Semi-Solid Nutrients for Prevention of Enteral Tube Feeding-Related Complications in Japanese Population: A Systematic Review and Meta-Analysis

The aim of this systematic review was to assess the best available evidence on semi-solid nutrients for prevention of complications associated with enteral tube feeding (ETF). PubMed (MEDLINE), EMBASE, Cochrane Central Register of Controlled Trial, Ichushi-web, and World Health Organization International Clinical Trials Registry Platform databases were searched for relevant articles. Randomized controlled trials (RCTs), cluster RCTs, and crossover trials comparing the effects of semi-solid nutrients with those of control interventions in patients on ETF were included in the review. The primary outcome was development of gastroesophageal reflux (GER). Eight RCTs and five crossover trials involving 889 study participants in total were examined via meta-analysis. The meta-analysis showed that semi-solid nutrients significantly decreased the risk of GER (risk ratio 0.39; 95% confidence interval (CI) 0.21 to 0.73) and the GER index (mean difference −2.93; 95% CI −5.18 to −0.68). Dwell time in the stomach was significantly shortened (standardized mean difference (SMD) −0.50; 95% CI −0.99 to −0.02), as was care time defined as the time needed to prepare and administer the nutrient solution (SMD −8.02; 95% CI −10.94 to −5.10). Semi-solid nutrients significantly decrease the risk of GER and the dwell time in the stomach in adult patients.


Introduction
Enteral tube feeding (ETF) plays a major role in the management of patients with poor voluntary food intake, those with chronic neurological or mechanical dysphagia or gut dysfunction, and those who are critically ill [1]. ETF is widely used in acute and subacute care, rehabilitation, long-term care, and home settings [2]. ETF is provided to maintain gut integrity, modulate stress and the systemic immune response, and attenuate disease severity [3][4][5][6]. The logistics of administering ETF may appear less complex than those involved in parenteral nutrition, but serious harm and even death can result from the adverse events that can occur during the process of ordering, administering, and monitoring ETF [7].
The potential adverse events associated with ETF include clinical complications such as gastrointestinal complications [2]. High gastric residuals have been reported to occur in 39% of patients receiving ETF, constipation in 15.7%, diarrhea in 2.0-95.0%, abdominal distention in 12.2%, vomiting in 13.2%, nausea in 10-20%, regurgitation in 0.4-6.0%, and pneumonia in 12.5-30.0% [1,6,[8][9][10]. Furthermore, withdrawal of ETF as a consequence of uncontrollable gastrointestinal complications has been reported to occur in 15.2% of patients [8]. These complications prevent completion of ETF in situations where nutritional management is necessary. Therefore, it is crucial to prevent gastrointestinal complications.
It is unclear whether use of semi-solid nutrients is effective for prevention of complications in patients with ETF. Semi-solid nutrients are used as a nutritional management method intended to prevent these complications by increasing the viscosity or changing the shape of the nutrient material injected through the tube. In Japan, semi-solid nutrient pharmaceuticals were launched in 2014. They are inexpensive, covered by insurance, and can be used in a variety of settings. In recent years, semi-solid nutrient formulations with higher viscosity than liquid nutrients have been reported to be effective in reducing gastroesophageal reflux (GER) in patients who have undergone percutaneous endoscopic gastrostomy [11][12][13].
A systematic review of trials examining the efficacy and the acceptability of semi-solid nutrients would be informative and useful for clinicians and researchers. Previous studies have reached inconsistent conclusions due to differing results, study quality, and effect sizes. A systematic review can provide comprehensive evidence through systematic searching, identification, selection, evaluation, and integration.
The aim of this systematic review was to assess the best available evidence on semi-solid nutrients for prevention of complications associated with ETF. We hope the findings will guide future directions of research in this field.

Materials and Methods
The protocol for this review was prospectively registered with the PROSPERO database for systematic reviews (CRD42018110004).

Types of Studies
Randomized controlled trials (RCT)s, cluster RCTs, and crossover trials were eligible for inclusion in the meta-analysis. Abstracts and non-English language publications were included.

Types of Participants
Studies involving patients aged ≥20 years were eligible for inclusion regardless of sex or indication for ETF.

Types of Interventions and Comparisons
We reviewed studies that included semi-solid nutrient interventions intended to prevent complications associated with ETF. Control interventions were defined as liquid nutrients. The major difference between the two interventions is the dynamic viscosity of the feeds. A semi-solid nutrient is defined as a nutritional management method that entails "increasing the viscosity or changing the shape of a nutrient material injected through a tube" [14].
Semi-solid nutrients can be divided into two types. The first are commercially available semi-solid nutrients. More than 10 of these products are available on the market in Japan, including semi-solid nutrient pharmaceuticals that were first launched in Japan in 2014. Semi-solid formulas and food additives are used to increase the viscosity of liquid formulas by 2000-20,000 mPa·s. The package insert for semi-solid nutrients pharmaceuticals typically states, "The standard dose for adults is 1200-2000 kcal/day administered several times a day in divided doses directly into the stomach via a gastrostomy tube. The administration time is 2-3 min per 100 g (6-9 min per 300 g), and the maximum single dose is 600 g".
The second type are nutrients whose viscosity has been adjusted by adding a thickener or gelling agent (agar, gelatin, pectin, carrageenan, starch, guar gum, or xanthan gum) to commercially available liquid nutrients. Semi-solid nutrients can be directly administered via a feeding tube or nutrients, and the thickener or gelling agent can be separately administered through the feeding tube such that partial solidification occurs in the stomach.

Outcomes
The primary outcome was development of GER. Secondary outcomes were rates of pneumonia, diarrhea, constipation, pressure ulcer, leak from the gastrostomy tube, dwell time in the stomach, care time (time taken to prepare and administer the nutrient solution), rehabilitation time, activities of daily living, and medical costs.

Search Strategy
All relevant published studies were identified by searching the following databases: PubMed (MEDLINE), Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE, and Ichushi-web, which is a Japanese journal database. All searches were performed for publications from the inception of each database until July 2019 ( Figures A1-A4). We also searched for ongoing studies using the World Health Organization International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp) ( Figure A5).

Selection of Studies
Two authors (Y.K., C.S.) independently reviewed all potentially eligible studies by examining the title and the abstract and, where necessary, the full-text version of the paper. If agreement could not be reached by discussion, a third author (R.M.) made the final decision about eligibility.

Data Extraction and Management
Two authors (Y.K., C.S.) worked independently and used a standardized form to extract study characteristics and outcome data from the included studies. The original data were checked if a discrepancy was found, and any disagreements were resolved by a third author (R.M.).

Risk of Bias Assessment
The methodological quality of the selected studies was assessed as recommended by the Cochrane Review Groups [15]. The same two authors (Y.K., C.S.) independently performed the quality assessment. Any disagreements between authors was resolved by discussion. We contacted the authors of the primary studies in the event of missing data. A risk of bias table was created that included a description and judgment (low risk, high risk, or unclear risk) of the following seven domains for each of the included studies: (1) random sequence generation; (2) allocation concealment; (3) blinding of participants and personnel; (4) blinding of outcome assessment; (5) incomplete outcome data; (6) selective reporting; and (7) other sources of bias.

Statistical Analysis
For all outcomes related to continuous data, we calculated a pooled estimate of the standardized mean difference (SMD) with a 95% confidence interval (CI). The mean difference (MD) was used for continuous data if the outcomes were measured in the same way between trials. Dichotomous data are presented as the summary risk ratio (RR) with a 95% CI. We used fixed-effect meta-analysis of the combined data where it was reasonable to assume that the studies were estimating the same underlying treatment effect. We used the I2 statistic to assess heterogeneity. An I2 of 50% was considered to reflect substantial heterogeneity. If the I2 was more than 50%, we used random-effects analysis to combine the data. The threshold for significance was set at p = 0.05. The EZR package was used for all statistical analyses [16].

Results
The protocol for this review was prospectively registered with the PROSPERO database for systematic reviews (CRD42018110004).

Outcomes
Of the four trials that assessed GER [11][12][13]18], three categorized it dichotomously as present or absent [11,13,18], and the remaining trial used the GER index [12]. Intragastric and esophageal distribution were monitored using a scintillation camera. Reflux of contrast agent into the esophagus was observed on radiographic examination. Thirty minutes after administration of the contrast agent, a computed tomography scan of the esophagus was performed with a slice thickness of 1 cm. GER was confirmed if the Hounsfield number exceeded 100 in each examined slice [11]. The upper gastrointestinal tract was observed radiologically from onset to 1 min after the end of administration of the contrast agent. GER was considered present if any reflux of contrast agent into the esophagus was observed [13]. Intragastric and esophageal distribution were monitored using a scintillation camera in the supine position. The radioactivity of the esophagus and the stomach was determined at a rate of one frame every 150 s using a scintillation camera for up to 90 min after bolus infusion of 200 mL radiolabeled liquid or semi-solid nutrients through percutaneous endoscopic gastrostomy. The GER index was defined as the maximal percentage of esophageal radioactivity count to total infused radioactivity [12].
Pneumonia, diarrhea, constipation, and leak from the gastrostomy site were also categorized dichotomously (present or absent). Pneumonia was diagnosed based on the following: the Japanese Respiratory Society guidelines [17]; clinical symptoms confirmed by radiologic findings or detection of enteral feed material in aspirate from the trachea26; body temperature ≥37.5 • C; respiratory symptoms; abnormal blood test results, including for white blood cell count and C-reactive protein level; and infiltrative shadow observed on chest radiography or chest computed tomography [27].
Diarrhea was defined as watery or muddy stools [18,24], watery or muddy stools more than three times a day [21]; a King's Stool Chart score of ≥15 [26], or watery or soft stools more than five times a day [27]. Dwell time in the stomach was assessed by Tlag (time to peak excretion) [25], T1/2 (half-emptying time) [25], and gastric emptying time determined as the time required for 50% of the initial radioactivity to empty from the stomach [12]. Care time included the amount of time needed to prepare and administer the nutrients and water [22,24,27,28].

Constipation
Only one trial [28] (112 participants) reported data on the incidence of constipation. We found that ETF with semi-solid nutrients did not significantly reduce the risk of constipation (RR 0.25; 95% CI 0.03-2.17; p = 0.208).

Leak from Gastrostomy Site
Only one trial [18] (with 14 participants) reported data on the incidence of leak from the gastrostomy site. Meta-analysis showed that ETF with semi-solid nutrients did not significantly decrease the risk of leak from the gastrostomy site (RR 0.20; 95% CI 0.01-3.50; p = 0.271).
3.6.8. Pressure Ulcer, Rehabilitation Time, Activities of Daily Living, and Medical Costs None of the studies reported on pressure ulcer, rehabilitation time, activities of daily living, or medical costs.

Summary of Results
Our literature search identified 15 eligible randomized trials with a total of 946 participants. Meta-analysis was possible for GER, pneumonia, diarrhea, constipation, leak from the gastrostomy site, and dwell time in the stomach. We found that administration of semi-solid nutrients significantly decreased the risk of GER, dwell time in the stomach, and care time but had no statistically significant effect on rates of pneumonia, diarrhea, constipation, or leak from the gastrostomy site. However, all outcomes were measured in a small number of trials. Therefore, our findings should be interpreted with caution because the quality of the evidence was often unclear and differed among the outcome measures.

Overall Completeness and Applicability of Evidence
The results of the review are limited by a number of factors. First, six of the studies were published in only abstract form, which explains why many risks of bias were evaluated as unclear. Second, all patients included in the meta-analysis were Japanese, which limits the generalizability of the data to other populations. Therefore, in the future, research targeting non-Japanese participants should be conducted. Third, there may have been differences in the types of patients included in the studies. Known indications for ETF include stroke and dementia, and almost all the patients had one of these conditions; however, comorbidities were less clear. Furthermore, the effects of medication could not be investigated. Fourth, the patients either already had an established gastrostomy or had a new percutaneous endoscopic gastrostomy, thus the backgrounds of the patients were not exactly the same. Fifth, in the trials that evaluated semi-solid nutrients, none of the products were identical in composition, formulation, or quantity. Stronger evidence would be obtained by performing trials with the same semi-solid nutrients. Sixth, the definition of complications was inconsistent and was unclear in some studies. We contacted the authors by post and e-mail to clarify definitions used but three corresponding authors could not be reached. The exact definition of complications should be standardized for future trials to obtain high-quality results. Seventh, the longest follow-up period was 28 days. Therefore, the long-term effects of this intervention remain unclear. Finally, the meta-analysis for GER as a primary outcome was performed for only four trials with a total of only 210 participants. Therefore, it may be premature to perform meta-analysis of the data for GER at this stage. However, it is important to consider the best available evidence in order to indicate the directions of future research.

Quality of Evidence
In general, the quality of reporting was poor. Most trials reported random assignment of patients, but the methods of randomization were not described in full detail. Although all trials used semi-solid nutrients in the intervention group, the success of blinding was not recorded. Studies of semi-solid nutrients are difficult to perform with blinding of participants and personnel because they have higher viscosity and are clearly different in appearance from liquid nutrients. There was large heterogeneity between the studies, particularly for care time. The potential problem associated with this heterogeneity relates to the definition of care time. The definition of care time was variously "infusion time" [27,28] or "time for preparing nutrients and for administering nutrients and water" [24] and differed from study to study.

Potential Biases in the Review Process
We were unable to assess potential reporting bias because of the small number of studies in the review, which prevented us from constructing a funnel plot. Second, the included patients were adults aged ≥20 years and all had received ETF in a hospital setting. Therefore, our results cannot be applied to younger patients or those using ETF at home. Finally, a potential source of bias of this review may originate from the search strategy, which detected only English and Japanese language publications.

Agreements and Disagreements with Other Studies or Reviews
To our knowledge, no review has previously examined the impact of semi-solid nutrients on complications of ETF.

Conclusions
This systematic review and meta-analysis found that semi-solid nutrients significantly decreased the risk of GER and also decreased the dwell time in the stomach and the care time for ETF in adult patients. However, use of semi-solid nutrients did not decrease the rates of pneumonia, diarrhea and constipation, or leak from the gastrostomy site. The limitations of this study were the small number of trials included and the lack of high-quality evidence. Large-scale, high-quality, prospective randomized studies are needed to further investigate the impact of semi-solid nutrients on complications in patients receiving ETF. Currently, some clinical trials are under way (JPRN-UMIN000006732, ClinicalTrials.gov. NCT03017079) and further evidence is awaited.
Author Contributions: Y.K., C.S., H.W., K.M., K.S. and R.M. contributed to the analysis of the data; Y.K., C.S., H.W., K.M., K.S. and R.M. contributed the reagents/materials/analysis tools; Y.K. and R.M. contributed to the statistical analysis and Y.K. drafted the manuscript. All authors critically revised the manuscript, agree to be fully accountable for ensuring the integrity and the accuracy of the work, and read and approved the final manuscript.