleski GX, Lorenz J, et al. Radiologic gastrostomy placement: pigtail- versus mushroom-retained catheters. AJR Am J Roentgenol 2000; 175:375–379. 5. Saini S, Mueller PR, Gaa J, et al. Percutaneous gastrostomy with gastropexy: experience in 125 patients. AJR Am J Roentgenol 1990; 154: 1003–1006. 6. Wollman B, D’Agostino HB. Percutaneous radiologic and endoscopic gastrostomy: a 3-year institutional analysis of procedure performance. AJR Am J Roentgenol 1997; 169:1551–1553. 7. Wollman B, D’Agostino HB, Walus-Wigle JR, Easter DW, Beale A. Radiologic, endoscopic, and surgical gastrostomy: an institutional evaluation and meta-analysis of the literature. Radiology 1995; 197: 699–704. 8. Ahmed O, Jilani D, Sheth S, Giger M, Funaki B. Radiologically guided placement of mushroom-retained gastrostomy catheters: long-term outcomes of use in 300 patients at a single center. Radiology 2015; 276:588–596.
Han et al ▪ JVIR
9. Clark JA, Pugash RA, Pantalone RR. Radiologic peroral gastrostomy. J Vasc Interv Radiol 1999; 10:927–932. 10. Rosenzweig TB, Palestrant AM, Esplin CA, Gilsdorf RB. A method for radiologic-assisted gastrostomy when percutaneous endoscopic gastrostomy is contraindicated. Am J Surg 1994; 168:587–590; discussion 590–591. 11. Szymski GX, Albazzaz AN, Funaki B, et al. Radiologically guided placement of pull-type gastrostomy tubes. Radiology 1997; 205:669–673. 12. Pitton MB, Herber S, Duber C. Fluoroscopy-guided pull-through gastrostomy. Cardiovasc Interv Radiol 2008; 31:142–148. 13. Yang Y, Schneider J, Duber C, Pitton MB. Comparison of fluoroscopyguided Pull-type percutaneous radiological gastrostomy (Pull-type-PRG) with conventional percutaneous radiological gastrostomy (Push-typePRG): clinical results in 253 patients. Eur Radiol 2011; 21:2354–2361. 14. Kuo YC, Shlansky-Goldberg RD, Mondschein JI, et al. Large or small bore, push or pull: a comparison of three classes of percutaneous fluoroscopic gastrostomy catheters. J Vasc Interv Radiol 2008; 19:557–563; quiz 564. 15. Bernstein OA, Campbell J, Rajan DK, et al. Randomized trial comparing radiologic pigtail gastrostomy and peroral image-guided gastrostomy: intra- and postprocedural pain, radiation exposure, complications, and quality of life. J Vasc Interv Radiol 2015; 26:1680–1686; quiz 1686. 16. Sacks D, McClenny TE, Cardella JF, Lewis CA. Society of Interventional Radiology clinical practice guidelines. J Vasc Interv Radiol 2003; 14: S199–S202. 17. Ho CS. Percutaneous gastrostomy for jejunal feeding. Radiology 1983; 149:595–596. 18. Cosentini EP, Sautner T, Gnant M, Winkelbauer F, Teleky B, Jakesz R. Outcomes of surgical, percutaneous endoscopic, and percutaneous radiologic gastrostomies. Arch Surg 1998; 133:1076–1083. 19. Given MF, Hanson JJ, Lee MJ. Interventional radiology techniques for provision of enteral feeding. Cardiovasc Interv Radiol 2005; 28:692–703. 20. Laasch HU, Wilbraham L, Bullen K, et al. Gastrostomy insertion: comparing the options—PEG, RIG or PIG? Clin Radiol 2003; 58:398–405. 21. Lowe AS, Laasch HU, Stephenson S, et al. Multicentre survey of radiologically inserted gastrostomy feeding tube (RIG) in the UK. Clin Radiol 2012; 67:843–854. 22. Akkersdijk WL, van Bergeijk JD, van Egmond T, et al. Percutaneous endoscopic gastrostomy (PEG): comparison of push and pull methods and evaluation of antibiotic prophylaxis. Endoscopy 1995; 27:313–316. 23. Yip D, Vanasco M, Funaki B. Complication rates and patency of radiologically guided mushroom gastrostomy, balloon gastrostomy, and gastrojejunostomy: a review of 250 procedures. Cardiovasc Interv Radiol 2004; 27:3–8. 24. Cruz I, Mamel JJ, Brady PG, Cass-Garcia M. Incidence of abdominal wall metastasis complicating PEG tube placement in untreated head and neck cancer. Gastrointest Endosc 2005; 62:708–711. 25. Cappell MS. Risk factors and risk reduction of malignant seeding of the percutaneous endoscopic gastrostomy track from pharyngoesophageal malignancy: a review of all 44 known reported cases. Am J Gastroenterol 2007; 102:1307–1311.
Randomized Controlled Trial Comparing Radiologic Pigtail-Retained Gastrostomy and Radiologic Mushroom-Retained Gastrostomy Kichang Han, MD, Man-Deuk Kim, MD, Joon Ho Kwon, MD, Yong Seek Kim, MD, Gyoung Min Kim, MD, Junhyung Lee, MD, Woosun Choi, MD, Jong Yun Won, MD, and Do Yun Lee, MD ABSTRACT Purpose: To prospectively compare 2 different types of percutaneous fluoroscopic gastrostomy procedures (pigtail-retained gastrostomy [PG] vs mushroom-retained gastrostomy [MG]). Materials and Methods: Between March 2014 and February 2016, 100 patients were randomly assigned to receive 14-F PG or 20-F MG. Block randomization (block size 4) was performed, and sample size was calculated to assess the difference in minor complications. One patient withdrew from the study after allocation. Baseline characteristics were not significantly different between groups (P > .05). Technical success, defined as successful placement of gastrostomy tube, and procedural complications were evaluated. Procedural complications were divided into major and minor complications according to the Society of Interventional Radiology criteria. Results: Technical success rate was 100%. In the PG group, the major complication rate was 2% (1 of 50); 1 patient had a misplaced PG in the peritoneal cavity between the gastric and abdominal walls and developed peritonitis that had to be surgically treated. The minor complication rate was 34% (17 of 50) in the PG group. In the MG group, the major complication rate was 0%, and the minor complication rate was 12.2% (6 of 49). The most common minor complication was tube occlusion. Minor complication rate was significantly higher in the PG group (P ¼ .016). Mean fluoroscopy time was significantly longer in the MG group (P ¼ .013). Conclusions: Both PG and MG demonstrated high technical success rates in all indications. MG had lower complication rates than PG at the cost of an increase in fluoroscopy times.
ABBREVIATIONS MG ¼ mushroom-retained gastrostomy, PG ¼ pigtail-retained gastrostomy, RIG ¼ radiologically inserted gastrostomy
Percutaneous fluoroscopically guided gastrostomy has been widely used as a safe and effective technique for feeding, medication, or gastric decompression (1–7). Radiologically inserted gastrostomy (RIG) is performed by placing a
From the Department of Radiology, Severance Hospital, Research Institute of Radiological Science, Yonsei University College of Medicine, 50 Yonsei-ro Seodaemun-gu, Seoul 120-752, Korea. Received May 12, 2017; final revision received June 21, 2017; accepted June 23, 2017. Address correspondence to M.-D.K.; E-mail: [email protected] None of the authors have identified a conflict of interest. From the SIR 2017 Annual Scientific Meeting. © SIR, 2017 J Vasc Interv Radiol 2017; ▪:1–6 http://dx.doi.org/10.1016/j.jvir.2017.06.031
pigtail-retained gastrostomy (PG) or balloon-retained gastrostomy catheter through the Seldinger technique with serial dilation of the insertion site. A decade after the introduction of RIG, peroral placement of a mushroomretained gastrostomy (MG) tube was reported by some investigators (8–13), in which the esophagus was catheterized after gastric puncture in a retrograde fashion, and then the gastrostomy was pulled from the mouth into the stomach. In previous studies comparing the 2 techniques, the technical success rates were uniformly high; however, different results were reported concerning complications. Although Kuo et al (14) demonstrated that there was no significant difference in the complication rate between the 2 techniques, Yang et al (13) reported that more complications occurred in PG than in MG. It is desirable that the differences between the 2 gastrostomy types be confirmed with randomized data. At the present time, only 1 randomized
2 ▪ Pigtail-Retained Gastrostomy vs Mushroom-Retained Gastrostomy
EDITORS’ RESEARCH HIGHLIGHTS Prospective randomized controlled trial compared 14-F pigtail-retained gastrostomy (PG; 50 patients) and 20-F mushroom-retained gastrostomy (MG; 49 patients) (level of evidence 1b). Technical success rate was 100% for both groups. Minor complication rate was higher in the PG group (34%) than in the MG group (12.2%; P ¼ .016) owing to higher rates of tube occlusion (16% vs 0%) or dislodgment (4% vs 0%) and peristomal leakage (6% vs 2%). Major complication rate was 2% in the PG group and 0% in the MG group (P > .05).
controlled trial comparing PG and MG exists; however, the study included only patients with head and neck cancer, and there was no significant difference in complication rates, technical success, and procedure times between the 2 groups (15). Therefore, complication rates and catheter performance need to be documented in a larger patient cohort with broader indications for gastrostomy. The purpose of this study was to prospectively compare 2 different types of gastrostomy (PG and MG) in terms of technical success, procedure-related and catheter-related complications, and mean fluoroscopy time.
MATERIALS AND METHODS Patients This was a parallel-group randomized trial approved by the institutional review board. Between March 2014 and February 2016, 127 patients were assessed for eligibility. Recruitment took place in a radiology outpatient clinic where radiologists interviewed the patients and/or their caregivers. The inclusion criteria were as follows: (i) informed consent; (ii) age > 18 years; and (iii) need for a gastrostomy for enteral feeding, medication or gastric decompression. The exclusion criteria included (i) patient refusal; (ii) previous gastrectomy; (iii) no treatment for head and neck cancer; (iv) radiologic evidence of residual or recurrent tumor after anticancer treatment; (v) ventriculoperitoneal shunt; and (vi) uncorrectable coagulopathy (Table 1). Of 127 patients, 100 patients who met the inclusion criteria for undergoing percutaneous RIG were enrolled in this study. Each patient provided a signed written informed consent form before study participation. One patient withdrew from the study after allocation to MG because of relocation to a city far from the study institution.
Sample Size and Randomization The sample size was based on the assumption that the minor complication rate would be 13% in the MG group (8) and 31% in the PG group (15). Using PASS (NCSS, LLC, Kaysville, Utah) statistical software, it was calculated that 50 patients including a dropout rate of 5% would be required in
Han et al ▪ JVIR
Table 1. Eligibility Criteria for Participants Inclusion Criteria
Exclusion Criteria
Signed informed consent
Patient refusal
Age > 18 y
Previous gastrectomy
Gastrostomy needed for enteral feeding, medication, or gastric decompression
Residual or recurrent tumor after anticancer treatment in patients with head and neck or esophageal cancer Ventriculoperitoneal shunt Uncorrectable coagulopathy No treatment for head and neck or esophageal cancer
each group to achieve a power of .80 and a significance level of .05 for assessing the difference between the 2 groups. Randomization was performed through block randomization (block size 4) to balance the gastrostomy assignment. The patients and interventionists could not be blinded to the gastrostomy type. The Consolidated Standards of Reporting Trials flow diagram shows the randomization and flow of patient evaluations throughout the trial (Fig).
Baseline Demographics Baseline patient characteristics are presented in Table 2. The mean age of all patients was 60.1 y ± 14.9, and 20.2% were women (P ¼ .653 between groups). The indications for gastrostomy were equally distributed between the 2 groups: head and neck cancer (57.6% of patients), cerebrovascular accident (19.2% of patients), neuromuscular disease (9.1% of patients), and other conditions (14.1% of patients) (P ¼ .726 between groups).
Procedure Details Gastrostomy procedures were performed by 2 boardcertified interventional radiologists with > 10 years of clinical experience. The puncture site was chosen under fluoroscopy taking into account the following 3 anatomic relationships: gastroesophageal junction, costal margin, and colon. All patients undergoing gastrostomy received antibiotic prophylaxis with intravenous administration of 1 g of cefazolin (Ildong Pharmaceutical, Seoul, Korea).
Pigtail-Retained Gastrostomy The stomach was inflated by using either a nasogastric tube or a 5-F end-hole catheter under fluoroscopic guidance. After subcutaneous administration of lidocaine, the mid to lower gastric body of the stomach was punctured with a 17-gauge needle. Contrast agent was injected to confirm the appropriate position of the needle in the stomach. Two gastropexy T-fasteners (Cope Gastrointestinal Suture Anchor Set; Cook, Inc, Bloomington, Indiana) were routinely placed around the intended catheter insertion site. A 5-mm incision was made between the 2 T-fastener sites. The stomach was accessed through the incision, and the tract was serially dilated to accept a 14-F PG catheter (Wills-Oglesby gastrostomy
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
3
Figure. Consolidated Standards of Reporting Trials flow diagram.
Mushroom-Retained Gastrostomy
Table 2. Baseline Patient Characteristics Characteristic
PG (n ¼ 50) MG (n ¼ 49) P Value
Sex Men
.653 39
40
Women
11
9
Mean age, y
60.9 ± 16.6
59.8 ± 15.9
Indications
.718 .726
Head and neck cancer
26
31
Cerebrovascular accident
10
9
Neuromuscular disease
6
3
Other malignancy
4
2
Others
4
4
MG ¼ mushroom-retained gastrostomy; PG ¼ pigtail-retained gastrostomy.
catheter; Cook, Inc). After gastrostomy placement, contrast agent was injected to confirm the correct position. The gastrostomy tubes were directed toward the gastric fundus to avoid the risk of bowel perforation. The catheters were secured to the patient with sutures, and feeding or medication was initiated 24 hours from the time of placement in the absence of any immediate complication. T-fasteners were cut 2 weeks after tube placement.
The stomach was insufflated by using either a nasogastric tube or a 5-F end-hole catheter under fluoroscopic guidance. After administration of local anesthesia, a 5- to 10-mm incision was made over the intended gastric puncture site. The stomach was punctured through the incision with a 17gauge needle and 180-cm, 0.035-inch guide wire (Terumo Corporation, Tokyo, Japan). Contrast medium was injected to confirm the appropriate position of the needle in the stomach. For MG, gastropexy was not used. Instead, an 8-F vascular sheath was placed over the guide wire, and a 5-F end-hole catheter was introduced to facilitate the selection of the esophagus in a retrograde manner. The interventionist then advanced the wire out of the mouth of the patient under fluoroscopy. The 5-F catheter was exchanged for a 7-F endhole catheter, and the thread included in the kit was advanced through the catheter from the abdominal wall to the mouth. The snare attached to the 20-F gastrostomy catheter (Ponsky Pull PEG Kit; Bard Access Systems, Inc, Salt Lake City, Utah) was tied to the thread, and the tube was pulled in anterograde fashion into the stomach. The catheter was secured to the patient with sutures, and the dilator on the end of the catheter was cut. After gastrostomy placement, contrast agent was injected to confirm the correct position of the gastrostomy tube. Feeding or medication was
4 ▪ Pigtail-Retained Gastrostomy vs Mushroom-Retained Gastrostomy
initiated 24 hours from the time of placement in the absence of any immediate complication. If the guide wire failed to enter the esophagus retrogradely, the snare was introduced into the stomach via the sheath, and the guide wire was advanced from the mouth to the stomach by using a 5-F end-hole catheter. The guide wire was captured by the snare in the stomach and pulled out of the stomach through the abdominal wall. The tube was then placed in exactly the same manner as described earlier.
Han et al ▪ JVIR
complication rate was significantly higher in the PG group (P ¼ .016) (Table 3). The 30-day mortality rates were 2% (1 of 50) and 4% (2 of 49) for the PG and MG groups, respectively (P ¼ .617). No patient deaths were related to gastrostomy placement. The mean fluoroscopy time was significantly longer in the MG group than in the PG group (2.7 min for PG vs 6 min for MG; P ¼ .013). In patients who received PG catheters, the major complication rate was 2% (1 of 50), and the minor complication rate was 34% (17 of 50). The major complication was peritonitis that developed 1 day after gastrostomy placement. Computed tomography revealed that the locking pigtail of the catheter was located between the abdominal and gastric walls, and the gastric contents were spilled into the peritoneum. This patient needed surgical repair and aggressive antibiotic treatment. Minor complications included tube occlusion (n ¼ 7), peristomal leakage (n ¼ 3), superficial cellulitis (n ¼ 3), tube dislodgment (n ¼ 2), transient bleeding (n ¼ 1), and pain in the insertion site (n ¼ 1). In 7 patients, occlusion was resistant to saline flushing and was treated by removing the catheter and inserting a new catheter via the existing gastrostomy tract. In 3 patients with pericatheter leakage, the PG catheter was exchanged for an 18-F balloon-retained gastrostomy catheter via the existing tract. Dislodgment was managed by placing a new catheter via the existing tract in 2 patients. Pain subsided with administration of the nonnarcotic analgesic ketorolac tromethamine (Keromin; Hana Pharmaceutical Co, Ltd, Seoul, Korea) 30 mg intravenously, and superficial cellulitis was managed with cefazolin. In patients who received MG catheters, the major complication rate was 0%, and the minor complication rate was 12.2% (6 of 49). Minor complications included pain and/or discomfort in the catheter site (n ¼ 3), superficial cellulitis (n ¼ 1), peristomal leakage (n ¼ 1), and partial tube fracture (n ¼ 1). Pain subsided with administration of ketorolac tromethamine 30 mg intravenously, and superficial cellulitis was managed with cefazolin. Pericatheter leakage and fractured catheters were managed by placing a 24-F balloon-retained gastrostomy catheter via the existing tract.
Follow-up The primary endpoint was procedural complications. Technical success was defined as successful placement of the gastrostomy tube. Procedural complications were divided into major and minor complications according to the Society of Interventional Radiology (SIR) classification system (16). Follow-up was conducted on a monthly basis for 6 months after catheter placement for gastrostomy-related complications or dysfunction or until gastrostomy removal if earlier than 6 months. For patients who were discharged from the hospital, phone calls were made every month to ask if they experienced any complications. The patients were not scheduled for routine catheter exchange; rather, tubes were changed only when they malfunctioned.
Statistical Analysis Comparisons between groups were performed using Student t test for continuous data and c2 or Fisher exact test for categorical data. All analyses were by intention to treat and performed with IBM SPSS Statistics for Windows Version 23 (IBM Corp, Armonk, New York). P values < .05 were considered statistically significant.
RESULTS After randomization, 50 patients underwent PG creation, and 49 patients underwent MG creation. All procedures (100%) were technically successful. The minor
Table 3. Complication Rates in Both Groups Parameters
PG (n ¼ 50)
Interval (d)
MG (n ¼ 49)
Interval (d)
1 (2%)
1
0
—
Catheter misplacement
7 (16%)
24–126
0
— 2
Minor complications Tube occlusion
.016
Skin infection
3 (8%)
1–9
1 (5%)
Peristomal leakage
3 (6%)
1–3
1 (2%)
1
Tube dislodgment Catheter site pain
2 (4%) 1 (2%)
36–92 1
0 3 (6%)
— 1–4
0
—
1 (2%)
48
1 (2%)
1
0
—
Tube fracture Transient bleeding
P Value > .999
Major complications
MG ¼ mushroom-retained gastrostomy; PG ¼ pigtail-retained gastrostomy.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2017
DISCUSSION Since RIG was first described in 1983, knowledge about the efficacy and safety of percutaneous gastrostomy has been increasing. Percutaneous gastrostomy has largely replaced surgical gastrostomy owing to its lower complication rates and less invasiveness (7,17–19). For the varied percutaneous access techniques for PG and MG, a wide range of efficacy and complication rates have been reported in retrospective and prospective studies (3,4,8,20,21). Comparative studies between PG and MG have been published; however, these studies reported conflicting results and had nonnegligible selection bias (13–15). In a recent randomized trial comparing the 2 techniques, the measured outcomes, including pain scores, technical success, and complication rates, were comparable in the 2 groups (15). However, the study included only patients with head and neck cancer who were undergoing prophylactic gastrostomy creation before radiation therapy. The present study demonstrated that the minor complication rates were significantly higher in the PG group, and the mean total fluoroscopy times were significantly longer in the MG group. In prior studies, PG and MG had equally high technical success rates (98%–100%) (3,4,15,20). In this study, the technical success rate was 100% in both groups. Concerning procedure time, MG had a significantly longer fluoroscopy time than PG, which was in accordance with a previous report (15). In some patients, retrograde catheterization of the guide wire into the esophagus or capture of the guide wire by using a snare within the insufflated stomach substantially prolonged the fluoroscopy time. The most common indication for gastrostomy was to provide nutrition for patients with head and neck cancer, esophageal cancer, cerebrovascular accident, or neuromuscular disease. This finding was also in agreement with findings of most prior studies (3,4,13). Other indications for gastrostomy included corrosive esophagitis, deep neck infection, and gastric decompression. Concerning complications, in 1 patient, the locking pigtail was located between the abdominal and gastric walls, and the gastric juice spilled into the peritoneal cavity, causing peritonitis. Although 2 gastropexy T-fasteners were routinely placed to prevent the gastric wall from retracting from the abdominal wall, misplacement of the locking pigtail still occurred. This could have been prevented if the patient had received MG because the catheter would have been deployed from the mouth to the stomach and thus would have provided better apposition between the 2 walls. Skin infection was reported to occur in 1.6%–4.4% of patients who underwent RIG (1,2,5,6). Compared with PG, the infection rates were higher in MG because of the oropharyngeal route of tube placement (14,22,23). However, in the present study, superficial cellulitis was more common in patients with PG than in patients with MG (4 in PG group vs 1 in MG group), and no major infection occurred. This was likely due to antibiotic prophylaxis and peristomal leakage.
5
Peristomal leakage was more common in the PG group, and it might have caused skin ulceration, increasing the risk for subsequent infection. Pain in the catheter site was more commonly seen in MG, and this could be attributed to the larger diameter of the MG tube. In general, catheter performance is important for patient and caregiver satisfaction, and catheter complications can adversely affect referrals for gastrostomy. In the present study, the catheter complication rate was significantly higher in the PG group. Funaki et al (4) and Laasch et al (20) also reported a significantly higher catheter complication rate in the PG group in their comparative studies. However, in the study by Bernstein et al (15), the catheter complication rate was not significantly different between the 2 techniques, and the discrepancy might be attributed to the selective patient cohort having head and neck cancer. The tube occlusion rate was the most distinctive difference between the 2 groups (7 tube occlusions in the PG group vs no tube occlusions in the MG group). Tube patency is determined as a function of diameter and length. It is plausible that MG is less likely to be occluded than PG because MG has a larger diameter and a shorter length than PG. Tube dislodgment was more common in the PG group (2 in PG group vs 0 in MG group). The stability of a gastrostomy catheter is determined by its retaining mechanism. In PG, the locking pigtail is retained only by a thread, and the thread is vulnerable to the acidic gastric juice. In MG, the internal and external retention plates provide more stability than PG, making MG more difficult to dislodge (4). Peristomal leakage was also more common in the PG group (3 in PG group vs 1 in MG group). The possible reason for this was that serial dilation of the gastrostomy tract may lead to overdilatation, making it easier for the gastric contents to leak through the catheter site (4). In MG, however, the dilator attached to the tube exactly matches the diameter of the gastrostomy catheter; thus, overdilatation rarely occurs. Seeding metastasis to the gastrostomy tract is a rare but catastrophic complication of MG with a reported incidence of 0.7%–1% (13,24,25). In a study by Cappell (25), squamous cell carcinoma, large tumor size, advanced tumor stage, tumor histology less well differentiated, and pharyngoesophageal cancer were associated with malignant seeding. In the present study, malignant tumor seeding did not occur. That may be because sterile patients with head and neck cancer or esophageal cancer were included, which means that only patients who underwent anticancer treatment and had no evidence of residual or viable tumor on cross-sectional imaging were enrolled in this study. In a study by Yang et al (13), malignant seeding developed 243 days after MG placement; thus, the relatively short-term follow-up period in the present study might have missed such a complication. However, the superiority of MG over PG in terms of overall complications seems to cancel out the low possibility of tumor seeding. This study has some limitations. First, aspiration pneumonia that was included in prior studies as a major complication was not evaluated owing to pre-existing risks
6 ▪ Pigtail-Retained Gastrostomy vs Mushroom-Retained Gastrostomy
of aspiration in this patient cohort. Second, this randomized study could not be conducted using gastrostomy catheters of the same diameter because all the available MG tubes in the country of this study were 20 F in diameter. As the tube diameter affects the likelihood of tube occlusion, excluding 7 tube occlusions leaves 10 complications in the PG group (20%) and 6 in the MG group (12%). Third, the follow-up period was 6 months, and there was a possibility of missing some complications that occurred > 6 months after the procedure. Fourth, radiation exposure in each group was not measured and compared in this study. In conclusion, both PG and MG demonstrated high technical success rates in all indications. However, this randomized controlled study demonstrates the advantages of MG over PG in terms of the overall complication rate.
REFERENCES 1. Bell SD, Carmody EA, Yeung EY, Thurston WA, Simons ME, Ho CS. Percutaneous gastrostomy and gastrojejunostomy: additional experience in 519 procedures. Radiology 1995; 194:817–820. 2. de Baere T, Chapot R, Kuoch V, et al. Percutaneous gastrostomy with fluoroscopic guidance: single-center experience in 500 consecutive cancer patients. Radiology 1999; 210:651–654. 3. Funaki B, Peirce R, Lorenz J, et al. Comparison of balloon- and mushroomretained large-bore gastrostomy catheters. AJR Am J Roentgenol 2001; 177:359–362. 4. Funaki B, Zaleski GX, Lorenz J, et al. Radiologic gastrostomy placement: pigtail- versus mushroom-retained catheters. AJR Am J Roentgenol 2000; 175:375–379. 5. Saini S, Mueller PR, Gaa J, et al. Percutaneous gastrostomy with gastropexy: experience in 125 patients. AJR Am J Roentgenol 1990; 154: 1003–1006. 6. Wollman B, D’Agostino HB. Percutaneous radiologic and endoscopic gastrostomy: a 3-year institutional analysis of procedure performance. AJR Am J Roentgenol 1997; 169:1551–1553. 7. Wollman B, D’Agostino HB, Walus-Wigle JR, Easter DW, Beale A. Radiologic, endoscopic, and surgical gastrostomy: an institutional evaluation and meta-analysis of the literature. Radiology 1995; 197: 699–704. 8. Ahmed O, Jilani D, Sheth S, Giger M, Funaki B. Radiologically guided placement of mushroom-retained gastrostomy catheters: long-term outcomes of use in 300 patients at a single center. Radiology 2015; 276:588–596.
Han et al ▪ JVIR
9. Clark JA, Pugash RA, Pantalone RR. Radiologic peroral gastrostomy. J Vasc Interv Radiol 1999; 10:927–932. 10. Rosenzweig TB, Palestrant AM, Esplin CA, Gilsdorf RB. A method for radiologic-assisted gastrostomy when percutaneous endoscopic gastrostomy is contraindicated. Am J Surg 1994; 168:587–590; discussion 590–591. 11. Szymski GX, Albazzaz AN, Funaki B, et al. Radiologically guided placement of pull-type gastrostomy tubes. Radiology 1997; 205:669–673. 12. Pitton MB, Herber S, Duber C. Fluoroscopy-guided pull-through gastrostomy. Cardiovasc Interv Radiol 2008; 31:142–148. 13. Yang Y, Schneider J, Duber C, Pitton MB. Comparison of fluoroscopyguided Pull-type percutaneous radiological gastrostomy (Pull-type-PRG) with conventional percutaneous radiological gastrostomy (Push-typePRG): clinical results in 253 patients. Eur Radiol 2011; 21:2354–2361. 14. Kuo YC, Shlansky-Goldberg RD, Mondschein JI, et al. Large or small bore, push or pull: a comparison of three classes of percutaneous fluoroscopic gastrostomy catheters. J Vasc Interv Radiol 2008; 19:557–563; quiz 564. 15. Bernstein OA, Campbell J, Rajan DK, et al. Randomized trial comparing radiologic pigtail gastrostomy and peroral image-guided gastrostomy: intra- and postprocedural pain, radiation exposure, complications, and quality of life. J Vasc Interv Radiol 2015; 26:1680–1686; quiz 1686. 16. Sacks D, McClenny TE, Cardella JF, Lewis CA. Society of Interventional Radiology clinical practice guidelines. J Vasc Interv Radiol 2003; 14: S199–S202. 17. Ho CS. Percutaneous gastrostomy for jejunal feeding. Radiology 1983; 149:595–596. 18. Cosentini EP, Sautner T, Gnant M, Winkelbauer F, Teleky B, Jakesz R. Outcomes of surgical, percutaneous endoscopic, and percutaneous radiologic gastrostomies. Arch Surg 1998; 133:1076–1083. 19. Given MF, Hanson JJ, Lee MJ. Interventional radiology techniques for provision of enteral feeding. Cardiovasc Interv Radiol 2005; 28:692–703. 20. Laasch HU, Wilbraham L, Bullen K, et al. Gastrostomy insertion: comparing the options—PEG, RIG or PIG? Clin Radiol 2003; 58:398–405. 21. Lowe AS, Laasch HU, Stephenson S, et al. Multicentre survey of radiologically inserted gastrostomy feeding tube (RIG) in the UK. Clin Radiol 2012; 67:843–854. 22. Akkersdijk WL, van Bergeijk JD, van Egmond T, et al. Percutaneous endoscopic gastrostomy (PEG): comparison of push and pull methods and evaluation of antibiotic prophylaxis. Endoscopy 1995; 27:313–316. 23. Yip D, Vanasco M, Funaki B. Complication rates and patency of radiologically guided mushroom gastrostomy, balloon gastrostomy, and gastrojejunostomy: a review of 250 procedures. Cardiovasc Interv Radiol 2004; 27:3–8. 24. Cruz I, Mamel JJ, Brady PG, Cass-Garcia M. Incidence of abdominal wall metastasis complicating PEG tube placement in untreated head and neck cancer. Gastrointest Endosc 2005; 62:708–711. 25. Cappell MS. Risk factors and risk reduction of malignant seeding of the percutaneous endoscopic gastrostomy track from pharyngoesophageal malignancy: a review of all 44 known reported cases. Am J Gastroenterol 2007; 102:1307–1311.