Patient Selection for Pediatric Gastrostomy Tubes: Are We Placing Tubes That Are Not Being Used?

1. IntroductionGastrostomy tube (GT) placement is one of the most commonly performed operations in children [Behr CA Hesketh AJ Akerman M Dolgin SE Cowles RA. Recent trends in the operative experience of junior pediatric surgical attendings: a study of APSA applicant case logs.]. GT placement is often performed in pediatric inpatients, typically in the neonatal intensive care unit or pediatric intensive care unit, when long-term enteral feeding access is anticipated. The indications for pediatric GTs vary widely from acquired needs and chronic conditions such developmental disorders causing inability to swallow, inadequate nutrient absorption, craniofacial abnormalities, gastrointestinal malformations, or failure to thrive secondary to prematurity [Backman E Karlsson AK Sjögreen L. Gastrostomy Tube Feeding in Children With Developmental or Acquired Disorders: A Longitudinal Comparison on Healthcare Provision and Eating Outcomes 4 Years After Gastrostomy., Percutaneous endoscopic gastrostomy in children.]. As opposed to GTs placed for these conditions, some patients have a more acute need for a GT, such as following renal or cardiac failure, after being diagnosed with a malignancy, or secondary to traumatic events [Percutaneous endoscopic gastrostomy in children.].While GT placement is regarded as a safe procedure, there are potential complications. These complications range from minor issues such as granulation tissue, local infections, or leakage at the GT site, to major complications such as early tube dislodgement and peritonitis [Franken J Mauritz FA Suksamanapun N Hulsker CC van der Zee DC van Herwaarden-Lindeboom MY. Efficacy and adverse events of laparoscopic gastrostomy placement in children: results of a large cohort study., Franken J Stellato RK Tytgat SHAJ van der Zee DC Mauritz FA Lindeboom MYA. Health-related quality of life in children after laparoscopic gastrostomy placement., Beres A Bratu I Laberge JM. Attention to small details: big deal for gastrostomies.]. The overall rate of minor complications ranges from 43-74% and the rate of major complications ranges from 2-32% [Franken J Mauritz FA Suksamanapun N Hulsker CC van der Zee DC van Herwaarden-Lindeboom MY. Efficacy and adverse events of laparoscopic gastrostomy placement in children: results of a large cohort study., Kutiyanawala MA Hussain A Johnstone JM Everson NW Nour S. Gastrostomy complications in infants and children., A retrospective survey of tube-related complications in patients receiving long-term home enteral nutrition., Wyrick DL Bozeman AP Smith SD Jackson RJ Maxson RT Kelley KR et al.Persistent gastrocutaneous fistula: factors affecting the need for closure.]. Even though the majority of these complications are minor and non-life threatening, they do result in high healthcare utilization, with increased phone calls, clinic and emergency department (ED) visits, and even returns to the operating room [A retrospective survey of tube-related complications in patients receiving long-term home enteral nutrition.]. The population of children where the potential risks and complications of GT placement may outweigh the benefits is an area of ongoing research [Gastrostomy tube use in children with cancer., Kim J Koh H Chang EY Park SY Kim S. Single Center Experience with Gastrostomy Insertion in Pediatric Patients: A 10-Year Review., Mahant S Jovcevska V Cohen E. Decision-making around gastrostomy-feeding in children with neurologic disabilities.].The aim of this study was to identify patients undergoing inpatient GT placement at a tertiary children's hospital to determine if there were patients who had short-term GT use (AbbreviationsGT

Gastrostomy tube

NGT

Nasogastric tube

LOS

Total hospital length of stay

ED

Emergency department

OR

Operating room

IQR

Interquartile range

2. Methods2.1 Study Design

This retrospective cohort study included all patients <18 years old who underwent gastrostomy tube (GT) placement during an inpatient stay at a tertiary children's hospital from 9/2014-2/2020. Charts were reviewed during 8/2020 to capture 6-month follow-up data post-operatively, therefore patients with GT placement after 2/2020 were excluded as there was not sufficient time to capture follow-up data. Patients were excluded if they had an elective GT placement (not during an inpatient hospital stay). Demographic information, surgical indication, date of GT placement, method of GT placement, total hospital length of stay (LOS), postoperative LOS, and returns to the emergency department (ED) or operating room (OR) for GT-related issues were collected. Patients who had GT placed by faculty of the Division of Pediatric Surgery were included. Institutional review board approval (935667-4) was obtained prior to electronic medical record review.

2.2 Outcomes

The primary outcome of this study was GT duration of use less than 6 months post-operatively. This was determined by obtaining the surgery date, time at which they stopped using the GT as reported by caregivers, and/or the time at which the GT was removed to analyze both the duration of use and the duration in place. The approximate date at which the GT was reported to no longer be used was collected as there may be a waiting period prior to GT removal in these patients, during which time the tube is not used. Short-term duration was defined use for less than 6 months post-operatively, and patients with reported discontinuation of GT use prior to 6 months post-operatively were included in this cohort even if their GT was not removed by 6 months. Long-term duration was defined as in use for greater than 6 months post-operatively. The indication for GT placement was determined by electronic medical record review of the operative report. Indications were further categorized into one of the following groups: cardiac, pulmonary, neurological, renal, gastrointestinal, genetic, anatomic, traumatic, failure to thrive, and malignancy. Total hospital charges were obtained for GT placements in the short-term cohort, which included operative and anesthesia charges. Additionally, hospital charges were collected for patients in the short-term cohort who underwent gastrocutaneous fistula closure, which included operative, anesthesia, and recovery charges as these were all outpatient procedures. Operative charges included not only procedure charges, but materials used such as sutures and OR charges. Anesthesia charges also included medications given.

2.3 Statistical Analysis

The short-term and long-term cohorts were compared using chi-square and Fisher's exact test for categorical data and Mann-Whitney-U test for continuous data. Values were considered significant at p<0.05. Analysis was performed using GraphPad Prism (version 8 for MacOS, San Diego, CA).

3. Results3.1 Overall ResultsInpatient gastrostomy tubes (GT) were placed in 142 patients during the study period. The median age overall at time of GT placement was 3 months (IQR 1.0-24.8 months) and 47.9% of patients were female. Most patients underwent laparoscopic gastrostomy tube placement (84.5%), with 14.8% undergoing open placement and 0.7% percutaneous endoscopic placement. Of the patients who underwent open gastrostomy placement, 61.9% (13/21) of them had GTs placed at the time of other abdominal surgeries. The median total hospital length of stay was 42.1 days (IQR 22.5-90.2 days) and the median postoperative LOS was 8.0 days (IQR 5.0-26.3). Demographic data is shown in Table 1.

Table 1Patient demographic information and characteristics for the overall inpatient gastrostomy tube cohort. IQR: Interquartile range; LOS: length of stay: ED: emergency department; GT: gastrostomy tube; OR: operating room; GJ: Gastrojejunostomy. For GT removal, this was evaluated at study endpoint.

Of the 142 total GT placements, 15.5% (n=22) were considered short-term as they were either no longer being used or had been removed within 6 months of their placement. The median age of the short-term patients at time of GT placement was 1 month old (IQR 1.0-100.0 months) compared to 4 months old (IQR 1.0-24.0 months) for long-term patients (p=0.16). Adolescent patients ≥12 years old were more likely to be in the short-term group (22.7% vs. 4.0%, p=0.005). There were less females in the short-term group (31.8% vs. 50.8%, p=0.11). When evaluating LOS, short-term GT patients had a significantly shorter postoperative LOS (5.0 days vs. 10.0 days, p=0.002). There were no other significant differences with regards to baseline characteristics (Table 2).

Table 2Demographic characteristics and comparison of outcomes, short-term vs. long-term patients. IQR: Interquartile range; LOS: length of stay: ED: emergency department; GT: gastrostomy tube; OR: operating room. Of note for the return to OR for gastrocutaneous fistula closure, in the short-term cohort there were 18 of the patients who had GT removed and in the long-term cohort 14 of the patients had their GT removed, therefore changing the denominator.

There was no statistically significant difference between cohorts with regards to the GT indication (p=0.30, Table 2). In the short-term group the two most common indications for placement were after a traumatic event (n=4, 18.0%) and for neurologic indication (n=5, 22.7%). All 4 patients who had GTs placed following trauma suffered traumatic brain injuries (TBIs) and the median GT duration of use for these patients was 1.4 months (IQR 0.4-1.8). The specific non-traumatic neurologic indications included neonatal hypoxic encephalopathy (n=2), holoprosencephaly (n=1), and anoxic brain injury after cardiac arrest (n=1) and diphenhydramine overdose (n=1) and the median GT duration of use for these patients was 0.9 months (IQR 0.2-1.6). Patients undergoing GT placement after a traumatic event were more likely to be in the short-term group (18.2% vs. 4.2%, p=0.03), but there were similar proportions of patients undergoing GT placement for non-traumatic neurologic indications in both cohorts (22.7% vs. 16.7%, p=0.54).

On further evaluation of the adolescent patients in the short-term cohort, the indications for placement were as follows: traumatic brain injury (n=3), anoxic brain injury after cardiac arrest (n=1), and diphenhydramine overdose (n=1). For these five patients, the median duration of gastrostomy tube use was 1.5 months. All five (100.0%) had their gastrostomy tubes removed, and the median time from insertion to removal was 2.1 months.

3.2 GT ComplicationsIn the short-term GT cohort, two patients returned to the emergency department within 30 days of discharge with GT related complications (10%); one patient for a dislodged GT and one patient for GT site infection requiring antibiotics (Table 2). In comparison, 27 patients in the long-term group returned to the ED within 30 days of discharge (22.5%), however there was no difference in the rate of return to the emergency department when comparing the two groups overall (p=0.25) or when evaluated by reason for ED visit (p=0.61).Eleven patients required a return to the operating room for repair of a gastrocutaneous fistula: six in the short-term cohort and five in the long-term cohort (Tables 2 & 3). In the short-term cohort, 18 patients had their GT removed during the study period; therefore, the percentage of patients in the short-term cohort who required return to the OR for gastrocutaneous fistula closure was 33.3% (6/18). Although gastrocutaneous fistula closure was slightly more common in the long-term cohort (5/14 patients who had their GTs removed, 35.7%), this was not statistically significant (p>0.99). Patient demographics for those who developed a gastrocutaneous fistula were similar to those who did not (Table 3). There was no difference in age, sex, or initial GT placement operation with regards to the need for return to OR for gastrocutaneous fistula closure. All patients requiring gastrocutaneous fistula closure initially underwent laparoscopic gastrostomy placement, compared to 83.3% of patients who did not require gastrocutaneous fistula closure (p=0.53). The median time from insertion to removal of the GT for those who required return to the OR for gastrocutaneous fistula closure was 5 months compared to 10.2 months for those who did not develop a gastrocutaneous fistula (p=0.58).

Table 3Demographic characteristics and comparison for those patients who developed a gastrocutaneous fistula and those who did not. To be included in this evaluation patients had to have had their GT removed. Age was at the time of initial GT placement. IQR: Interquartile range; GT: gastrostomy tube.

The median operative and anesthesia hospital charges for gastrostomy tube placements in the short-term cohort were $83,857 per patient (IQR $48,104-$153,307). Additionally, for short-term patients, the median total hospital charges for gastrocutaneous fistula closure were $29,989 per patient (IQR $27,539-$46,303). For these outpatient gastrocutaneous fistula closures, the median operative charges were $13,554 (IQR $10,913-22,849), median anesthesia charges were $10,983 (IQR $8,868-12,464), and median recovery charges were $6,703 (IQR $27,539-46,303).

4. Discussion

In this single center retrospective study, we found that 15.5% of inpatient pediatric GTs placed were in use for <6 months post-operatively with a short median duration of actual use (1.6 months). Short-term GT patients were more likely to have a GT placed following a traumatic brain injury and to be adolescents (≥ 12 years old). There were no differences in total hospital LOS, ED visits for GT-related issues, or re-operations for GT-related complications between short-term and long-term GT users. However, 33.3% of short-term patients did require a return to the OR for gastrocutaneous fistula closure following GT removal, which incurred about $30,000 per patient of potentially avoidable total operative and anesthesia charges, and an additional anesthetic and operative risk.

Few previous studies have evaluated the average indwelling time for pediatric gastrostomy tubes or the proportion of patients who use their gastrostomy tubes for a short period of time. In a study of 325 children with GTs placed, Benoit et al found that patients with cerebral palsy were least likely to have tubes removed in the first year after placement and conversely, patients with malignancies were more likely [Benoit D Wang EE Zlotkin SH. Characteristics and outcomes of children with enterostomy feeding tubes: A study of 325 children.]. Ricciuto et al in a retrospective study of 166 pediatric patients receiving tube feeding found that GTs were used for shorter durations in patients with renal failure (median 2.2 months) and malignancy (median 12.1 months). In contrast, longer durations were observed with patients who had inborn errors of metabolism, genetic disorders, or short bowel syndrome [Ricciuto A Baird R Sant'Anna A. A retrospective review of enteral nutrition support practices at a tertiary pediatric hospital: A comparison of prolonged nasogastric and gastrostomy tube feeding.]. We found that pediatric patients having a GT placed after a TBI were more likely to no longer require their gastrostomy tube within 6 months after insertion. Indications which are transient and likely to improve more commonly require only short-term duration of gastrostomy tube use, including post-traumatic states. These patients may potentially benefit from nasogastric tube feeding rather than surgical gastrostomy placement.We also found that adolescent patients with acute brain injury, either from trauma or anoxia, were more likely to require only short-term GT use. Prior to gastrostomy tube removal, patients must be able to swallow safely, have no aspiration when eating, and be able to maintain their nutrition and hydration needs [Feeding problems in infancy and early childhood: Identification and management.], but the ability to predict a time at which these will occur is difficult. Older age at time of gastrostomy tube placement may potentially make patients more likely to be able to return to a baseline ability to swallow safely when compared to younger patients, such as premature infants requiring gastrostomy tube placement, who may have never been able to tolerate oral feedings. The effect of both age and indication are compounded in this subgroup, as they had a transient indication for gastrostomy tube placement as well as older age. This subgroup of adolescent acute brain injury patients may benefit from consideration of nasogastric tube feeds even up to 1-2 months post-injury before considering surgical gastrostomy tube placement as our data showed that this subgroup used their GT for a median of 1.5 months.Overall gastrostomy tube complication rates range from 4-73% in the literature, which includes both early and late complications and stoma-related complications [Ricciuto A Baird R Sant'Anna A. A retrospective review of enteral nutrition support practices at a tertiary pediatric hospital: A comparison of prolonged nasogastric and gastrostomy tube feeding., Fortunato JE Troy AL Cuffari C Davis JE Loza MJ Oliva-Hemker M et al.Outcome after percutaneous endoscopic gastrostomy in children and young adults., Liu R Jiwane A Varjavandi A Kennedy A Henry G Dilley A et al.Comparison of percutaneous endoscopic, laparoscopic and open gastrostomy insertion in children., Ségal D Michaud L Guimber D Ganga-Zandzou PS Turck D Gottrand F. Late-onset complications of percutaneous endoscopic gastrostomy in children.]. The development of a persistent gastrocutaneous fistula after GT removal is a well-known possible complication. In a retrospective review of 44 patients, El-Rifai et al found that 16% of patients required operative intervention for a gastrocutaneous fistula. Additionally, those who developed a gastrocutaneous fistula had a significantly longer mean duration of gastrostomy placement (39 vs. 22 months), however they did not find any other significant associations between the presence of gastrocutaneous fistula and the characteristics of patients or type of gastrostomy [El-Rifai N Michaud L Mention K Guimber D Caldari D Turck D et al.Persistence of gastrocutaneous fistula after removal of gastrostomy tubes in children: prevalence and associated factors.]. In a retrospective review of 950 patients, Wyrick, et al found that 32% required gastrocutaneous fistula closure. Open GT placement operations and younger age were significantly associated with the development of persistent gastrocutaneous fistula. Lastly, Gordon et al found that the length of time the GT was in place was the most important factor predisposing children to gastrocutaneous fistulas, with 87% of patients with a GT in place for > 9 months requiring gastrocutaneous fistula closure, compared to only 6% of patients who had their GT removed within 8 months of insertion [Gastrocutaneous fistula in children after removal of gastrostomy tube: incidence and predictive factors.]. Shorter duration GTs have a lower risk of gastrocutaneous fistula development as there is less time for a gastrocutaneous fistula tract to epithelialize [Wyrick DL Bozeman AP Smith SD Jackson RJ Maxson RT Kelley KR et al.Persistent gastrocutaneous fistula: factors affecting the need for closure., Gastrocutaneous fistula in children after removal of gastrostomy tube: incidence and predictive factors., Janik TA Hendrickson RJ Janik JS Landholm AE. Analysis of factors affecting the spontaneous closure of a gastrocutaneous fistula., Gastrocutaneous fistula after tube gastrostomy. Incidence in infants and children., Haws EB Sieber WK Kiesewetter WB. Complications of tube gastrostomy in infants and children. 15-year review of 240 cases.]. We found a 34.4% risk of requiring return to the operating room for gastrocutaneous fistula closure after GT removal. Age, sex, type of GT placement operation, time to GT removal, or duration of GT use did not affect this outcome. However, we did identify 6 of 18 (33.3%) patients who had their GT removed within 6 months of placement and still developed a gastrocutaneous fistula requiring operative intervention to close. This data in conjunction with an overall rate of 15.5% of patients using their GT for less than 6 months, highlights the importance of carefully evaluating and determining the need for long-term enteral feeding. This data may be useful for providers when advising families and caregivers regarding the insertion and removal of GT, including the potential for a re-operation even if the GT is used or in place for less than 6 months.

The financial impact of gastrocutaneous fistula closure has not previously been reported. Enteral feeding via GTs will remain necessary for some neonatal and pediatric patients and pediatric surgeons will continue to be asked to evaluate for and provide this surgical enteral access for patients. However, it is imperative that pediatric surgeons are aware of the potential economic burden for the healthcare system. Our data showed that the median per patient excess hospital charges for gastrocutaneous fistula closure operation and anesthesia were almost $30,000. When evaluating the short-term GT cohort alone, excess hospital charges totaled over $200,000. This highlights the need for careful patient selection for GT placement.

This study has several limitations, including that it is retrospective in nature and completed with single center data and therefore a relatively small cohort. It is possible that institutional culture regarding timing of GT placement could be a confounding factor. Additionally, only data able to be extracted from the electronic medical records was available. We were not able to reliably evaluate the duration of nasogastric tube feeds prior to gastrostomy tube placement. Additionally, although discharging a patient with home nasogastric tube feeds potentially avoids a trip to the operating room for GT placement, there are complications associated with nasogastric feedings, including NGT dislodgments at home which may sometimes require ED visits, and accidental NGT placement into the lung by caregivers, with potential complications. We were unable to assess caregiver attitudes toward potential home NGT feedings instead of GT feedings, and this is an important aspect of the decision-making process. Furthermore, regarding NGT feeds, a prospective study to evaluate all patients placed on long-term NGT feeds or home NGT feeds and determine if there are any patient factors that make a patient more likely to require surgical GT placement is warranted. In addition, a study evaluating caregiver attitudes and perceptions of home NGT feeds compared to early surgical GT placement would be of great utility in providing guidance to caregivers facing this decision. Despite these limitations, this study is the first to examine factors that may be associated with short-term gastrostomy tube use or removal within 6 months and the first to evaluate this economic impact. Future studies should focus on assessing provider and patient perspectives on short-term NGT feedings to guide shared decision-making in this area.

留言 (0)

沒有登入
gif