Monitoring of gastric residual volume during enteral nutrition

Abstract

Background

The main goal of enteral nutrition (EN) is to manage malnutrition in order to improve clinical outcomes. However, EN may increase the risks of vomiting or aspiration pneumonia during gastrointestinal dysfunction. Consequently, monitoring of gastric residual volume (GRV), that is, to measure GRV periodically and modulate the speed of enteral feeding according to GRV, has been recommended as a management goal in many intensive care units. Yet, there is a lack of robust evidence that GRV monitoring reduces the level of complications during EN. The best protocol of GRV monitoring is currently unknown, and thus the precise efficacy and safety profiles of GRV monitoring remain to be ascertained.

Objectives

To investigate the efficacy and safety of GRV monitoring during EN.

Search methods

We searched electronic databases including CENTRAL, MEDLINE, Embase, and CINAHL for relevant studies on 3 May 2021. We also checked reference lists of included studies for additional information and contacted experts in the field.

Selection criteria

We included randomized controlled trials (RCTs), randomized cross‐over trials, and cluster‐RCTs investigating the effects of GRV monitoring during EN. We imposed no restrictions on the language of publication.

Data collection and analysis

Two review authors independently screened the search results for eligible studies and extracted trial‐level information from each included study, including methodology and design, characteristics of study participants, interventions, and outcome measures. We assessed risk of bias for each study using Cochrane's risk of bias tool. We followed guidance from the GRADE framework to assess the overall certainty of evidence across outcomes. We used a random‐effects analytical model to perform quantitative synthesis of the evidence. We calculated risk ratios (RRs) with 95% confidence intervals (CIs) for dichotomous and mean difference (MD) with 95% CIs for continuous outcomes.

Main results

We included eight studies involving 1585 participants. All studies were RCTs conducted in ICU settings.

Two studies (417 participants) compared less‐frequent (less than eight hours) monitoring of GRV against a regimen of more‐frequent (eight hours or greater) monitoring. The evidence is very uncertain about the effect of frequent monitoring of GRV on mortality rate (RR 0.91, 95% CI 0.60 to 1.37; I² = 8%; very low‐certainty evidence), incidence of pneumonia (RR 1.08, 95% CI 0.64 to 1.83; heterogeneity not applicable; very low‐certainty evidence), length of hospital stay (MD 2.00 days, 95% CI –2.15 to 6.15; heterogeneity not applicable; very low‐certainty evidence), and incidence of vomiting (RR 0.14, 95% CI 0.02 to 1.09; heterogeneity not applicable; very low‐certainty evidence).

Two studies (500 participants) compared no GRV monitoring with frequent (12 hours or less) monitoring. Similarly, the evidence is very uncertain about the effect of no monitoring of GRV on mortality rate (RR 0.87, 95% CI 0.62 to 1.23; I² = 51%; very low‐certainty evidence), incidence of pneumonia (RR 0.70, 95% CI 0.43 to 1.13; heterogeneity not applicable; very low‐certainty evidence), length of hospital stay (MD –1.53 days, 95% CI –4.47 to 1.40; I² = 0%; very low‐certainty evidence), and incidence of vomiting (RR 1.47, 95% CI 1.13 to 1.93; I² = 0%; very low‐certainty evidence).

One study (322 participants) assessed the impact of GRV threshold (500 mL per six hours) on clinical outcomes. The evidence is very uncertain about the effect of the threshold for GRV at time of aspiration on mortality rate (RR 1.01, 95% CI 0.74 to 1.38; heterogeneity not applicable; very low‐certainty evidence), incidence of pneumonia (RR 1.03, 95% CI 0.72 to 1.46; heterogeneity not applicable; very low‐certainty evidence), and length of hospital stay (MD –0.90 days, 95% CI –2.60 to 4.40; heterogeneity not applicable; very low‐certainty evidence).

Two studies (140 participants) explored the effects of returning or discarding the aspirated/drained GRV. The evidence is uncertain about the effect of discarding or returning the aspirated/drained GRV on the incidence of vomiting (RR 1.00, 95% CI 0.06 to 15.63; heterogeneity not applicable; very low‐certainty evidence) and volume aspirated from the stomach (MD –7.30 mL, 95% CI –26.67 to 12.06, I² = 0%; very low‐certainty evidence)

We found no studies comparing the effects of protocol‐based EN strategies that included GRV‐related criteria against strategies that did not include such criteria.

Authors' conclusions

The evidence is very uncertain about the effect of GRV on clinical outcomes including mortality, pneumonia, vomiting, and length of hospital stay.

Plain language summary

Periodic measurement of the containment volume of the stomach during tube feeding

Review question

Is it necessary to measure the containment volume of the stomach periodically during tube feeding? If so, what is the best way to monitor the volume? How frequently does the monitoring need to be? How large would a volume be regarded as safe?

Background

People with acute illnesses may not be able to consume food due to several reasons (e.g. unconsciousness, need for mechanical ventilation). To maintain sufficient levels of energy and nutrients for this population, tube feeding, that is, administering liquidized dense nutrients through a flexible tube that reaches the stomach via the nose, is commonly used. Tube feeding is currently recommended as a first‐line treatment for critically ill people with acute illnesses because such a technique can provide non‐nutritional (e.g. protecting the immune system against suppression due to acute disease) as well as nutritional benefits.

However, people with acute illnesses often have a dysfunction of the stomach and intestines, and thus are unable to empty stomach contents. When increasing amounts of liquid nutrients are fed into the stomach via a tube, it can cause reflux (where contents travel back up the food pipe) or vomiting and may lead to aspiration pneumonia (when contents are breathed into the lungs or airways leading to the lungs).

One method to avoid these complications of tube feeding is to periodically monitor the gastric residual volume (GRV), which is the amount of liquid contents drained from the stomach. The speed of tube feeding can then be adjusted according to the volume.

Although monitoring of GRV may minimize the complications of tube feeding, and this technique has been recommended in many intensive care units (ICUs) for decades, we lack sufficient data to support a universal approach. Some research findings show that monitoring of GRV had no effects on tube‐feeding complications; moreover, the technique was found to reduce the amount of nutrients delivered, thereby affecting the overall treatment goals of tube feeding.

We designed this review in the hope of answering the following questions. Is monitoring of GRV effective and safe? What is the best way to monitor GRV (how often should GRV be measured per day; how large a threshold should be set for GRV)?

Key findings

We included evidence published up to 3 May 2021. We included findings from eight studies involving 1585 adults, with most being men (1019 men versus 506 women) with average ages 60 to 69 years. All studies were conducted in the ICU settings, and many people were severely ill and required mechanical ventilation and tube feeding for more than 48 hours. The duration of the studies ranged from three to 90 days.

Two studies (417 participants) compared less‐frequent GRV monitoring with a more‐frequent regimen. Two studies (500 participants) compared no GRV monitoring with frequent monitoring. One trial (329 participants) assessed the effects of GRV threshold by comparing a higher threshold at the time of aspiration against a lower threshold. Two studies (140 participants) compared the technique of returning versus discarding the aspirated/drained GRV.

We found that the evidence is uncertain about GRV monitoring (less frequent versus more frequent; no monitoring versus frequent monitoring) on mortality, pneumonia, vomiting, and length of hospital stay.

Reliability of the evidence

Five of the eight included studies assessed mortality as an outcome measure. Most studies were poorly conducted with sparse data, which made interpretation difficult. Thus, the overall reliability of the included evidence for our review outcomes was very low, and our findings should be treated with caution.

Background

Description of the condition

Malnutrition often leads to increased medical expenses, increased length of hospital stays, and poor prognosis. It has been reported that up to 40% of people from inpatient settings were affected by some form of disease‐related malnutrition, which is a specific type of malnutrition caused by concomitant diseases (Cederholm 2017; nutritionDay 2004). To achieve desired target energy levels for hospitalized people at risk of malnutrition, it is important to take pre‐emptive measures to prevent malnutrition. However, in some people with acute illnesses, oral food intake may not provide the necessary nutrients due to various reasons such as appetite loss due to acute illness, nausea, vomiting, early satiety, and difficulty in swallowing (Gomes 2018; Weimann 2017). This issue is particularly relevant to critically ill people since oral intake is severely affected for situations including mechanical ventilation, gastrointestinal surgery, or unconsciousness. Therefore, under such conditions, high mortality rates are often observed (Esteban 2013; Rubenfeld 2005), and medical care providers must explore the best interventions to maintain proper nutritional status. Indeed, nutritional management has been emphasized as an important determinant of overall survival and prognosis (Reintam Blaser 2017). Critically ill populations are often associated with highly variable metabolic and immune responses to injury or illness (Shaw 1993; Wanzer 1989). Inflammatory conditions result in increased glycogenolysis, protein catabolism, fatty acid degradation, and insufficient nutrient intake causes the depletion of organ proteins, ultimately leading to malnutrition and increased risk of infection and death (Reintam Blaser 2017).

Under these circumstances, enteral nutrition (enteral tube feeding; EN) or parenteral nutrition (delivery of calories and nutrients into a vein; PN) can compensate for nutritional intake until oral administration becomes satisfactory (Bounoure 2016). People with reduced nutrient intake from the gastrointestinal tract are at an increased risk of infection because of reduced gut integrity and the physiologic stress response (McClave 2009a). The most recent clinical practice guidelines have collectively suggested the use of EN over PN for hospitalized people requiring non‐oral nutrition therapy unless EN is contraindicated (Critical Care Nutrition 2015; JSICM 2017; McClave 2016a; McClave 2016b; Reintam Blaser 2017). One Cochrane Review reported that treatment of EN resulted in a risk ratio reduction of serious adverse events (Feinberg 2017).

A wide array of micro‐organisms exists in the gastrointestinal tract, and the gastrointestinal mucosa acts as a barrier against microbial infection. Immune tissue, known as the Peyer's patches, located in the gastrointestinal mucosa, plays a preventive role against bacterial contamination of the body (Reintam 2012). It has been proposed that when nutrients do not flow in the gastrointestinal tract, the gastrointestinal mucosa becomes atrophied, leaving the individual susceptible to infection due to the reduced interaction between the gut and the systemic immune response, and leads to poor prognosis in critically ill populations (McClave 2009a). People who are critically ill may have highly variable metabolic and immune responses to injury or illness, and thus early nutrition intake via the intestinal tract is highly important. Previous studies have suggested that in such subpopulations, early EN may reduce the rate of infection and mortality (McClave 2009a). Current clinical practice guidelines also recommend that early EN should be administrated, especially to people in the intensive care unit (ICU) setting (Reintam Blaser 2017).

Enteral feeding is a relatively low‐risk and well‐tolerated approach for people with normal gastric functions (Zanetti 2016), and in the context of normal functioning, perfusion, secretion, movement, and co‐ordinated intestinal microbial interaction are essential (Reintam 2012).

People who are hospitalized often have their gastric functions impaired due to pre‐existing diseases (e.g. diabetes mellitus, vagotomy, myopathies, shock, pancreatitis, spinal cord injury, trauma, abdominal surgery, burn); use of medications (e.g. sedatives, opioids, anticholinergics, vasopressors); or electrolyte abnormalities (e.g. hyperglycemia, hypokalemia) (Deane 2007; Zanetti 2016). Among these risk factors, the severity of pre‐existing diseases is the main reason for gastric dysfunction and may influence the occurrence and degree of complications due to EN (Deane 2007; Nguyen 2008); the sympathetic nervous system predominates over the parasympathetic nervous system, leading to reduced gastrointestinal peristalsis and reduced absorption capacity in the digestive tract, with a concomitant increase in enteral nutrient stagnation time in the stomach. Gastrointestinal dysfunction is a common event during critical illness, with an incidence rate of 63%, and can emerge as part of multi‐organ failure (Montejo 1999; Reintam 2012).

Gastrointestinal dysfunction is often an obstacle to EN. Feeding intolerance signifies gastric dysfunction and manifests due to motility and absorption disorders of the gastrointestinal tract, frequently leading to reduced EN intake (Elke 2015; Zanetti 2016). The incidence of feeding intolerance is about 27% in general‐ward settings and about 36% in ICUs (Gungabissoon 2015; Wang 2017). One systematic review of observational studies showed that compared to people without feeding intolerance, those with feeding intolerance presented with higher infectious complications and ICU mortality, and longer ICU stays (Blaser 2014).

The leading cause of feeding intolerance is delayed gastric emptying. Gastric emptying can be assessed by various methods, such as scintigraphy, paracetamol absorption test, ultrasound, refractometry, breath test, and gastric impedance monitoring (Moreira 2009). However, in real‐world clinical practice, it is usually assessed by measuring the gastric residual volume (GRV). GRV is the amount of liquid drained from a stomach following administration of enteral feed; this liquid consists mainly of infused nutritional formula or water and secreted gastric juice. Measurement of GRV is often via aspiration using a syringe or by gravity drainage to a reservoir (Elke 2015). Though GRV can vary depending on the method of drainage, body position, amount of gastric juice, type of tube (large/small diameter, pored/non‐pored), and position of tube tip (Bartlett 2015; Metheny 2005), it is the preferred clinical indicator of gastric emptying because of its simplicity.

Description of the intervention

Monitoring of GRV involves obtaining frequent measurements and employing appropriate interventions in people with large GRVs. It is an essential component of the EN care pathway and aids in preventing complications due to EN (McClave 2009b; Metheny 2012). In healthy adults and people with mild illness, seven to nine liters of digestive juices are secreted daily (Jeejeebhoy 1977; Jeejeebhoy 2002); most of these juices are absorbed by the small bowel, while approximately 500 mL reaches the colon, and 150 g remains in the stools. Administering additional enteral nutrients in people with increased GRV may cause aspiration and lead to an increase in intra‐abdominal pressure, which, in turn, increases the risk of respiratory and circulatory failure as well as intestinal necrosis. For this reason, it is particularly important to monitor GRV in the early stages of EN feeding, especially for people who are critically ill. Frequency of GRV measurement (e.g. every six hours) and the intervention strategy for large GRVs (e.g. GRV above 500 mL, hold feeding for two hours and recheck GRV) are usually decided as per institution‐specific protocols and needs of the inpatient population (Bounoure 2016). GRV is usually monitored in the ICU during nasogastric feeding or gastrostomy tube. GRV monitoring is a well‐established and common nursing practice in the ICU. Metheny 2012 reported that about 97.1% of critical care nurses reported GRV measurements in the US. Optimal GRV monitoring involves standardization of several parameters, and the following aspects have been studied so far: frequency of monitoring (Reignier 2013; Williams 2014), comparison of the methods of managing GRV to prevent complications (Booker 2000), and whether the remaining contents of the stomach should be returned to the stomach or discarded (Julien 2009; Williams 2010).

How the intervention might work

Feeding intolerance is highly prevalent and is associated with worsened outcomes, especially in critically ill subpopulations. Thus, it is imperative to standardize assessment of gastric functions together with a bedside exam of the abdomen (JSICM 2017; McClave 2016a). Monitoring of GRV is considered a simple and effective method of monitoring feeding intolerance. One systematic review showed that most included studies (83%) used a large GRV to define feeding intolerance in intensive care (Blaser 2014). The main purpose of monitoring GRV is to improve safety and minimize complications in people receiving EN. Administration of more enteral nutrients via the feeding tube, while the stomach is already full (high GRV) is not recommended for people with reduced gastric tolerance. In such cases, the residual gastric fluid is refluxed from the stomach into the esophagus, causing vomiting and ultimately increasing the risk of aspiration pneumonitis caused by aspiration of the vomit (McClave 2009b). Aspiration pneumonitis is a severe complication that prolongs hospital stays and increases mortality rates and the expense of hospitalization (Hayashi 2014). Hence, monitoring of GRV is considered an important procedure.

Furthermore, GRV monitoring may enable clinicians to effectively identify people with delayed gastric emptying and apply management strategies to minimize the adverse effects of feeding intolerance. These strategies include the use of prokinetic agents, postpyloric feeding, and pausing/reducing EN (Elke 2015). Nutrition treatment protocols involving interventions for cases with large GRVs may help achieve goal rates and prevent aspiration (Metheny 2010; Racco 2012). According to one national survey in the US, almost 70% of critical care nurses used 200 mL or 250 mL as threshold levels for interrupting EN, and 80% of the nursing personnel measured GRV every four hours (Metheny 2012).

However, apart from the benefits mentioned above, monitoring of GRV has several disadvantages (Edwards 2000). It can be regarded as an unnecessary intervention due to insufficient evidence to support its efficacy profile in some cases and may result in an increase in the time taken to reach the target amount of enteral nutrients because of interrupted feeding (Edwards 2000). Digestive juices are included in the residual contents of the stomach, and important electrolytes and digestive enzymes may be discarded along with the residual contents of the stomach, which might lead to electrolyte imbalances and poor digestion. Furthermore, GRV monitoring must be confirmed manually primarily by a nurse and increases other clinical care costs (e.g. requirement of additional syringes). Therefore, unnecessary monitoring of GRV contrarily increases nursing burden, which may lower the quality of medical care.

Why it is important to do this review

Although GRV monitoring has been part of practice guideline recommendations in critical care for decades, there are still concerns and conflicting information regarding its clinical importance and relevance. In recent years, GRV monitoring has been routinely performed as per institution‐specific protocols in people receiving EN (Reintam Blaser 2017). However, it has been suggested that routine monitoring of GRV may increase nurses' workload, thereby delaying treatment for other target groups. Insufficient nursing care may also be a consequence of increased nursing burden. Furthermore, it is thought that interruption of feeding in order to account for GRV monitoring increased the time taken to reach the feeding goal, increased the risk of infection caused by malnutrition from insufficient energy intake, and reduced usage of the gastrointestinal tract (Reintam Blaser 2017).

It is worth noting that in a number of critical care guidelines, recommendations on GRV monitoring vary. The American College of Gastroenterology and Society of Critical Care Medicine and American Society for Parenteral and Enteral Nutrition (SCCM/ASPEN) recommend against GRV monitoring (McClave 2016a; McClave 2016b). However, the Canadian Critical Care Society and Canadian Critical Care Trials Group recommend GRV monitoring once every four to eight hours (Critical Care Nutrition 2015). Both the SCCM/ASPEN and the Japanese Society of Intensive Care Medicine suggest avoiding holding off EN for GRVs less than 500 mL in the absence of other signs of intolerance (JSICM 2017; McClave 2016a). Guidelines of the European Society of Intensive Care Medicine suggest delaying EN if GRV is greater than 500 mL per six hours (Reintam Blaser 2017).

There are no adequately powered robust clinical studies to demonstrate how best to assess GRV in clinical practice, and high‐quality systematic reviews to explore the risk–benefit ratio of GRV monitoring are warranted. Specifically, there is an urgent need to better define the best available methods to monitor GRV, for an overall aim to reduce overtreatment and workload, and assure safety.

Objectives

To investigate the clinical efficacy and safety of monitoring GRV during EN.

We sought to answer the following questions.

• Is GRV monitoring necessary for reducing mortality and EN‐related complications?

• If GRV monitoring is performed, how frequently should it be done to minimize the risks of mortality and EN‐related complications?

• Would a lower GRV threshold affect the incidence of mortality and complications during EN?

• Should residual gastric contents be discarded or returned after GRV monitoring in a view to reduce mortality and EN‐related complications?

Methods

Criteria for considering studies for this review

Types of studies

We included randomized controlled trials (RCTs) and randomized cross‐over trials (data from the first period, i.e. before crossover only). For updates, we will also consider cluster‐RCTs if the following characteristics are known and reported: the number of clusters or the mean size of each cluster, outcome data for the total number of individuals, and an estimate of intracluster (or intraclass) correlation coefficient (ICC).

Types of participants

We included adults (ages 18 years or older) receiving EN via a nasogastric tube or a gastrostomy tube, with no restrictions on the type of clinical diagnoses or settings (e.g. ICU, hospital outpatients). We excluded studies involving EN feeding via tubes placed beyond the pylorus (postpyloric feeding).

Types of interventions

We included trials assessing the following treatment regimens, for which intervention group was defined as any type of GRV monitoring during EN:

• regimens that monitored GRV at different intervals (since GRV during EN may be influenced by monitoring frequency);

• regimens that included the upper limit/threshold of GRV before intervention;

• regimens that involved a strategy for handling aspirated residual gastric fluid, such as returning it to the stomach or discarding it.

We included all methods of GRV monitoring (i.e. by aspiration/drainage from a nasogastric or a gastrostomy tube, by ultrasound exam, or by computed tomography (CT) scan). If a study used different methods of measuring GRV for different interventions, we excluded that study from this review. We defined a minimum intervention period of 24 hours and a maximum of 14 days. It was expected that the dose of EN that did not cause complications during the acute stage would be approximately seven days and that the observation period of 14 days would be considered appropriate when administering EN in the chronic phase.

We considered the following comparisons:

• more frequent (less than eight hours) versus less frequent (eight hours or greater) monitoring of GRV (Williams 2014);

• frequent (12 hours or less) versus no monitoring (Ozen 2016);

• higher GRV threshold (500 mL per six hours or greater) versus the lower threshold (less than 500 mL per six hours) at the time of aspiration (Reintam Blaser 2017);

• returning versus discarding the aspirated/drained GRV;

• protocol‐based nutrition strategy with GRV‐relevant criteria versus protocol‐based EN strategy without GRV‐related criteria or non‐protocol‐based nutrition strategy.

For this review, we defined a protocol‐based nutrition strategy with GRV‐relevant criteria as a regimen where the attending physicians were required to monitor the GRV for increasing/decreasing the EN volume. For protocol‐based EN strategy without GRV‐related criteria or any non‐protocol‐based nutrition strategies, such physician‐led monitoring of GRV was not implemented.

Types of outcome measures

Primary outcomes

• Mortality (at the end of follow‐up or up to 28 days).

Secondary outcomes

• Pneumonia (as per the study authors' definitions, including ventilator‐associated pneumonia and hospital‐acquired pneumonia; follow‐up from the day EN was initiated until it was discontinued or up to 28 days).

• Length of hospital stay (days).

• Vomiting (as per the trialists' definition; follow‐up from the day EN was initiated until discontinued).

• Duration (hours) for reaching the target calories per day during EN feeding.

• Volume aspirated from the stomach via a nasogastric tube or a gastrostomy tube (milliliters) (as per the study investigators' definition).

• Adverse events as reported in individual studies.

Volume aspirated from the stomach and duration for reaching the target calories per day during EN feeding are considered surrogate outcomes, but we decided to include them as secondary outcomes because they are likely to be related to clinically important outcomes.

Reporting of the outcomes listed here was not an inclusion criterion for the review.

Search methods for identification of studies

There were no restrictions on the language of publication when searching the electronic databases.

Electronic searches

We conducted a comprehensive search to identify all eligible studies. We placed no restrictions on the language of publication. We translated non‐English language papers and assessed the full texts for potential inclusion in the review where necessary. We included studies available as abstracts only as well as unpublished data.

We screened the following electronic databases on 3 May 2021 based on systematic search strategies illustrated in Appendix 1, Appendix 2, Appendix 3, and Appendix 4:

• Cochrane Central Register of Controlled Trials (CENTRAL) in the Cochrane Library (via Ovid Evidence‐Based Medicine Reviews Database (EBMR), from inception);

• MEDLINE (via Ovid from 1946);

• Embase (via Ovid from 1974);

• CINAHL (Cumulative Index to Nursing and Allied Health Literature) (via EBSCO from 1982).

Searching other resources

We screened the reference lists of all primary studies and relevant review articles for additional information. We contacted authors of identified trials as well as experts in the field to locate other published or unpublished studies. 

For grey literature, we searched the following resources:

• Health Management Information Consortium (HMIC) database (www.ovid.com/site/catalog/databases/99.jsp);

• National Technical Information Service (NTIS) database (classic.ntis.gov/products/ntis-database/);

• OpenGrey (www.opengrey.eu);

• PsycEXTRA (www.apa.org/pubs/databases/psycextra).

We also searched for errata or retractions of eligible trials on PubMed (www.ncbi.nlm.nih.gov/pubmed), and reported the date this was done in the review.

We conducted a search of clinical trial registers/trial result registers:

• ClinicalTrials.gov (clinicaltrials.gov);

• EU Clinical Trials register (www.clinicaltrialsregister.eu);

• World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (apps.who.int/trialsearch/).

Data collection and analysis

Selection of studies

Two review authors (NK, RY) independently screened the titles and abstracts of all studies identified via the electronic searches and coded them as 'retrieve' (eligible, potentially eligible, or unclear) or 'do not retrieve.' We retrieved the full‐text reports of the former, and two review authors (NK, RY) independently screened these and identified studies for inclusion as well as record reasons for excluding ineligible studies. We resolved any disagreements through discussion by consulting a third review author (SA). We removed duplicate records and collated multiple reports of the same study, so that each study, rather than each report, was the unit of interest. Our study selection process is illustrated as a PRISMA flow diagram. Further study‐level information is provided in the Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification; and Characteristics of ongoing studies tables.

Data extraction and management

For extraction of study characteristics and outcome data, we used a prestandardized data collection form that had been piloted on at least one study in the review. One review author (HY) extracted the following study characteristics from the included studies.

• Methods: study design, total duration of the study and run‐in period, number and locations of study centers, study setting, withdrawals, and date of study.

• Participants: number, mean age, age range, sex, severity of condition, diagnostic criteria, baseline lung function, smoking history, inclusion criteria, and exclusion criteria.

• Interventions: intervention method, comparison method, concomitant medications, and excluded medications.

• Outcomes: specified and collected primary and secondary outcomes, and time points reported.

• Notes: funding details for the trial and notable conflicts of interest of trial authors.

Two review authors (RY, SA) independently extracted outcome data from the included studies. In the Characteristics of included studies table, we recorded situations where outcome data were reported in an unusable manner. We resolved any disagreements by consensus or by involving a third review author (NK). One review author (HY) copied the data from the data collection form into Review Manager 5 (Review Manager 2014). We double‐checked that data had been entered correctly by comparing the study reports with the presentation of data in the systematic review. A second review author (RY or SA) spot‐checked study characteristics for accuracy against the trial report.

Assessment of risk of bias in included studies

Two review authors (NK, RY) independently assessed risk of bias in each included study using Cochrane's risk of bias tool outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2017). We resolved any disagreements by discussion or by involving a third review author (HY). As per the Cochrane risk of bias tool, we assessed risk of bias according to the following domains:

• random sequence generation;

• allocation concealment;

• blinding of participants and personnel;

• blinding of outcome assessment;

• incomplete outcome data;

• selective outcome reporting;

• other bias.

We judged each potential source of bias as high, low, or unclear, and provide a quote from the study report and justification for our judgment in the Characteristics of included studies table. We summarized our judgments for each included study in a risk of bias graph and risk of bias summary. We considered blinding separately for different key outcomes where necessary, for example, risk of bias for unblinded outcome assessment might be very different for all‐cause mortality than for a patient‐reported pain scale. Where information on risk of bias related to unpublished data or correspondence with a trialist, we noted this in the Characteristics of included studies table.

Assessment of bias in conducting the systematic review

We conducted the review according to the published protocol (Yasuda 2019), and reported any deviations from it in the Differences between protocol and review section.

Measures of treatment effect

We analyzed dichotomous data (mortality, pneumonia, vomiting, and adverse events) as risk ratios (RR) with 95% confidence intervals (CIs), and continuous data (length of hospital stay, number of hours to reach the target calorie levels, and volume aspirated from stomach) as mean differences (MD), with 95% CIs. For rate outcomes, results were expressed as rate ratios with 95% CIs. We ensured that higher scores for continuous outcomes had the same meaning for the particular outcome, explained the direction to the reader, and reported where the directions were reversed if this was necessary (Review Manager 2014).

We undertook meta‐analyses only where this was meaningful: if the treatments, participants, and the underlying clinical question had a high degree of similarity and were conducive to data pooling.

A common way that trialists indicated they had skewed data was by reporting medians and interquartile ranges. When we encountered this, we noted that the data were skewed and considered the implication of this. If the data were skewed, we did not perform a meta‐analysis, but provided a narrative summary instead.

We planned to include only the relevant arms if there were multiple trial arms in a single trial. If two comparisons (e.g. frequent monitoring of GRV less than four hours versus between four and eight hours and more than eight hours) must have been entered into the same meta‐analysis, we halved the control group to avoid double‐counting.

Unit of analysis issues

We identified no relevant cluster‐RCTs for inclusion in this review. For review updates, for dichotomous data extracted from cluster‐RCTs, we plan to account for the design effect, and calculate effective sample size and number of events using the ICC, the mean cluster size for dichotomous data, and adjusted standard errors if they have been reported, as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2021). If we are unable to obtain the ICC from study reports, we will use the ICC of similar studies as a substitute. For continuous data only, the sample size will be reduced, while means and standard deviations will remain unchanged (Higgins 2021).

Similarly, identified no randomized cross‐over trials in this review. For review updates, we planned to only include data from the first period (i.e. before crossover). If we include multiple‐arm studies in the future, we intend to only extract and analyze data from the relevant study arms.

To avoid double counting of events, we considered how to report adverse events (i.e. as single events or included in a group of events). For count data (for events that occur more than once in one participant), we used the counts of rare events as a rate ratio and the counts of more common events as continuous data. Otherwise, we used dichotomous data with participants as the unit of analysis.

Dealing with missing data

We contacted the study investigators to obtain missing outcome data. In cases where that failed, we did not impute for dichotomous outcomes; for continuous outcomes, we imputed the mean from the median (i.e. considered median as the mean) and the standard deviation from the standard error, interquartile range, or P values, according to the guidelines in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2021). 

Assessment of heterogeneity

We used the I² statistic to measure heterogeneity among trials in each analysis (Higgins 2003). If we identified substantial heterogeneity as per the Cochrane Handbook for Systematic Reviews of Interventions (I² > 50%) (Higgins 2021), we explored the extent of such statistical heterogeneity by conducting subgroup analysis (Subgroup analysis and investigation of heterogeneity). We also assessed heterogeneity by evaluating whether there was an acceptable overlap of CIs.

Assessment of reporting biases

For review updates, if we are able to include more than 10 trials, we will create and examine a funnel plot for possible publication bias. We will use the Egger's test to determine the statistical significance of the reporting bias, with P < 0.05 to be set for statistical significance (Egger 1997).

Data synthesis

We used the random‐effects model for data synthesis by default. To test the robustness of our findings regardless of which analytical method was chosen, we conducted a sensitivity analysis for primary outcomes using fixed‐effect models. In case of divergence between the two models, we presented both results; otherwise, we presented only results from the random‐effects model.

Subgroup analysis and investigation of heterogeneity

We planned to carry out the following subgroup analyses using the Cochrane's Q test for subgroup interactions:

• population subsets/settings (ICU, non‐ICU (general hospital wards), or outpatients);

• participants with a high severity of illness score as defined by a validated severity scale specific to critical care (such as the Acute Physiology and Chronic Health Evaluation (APACHE), Simplified Acute Physiology Score (SAPS), or the Sequential Organ Failure Assessment (SOFA));

• obesity status (as defined in each of the included studies);

• methods used for monitoring GRV, such as aspiration from a nasogastric tube or a gastrostomy tube, ultrasound exam, and CT scans.

The a priori subgroups were designed to assess the following primary outcomes:

• mortality (at end of follow‐up);

• pneumonia (by study investigators' definition; follow‐up from EN initiation until discontinued);

• vomiting (as per study authors' definition; follow‐up from EN initiation to discontinuation).

Sensitivity analysis

We performed the following sensitivity analyses to assess the robustness of our conclusions for the primary outcomes:

• comparison based on our risk of bias assessment of included studies where we excluded low‐certainty (high risk of selection bias) studies;

• comparison of results synthesized from a fixed‐effect versus those from a random‐effects model;

• excluding trials for which means or standard deviations or both were imputed.

Reaching conclusions

We based our conclusions on findings from quantitative/narrative synthesis. We avoided making recommendations for practice; our implications for research were intended to provide a clear sense of our suggestions regarding future research directions and uncertainties in the field.

Summary of findings and assessment of the certainty of the evidence

We created summary of findings tables with the following outcomes: mortality, pneumonia, and length of hospital stay. These tables provide important information on the certainty of the evidence, the magnitude of the intervention effects, and the total available data on the main outcomes (Schünemann 2011a). We used the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness, and publication bias) to assess the certainty of the evidence on the studies that contributed data for each outcome, classifying the certainty as 'high', 'moderate', 'low', or 'very low' (Guyatt 2008; Guyatt 2011; Schünemann 2011b). When considering treatment effects, we took into account the risk of bias for the studies that contributed to that outcome (Assessment of risk of bias in included studies). Two review authors (NK, RY) independently assessed the certainty of the evidence for each study using the methods and recommendations described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2021), using GRADEpro GDT software (GRADEpro GDT). We justified all decisions to downgrade or upgrade the certainty of the evidence in the footnotes and provided comments to aid the reader's understanding of the review where necessary. We considered whether there was additional outcome information that was not incorporated into the meta‐analyses, noted this in the comments, and stated if it supported or contradicted the information from the meta‐analyses.

Results

Description of studies

Results of the search

We identified 4725 records through electronic searches. After removing duplicates, we screened the titles and abstracts of 3707 records; of these, 3659 were clearly irrelevant and were excluded. We retrieved the full texts for the remaining 48 articles for further assessment. Eventually, we included eight studies in this review, excluded 39 articles (we provided a representative selection in the Excluded studies section), three studies are awaiting classification, and one study is ongoing (IRCT20170118032029N3). Our study selection progress is illustrated in Figure 1.

1.

Study flow diagram.

Included studies

The eight included studies were individually randomized, parallel‐group trials by design (see Characteristics of included studies table). Samples sizes of ranged from 31 to 452. Three studies had more than 300 participants (Montejo 2010; Reignier 2013; Williams 2014).

Setting

All eight studies were conducted in ICUs or stroke units, of which four were multi‐center studies (Booker 2000; Ozen 2016; Montejo 2010; Reignier 2013), and four were single‐center studies (Büyükçoban 2016; Chen 2015; Juvé‐Udina 2009; Williams 2014). Six studies were conducted in medical and surgical ICUs (Booker 2000; Büyükçoban 2016; Juvé‐Udina 2009; Montejo 2010; Reignier 2013; Williams 2014); one was conducted in a medical ICU (Ozen 2016), and one was conducted in a stroke care unit (Chen 2015).

Participants

We included 1585 adults, with a gender distribution of 1019 men and 506 women (genders of 60 participants were not reported). The mean age of the included participants was 60 to 69 years. All participants were fed with a nasogastric tube. Most of the participants needed mechanical ventilation and were likely to receive enteral feeding for longer than 48 hours.

Seven studies reported disease severity scores; six studies reported APACHE II (Büyükçoban 2016; Chen 2015; Juvé‐Udina 2009; Montejo 2010; Ozen 2016; Williams 2014); three studies reported SOFA (Büyükçoban 2016; Montejo 2010; Reignier 2013), one study reported SAPS II (Reignier 2013). Six studies reported information on

index (BMI) (Booker 2000; Chen 2015; Juvé‐Udina 2009; Montejo 2010; Ozen 2016; Williams 2014). The APACHE II score of the participants ranged from 17.8 to 25.3.

Interventions and comparisons

Two studies involving 417 participants compared less‐frequent with more‐frequent monitoring of GRV (Büyükçoban 2016; Williams 2014). Two studies involving 500 participants compared no monitoring with frequent monitoring (Ozen 2016; Reignier 2013). One study compared no monitoring with frequent monitoring, but none of our a priori outcomes of interest were reported by the study investigators (Chen 2015). One study involving 329 participants compared a higher GRV threshold with a lower threshold at the time of aspiration (Montejo 2010). Two studies involving 140 participants compared the strategy of returning versus discarding the aspirated/drained GRV (Booker 2000; Juvé‐Udina 2009). We found no studies comparing protocol‐based EN strategies (strategy with GRV‐related criteria versus strategy without GRV‐relevant criteria versus non‐protocol‐based EN strategy).

Outcome measures

The studies assessed the following outcomes of interest.

• Mortality: five studies (1239 participants) (Büyükçoban 2016; Montejo 2010; Ozen 2016; Reignier 2013; Williams 2014).

• Pneumonia: three studies (1128 participants) (Montejo 2010; Reignier 2013; Williams 2014).

• Length of hospital stay: four studies (1179 participants) (Montejo 2010; Ozen 2016; Reignier 2013; Williams 2014).

• Vomiting: five studies (900 participants) (Booker 2000; Büyükçoban 2016; Montejo 2010; Ozen 2016; Reignier 2013).

• Number of hours to reach the target calories per day in EN: two studies (111 participants) (Büyükçoban 2016; Ozen 2016).

• Volume aspirated from the stomach: two studies (140 participants) (Booker 2000; Juvé‐Udina 2009).

• Adverse events: no studies.

Excluded studies

We excluded 39 records from the review mainly for the following reasons: 19 studies were not RCTs; for one study, the participant of one record and the intervention/comparison of 21 records did not meet the predefined analysis plan; one abstract that was found to be superseded by a full publication. Reasons for study exclusion are provided in the Characteristics of excluded studies table.

Studies awaiting classification

We identified three studies awaiting classifications since it is currently unclear whether the comparison group in each study included elements of GRV monitoring (Anandika 2018; Nasiri 2017; Yaghoubinia 2017).

Ongoing studies

We retrieved one ongoing study that might be relevant to this review scope and the findings, once published, will be included in the next review update (IRCT20170118032029N3).

Risk of bias in included studies

We contacted, via e‐mail, all first authors or corresponding authors for additional study information. However, only one author replied (Chen 2015).

Our overall judgments found that only one trial was at low risk of bias across all domains except for blinding of participants and personnel (Reignier 2013). The remaining seven trials were at high or unclear risk of bias in one or more domains other than blinding of participants and personnel. Findings of our risk of bias assessment are summarized in Figure 2 and Figure 3.

2.

Risk of bias graph: judgments about each risk of bias item presented as percentages across all included studies.

3.

Risk of bias summary: judgments about each risk of bias item for each included study.

Allocation

Three studies had insufficient information regarding allocation methods (Booker 2000; Büyükçoban 2016; Montejo 2010). However, it is usual that the randomization process in multi‐center RCTs is achieved through a centralized system, and therefore we judged that the risk of selection bias was low in Montejo 2010, which was conducted in 28 ICUs across Spain. For Booker 2000 and Büyükçoban 2016, we judged the risk of selection bias to be unclear.

Four studies used a computer‐generated randomization process (Juvé‐Udina 2009; Ozen 2016; Reignier 2013; Williams 2014). Another study used a random number table (Chen 2015).

For allocation concealment, four studies provided insufficient information (Booker 2000; Büyükçoban 2016; Chen 2015; Ozen 2016). Two studies mentioned a centralized method (Montejo 2010; Reignier 2013). Two studies reported using sealed envelopes (Juvé‐Udina 2009; Williams 2014).

Blinding

Due to the nature of the types of interventions considered in this review, blinding of participants and study personnel could not be performed. Thus, we considered the risk of performance bias to be high. Similarly, in relation to blinding of outcome assessment, there was no blinding in six studies (Booker 2000; Büyükçoban 2016; Chen 2015; Juvé‐Udina 2009; Montejo 2010; Ozen 2016). We considered the risk of bias was high for subjective outcomes such as pneumonia, adverse events, vomiting; for objective outcomes including mortality, hospital length of stay, and volume aspirated from the stomach, we judged the level of risk of bias to be low.

Two studies originally planned to blind the outcome assessors who evaluated the diagnosis of pneumonia (Reignier 2013; Williams 2014). However, Williams 2014 eventually failed to pursue blinding of outcome assessment, and we judged it at high risk of detection bias (Williams 2014). For one study, we considered the risk of bias was low (Reignier 2013). 

Incomplete outcome data

Three studies performed data analysis based on the intention‐to‐treat (ITT) principle (Büyükçoban 2016; Reignier 2013; Williams 2014).

Five studies had incomplete outcome data (Booker 2000; Chen 2015; Juvé‐Udina 2009; Montejo 2010; Ozen 2016). Booker 2000 omitted data for 17 participants from the analysis. Chen 2015 excluded four participants allocated to the control group and who failed to continue EN for more than 72 hours after randomization. Juvé‐Udina 2009excluded two participants in the intervention group and one in the control group from the final analysis. Montejo 2010 excluded seven participants from the final analysis. Ozen 2016 excluded nine people from the analysis from the GRV monitoring group as they were lost to follow‐up. We assessed this domain at high risk of bias since the exclusions were executed after randomization and probably affected the balance of the two groups.

Selective reporting

One study was registered in ClinicalTrials.gov (Reignier 2013; NCT01137487), and we found no evidence suggestive of selective reporting.
We could not locate study protocols or pre‐registration details for the rest of the included studies (Booker 2000; Büyükçoban 2016; Chen 2015; Juvé‐Udina 2009; Montejo 2010; Ozen 2016; Williams 2014). Although study protocols or preregistration information were not available, the published study reports included all planned/expected outcomes. Therefore, we judged that the risk of selective reporting bias was low. 

Other potential sources of bias

One study reported that there was a lack of full‐time clinical research monitoring by the trialists and was at high risk of other bias (Booker 2000).

Effects of interventions

See: Table 1; Table 2; Table 3; Table 4

Summary of findings 1. More frequent (less than eight hours) versus less frequent (eight hours or greater) monitoring of gastric residual volume.

More frequent (< 8 hours) versus less frequent (≥ 8 hours) monitoring of GRV
Population: adults receiving EN via nasogastric tube or gastrostomy tube Setting: ICU department Intervention: more frequent GRV monitoring (< 8 hours) Comparison: less frequent GRV monitoring (≥ 8 hours)
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with more frequent monitoring	Risk with less frequent monitoring
Mortality (at the end of follow‐up; up to 28 days)	Study population		RR 0.91 (0.60 to 1.37)	417 (2 RCTs)	⊕⊝⊝⊝ Very lowa,b,c	Büyükçoban 2016 randomized participants into 2 groups in which there were sequential 30 patients in each group. Details of randomization not described.
	202 per 1000	184 per 1000 (121 to 277)
Pneumonia (until discontinued or up to 28 days)	Study population		RR 1.08 (0.64 to 1.83)	357 (1 RCT)	⊕⊝⊝⊝ Very lowa,c	Büyükçoban 2016 did not assess pneumonia.
	129 per 1000	140 per 1000 (83 to 236)
Length of hospital stay (days) (at the end of follow‐up)	Study population		—	357 (1 RCT)	⊕⊝⊝⊝ Very lowa,c	—
	The mean length of hospital stay was 0 days	MD 2 days higher (2.15 lower to 6.15 higher)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; EN: enteral nutrition; GRV: gastric residual volume; ICU: intensive care unit; MD: mean difference; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

^aDowngraded two levels due to very serious imprecision: wide 95% confidence intervals, and optimal information size criterion was not met.
^bDowngraded one level due to serious inconsistency: different direction of effect in the study (null and effective).
^cDowngraded one level due to serious risk of bias: lack of details of randomization or allocation concealment; no blinding of participants, study personnel, or data assessors.

Summary of findings 2. No monitoring versus frequent monitoring (12 hours or less) of gastric residual volume.

No monitoring versus frequent monitoring (≤ 12 hours) of GRV
Population: adults receiving EN via nasogastric or gastrostomy tube Setting: ICU department Intervention: no monitoring of GRV Comparison: frequent monitoring of GRV (≤ 12 hours)
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with frequent monitoring	Risk with no monitoring
Mortality (at the end of follow‐up; up to 28 days)	Study population		RR 0.87 (0.62 to 1.23)	500 (2 RCTs)	⊕⊝⊝⊝ Very lowa,b,c	—
	328 per 1000	285 per 1000 (203 to 403)
Pneumonia (until discontinued or up to 28 days)	Study population		RR 0.70 (0.43 to 1.13)	449 (1 RCT)	⊕⊝⊝⊝ Very lowa,d	Ozen 2016 did not assess pneumonia.
	158 per 1000	110 per 1000 (68 to 178)
Length of hospital stay (days) (at the end of follow‐up)	Study population		—	500 (2 RCTs)	⊕⊝⊝⊝ Very lowa,b,c	—
	The mean length of hospital stay was 0 days	MD 1.53 days lower (4.47 lower to 1.4 higher)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; EN: enteral nutrition; GRV: gastric residual volume; ICU: intensive care unit; MD: mean difference; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

^aDowngraded two levels due to very serious imprecision: wide 95% confidence interval, and optimal information size criterion was not met.
^bDowngraded one level due to serious inconsistency: different direction of effect in the study (null and effective).
^cDowngraded one level due to serious risk of bias: lack of details of concealment process and incomplete outcome data.
^dDowngraded one level due to serious risk of bias: lack of details of concealment process.

Summary of findings 3. Higher threshold (500 mL per six hours or greater) versus lower threshold (less than 500 mL per six hours) at time of aspiration.

Higher threshold (≥ 500 mL per 6 hours) versus lower threshold (< 500 mL per 6 hours) at time of aspiration
Population: adults receiving EN via nasogastric or gastrostomy tube Setting: ICU department Intervention: higher GRV threshold at the time of aspiration (≥ 500 mL per 6 hours) Comparison: lower GRV threshold at the time of aspiration (> 500 mL per 6 hours)
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with lower threshold	Risk with higher threshold
Mortality (at the end of follow‐up; up to 28 days)	Study population		RR 1.01 (0.74 to 1.38)	322 (1 RCT)	⊕⊝⊝⊝ Very lowa,b	Only 1 study assessed all outcomes for this comparison. 7 patients (4 in the control group and 3 in the study group) were inappropriately included (non‐mechanically ventilated participants at the time of EN) and were excluded from the final analysis.
	333 per 1000	337 per 1000 (247 to 460)
Pneumonia (until discontinued or up to 28 days)	Study population		RR 1.03 (0.72 to 1.46)	322 (1 RCT)	⊕⊝⊝⊝ Very lowa,b	Only 1 study assessed all outcomes for this comparison. 7 patients (4 in the control group and 3 in the study group) were inappropriately included (non‐mechanically ventilated participants at the time of EN) and were excluded from the final analysis.
	273 per 1000	281 per 1000 (196 to 398)
Length of hospital stay (days) (at the end of follow‐up)	Study population		—	322 (1 RCT)	⊕⊝⊝⊝ Very lowa,b	Only 1 study assessed all outcomes for this comparison. 7 patients (4 in the control group and 3 in the study group) were inappropriately included (non‐mechanically ventilated patients at the time of EN) and were excluded from the final analysis.
	The mean length of hospital stay was 0 days	MD 0.9 days higher (2.6 lower to 4.4 higher)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; EN: enteral nutrition; GRV: gastric residual volume; ICU: intensive care unit; MD: mean difference; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

^aDowngraded two levels due to very serious imprecision: wide 95% confidence interval, and optimal information size criterion was not met.

^bDowngraded one level due to serious risk of bias: not blinded and incomplete outcome data.

Summary of findings 4. Returning versus discarding the aspirated/drained gastric residual volume.

Returning the aspirated/drained GRV compared to discarding it for health problem or population
Population: adults receiving EN via nasogastric or gastrostomy tube Setting: ICU department Intervention: returning the aspirated/drained GRV Comparison: discarding the aspirated/drained GRV
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with discarding the aspirated/drained GRV	Risk with returning the aspirated/drained GRV
Mortality (at the end of follow‐up; up to 28 days)	Not reported	Not reported	—	—	—	—
Pneumonia (at the end of follow‐up)	Not reported	Not reported	—	—	—	—
Length of hospital stay (days) (at the end of follow‐up)	Not reported	Not reported	—	—	—	—
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; EN: enteral nutrition; GRV: gastric residual volume; ICU: intensive care unit.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

See: Table 1, Table 2, Table 3, and Table 4 for main comparisons

More frequent (less than eight hours) versus less frequent (eight hours or greater) monitoring of gastric residual volume

Two studies compared more frequent versus less frequent monitoring of GRV (Table 1; Booker 2000; Juvé‐Udina 2009). The monitoring frequencies of the intervention groups were four hours and control groups were eight hours in both studies.

Mortality

Two studies (417 participants) found very low‐certainty evidence about the effect of the frequent monitoring strategy on mortality rate (RR 0.91, 95% CI 0.60 to 1.37; I² = 8%; Analysis 1.1; Büyükçoban 2016; Williams 2014).

1.1. Analysis.

Comparison 1: More frequent (less than eight hours) versus less frequent (eight hours or greater hours) monitoring, Outcome 1: Mortality (at the end of follow‐up; up to 28 days)

Pneumonia

One study (357 participants) found very low‐certainty evidence about the effect of the frequent monitoring strategy on the incidence of pneumonia (RR 1.08, 95% CI 0.64 to 1.83; Analysis 1.2; Williams 2014).

1.2. Analysis.

Comparison 1: More frequent (less than eight hours) versus less frequent (eight hours or greater hours) monitoring, Outcome 2: Pneumonia

Length of hospital stay

One study (357 participants) found very low‐certainty evidence about the effect of the frequent monitoring on the length of hospital stay (MD 2.00 days, 95% CI –2.15 to 6.15; Analysis 1.3; Williams 2014).

1.3. Analysis.

Comparison 1: More frequent (less than eight hours) versus less frequent (eight hours or greater hours) monitoring, Outcome 3: Length of hospital stay (days)

Vomiting

One study (60 participants) found very low‐certainty evidence about the effect of the frequent monitoring strategy on the incidence of vomiting (RR 0.14, 95% CI 0.02 to 1.09; Analysis 1.4; Büyükçoban 2016).

1.4. Analysis.

Comparison 1: More frequent (less than eight hours) versus less frequent (eight hours or greater hours) monitoring, Outcome 4: Vomiting

Duration (hours) for reaching the target calories per day during enteral nutrition feeding

One study (60 participants) found very low‐certainty evidence about the effect of the frequent monitoring on the number of hours to reach the target calories per day in EN (MD –0.80 hours, 95% CI –4.88 to 3.28; Analysis 1.5; Büyükçoban 2016).

1.5. Analysis.

Comparison 1: More frequent (less than eight hours) versus less frequent (eight hours or greater hours) monitoring, Outcome 5: Duration for reaching the target calories per day during enteral nutrition feeding (hours)

Volume aspirated from the stomach

Neither study reported volume aspirated from the stomach.

Adverse events

Neither study reported adverse events.

No monitoring versus frequent monitoring (12 hours or less) of gastric residual volume

Two studies compared no monitoring versus frequent monitoring (Table 2). The monitoring frequencies of control groups in each study were six hours in Reignier 2013 and eight hours in Ozen 2016.

Mortality

Two studies (500 participants) found very low‐certainty evidence about the effect of no monitoring of GRV on morality rate (RR 0.87, 95% CI 0.62 to 1.23; I² = 51%; Analysis 2.1; Ozen 2016; Reignier 2013).

2.1. Analysis.

Comparison 2: No monitoring versus frequent (12 or less hours) monitoring of GRV, Outcome 1: Mortality (at the end of follow‐up; up to 28 days)

Pneumonia

One study (449 participants) found very low‐certainty evidence about the effect of no monitoring of GRV on the incidence of pneumonia (RR 0.70, 95% CI 0.43 to 1.13; Analysis 2.2; Reignier 2013).

2.2. Analysis.

Comparison 2: No monitoring versus frequent (12 or less hours) monitoring of GRV, Outcome 2: Pneumonia

Length of hospital stay

Two studies (500 participants) found very low‐certainty evidence about the effect of no monitoring of GRV on the length of hospital stay (MD –1.53 days, 95% CI –4.47 to 1.40; I² = 0%; Analysis 2.3; Ozen 2016; Reignier 2013).

2.3. Analysis.

Comparison 2: No monitoring versus frequent (12 or less hours) monitoring of GRV, Outcome 3: Length of hospital stay (days)

Vomiting

Two studies (500 participants) found very low‐certainty evidence about the effect of no monitoring of GRV on the incidence of vomiting (RR 1.47, 95% CI 1.13 to 1.93; I² = 0%; Analysis 2.4; Ozen 2016; Reignier 2013).

2.4. Analysis.

Comparison 2: No monitoring versus frequent (12 or less hours) monitoring of GRV, Outcome 4: Vomiting

Duration (hours) for reaching the target calories per day during enteral nutrition feeding

One study (51 participants) found very low‐certainty evidence about the effect of no monitoring of GRV on the duration required to reach the target calories per day in EN (MD –3.07 hours, 95% CI –5.75 to –0.39; Analysis 2.5; Ozen 2016).

2.5. Analysis.

Comparison 2: No monitoring versus frequent (12 or less hours) monitoring of GRV, Outcome 5: Duration for reaching the target calories per day during enteral nutrition feeding (hours)

Volume aspirated from the stomach

Neither study reported volume aspirated from the stomach.

Adverse events

Neither study reported adverse events.

Higher (500 mL or greater per six hours) versus lower (less than 500 mL per six hours) gastric residual volume threshold at the time of aspiration

One study compared higher (500 mL or greater per six hours) versus lower (less than 500 mL per six hours) GRV threshold at the time of aspiration (Table 3). The monitoring frequencies of both intervention and control groups was six hours (Montejo 2010).

Mortality

One study (322 participants) found very low‐certainty evidence about the effect of the GRV threshold at the time of aspiration on mortality rate (RR 1.01, 95% CI 0.74 to 1.38; Analysis 3.1; Montejo 2010).

3.1. Analysis.

Comparison 3: Higher threshold (500 mL or greater per 6 hours) versus lower threshold (less than 500 mL per 6 hours) for GRV at the time of aspiration, Outcome 1: Mortality (at the end of follow‐up; up to 28 days)

Pneumonia

One study (322 participants) found very low‐certainty evidence about the effect of the GRV threshold at the time of aspiration on the incidence of pneumonia (RR 1.03, 95% CI 0.72 to 1.46; Analysis 3.2; Montejo 2010).

3.2. Analysis.

Comparison 3: Higher threshold (500 mL or greater per 6 hours) versus lower threshold (less than 500 mL per 6 hours) for GRV at the time of aspiration, Outcome 2: Pneumonia

Length of hospital stay

One study (322 participants) found very low‐certainty evidence about the effect of the GRV threshold at the time of aspiration on the length of hospital stay (MD –0.90 days, 95% CI –2.60 to 4.40; Analysis 3.3; Montejo 2010).

3.3. Analysis.

Comparison 3: Higher threshold (500 mL or greater per 6 hours) versus lower threshold (less than 500 mL per 6 hours) for GRV at the time of aspiration, Outcome 3: Length of hospital stay (days)

Vomiting

One study (322 participants) found very low‐certainty evidence about the effect of the GRV threshold at the time of aspiration on the incidence of vomiting (RR 0.74, 95% CI 0.42 to 1.33; Analysis 3.4; Montejo 2010).

3.4. Analysis.

Comparison 3: Higher threshold (500 mL or greater per 6 hours) versus lower threshold (less than 500 mL per 6 hours) for GRV at the time of aspiration, Outcome 4: Vomiting

Duration (hours) for reaching the target calories per day during enteral nutrition

The study did not report duration for reaching the target calories per day during EN.

Volume aspirated from the stomach

The study did not report volume aspirated from the stomach.

Adverse events

The study did not report adverse events.

Returning versus discarding the aspirated/drained gastric residual volume

Two studies compared returning versus discarding the aspirated/drained GRV (Table 4). The monitoring frequencies of both intervention and control groups was six hours in Juvé‐Udina 2009. However, there were no data on monitoring frequency in Booker 2000.

Mortality

Neither study reported mortality.

Pneumonia

Neither study reported pneumonia.

Length of hospital stay

Neither study reported length of hospital stay.

Vomiting

Two studies (140 participants) found very low‐certainty evidence about the effect of returning or discarding the aspirated/drained GRV on the incidence of vomiting as compared to discarding it (RR 1.00, 95% CI 0.06 to 15.63; Analysis 4.1; Booker 2000; Juvé‐Udina 2009).

4.1. Analysis.

Comparison 4: Returning versus discarding the aspirated/drained GRV, Outcome 1: Vomiting

Duration (hours) for reaching the target calories per day during enteral nutrition feeding

Neither study reported duration (hours) for reaching the target calories per day during EN feeding.

Volume aspirated from stomach

Two studies (140 participants) found very low‐certainty evidence about the effect of returning or discarding the aspirated/drained GRV on the volume aspirated from stomach (MD –7.30 mL, 95% CI –26.67 to 12.06; I² = 0%; Analysis 4.2; Booker 2000; Juvé‐Udina 2009).

4.2. Analysis.

Comparison 4: Returning versus discarding the aspirated/drained GRV, Outcome 2: Volume aspirated from the stomach

Adverse events

Neither study reported adverse effects.

Subgroup analysis

We did not carry out any subgroup analyses for the current review due to limited data. However, for future updates of this review, we will reassess this should we identify and include further relevant randomized evidence.

Sensitivity analysis

We performed a sensitivity analysis exploring the impact of the chosen analytical model (fixed‐effect or random‐effects) on our overall review findings and found the models yielded similar results (results were not shown). We were unable to further explore the influence of selection bias in included studies as a sensitivity analysis since all the included studies were at low or unclear risk of selection bias.

Discussion

Summary of main results

Our review findings indicated that the evidence is uncertain about the effect of any type of monitoring for GRV during EN feeding on the reduction of mortality rate, the occurrence of pneumonia, and length of hospital stay. The certainty of evidence was very low. Most analyses included only two studies, with some outcomes only reported by one RCT.

Overall completeness and applicability of evidence

This Cochrane Review included relevant randomized evidence regarding four GRV monitoring relevant comparisons, and we imposed no restrictions on search methods. The eight included studies were conducted in the ICU settings, where participants receiving EN were critically ill requiring mechanical ventilation. In the comparison between no monitoring and frequent monitoring, the cutoff was set at 12 hours based on existing studies and clinical perspectives, but the actual timing of GRV confirmation was six hours (Reignier 2013) and eight hours (Ozen 2016). Although there was a difference in the GRV cutoff values for these two comparisons, all the studies actually included in each comparison were within eight hours (Booker 2000; Büyükçoban 2016; Juvé‐Udina 2009; Montejo 2010; Williams 2014), and the results would have been the same if the cutoff for the no monitoring comparison had been eight hours. Therefore, it is acceptable to consider these cutoffs to be eight hours. In addition, the severity of illness of the participants in the studies was moderate to severe according to the APACHE II score. We are aware that our study findings are somewhat difficult to apply to other settings/types of populations. Regarding a comparison of GRV monitoring frequency, two included studies compared GRV measurement every four hours and eight hours. In addition, comparing returning and discarding of GRV, the two included studies did not determine their threshold of GRV, and the heterogeneity of study methods between the two studies is considered high. Only one study was concerned with the GRV threshold, comparing 500 mL and 250 mL of GRV in six hours. However, in the comparison of no monitoring and frequent monitoring, there were two studies in the frequent monitoring group, and the frequencies of GRV monitoring were six and eight hours. Therefore, it was difficult to compare the frequency of GRV monitoring other than six to eight hours with no monitoring. Therefore, the applicability of the findings to GRV monitoring that used higher or lower frequencies may be limited. Furthermore, since the severity of the illness of the participants included in each study was mostly moderate to severe, it is uncertain whether the results of this review can be applied to people with illness of mild severity.

We identified three studies awaiting classification, which may provide further insights on protocol‐based EN strategy with GRV‐related criteria versus protocol‐based EN strategy without GRV‐relevant criteria versus non‐protocol‐based EN feeding strategy. We await these study findings and are certain that, once they are assessed for eligibility for inclusion, our review findings will be updated substantially, thereby increasing the relevance and value of our conclusions.

We also identified one ongoing study comparing no GRV monitoring with frequent monitoring. Details from the study register indicate that the study intends to include 138 participants and the study plan lists only one outcome measure of interest: pneumonia. Since the total number of the studies and participants included in this review were two RCTs and 449 participants, this ongoing study (138 participants) may affect the effect estimate and 95% CI in this review.

Quality of the evidence

We used the GRADE approach to assess the overall certainty of the evidence for the outcomes of mortality, pneumonia, length of hospital stays, vomiting, number of hours to reach the target calories per day in EN, and volume aspirated from the stomach. In our review, we did not downgrade based on blinding status, since the nature of the interview prohibited blinding of participants and study personnel.

In a comparison of frequent monitoring to less monitoring of GRV (Table 1), the certainty of the evidence for each outcome was very low according to the GRADE approach, with the following elaborations:

• mortality because of serious risk of bias, serious inconsistency, and serious imprecision;

• pneumonia and length of hospital stay because of serious risk of bias, serious imprecision, and insufficient optimal information size;

• vomiting: wide 95% CI due to small sample size (Büyükçoban 2016), limitations in study design (i.e. selection bias due to lack of details of randomization process and allocation concealment), and imprecision of results (wide 95% CI);

• number of hours to reach target calories per day in EN: study design (lack of details on randomization and concealment methods), imprecision, and indirectness.

Details of other comparisons are available in Table 2, Table 3, and Table 4.

Potential biases in the review process

Throughout the entire review process, we followed guidance from the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2021). In our review, there is only one ongoing study. However, this review has several limitations. First, among the four comparisons studied in this study, the cutoff value for GRV threshold was set to a value considered to be clinically valid. Consequently, existing studies that did not apply this threshold cutoff as part of their study design were eventually excluded from the review. Thus, our review findings are likely to vary depending on the choice of GRV threshold cutoff. Second, among the outcome measures assessed in this study, pneumonia and vomiting were both endpoints as defined by study investigators, and it is common for studies to employ different clinical definitions. Finally, due to the small number of included studies, we could not assess publication bias.

Agreements and disagreements with other studies or reviews

Existing reviews investigated the effects of GRV monitoring during EN feeding (Guo 2015; Kim 2014; Kuppinger 2013; Pham 2019; Wang 2019; Wen 2019). However, several of these published reviews did not pursue meta‐analysis as a method for quantitative analysis. Only two systematic reviews with meta‐analyses provided further insights on the four comparisons in our review. Wang 2019 included data from RCTs and observational studies, and the main comparison was no monitoring versus frequent monitoring of GRV. This review showed that, when compared with monitoring GRV, no monitoring of GRV led to an increased rate of vomiting in critically ill people. Wen 2019 explored findings from only RCTs comparing the strategy of returning versus discarding the aspirated/drained GRV and reported no significant differences on GRV within 48 hours or incidence of vomiting. Our results are in agreement with those from Wang 2019 and Wen 2019, thereby indicating the robustness of our review.

Authors' conclusions

Implications for practice.

The evidence is uncertain about the effect of gastric residual volume (GRV) monitoring on mortality rates, pneumonia, and length of hospital stay. From a clinical perspective, GRV monitoring includes frequent monitoring, the threshold of GRV, and discarding the aspirated GRV. The precise strategy/frequency of GRV monitoring remains unclear. 

Implications for research.

We investigated four clinically significant comparisons of GRV monitoring strategies, although the number of included studies was small and thus our conclusions are somewhat limited due to the apparent lack of statistical power. Future studies with large sample sizes and sufficient follow‐up durations may produce different yet clinically relevant results. Since high‐certainty evidence is currently lacking for all the five comparisons included in this review, future research examining the effectiveness of GRV monitoring itself as well as the extent/degree of monitoring (i.e. how frequent?) and the GRV thresholds are of great relevance and importance. Furthermore, researchers in the field of nutritional supplements and gastric disorders should strive to clearly report the definitions of selected clinically relevant outcome measures, which would allow subsequent systematic reviewers to conduct meaningful data synthesis.

History

Protocol first published: Issue 5, 2019

Appendices

Appendix 1. CENTRAL search strategy (via Ovid Evidence‐Based Medicine Reviews Database (EBMR))

1. exp enteral nutrition/

2. exp Intubation, Gastrointestinal/

3. ((stomach or gastric or gastro* or intragastric or nasogastr* or naso‐gasstric or nasal or nose or duoden* or nasoduoden* or jejun* or nasojejun* or esophag* or oesophag* or fine bore or Ryles or "PEJ" or "PEG" or bowel* or intestine* or intestinal or gastrointestinal or postpylor* or post‐pylor* or transpylor* or trans‐pylor* or nasoenter* or gavage or enteral or enteric) adj3 (feed* or fed or feeding or tube* or intubat* or tubal or nutrition* or immunonutrition* or nutrient* or micronutrient*)).tw,kw.

4. ((feeding or fed or feed) adj3 (tube* or intubat* or tubal)).tw,kw.

5. (g‐tube* or ng‐tube* or j‐tube* or gj‐tube* or nj‐tube*).tw.

6. or/1‐5

7. (((gastric or gastro* or stomach) and (residual or residue or volume* or capacit*)) or "GRV").af.

8. exp Gastric Emptying/

9. ((gastric or gastro* or stomach) adj5 (empty* or evacuat*)).tw,kw.

10. or/7‐9

11. 6 and 10

Appendix 2. MEDLINE search strategy (via Ovid)

1. exp enteral nutrition/

2. exp Intubation, Gastrointestinal/

3. ((stomach or gastric or gastro* or intragastric or nasogastr* or naso‐gasstric or nasal or nose or duoden* or nasoduoden* or jejun* or nasojejun* or esophag* or oesophag* or fine bore or Ryles or "PEJ" or "PEG" or bowel* or intestine* or intestinal or gastrointestinal or postpylor* or post‐pylor* or transpylor* or trans‐pylor* or nasoenter* or gavage or enteral or enteric) adj3 (feed* or fed or feeding or tube* or intubat* or tubal or nutrition* or immunonutrition* or nutrient* or micronutrient*)).tw,kw.

4. ((feeding or fed or feed) adj3 (tube* or intubat* or tubal)).tw,kw.

5. (g‐tube* or ng‐tube* or j‐tube* or gj‐tube* or nj‐tube*).tw.

6. or/1‐5

7. (((gastric or gastro* or stomach) and (residual or residue or volume* or capacit*)) or "GRV").af.

8. exp Gastric Emptying/

9. ((gastric or gastro* or stomach) adj5 (empty* or evacuat*)).tw,kw.

10. or/7‐9

11. 6 and 10

12. randomized controlled trial.pt.

13. controlled clinical trial.pt.

14. random*.mp.

15. placebo.ab.

16. trial.ab.

17. groups.ab.

18. drug therapy.fs.

19. or/12‐18

20. exp animals/ not humans.sh.

21. 19 not 20

22. 11 and 21

Note: Lines 12‐21. RCT filter for MEDLINE: Cochrane sensitivity‐maximizing version (2008 revision); Ovid format. We made the following minor revision: we used “random*” instead of “randomized.ab” or “randomly.ab.” to capture word variations such as “randomised, randomization, random”.

Appendix 3. Embase search strategy (via Ovid)

1. exp enteric feeding/

2. exp stomach tube/

3. exp nasogastric tube/

4. exp nose feeding/

5. tube feeding/

6. ((stomach or gastric or gastro* or intragastric or nasogastr* or naso‐gasstric or nasal or nose or duoden* or nasoduoden* or jejun* or nasojejun* or esophag* or oesophag* or fine bore or Ryles or "PEJ" or "PEG" or bowel* or intestine* or intestinal or gastrointestinal or postpylor* or post‐pylor* or transpylor* or trans‐pylor* or nasoenter* or gavage or enteral or enteric) adj3 (feed* or fed or feeding or tube* or intubat* or tubal or nutrition* or immunonutrition* or nutrient* or micronutrient*)).tw,kw.

7. ((feeding or fed or feed) adj3 (tube* or intubat* or tubal)).tw,kw.

8. (g‐tube* or ng‐tube* or j‐tube* or gj‐tube* or nj‐tube*).tw.

9. or/1‐8

10. exp residual volume/

11. (((gastric or gastro* or stomach) and (residual or residue or volume* or capacit*)) or "GRV").af.

12. exp stomach emptying/

13. ((gastric or gastro* or stomach) adj5 (empty* or evacuat*)).tw,kw.

14. or/10‐13

15. 9 and 14

16. random: tw.

17. placebo:.mp.

18. double‐blind:.tw.

19. 16 or 17 or 18

20. 15 and 19

Note: Lines 16‐18, Hedge Best balance of sensitivity and specificity filter for identifying randomized trials in Embase. hiru.mcmaster.ca/hiru/HIRU_Hedges_EMBASE_Strategies.aspx

Appendix 4. CINAHL search strategy (via EBSCO)

1. (MH "Enteral Nutrition")

2. (MH "Intubation, Gastrointestinal")

3. TI ( stomach or gastric or gastro* or intragastric or nasogastr* or naso‐gasstric or nasal or nose or duoden* or nasoduoden* or jejun* or nasojejun* or esophag* or oesophag* or fine bore or Ryles or "PEJ" or "PEG" or bowel* or intestine* or intestinal or gastrointestinal or postpylor* or post‐pylor* or transpylor* or trans‐pylor* or nasoenter* or gavage or enteral or enteric) OR AB ( stomach or gastric or gastro* or intragastric or nasogastr* or naso‐gasstric or nasal or nose or duoden* or nasoduoden* or jejun* or nasojejun* or esophag* or oesophag* or fine bore or Ryles or "PEJ" or "PEG" or bowel* or intestine* or intestinal or gastrointestinal or postpylor* or post‐pylor* or transpylor* or trans‐pylor* or nasoenter* or gavage or enteral or enteric)

4. TI ( feed* or fed or feeding or tube* or intubat* or tubal or nutrition* or immunonutrition* or nutrient* or micronutrient*) OR AB ( feed* or fed or feeding or tube* or intubat* or tubal or nutrition* or immunonutrition* or nutrient* or micronutrient*)

5. S3 AND S4

6. TI ( feeding or fed or feed ) OR AB ( feeding or fed or feed )

7. TI ( tube* or intubat* or tubal ) OR AB ( tube* or intubat* or tubal)

8. S6 AND S7

9. TI (g‐tube* or ng‐tube* or j‐tube* or gj‐tube* or nj‐tube* ) OR AB ( g‐tube* or ng‐tube* or j‐tube* or gj‐tube* or nj‐tube* )

10. S1 OR S2 OR S5 OR S8 OR S9

11. TI (((gastric or gastro* or stomach) and (residual or residue or volume or capacity)) or "GRV") OR AB (((gastric or gastro* or stomach) and (residual or residue or volume or capacity)) or "GRV")

12. TI ( (gastric or gastro* or stomach) and (empty* or evacuat*) ) OR AB ( (gastric or gastro* or stomach) and (empty* or evacuat*) )

13. S11 OR S12

14. S10 AND S13

15. (MH "Randomized Controlled Trials")

16. TI ( random* or placebo* or blind* or double blind* ) OR AB ( random* or placebo* or blind* or double blind* )

17. S15 OR S16

18. S14 AND S17

Data and analyses

Comparison 1. More frequent (less than eight hours) versus less frequent (eight hours or greater hours) monitoring.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Mortality (at the end of follow‐up; up to 28 days)	2	417	Risk Ratio (M‐H, Random, 95% CI)	0.91 [0.60, 1.37]
1.2 Pneumonia	1	357	Risk Ratio (M‐H, Random, 95% CI)	1.08 [0.64, 1.83]
1.3 Length of hospital stay (days)	1	357	Mean Difference (IV, Random, 95% CI)	2.00 [‐2.15, 6.15]
1.4 Vomiting	1	60	Risk Ratio (M‐H, Random, 95% CI)	0.14 [0.02, 1.09]
1.5 Duration for reaching the target calories per day during enteral nutrition feeding (hours)	1	60	Mean Difference (IV, Random, 95% CI)	‐0.80 [‐4.88, 3.28]

Comparison 2. No monitoring versus frequent (12 or less hours) monitoring of GRV.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Mortality (at the end of follow‐up; up to 28 days)	2	500	Risk Ratio (M‐H, Random, 95% CI)	0.87 [0.62, 1.23]
2.2 Pneumonia	1	449	Risk Ratio (M‐H, Random, 95% CI)	0.70 [0.43, 1.13]
2.3 Length of hospital stay (days)	2	500	Mean Difference (IV, Random, 95% CI)	‐1.53 [‐4.47, 1.40]
2.4 Vomiting	2	500	Risk Ratio (M‐H, Random, 95% CI)	1.47 [1.13, 1.93]
2.5 Duration for reaching the target calories per day during enteral nutrition feeding (hours)	1	51	Mean Difference (IV, Random, 95% CI)	‐3.07 [‐5.75, ‐0.39]

Comparison 3. Higher threshold (500 mL or greater per 6 hours) versus lower threshold (less than 500 mL per 6 hours) for GRV at the time of aspiration.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
3.1 Mortality (at the end of follow‐up; up to 28 days)	1	322	Risk Ratio (M‐H, Random, 95% CI)	1.01 [0.74, 1.38]
3.2 Pneumonia	1	322	Risk Ratio (M‐H, Random, 95% CI)	1.03 [0.72, 1.46]
3.3 Length of hospital stay (days)	1	322	Mean Difference (IV, Random, 95% CI)	0.90 [‐2.60, 4.40]
3.4 Vomiting	1	322	Risk Ratio (M‐H, Random, 95% CI)	0.74 [0.42, 1.33]

Comparison 4. Returning versus discarding the aspirated/drained GRV.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
4.1 Vomiting	2	140	Risk Ratio (M‐H, Random, 95% CI)	1.00 [0.06, 15.63]
4.2 Volume aspirated from the stomach	2	140	Mean Difference (IV, Random, 95% CI)	‐7.30 [‐26.67, 12.06]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Booker 2000.

Study characteristics
Methods	Study design: parallel‐group RCT
Participants	Country: USA Setting: multi‐center; medical and surgical ICU Inclusion criteria: receiving nutritional support through a gastric feeding tube with anticipated duration of tube feeding > 48 hours Exclusion criteria: not reported Baseline characteristics Number randomized: 35 Number analyzed: 18 Intervention arm Age (mean): 62.1 (range 18–77) years Number of participants: 10 Gender (men/women; n): 5/5 Control arm Age (mean): 58.9 (range 17–83) years Number of participants: 8 Gender (men/women; n): 4/4 Baseline imbalances: comparable
Interventions	Intervention arm All GRV discarded (frequency of monitoring: no data) Control arm All GRV returned (frequency of monitoring: no data) Loss to follow‐up: 17 participants withdrew from the trial: death (n = 2); transfer from ICU before completion of 3 days (n = 6); early removal of feeding tube (n = 5); missing data (n = 4). There was no mention of which arm these 17 participants belonged to. Duration of follow‐up: 3 days Duration of trial: September 1995 to June 1997
Outcomes	Volume aspirated from stomach
Notes	Funding source: Sigma Theta Tau, Delta Lambda Chapter, St Louis University Conflict of interest: unknown Clinical trial registration: unknown Duration of follow‐up was changed from 7 days due to serious numbers of missing data. Any disease severity scores of participants were not reported.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	People who consented were assigned to treatment group randomly. Details of randomization methods not provided.
Allocation concealment (selection bias)	Unclear risk	Multi‐center (3 centers) study and often for such a study design, a centralized allocation method would be performed. However, for this study there were no further details on allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Unblinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Unblinded. For subjective outcomes, such as pneumonia and adverse events, the risk of detection bias would be high; for objective outcome measures, such as mortality, the risk of detection of bias would be low.
Incomplete outcome data (attrition bias) All outcomes	High risk	Data for 17 participants were omitted from analysis.
Selective reporting (reporting bias)	Low risk	All outcomes were reported.
Other bias	High risk	Lack of full‐time clinical research monitoring by study investigators.

Büyükçoban 2016.

Study characteristics
Methods	Study design: parallel‐group RCT
Participants	Country: Turkey Setting: single center; medical and surgical ICU Inclusion criteria: aged > 18 years, who were planned to be under mechanical ventilation support and planned to be fed with EN for following ≥ 3 days Exclusion criteria: mechanical bowel obstruction, paralytic ileus, generalized peritonitis, inflammatory bowel disease, fistula in distal duodenum (if output > 500 mL/day), gastrointestinal bleeding, short bowel syndrome, morbid obesity (BMI > 40), having gastrostomy/jejunostomy process, and drug use affecting gastrointestinal motility Baseline characteristics Number randomized: not reported Number analyzed: 60 Intervention arm Age (mean): 54.8 (SD 22.2) years Number of participants: 30 Gender (men/women; n): not reported APACHE II score (mean): 19.6 (SD 9.2) points Control arm Age (mean): 54.1 (SD 22.8) years Number of participants: 30 Gender (men/women; n): not reported APACHE II score (mean): 17.8 (SD 8.0) points Baseline imbalances: comparable
Interventions	Intervention arm GRV monitoring: threshold 100 mL; interval 4 hours Control arm GRV monitoring: threshold 200 mL; interval 8 hours Loss to follow‐up: number of participants who withdrew from trial not reported Duration of follow‐up: 3 days Duration of trial: unknown
Outcomes	ICU mortality Vomiting Hours to reach the goal rate of feeding
Notes	Funding source: none Conflict of interest: none Clinical trial registration: unknown Duration of follow‐up for ICU mortality was not clearly mentioned.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Participants randomized into 2 groups in which there were sequential 30 per group. However, details of randomization not reported.
Allocation concealment (selection bias)	Unclear risk	No details about allocation concealment provided.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Unblinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Subjective endpoints were of high risk of bias due to the open‐label nature of study. However, objective outcomes such as mortality were assessed at low risk of bias.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All randomized participants were analyzed.
Selective reporting (reporting bias)	Low risk	Although the study protocol was not available, published reports included all expected outcomes.
Other bias	Low risk	Review authors believed the study free of other sources of bias.

Chen 2015.

Study characteristics
Methods	Study design: parallel‐group RCT
Participants	Country: China Setting: single‐center stroke care unit Inclusion criteria: people with stroke aged > 18 years, who were hospitalized for > 96 hours; had a definite indication for EN; planned ICU length of stay > 5 days Exclusion criteria: immunocompromised; hepatic failure; gastrointestinal issues postoperatively Baseline characteristics Number randomized: 210 (authors allocated participants into intervention and control arms with the ratio of 3:2) Number analyzed: 206 Intervention arm Age (mean): 68.1 (SD 17.5) years Number of participants: 126 Gender (men/women; n): 71/55 APACHE II score (mean): 19.1 (SD 3.4) points Control arm Age (mean): 61.3 (SD 14.3) years Number of participants: 80 Gender (men/women; n): 44/36 APACHE II score (mean): 18.7 (SD 4.1) points Baseline imbalances: comparable
Interventions	Intervention arm Without monitoring GRV Control arm GRV monitoring: threshold 200 mL; interval 4 hours Loss to follow‐up: 4 participants withdrew from intervention arm due to early discontinuation of EN. Duration of follow‐up: 3 days Duration of study: January 2010 to December 2012
Outcomes	No outcomes of interest reported
Notes	Funding source: Guangdong Science and Technology Foundation Conflict of interest: none Clinical trial registration: unknown
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants were randomly assigned into 2 groups. However, detail of randomization not described. We contacted the authors who replied that they randomized using a random number table.
Allocation concealment (selection bias)	Unclear risk	Details not described.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Unblinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Blinding was not done for outcome assessors. Because of unblinding, subjective outcomes are high risk of bias. However, objective outcome such as mortality are not affected by detection bias.
Incomplete outcome data (attrition bias) All outcomes	High risk	4 participants in the control group who did not continue EN for > 72 hours were excluded after randomization and were not included analysis.
Selective reporting (reporting bias)	Low risk	Although protocol was not available, published reports included all expected outcomes.
Other bias	Low risk	Review authors believed the study to be free of other sources of bias.

Juvé‐Udina 2009.

Study characteristics
Methods	Study design: parallel‐group RCT
Participants	Country: Spain Setting: single center; medical and surgical ICU Inclusion criteria: admitted to the ICU, aged ≥ 18 years, hemodynamically monitored, enterally or parenterally fed, needing GRV controls due to their condition and treatment, length of stay > 48 hours Exclusion criteria: connected to an intermittent gastric aspiration system because of paralytic ileum, bowel obstruction, gastric fistula, or gastric surgery Baseline characteristics Number randomized: 125 Number analyzed: 122 Intervention arm Age (mean): 60.2 (SD 16.7) years Number of participants: 61 Gender (men/women; n): 43/18 APACHE II score (mean): 18.8 (SD 6.4) points Control arm Age (mean): 55.6 (SD 16.4) years Number of participants: 61 Gender (men/women; n): 43/18 APACHE II score (mean): 20.2 (SD 7.0) points Baseline imbalances: comparable
Interventions	Intervention arm All GRV returned (frequency of monitoring: 6 hours) Control arm All GRV discarded (frequency of monitoring: 6 hours) Loss to follow‐up: 3 participants withdrew from the trial: 2 in the intervention arm because of shorter than expected length of ICU stay; 1 in the control arm because of major protocol error Duration of follow‐up: 10 days Duration of trial: unknown
Outcomes	Volume aspirated from stomach
Notes	Funding source: Collegi Oficial d'Infermeria de Barcelona Conflict of interest: none Clinical trial registration: unknown
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Used a computer‐generated random list and sealed envelope to randomize people to the GRV return group or the GRV discard arm.
Allocation concealment (selection bias)	Low risk	Used a computer‐generated random list and sealed envelope to randomize people to the GRV return group or the GRV discard arm.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Unblinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Blinding was not done for outcome assessors. Because of unblinding, subjective outcomes are high risk of bias. However, objective outcome such as mortality are not affected by detection bias.
Incomplete outcome data (attrition bias) All outcomes	Low risk	2/63 participants in the intervention arm and 1/62 assigned to the control group could not be included in the final analysis because of a shorter than expected length of stay and a major protocol error (GRV returned). This meant per‐protocol analysis.
Selective reporting (reporting bias)	Low risk	Although protocol was not available, published reports included all expected outcomes.
Other bias	Low risk	Review authors believed the study to be free of other sources of bias.

Montejo 2010.

Study characteristics
Methods	Study design: parallel‐group RCT
Participants	Country: Spain Setting: multi‐center; medical and surgical ICUs Inclusion criteria: adults in the ICU with mechanical ventilation and indication for EN for ≥ 5 days Exclusion criteria: people with duodenal or jejunal feeding Baseline characteristics Number randomized: 329 Number analyzed: 322 Intervention arm Age (mean): 65.0 (SD 27.5) years Number of participants: 157 Gender (men/women): 70.6%/29.4% APACHE II score (mean): 19.4 (SD 7.4) points Control arm Age (mean): 60.0 (SD 25.0) years Number of participants: 165 Gender (men/women): 63.4%/36.6% APACHE II score (mean): 18.9 (SD 7.5) points Baseline imbalances: comparable
Interventions	Intervention arm GRV threshold 500 mL; frequency of monitoring 6 hours Control arm GRV threshold 200 mL; frequency of monitoring 6 hours Loss to follow‐up: 7 participants (3 in intervention arm and 4 in control arm) were inappropriately included and were excluded. Duration of follow‐up: 28 days Duration of trial: February–September 2006
Outcomes	ICU acquired pneumoniaI ICU length of stay Hospital mortality
Notes	Funding source: Nestle Nutrition (Spain) for the statistical analysis Conflict of interest: unknown Clinical trial registration: unknown
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Co‐ordinating center performed a central randomization procedure by telephone call after each participant was included in each study center. Details of randomization process not described. However, this study was a multi‐center (28 ICUs) study and we judged that such a design would mean randomization process would be adequate.
Allocation concealment (selection bias)	Low risk	Randomization concealed. Co‐ordinating center performed a central randomization procedure by telephone call after each participant was included in each study center.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Unblinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	No blinding for outcome assessors. Because of unblinding, subjective outcomes are high risk of bias. However, objective outcome such as mortality are not affected by detection bias.
Incomplete outcome data (attrition bias) All outcomes	Low risk	7 participants (4 in control group and 3 in study group) were inappropriately included (non‐mechanically ventilated people at time of EN) and were excluded from the final analysis.
Selective reporting (reporting bias)	Low risk	Although protocol was not available, published reports included all expected outcomes.
Other bias	Low risk	Review authors believed the study to be free of other sources of bias.

Ozen 2016.

Study characteristics
Methods	Study design: parallel‐group RCT
Participants	Country: Turkey Setting: 2‐center medical ICU Inclusion criteria: aged > 18 years, planned to receive > 5 days of invasive mechanical ventilator treatment; where enteral feeding treatment would be started with a nasogastric tube Exclusion criteria: people with paralytic ileus, acute pancreatitis, pregnancy, inflammatory bowel disease, short bowel syndrome, Crohn disease, gastrointestinal bleeding, esophageal and fundic varices, morbid obesity (BMI > 40 kg/m²), or gastrostomy/jejunostomy; those receiving thoracic or abdominal radiotherapy Baseline characteristics Number randomized: 61 Number analyzed: 51 Intervention arm Age (mean): 68.08 (SD 18.04) years Number of participants: 26 Gender (men/women; n): 17/9 APACHE II score (mean): 25.29 (SD 7.43) points Control arm Age (mean): 64.92 (SD 14.95) years Number of participants: 25 Gender (men/women; n): 15/10 APACHE II score (mean): 23.58 (SD 6.97) points Baseline imbalances: comparable
Interventions	Intervention arm Without GRV monitoring Control arm GRV monitoring: threshold 250 mL; interval 8 hours Loss to follow‐up: 1 participant withdrew from the intervention arm due to death and 9 participants withdrew from the control arm: GRV amount > 250 mL twice (n = 1); continuous vomiting (n = 1); death (n = 3); cholestasis (n = 1); ileus (n = 1); extubation (n = 1); postoperative hemodynamic instability (n = 1) Duration of follow‐up: 5 days Duration of trial: March 2014 to April 2015
Outcomes	Vomiting Mortality Time in ICU Time to reach targeted volume
Notes	Funding source: Gulhane Military Medical Academy Scientific Research Council Conflict of interest: none Clinical trial registration: unknown
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomization sequence was "computer derived."
Allocation concealment (selection bias)	Unclear risk	Allocation concealment not described in detail.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Unblinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	No blinding of outcome assessors. Such detection bias would affect validity of subjective outcomes; for objective outcome such as mortality, our judgment was that their validity would not be affected by detection bias.
Incomplete outcome data (attrition bias) All outcomes	High risk	In GRV monitoring group, 9 participants were lost to follow‐up and excluded after randomization and thus not included in the overall analysis.
Selective reporting (reporting bias)	Low risk	Although protocol was not available, published reports included all expected outcomes.
Other bias	Low risk	Review authors believed the study to be free of other sources of bias.

Reignier 2013.

Study characteristics
Methods	Study design: parallel‐group RCT
Participants	Country: France Setting: multi‐center ICU Inclusion criteria: adults aged > 18 years admitted to ICUs between May 2010 and March 2011, expected to require > 48 hours of invasive mechanical ventilation, and started on EN via a nasogastric tube within 36 hours after intubation Exclusion criteria: abdominal surgery within the past month; history of esophageal, duodenal, pancreatic, or gastric surgery; bleeding from the esophagus, stomach, or bowel; contraindications to prokinetic agents; EN via a jejunostomy or gastrostomy; pregnancy; treatment‐limitation decisions; and current inclusion in a trial of VAP prevention, EN tolerance, or both Baseline characteristics Number randomized: 452 Number analyzed: 449 (in the modified ITT analysis) Intervention arm Age (mean): 61 (SD 15) years Number of participants: 227 Gender (men/women; n): 159/68 SAPS II score (mean): 49 (SD 17) points Control arm Age (mean): 62 (SD 14) years Number of participants: 222 Gender (men/women; n): 156/66 SAPS II score (mean): 51 (SD 16) points Baseline imbalances: comparable except for acute organ/system failure at ICU admission
Interventions	Intervention arm Without GRV monitoring Control arm GRV monitoring: threshold 250 mL; interval 6 hours Loss to follow‐up: 3 participants withdrew from the intervention arm due to withdrawal of consent. Duration of follow‐up: 90 days Duration of trial: May 2010 to March 2011
Outcomes	VAP Hospital length of stay Mortality Vomiting
Notes	Funding source: The Centre Hospital Departmental de la Vendee Conflict of interest: none Clinical trial registration: ClinicalTrials.gov Similar results were obtained in the per‐protocol analysis.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomization and concealment were achieved using a secure, computer generated, interactive, web‐response system managed by the biometrical unit of the Tours University Hospital, which had no role in recruitment."
Allocation concealment (selection bias)	Low risk	Quote: "Randomization and concealment were achieved using a secure, computer generated, interactive, web‐response system managed by the biometrical unit of the Tours University Hospital, which had no role in recruitment."
Blinding of participants and personnel (performance bias) All outcomes	High risk	Unblinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Primary outcome (VAP) was assessed by an adjudication committee blinded to the participant group. Assessment of other outcomes might not be blinded. Since mortality is an objective outcome, we judged that it would not be affected by detection bias and thus the overall judgment was low risk.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All analyses were conducted in both a modified ITT population and a per‐protocol population. The modified ITT population comprised all randomized participants except those who withdrew consent.
Selective reporting (reporting bias)	Low risk	This study was registered in ClinicalTrials.gov (identifier: NCT01137487). Published reports included all expected outcomes.
Other bias	Low risk	Review authors believed the study to be free of other sources of bias.

Williams 2014.

Study characteristics
Methods	Study design: parallel‐group RCT
Participants	Country: Australia Setting: single center; medical and surgical ICU Inclusion criteria: admitted to ICU and stayed > 48 hours, with a gastric tube inserted, and were likely to receive enteral feeding for ≥ 3 days Exclusion criteria: none Baseline characteristics Number randomized: 360 Number analyzed: 357 Intervention arm Age (median): 51 (IQR 36–65) years Number of participants: 178 Gender (men/women): 68%/32% APACHE II score (mean): 23.4 (SD 7.8) points Control arm Age (median): 54 (IQR 37–69) years Number of participants: 179 Gender (men/women): 70%/30% APACHE II score (mean): 23.1 (SD 7.7) points Baseline imbalances: comparable
Interventions	Intervention arm Variable regimen of up to 8 hourly GRV monitoring Control arm GRV monitoring every 4 hours Loss to follow‐up: 1 participant withdrew from the intervention arm due to breakage of allocation concealment and discharge; 2 participants withdrew from the control arm due to being randomized twice. Duration of follow‐up: 16 days (for VAP/pneumonia) Duration of trial: 13 October 2010 to 16 December 2011
Outcomes	VAP/pneumonia Hospital length of stay Hospital survival
Notes	Funding source: the Western Australian Nurses Memorial Charitable Trust, Australian College of Critical Care Nurses, Royal Perth Hospital Medical Research Fund, and the Royal Perth Hospital Foundation for Nursing Research Conflict of interest: unknown Clinical trial registration: Australian Clinical Trials Registry
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants randomized using computer‐generated randomization to receive either tube aspiration every 4 hours (standard practice) or a variable regimen of up to 8‐hourly gastric tube aspirations.
Allocation concealment (selection bias)	Low risk	The participant's allocation was concealed in sealed, opaque envelopes that were numbered sequentially and locked in a cupboard in a secure area until the participant was enrolled.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Study was not blinded due to the nature of the intervention.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Study was single‐blinded. For the primary outcome (pneumonia), blinding of outcome assessment was planned. An independent radiologist had been employed to conduct this review during the study. Other outcomes were assessed without blinding.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All analyses were performed on an ITT analysis (all participants randomized at baseline were included in the final analyses).
Selective reporting (reporting bias)	Low risk	Although protocol was not available, published reports included all expected outcomes.
Other bias	Low risk	Review authors believed the study to be free of other sources of bias.

APACHE: Acute Physiology and Chronic Health Evaluation; EN: enteral nutrition; GRV: gastric residual volume; ICU: intensive care unit; IQR: interquartile range; ITT: intention to treat; n: number; RCT: randomized controlled trial; SD: standard deviation; VAP: ventilator‐associated pneumonia.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Cao 2017	Wrong interventions: participants randomized to the intervention group were monitored by ultrasound examination while those from the control group were monitored by syringe withdrawal.
Metheny 2005	Wrong interventions: the GRV threshold in both groups were < 500 mL per 6 hours. Both groups fall into the control group of comparison 3.
Pinilla 2001	Wrong interventions: the GRV threshold of GRV in both groups were > 500 mL per 6 hours. Both groups fall into the intervention group of comparison 3.

Characteristics of studies awaiting classification [ordered by study ID]

Anandika 2018.

Methods	Study design: RCT Follow‐up period: unclear Duration of study: unclear
Participants	Country: India Setting: ICU Inclusion criteria: people with NG, people with orogastric tube feeding, normal gastric function, aged 18–75 years Exclusion criteria: people with short gut with jejunostomy feed, admitted with gastric intolerance, admitted with acute pancreatitis after intestinal surgery, with high ionotropic support Ileostomy, NJ tube feeding, partial intestinal failure, GI intolerance, death/discharge before 48 hours of observation Enrolled: not reported Age: 18–75 years Sex: men and women
Interventions	Intervention arm Feeding pump method of enteral feeding Control arm Bolus method of enteral feeding
Outcomes	Diarrhea Volume of aspirated gastric content Abdominal girth Vomiting Aspirational pneumonia
Notes	We could not determine whether the protocol of control group included GRV monitoring. We contacted to author, but received no response.

Nasiri 2017.

Methods	Study design: RCT Follow‐up period: unclear Duration of study: from October 2015 to September 2016.
Participants	Country: Iran Setting: ICU Inclusion criteria: aged 18–65 years; score of level of consciousness < 10; inability to swallow food orally and having indications for NGT feeding; have indications for both bolus or intermittent feeding methods; have healthy digestive system, which was confirmed by critical care medicine fellowships; lack of fistula, necrosis, obstruction, and surgery of GI system; lack of peritonitis; no history of addiction, cigarette, and alcohol abuse Exclusion criteria: development of ability to eat orally during the study; increasing level of consciousness from the baseline; discharge from ICU before 10 days of hospital admission; development of hemodynamic instability; need for urgent diagnostic or therapeutic procedures; presence of any signs of dehydration or addiction; need for changing diet or prescribing a specific diet during the study Enrolled: 60 Age: 18–65 years Sex: men and women
Interventions	Intervention arm Feeding started after confirmation of NGT position and control of input and output including the monitoring of GRV Control arm Based on the routine procedure of the recruitment center
Outcomes	Enteral feeding intolerance
Notes	We could not confirm whether the protocol of control group included GRV monitoring. No response upon contacting the study investigators.

Yaghoubinia 2017.

Methods	Study design: RCT Follow‐up period: unclear Duration of study: unclear
Participants	Country: Iran Setting: ICU Inclusion criteria: age 16–65 years; passing of 2 days after the intubation time; no history of pneumonia, acute respiratory distress syndrome and chronic obstructive pulmonary disease; standard diet formula based on protocol; lack of wound on abdominal region; lack of motion limits in neck and vertebral column; not pregnant Exclusion criteria: discharge, transfer, or death; weaning from mechanical ventilation within 96 hours; undergoing tracheostomy procedure during the study; initiation of prokinetic agents during the study; removal or changing of the tracheal or the NG tubes during the study; receiving nil by mouth; occurrence of diarrhea Enrolled: 35 Age: 16–65 years Sex: men and women
Interventions	Intervention arm Designed care program for reducing pneumonia incidence in addition to the routine care, including the monitoring of GRV Control arm Routine care pathway of each ICU
Outcomes	Pneumonia
Notes	Whether the protocol of control group included GRV monitoring is unclear. No response from study authors.

GI: gastrointestinal; GRV: gastric residual volume; ICU: intensive care unit; NG: nasogastric; NGT: nasogastric tube; NJ: nasojejunal; RCT: randomized controlled trial.

Characteristics of ongoing studies [ordered by study ID]

IRCT20170118032029N3.

Study name	Comparison the effect of two nasogastric feeding methods, monitoring and not monitoring residual gastric volume on risk of ventilator‐associated pneumonia
Methods	Study design: parallel‐group RCT
Participants	Country: Iran Setting: single‐center ICU Inclusion criteria: hospitalization in ICU; connected to ventilator; nutrition by NGT Exclusion criteria: people with renal failure, pneumonia, history of abdominal surgery; pregnant women
Interventions	Intervention arm Received NG feeding after the measurement of gastric secretion volume based on international guidelines Control arm Received NG feeding in a routine manner without the volume of residual gastric secretion measurement
Outcomes	Pneumonia
Starting date	22 June 2018
Contact information	Simin Jahani, PhD; Jahanisimin50@yahoo.com
Notes	www.irct.ir/search/result?query=IRCT20170118032029N3

ICU: intensive care unit; NG: nasogastric; NGT: nasogastric tube; RCT: randomized controlled trial.

Differences between protocol and review

• The objective of the protocol was "To investigate the clinical efficacy and safety of monitoring GRV during enteral nutrition" (Yasuda 2019). During the process of peer‐review, it was changed to a clarified expression as follows: "To investigate the clinical efficacy and safety of monitoring GRV during EN. In particular, we sought to answer the following questions. Is GRV monitoring necessary for reducing mortality and EN‐related complications? If GRV monitoring is performed, how frequently should it be done to minimize the risks of mortality and EN‐related complications? Would a lower GRV threshold affect the incidence of mortality and complications during EN? Should residual gastric contents be discarded or returned after GRV monitoring in a view to reduce mortality and EN‐related complications?

• The type of EN feeding was restricted to gastric feeding and thus we have excluded involving EN feeding via tubes placed beyond the pylorus (postpyloric feeding).

• Because studies with different methods of measuring GRV in one RCT were excluded from this review, the following sentence was added to the types of intervention section: if GRV was measured in a different way in each group, we excluded that study from this review.

• Because the definition of one of the four comparisons, protocol‐based nutrition strategy with GRV‐relevant criteria, was ambiguous, the following sentence was added to 'Types of interventions': "For the purpose of this review, we defined a protocol‐based nutrition strategy with GRV‐relevant criteria as a regimen where the attending physicians were required to monitor the GRV for increasing/decreasing the EN volume. For protocol‐based EN strategy without GRV‐related criteria or any non‐protocol‐based nutrition strategies, such physician‐led monitoring of GRV was not implemented."

• Since we did not search Current Controlled Trials metaRegiser of Controlled Trials (mRCT), one of the search databases, we deleted the description.

Contributions of authors

Conceiving the protocol: HY, NK, RY, HT, YT, TA, SA, and YK.

Designing the protocol: HY, NK, RY, HT, YT, TA, SA, and YK.

Co‐ordinating the protocol: HY, HT, YT, and YK.

Writing the protocol: HY, SA, HT, YT, and YK.

Providing general advice on the protocol: HY, SA, HT, YT, and YK.

Securing funding for the protocol: HY, HT, YT, and YK.

Performing previous work that was the foundation of the current study: HY, SA, HT, YT, and YK.

Sources of support

Internal sources

• No sources of support provided

External sources

• No sources of support provided

Declarations of interest

HY: None known.

NK: None known.

RY: None known.

SA: None known.

TA: None known.

HT: None known.

YT: None known.

YK: None known.

New