If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Reduced complication rate after implementation of a detailed treatment protocol for percutaneous endoscopic gastrostomy with T-fastener fixation in pediatric patients: A prospective study
Department of Pediatrics, Oslo Universitetssykehus, Ullevål Sykehus, Barne- og Ungdomsklinikken, Oslo University Hospital, PB 4956 Nydalen, Oslo 0424, Norway
Department of Pediatrics, Oslo Universitetssykehus, Ullevål Sykehus, Barne- og Ungdomsklinikken, Oslo University Hospital, PB 4956 Nydalen, Oslo 0424, Norway
Department of Gastrointestinal and Pediatric Surgery, Oslo University Hospital, NorwayDepartment of Gynecology and Obstetrics, Sørlandet Sykehus Kristiansand, Norway
Department of Pediatrics, Oslo Universitetssykehus, Ullevål Sykehus, Barne- og Ungdomsklinikken, Oslo University Hospital, PB 4956 Nydalen, Oslo 0424, Norway
Department of Pediatrics, Oslo Universitetssykehus, Ullevål Sykehus, Barne- og Ungdomsklinikken, Oslo University Hospital, PB 4956 Nydalen, Oslo 0424, Norway
Percutaneous endoscopic gastrostomy with push technique is more complex and has a higher number of early complicatons compared to pull technique.
•
What new information is contained in this article?
•
A detailed treatment protocol and limited number of physicians performing the procedure reduced the complication rate related to tube dislocations and migrated T-fasteners.
Abstract
Background/aims
Percutaneous endoscopic gastrostomy with push technique (PEG-T) is increasingly used in pediatric patients. In a retrospective study of PEG-T (cohort 1) we reported frequent complications related to T-fasteners and tube dislodgment. The aim of this study was to assess complications after implementation of a strict treatment protocol, and to compare these with the previous retrospective study.
Materials and methods
The study is a prospective study of PEG-T placement performed between 2017 and 2020 (cohort 2) in pediatric patients (0–18 years). Complications were recorded during hospital stay, fourteen days and three months postoperatively, graded according to the Clavien-Dindo classification and categorized as early (<30 days) or late (>30 days).
Results
In total 82 patients were included, of which 52 (60%) had neurologic impairments. Median age and weight were 2.0 years [6 months-18.1 years] and 13.4 kg [3.5–51.5 kg], respectively. There was a significant reduction in median operating time from 28 min [10–65 min] in cohort 1 to 15 min [6–35 min] in cohort 2 (p<0.001), number of patients with early tube dislodgement (cohort 1: 9 (10%) vs cohort 2: 1 (1%), p = 0.012), and number of patients with late migrated T-fasteners (cohort 1: 11 (13%) vs cohort 2: 1 (1%), p = 0.004).
Conclusion
We experienced less migrated T-fasteners and tube dislodgment after implementation of strict treatment protocol.
]. This push technique allows one-step insertion of a balloon gastrostomy tube or button, and reduces the need for sedation or anesthesia which is often required to replace a PEG-P tube with a low-profile button [
]. Furthermore, removal of a PEG-P gastrostomy may cause trauma to the stoma channel with post-procedural pain and increased risk of stoma infection, which is avoided with the PEG-T. In a study comparing PEG-T and PEG-P insertion in children, there was a marked reduction in peristomal infections and costs with the PEG-T [
In 2010 the push technique was introduced at our hospital. The PEG-Saf-T-pexy was placed according to the manufacturer's instructions. The first 87 children experienced a high number of postoperative complications including tube dislodgments and T-fastener related complications, of which many required general anesthesia [
]. As a result, we developed a detailed treatment protocol (Table 1). The aim of the present study was to prospectively record if implementation of this new protocol reduced complication rate. We hypothesized that there would be less tube dislodgments and T-fastener related complications with the new treatment protocol.
Table 1.Detailed treatment protocol used in cohort 2 during the percutaneous endoscopic gastrostomy Saf-T-Pexy procedure with direct gastrostomy-button insertion. Changes from cohort 1 are marked with bold text.
1.
Iv. prophylactic antibiotic (Cefalotin) is given prior to the procedure.
2.
Mark with a pen the midline and the costal arch on the patient.
3.
Identify an area of definite good illumination and good indentation.
4.
Inject local anesthesia around the gastrostomy site and measure the channel length by the “needle method”.
5.
Insert the suture needles one by one in a trigone around the gastrostomy site under gastroscopic control, and avoid traction. Suture needles should be placed approximately 1.5 cm away from the gastrostomy, to avoid the titanium rods to perforate the balloon. The needle insertion is easier if the stomach is moderately distended by air.
6.
Close the suture stopper around 2-3 mm above skin level under gastroscopic vision assuring the T-bar is in level with undisturbed mucosa and avoid T-bar being drawn into the mucosa.
7.
Incise the skin and then insert the red needle centrally in the trigone (i.e. gastrostomy site).
8.
Insert the guidewire with the smooth end into the stomach.
9.
Sequential dilatation in 5 steps, avoiding dilating the stoma channel too much and to avoid injury to the posterior wall structures.
10.
Ensure the peel-off is in place, take out the dilatator without removing the guide wire and inject a small amount of lidocaine gel in order to stop air leakage from the stomach.
11.
Insert the 14 Fr gastrostomy with straight extension tube (which opens the valve) over the guide wire and fill the balloon with sterile water according to the manufacturer’s recommendations under gastroscopic control. We block the inlet of the balloon with tape to prevent damage to the balloon by subsequent accidental injection of nutrition in the balloon.
12.
Control by gastroscopy that there is no leakage of water or bleeding.
13.
Change the extension tube to the angled type which will be left open after the procedure.
14.
Put on split bandaid and secure the button and extension tube by systematic Micropore taping in order to avoid accidental dislodgement.
Patients referred to Oslo University Hospital, Ullevål, for gastrostomy placement include all local patients and children from centers in the south eastern region of Norway which either do not perform procedures in children with severe comorbidities, or do not place gastrostomy in children under two years. PEG-T is the preferred technique for gastrostomy placement at the hospital, except in children with previous abdominal surgery, ventriculoperitoneal shunt, severe scoliosis, and hepatomegaly. In the current study, all patients less than18 years undergoing PEG-T at Oslo University Hospital, Ullevål between March 2017 and March 2020, were eligible for inclusion and referred to as cohort 2. The only exclusion criterion was need for an interpreter. Patients were included when admitted for the procedure. The department's surgical logbook was reviewed May 2021 to crosscheck the completeness of inclusion. Retrospective data from the first 87 patients (cohort 1) were used for comparison [
In cohort 1, insertion of PEG-T with three T-fasteners was performed according to oral and video instructions from the manufacturer, and the recommendations from the paper by Gothberg and Bjornsson [
]: The position of the gastrostomy button was determined by transillumination, and the external suture locks were closed 2 mm from skin level with Saf-T-Pexy T-fasteners. A skin incision was made and subsequent stoma tract dilatation was performed using a serial tract dilator with integrated peel-away sheath. The size of the gastrostomy button was determined by stoma trach measurement performed by an over-the-wire stoma measuring device. If suture locks had not detached spontaneously, parents were instructed to contact the local hospital for removal of the suture locks after 3,4 weeks.
In cohort 2 the PEG-T technique was refined and detailed into several steps during surgery (Table 1). Particularly, there was a particular focus on not dilating the stoma channel and tightening the T-fasteners too much. The stem length was chosen after measuring length by the needle technique (Table 1, image 2).
One surgeon and one gastroenterologist are the traditional combination of staff when performing PEG in our hospital. In the first cohort six surgeons and three endoscopists performed the procedure. In cohort 2 a dedicated team (two surgeons, three endoscopists) with previous experience with the technique performed the procedures. If the team members were absent, patients were allocated to PEG-P. During 2017–2020, a total of nine patients underwent PEG-P, because the dedicated team was absent.
Patients in both cohorts received prophylactic antibiotic (cefalotin or cefazolin) prior to or during the procedure, and all procedures were performed under general anesthesia. Local anesthesia (bupivacaine) was injected around the stoma channel. Two types of buttons were used during both study periods (MicKey and MiniOne buttons). The standard gastrostomy device was a 14 Fr. low-profile button. In a few children a 14 Fr. balloon gastrostomy tube (Mic Bolus) was placed if assessed appropriate by the gastroenterologist or owing to lack of a well-fitting low-profile tube. Written information and practical training on stoma care were given during the hospital stay. Parents and the local hospitals were instructed to remove the suture locks 14 days postoperatively.
2.3 Postoperative treatment and complications
All patients in both cohorts followed a standardized protocol for analgesia and initiation of tube feeding. Paracetamol were given to all patients, while non-steroidal anti-inflammatory drugs (NSAID) and opioids were given if needed. The gastrostomy was kept open for the first four hours, and after six hours liquids were started. Tube feeding started the first postoperative day. Patients were discharged to their local hospital or home, depending on whether the patient was a local patient or referred from another hospital. In cohort 1, sutures and external suture locks were removed after 3,4 week. In cohort 2, suture locks were cut after two weeks. When stoma competent personnel contacted all parents after 14 days to record outcome, they also checked that sutures had been cut. Advice and initiation of treatment for PEG-related complications were started per phone or patients were taken in for a clinical consultation, if needed.
2.4 Collection of data
Patient demographics, presence of a nasogastric tube at admittance, feeding route, and underlying diagnosis most likely contributing to the patients’ feeding problems were registered at admission. Duration, type and size of inserted gastrostomy, perioperative complications, and any concomitant procedures were recorded at the end of the surgery. Operative time was not registered in patients undergoing concomitant procedures. Postoperative complications, analgesia, feeding tolerance, and length of stay were recorded and summarized at discharge. Postoperative follow-up was done 14 days and 3 months after the procedure, either by clinical visit or phone interview. Except for the perioperative data, medical professionals who did not participate in the PEG procedure performed data collection. All postoperative complications were recorded regardless of cause, and complications were graded according to the Clavien-Dindo classification for surgical complications [
]. Stoma infections were categorized as grade I if treated with topical agents or as grade II if systemic antibiotics were given. Complications related to the gastrostomy stoma and T-fasteners were registered as separate items. All types of complications were recorded during the hospital stay, also in patients undergoing additional procedures. After the first 30 postoperative days, only gastrostomy related complications were registered.
2.5 Ethical approval
The study was approved by the hospital's Commission for Personal Security; study number 2017/1584. All parents signed a letter of consent.
2.6 Outcome measures and power calculations
In cohort 1, 23% of patients experienced early tube dislodgement or migrated T-fasteners, and these two complications were the main causes for Clavien-Dindo grade IIIb complications. The main aim of this study was to assess if a detailed treatment protocol could reduce these complications, and consequently the number of patients undergoing a second anesthesia. A complete elimination of these complications was desired, but from a clinical point of view, a reduction of more than two-thirds in the rate of early tube dislodgements and migrated T-fasteners was considered acceptable. It was calculated that a sample size of 77 patients in each group was necessary to detect a reduction in these two complications of two-thirds or more (23% vs 7%), given a significance level of P < 0.05, and statistical power of 0.8.
2.7 Statistics
In the retrospective study (cohort 1) complications were categorized as early (<30 days) or late (>30 days) [
]. For comparison we merged all complications registered until 14 days postoperatively and categorized them as early. Complications registered at three months were categorized as late complications. Chi-square test was used to compare complication rates between the two cohorts. Mann Whitney U test was used for comparison of operation time. All analyses were performed by IBM SPSS software for Windows version 26 (IBM Corp, Armonk, NY). P values <0.05 were considered statistically significant.
3. Results
3.1 Patients
Ninety-one patients underwent PEG-T in the inclusion period, and three patients were excluded owing to the need of an interpreter. Of the eligible 88 patients, 6 patients were not included because the study coordinator was not present. Consequently, 82 (93%) were included. There were no differences in patient demographics between cohorts 1 and 2 (Table 2).
Table 2.Demographic data of patients included in the present study (2017–2020) compared with the first series of patients (2010–2014). Cohort 1 is data from Kvello M et al 2018 [
In addition to neurologic impairment, underlying conditions in cohort 2 included ex-prematurity without neurological impairment (n = 7; 9%), respiratory tract disorders (n = 5; 6%), and other diseases (n = 14; 17%). In cohort 2 13 (16%) children were completely tube fed, 33 (40%) had troublesome retching and 24 (30%) used proton pump inhibitor (PPI) and/or histamine type 2 receptor antagonist (H2-blocker) preoperatively. These data were not recorded in cohort 1.
3.2 Perioperative data
Twenty-two (27%) patients in cohort 2 underwent a concomitant procedure; biopsies from the upper gastrointestinal tract (n = 16), various urogenital procedures (n = 2), insertion of central venous catheter (n = 1), frenulotomy of tongue (n = 1), ano-/rectoscopy (n = 1), and inguinal hernia repair (n = 1). The duration of the PEG-T procedure in the remaining 55 patients was median 15 min [6–35 min], compared to 28 min [10–65 min] in cohort 1 (p<0.001). All but one child in cohort 2 had three T-fasteners placed. This child was 6.3 kg and got only two fasteners because of limited area with transillumination. A 14 French gastrostomy button was placed in 76 (93%) and a 12 Fr. button in 2 (2%) children. In 4 (5%) children a 14 Fr. balloon gastrostomy tube was placed. One child on ketogenic diet experienced hypoglycaemia perioperatively owing to fasting before the procedure.
3.3 Early complications
Median length of hospital stay was 2.0 [1–14] days. There were 68 early complications in 44 children (54%), of which 7 complications were related to other procedures and underlying comorbidity in the patients (Table 3). Granulation tissue and leakage were the most frequently recorded grade I complications. Infection was the most common grade II complication, and nearly half of the infections were related to the suture locks. One child had a length of stay of 14 days owing to stoma leakage and peristomal infection. One child had the tube accidentally dislodged and needed general anesthesia to dilate the stoma channel and reinsert the gastrostomy button.
Table 3.Complications in children undergoing percutaneous endoscopic gastrostomy (PEG) using the push-PEG-T-Saf-pexy procedure. Some patients had more than one complication. Complications were graded according to Clavien-Dindo classification, and categorized as early (<30 days) or late (>30 days). Cohort 1 is data from Kvello M et al 2018 [
A total of 80 complications were recorded in 52 (63%) children at the last follow-up (Table 3). The majority were grade I, of which hypergranulation and stoma leakage were the most frequent. Seventeen children (21%) had infections requiring systemic antibiotics. There were 5 grade IIIb complications recorded in 4 (5%) children, of which only one had suspected migrated T-fasteners. However, no dislodged titanium rods were identified. One child who had the gastrostomy tube inserted through the colon underwent laparotomy when the misplacement was revealed during elective tube change.
3.5 Outcome after change of treatment protocol
The rate of migrated T-fasteners were significantly reduced, from 11 (13%) patients in cohort 1 to 1 (1%) patient in cohort 2 (p = 0.004) [
]. The frequency of early tube dislodgment was also significantly reduced from 9 (10%) in cohort 1 to 1 (1%) in cohort 2 (p = 0.012). There were no differences in the number of patients experiencing early or late grade II complications (early: cohort 1; n = 19, 22% vs cohort 2; n = 18, 22%, p = 0.986, late: cohort 1; n = 10, 14% vs cohort 2; n = 17, 21%, p = 0.101). The number of patients with early grade IIIb complications was not significantly reduced between cohort 1 (n = 6, 7%) and cohort 2 (n = 1, 1%) (p = 0.064). There was also no difference between the total number of patients with early and late grade IIIB complications between cohort 1 (n = 12, 14%) and cohort 2 (n = 5, 6%) (p = 0.096).
4. Discussion
In this prospective study, we found that a detailed treatment protocol significantly reduced the rate of some severe complications after PEG-T insertion. Importantly, the frequency of early tube dislodgments, late T-fastener related complications and need for a second anesthesia owing to postoperative complications were reduced. We believe that increased attention to certain steps in the procedure and an experienced team were important for better results, but it is impossible to outline which one were most important. Certain steps of importance included less dilatation of the stoma channel, minimal traction on the T-fasteners when dilating the stoma channel and tightening them, and cutting the T-fasteners after 2 weeks in all patients. In the previous cohort, T-fasteners were cut 3,4 weeks postoperatively, unless they already had detached spontaneously [
]. The decision to cut T-fasteners earlier was based on the assumption that longer traction increased the risk of migration, as one study found significantly fewer migrated T-fasteners when T-fasteners were cut five days after the procedure [
]. However, the optimal timing to cut the T-fasteners remains unclear. There are no data showing how quick the stomach adheres to the abdominal wall and how early the tube may be safely replaced after the T-fasteners are cut. The rate of T-fastener related complications now is in line with others [
The rate of early tube dislodgement was significantly reduced after introduction of the detailed treatment protocol. Although the push technique requires more dilatation of the stoma channel to fit the balloon gastrostomy tube than PEG-P, one of the major changes in the new treatment protocol was to avoid dilating the stoma channel too much. Another change was making sure the T-fasteners were not placed too close to the stoma, to avoid the titanium rods to perforate the balloon. In cohort 2 only one child (1%) experienced tube dislodgement, which is comparable to other studies [
Early peristomal infection needing antibiotics was common, but the frequency was not significantly changed after the new treatment protocol. The occurrence of early infections in cohort 2 is higher than in previous PEG-T studies [
]. Grade 1 complications of granulation tissue and leakage were more frequently reported in cohort 2 than in cohort 1. We believe that many minor complications were underreported in cohort 1. Hence, this change in frequencies is probably a result of study design and not a true increase.
There are many studies showing that complex procedures take longer time and have a higher complication rate before sufficient experience is achieved and a plateau is reached [
Systematic review of learning curves for minimally invasive abdominal surgery: a review of the methodology of data collection, depiction of outcomes, and statistical analysis.
]. The surgeons prepared by oral and video instructions from the manufacturer and becoming familiar with the procedure kit. Shortly after we discovered our results from cohort 1. Gothberg et al. published a paper showing that duration of surgery was nearly halved when comparing the first 100 procedures with the second 100 PEG-T insertions [
]. In our first cohort, we did not find any difference in duration of surgery or early postoperative complications when splitting the 80 patients into two groups [
]. In cohort 2, caseload per surgeon was increased and followed a protocol to guide attention to specific pit falls. In this cohort, procedure time and rate of some procedure specific complications were significantly reduced compared to the first 80 patients. We assume that our results show the same learning curve as Gothberg and Bjornsson that our surgeons became proficient during cohort 1 [
]. How fast the plateau on the learning curve is reached is determined by many factors, including the individual surgeon's skills and experience as well as the complexity of the procedure. Until surgical proficiency is reached, the patient is at increased risk for advert events including anesthesia related complications owing to prolonged operation time, and postoperative complications [
]. Consequently, one should carefully consider the severity of complications which may occur during the learning process before introducing a new procedure [
We believe that the most important factors to quickly reach proficiency and thereby avoid complications when introducing the PEG-T technique, is to keep it on few, dedicated hands, prepare well by watching instruction videos, become familiar with the surgical kit, and be particularly attentive when performing the steps listed in Table 1. Furthermore, we think that new teams should be supervised by an experienced team when performing PEG-T in children until the skills have been achieved. Although PEG-T has some important benefits compared to PEG-P, each center must consider whether they have sufficient volume of this procedure to get experience in a reasonable time, even when efforts to minimize the effects of the learning curve on patient outcome are introduced. Lastly, outcomes should always be carefully monitored when introducing new techniques.
Strengths of this study are the prospective design and the meticulous recording of all complications. We consider the results and the study population to be representative, as over ninety percent of eligible patients were included in both cohorts. To reduce the risk of bias, postoperative data collection, assessment, and statistical analyses were performed by medical professionals who did not participate in the PEG-T procedure themselves. The use of the Clavien-Dindo classification allows a more simplified and objective comparison between study groups, and it is also easier to compare outcome between various studies. Limitations of the study include different study designs between the two cohorts, where minor complications may have been missed in cohort 1.
To conclude, introduction of a detailed treatment protocol and limiting the number of physicians inserting the gastrostomies significantly reduced the complication rate. Our results highlight the importance of meticulous follow-up when new techniques are introduced.
Financial disclosure
The authors declare no financial relationships relevant to this article.
Declaration of Competing Interest
The authors declare no conflicts of interest relevaSconnt to this article.
Acknowledgments
A special thanks to the pediatric clinical research department at Oslo University Hospital for supporting the practical parts of this study, and to stoma nurses Marie Erdal Foss and Hilde Nevland for collecting some of the follow-up data. Lastly, we want to thank all parents and children for participating in this study.
Systematic review of learning curves for minimally invasive abdominal surgery: a review of the methodology of data collection, depiction of outcomes, and statistical analysis.
00258-5&id=gr1.jpg){kind=link}
00258-5&id=gr2.jpg){kind=link}
00258-5&id=gr3.jpg){kind=link}
00258-5&id=gr4.jpg){kind=link}
00258-5&id=gr5.jpg){kind=link}