Abstract
Rationale:
Measuring gastric residual volume (GRV) remains common in intensive care units (ICUs), despite weak evidence linking it to enteral feeding intolerance (EFI), gastric emptying, aspiration, or ventilator-associated pneumonia. Thresholds defining high GRV vary widely (75–1200 ml). This scoping review aimed to: (1) identify GRV thresholds indicating feeding intolerance in adult ICU patients; and (2) describe GRV-related management practices.
Methods:
This review was conducted following the Population–Concept–Context (PCC) framework. A comprehensive search was performed in PubMed and Web of Science, extracting studies published in English and Vietnamese from 2010 to 2024. Studies were eligible if they were original research, trials, pre-and-post intervention study, and case reports/case series, and reported both GRV threshold and specific management strategies, while reviews or other studies that only had either GRV cutoff or its interventions were excluded.
Results:
Among 73 records for screening, 25 studies were eligible. Most involved hemodynamically stable, mechanically ventilated ICU patients receiving enteral nutrition (EN). GRV monitoring was typically performed every 4–6 h, with the cutoff ranging from 150 to 500 mL. Most studies used 200–250 mL as the most frequently used reference point. Some other studies combined GRV and clinical symptoms–such as nausea, vomiting, or distension–to determine intolerance. When GRV was below the lower threshold, feeding was typically continued or increased by 10–30 mL/h. Elevated GRV prompted various interventions, including temporary suspension of EN, prokinetic administration, or transition to post-pyloric nutrition.
Conclusion:
Most studies used 200–250 mL as a cautious GRV threshold for an early recognition of feeding intolerance, and the time to check GRV was normally every 4–6 h. Management of high GRV for lower thresholds tended to be less absolute, by reducing feeding rate and prescribing prokinetics, while stricter management was performed if GRV exceeded the higher GRV threshold.
Introduction
Applying gastric residual volume (GRV) as an indicator of gastrointestinal intolerance is a common practice in adult intensive care unit (ICU) settings. This practice continues despite evidence suggesting that GRV measures do not correlate with enteral feeding tolerance, gastric emptying, the incidence of regurgitation, aspiration, or ventilator-associated pneumonia. 1 The continued use of GRV to assess gastrointestinal dysfunction and enteral feeding intolerance (EFI) may stem from the lack of alternative reliable functional markers for evaluating feed tolerance at the bedside. A 2022 scoping review examining potential biomarkers and functional indicators of gastric intolerance (GI) dysfunction and feeding intolerance in ICU patients found that no markers are currently appropriate for routine clinical application. 2 In the COSMOGI consensus, elevated GRV is also used as a symptom of intolerance to EN. Additionally, high GRV (GRV > 500 mL) is one way to define gastroparesis, apart from vomiting due to delayed gastric emptying (92.7% consensus at final). Experts concluded that GRV is countable, but cut-offs may vary, and interpretation of signs and symptoms may be subjective. 3
In contrast, there are many guidelines advising that GRV measures should no longer be monitored routinely in patients in the ICU. The American Society of Parenteral and Enteral Nutrition (ASPEN) recommended that holding EN for GRVs < 500 mL in the absence of other signs of intolerance should be avoided. 4 In some studies, the routine GRV measurement could delay the time to reach patients’ EN requirements. 5 However, the European Society of Parenteral and Enteral Nutrition (ESPEN) guideline on feeding intolerance in critically ill adults suggests delaying EN if the GRV exceeds 500 mL over 6 h, 6 and highlights the potential need to monitor GRV to ensure safe nutrition delivery in cases of feed intolerance. However, no direct correlation between measured GRV and incidences of pulmonary aspiration, ICU length of stay (LOS), and mortality or pneumonia was found. 7
Despite the above-mentioned recommendations, there is no clear agreement in research or clinical settings on what defines a high GRV, with reported thresholds varying widely from 75 to 1200 mL. 8 However, there is limited data available on current practices regarding GRV monitoring in adults. Inadequate achievement of nutritional goals during an ICU stay leads to malnutrition in the ICU, which has been linked to poorer clinical outcomes, such as an increased rate of mortality, 9 organ failure, and mechanical ventilation (MV). 10 Therefore, it is crucial that ICU practices do not interfere with the delivery of sufficient nutritional support.11,12
Due to the variability of opinions and practice of GRV, we performed a scoping review with the aim of identifying the thresholds of GRV indicating EFI among adult patients admitted to the ICU, and defining management of GRV appropriate for different thresholds.
Materials and methods
Search strategy
A scoping review protocol was developed a priori by the research team to ensure a systematic and transparent process. This review was conducted in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews) guidelines. The study aimed to address the following research questions: (1) What are the most commonly used GRV cut-off values for determining gastrointestinal intolerance in adult ICU patients? and (2) the current management strategies for high GRV according to specific medical institutions?
To define the eligibility criteria, the PCC (Population, Concept, and Context) framework was employed: Population – Adult patients aged ⩾ 18 years; Concept - GRV management and monitoring; Context
Eligibility criteria
We included studies that met the following criteria: (1) Reported the cut-off value of GRV and/or its management for the current review; (2) the population of interest involves adult patients aged ⩾ 18 years old and being admitted to the ICU; (3) published for 15 years since 2010; (4) original research, trials, pre-and-post intervention study, and case report/case series. In relevant studies of mixed populations, we accepted results presented separately for adult patients in the ICU as a subgroup. We excluded research if they (1) did not use GRV as a method for measuring gastrointestinal intolerance among ICU patients; (2) were systematic review, scoping review, or narrative review; (3) provided insufficient data regarding the thresholds or specific management protocols for high GRV.
Study selection
The process of selecting studies was performed by screening progress by two reviewers, both of whom screened articles from the PubMed and Web of Science databases. The process is described as follows: (i) review of title and abstracts against inclusion and exclusion criteria, and (ii) review of the full text. First, two reviewers independently screened titles and abstracts from two different databases, using the inclusion and exclusion criteria for potentially eligible articles. The results of each electronic search were exported to the reference manager Zotero, and duplicates were removed. In the second stage, full texts were retrieved, and then the results and methods were further screened to select eligible studies. The final list of articles was determined via discussion between two independent researchers.
Data extraction and management
The information was recorded on a Microsoft Excel spreadsheet. Two reviewers extracted the data independently according to the following items:
(i) General information: Author, year of publication, country of origin, and journal
(ii) Setting and population: Study design, inclusion and exclusion criteria, and sample size
(iii) Outcomes: Threshold values, frequency of GRV monitoring, and management strategies
Any discrepancies between reviewers were resolved through discussion to ensure consistency and reliability of the extracted data.
Results
Search results
A total of 94 records were identified, including 83 from database searches and 11 from additional sources. After removing duplicates, 73 records were retained for screening. Based on title and abstract review, 33 studies were selected for full-text assessment. Following a detailed eligibility evaluation, 25 studies met the inclusion criteria and were included in the final synthesis. The study selection process is illustrated in Figure 1.
The study procedure of article screening and inclusion.
Summary of included studies in the review
Characteristics and findings of the included studies are summarized in Table 1. Studies originated from the United Kingdom, Europe (the Netherlands, Czech Republic, Spain, Italy, Belgium, France), Asia (Thailand, Iran, Taiwan, Pakistan, Israel), and Australia. The studies were published between 2010 and 2024, with nearly one-third of these published between 2020 and 2024 (n = 7).5,7,8,13–16 Sample sizes ranged from 1617 to 188818 ICU patients.
General characteristics of the included studies.
The majority of the eligible studies were observational designs (n = 7).14,19–24 Among these, one study was a web-based survey, 8 and another one was designed as a mixed-method study. 25 There were 5 trials included,13,15,16,26,27 with 2 randomized controlled trials.16,26 Six research papers were designed as pre-and-post intervention/implementation study.5,28–32 Inclusion criteria commonly involve adult ICU patients requiring MV and EN. Several studies used additional criteria such as age range, 15 BMI,13,15 mNUTRIC score,13,15 or specific timing of EN initiation.5,31 Exclusion criteria frequently included gastrointestinal surgeries, bowel obstruction, acute pancreatitis, pregnancy, immunosuppression, high-dose sedatives, and end-stage disease.
GRV thresholds and their assessment frequency
Practical guidelines for GRV thresholds and the duration for checking GRV were shown in Table 2. The reviewed protocols showed a wide range of GRV thresholds. Many studies set the cut-off around 250 mL,7,13,21,29,31,33 while others allowed larger volumes, such as Compton et al. 28 used a threshold of 500 mL and Friesecke et al. 30 applied the 400 mL point.19,23 Jenkins et al. 8 concluded a large range of GRV cut-off values, between 200 and 1000 mL. Some protocols applied composite criteria: Destrebecq et al. 20 required either two measurements ⩾ 200 mL or one ⩾ 500 mL to trigger intervention, and Lakenman et al. 14 defined intolerance as two consecutive recordings of ⩾150 mL per day.
GRV threshold and its management.
The frequency of GRV monitoring also differed. Most studies reported intervals of 4–6 h.15,19,22,23,25,32 Some studies employed a shorter interval of every 3 h,7,28 while others used longer intervals, such as once daily. 14 Several protocols specified an initial check at 4 h followed by regular reassessments every 6–8 h, depending on tolerance.29,31 The study of Jenkins et al., which was a web-based survey in the UK, reported that GRV monitoring frequency varied between 2 and 8 h, depending on individual ICU practice at different institutions. 8
Management of high GRV
In terms of articles using only one threshold, all articles indicated that if the gastric aspirate volume was below the identified GRV for feeding intolerance, then the nurses would re-aspirate the aspiration and increase the infusion rate. The volume of increase and frequency of increase depended on individual protocol, normally an increase of 10–30 mL/h per 4–6 h, or even 8 h, or every day. Meanwhile, if the aspirate volume exceeded the GRV threshold, it depends on the value of the GRV threshold, given the management of large GRV. For example, if the threshold was set at a lower value (ex, lower GRV threshold at 200–250 mL), the current rate would be maintained and continued, aspirate would be discarded, and physicians/MD would prescribe prokinetics (metoclopramide I.V and erythromycin I.V). In contrast, in case the GRV threshold was set far higher than 200–250 (frequently 400 mL 30 or 500 mL), 28 EN would be held temporarily for the exact time as the frequency of the GRV check. If GRV was larger than the threshold once or twice, jejunal feeding would be considered, along with other management such as prokinetics prescription. According to Wiese et al. (2020). 5 If GRV was above 200 mL, two gradual steps of management should be implemented as follows: Prescribing prokinetics; reducing feeding rate to the last tolerated rate; considering jejunal feeding. A special protocol of Lakenman et al. (2022) 14 used 150 mL twice a day as the indicator of feeding intolerance, in which switching from nasogastric tube feeding to nasoduodenal tube feeding was applied (Table 2).
At the same time, articles using two threshold values might possess more complex management. Hrdy et al. 15 and Hsu et al. 23 not only used GRV to assess feeding intolerance of critically ill patients, but also applied signs and symptoms of intolerance, such as abdominal distention, nausea, vomiting, etc. Hsu et al. 23 revealed that GRV > 500 mL; Overt regurgitation or aspiration; 250 mL < GRV < 500 mL with presence of abdominal distention, nausea, or vomiting: nurses should delay EN for 1 h and recheck. Hrdy et al. 15 combined intolerance signs and GRV > 500 to decrease the infusion rate by 50%. In contrast, other authors only applied GRV to assess feeding intolerance, in various values and management accordingly. Kao et al. 17 used 300 mL as the lower threshold and 500 mL as the upper one. If 300mL < GRV < 500 ml once, intravenous metoclopramide would be administered. Meanwhile, Destrebecq et al. 20 used 200 and 500 mL, with twice the measurement of GRV as 200 ml equivalent to 500 mL in a single assessment. Different from the two studies above, in the study of Sumritpradit et al., 16 EN would be stopped, and medications would be prescribed when GRV > 400 mL. If 250mL ⩽ GRV ⩽ 400 mL once, delay EN for 1 h and then recheck. If the amount is still over 250 mL, the feeding was stopped for another hour before reassessing the gastric content, and the prokinetic agents were administered (Table 2).
When the GRV goes beyond the lower threshold at least once or twice, in several protocols, the infusion rate of the feeding formula would be reduced by half, 33 or administer intravenous prokinetics 31 or combine both methods of management.7,13,15,19,21
Apart from using the cut-off values of GRV as the main justification for clinical management, several studies show that the trends in volume and other signs of intolerance could be key factors for the decision of actions on EFI. Most studies used other signs and symptoms of gastrointestinal intolerance, such as vomiting, abdominal distension, diarrhea, overt regurgitation, or aspiration. Specifically, the web-based survey in the ICUs in the UK suggested that the majority of decisions about whether to hold or to continue EN were based on the trends in GRV volume rather than a one-off measurement. 8
Clinical outcomes, barriers, and possible associated factors of using GRV
Regarding clinical outcomes, most studies mentioned ventilator-associated pneumonia (VAP), ICU/hospital LOS, sequential organ failure assessment (SOFA) score, acute physiology and chronic health evaluation II (APACHE II) score, duration of MV, ICU/hospital mortality, and nutrition outcomes (Table 3). Notably, several studies indicated that higher GRV could increase APACHE II, SOFA score, LOS, length of MV, and mortality rate.7,23 Meanwhile, the studies of Montejo et al. 27 and Soroksky et al. 34 showed no statistical significance of the above-mentioned clinical outcomes. Poorer nutrition outcomes and late feeders were found in critically ill adult patients with GRV ⩾ 200–250 mL.18,22 There is a contradictory result of Ozen et al., 26 with the finding that the proportion of gastrointestinal intolerance was statistically lower in patients without GRV measurement, in comparison with the GRV measurement group.
Possible clinical barriers, associated factors, and clinical outcomes regarding high GRV.
We also found several barriers to applying GRV. This problem could be attributed to the insufficiency of evidence to support the practice of GRV monitoring, poor validity and reproducibility of GRV measurement, variability in threshold values for high GRV, variability in frequency of measurement (2–12 hourly), and lack of guidelines for the technique of monitoring GRV (only 28% of units have guidelines). 8 Furthermore, regular GRV measurement prevented ICU patients from achieving ⩾90% of their prescribed EN volume during ICU stay. 5
A prospective study by Saez de la Fuente et al. has reported that the prone position (PP) does not significantly compromise gastrointestinal function or increase the incidence of EFI compared to the supine position. 19 The switch from standard formula to peptide-based formula did not significantly change average GRV.16,19
Discussion
The use of GRV can be divided into two threshold levels: upper and lower. Some studies only use one GRV threshold, usually the absolute GRV, the threshold at which EN is often held. In studies in which dual thresholds were used, the cut-off values of GRV varied. Reducing EN rate, temporary postponement of EN, prokinetic prescription, and post-pyloric initiation were the most common methods for solving high GRV. GRV monitoring has commonly been used as a bedside indicator of EFI. GRV monitoring may provide clinicians with supportive information for identifying delayed gastric emptying earlier and guiding interventions to minimize the adverse effects. 35
In this scoping review, most studies applied a single fixed GRV cut-off at 250 mL7,13,21,22,26,29,33 to define elevated GRV or suspected feeding intolerance. Other protocols used range-based thresholds, most commonly between 200 and 500 mL19,20,24 or 250–500 mL. 15 One study adopted a very conservative threshold, 150 mL in two consecutive measurements, 14 to define elevated GRV. In contrast, a survey-based study described an extremely wide GRV range, varying from 200 to 1000 mL, 8 to characterize thresholds used across different institutions, highlighting substantial between-unit variability in routine practice. The wide dispersion of reported GRV cutoffs indicates that GRV has been conceptualized inconsistently across studies. Beyond the numerical values, the lack of standardization in measurement techniques - such as the choice between manual syringe aspiration (which may exert excessive negative pressure) and gravity drainage–further complicates the reliability of GRV as a universal metric.
Recommendations on routine GRV monitoring remain inconsistent. This inconsistency not only impacts the accuracy of EFI assessment but also introduces variability in actual nutritional delivery, potentially leading to underfeeding or unnecessary feeding interruptions. The American College of Gastroenterology and the Society of Critical Care Medicine/American Society for Parenteral and Enteral Nutrition (SCCM/ASPEN) recommend against routine GRV monitoring. 4 Overly frequent GRV monitoring not only increases the workload for healthcare staff but may also lead to unnecessary interruptions in enteral feeding, thereby reducing the actual caloric and protein delivery to the patient. Some evidence indicates that omitting routine GRV monitoring does not increase ventilator-associated pneumonia or mortality and may improve ENdelivery.1,27,36,37 Lakenman et al. 14 proposed a once-daily GRV check with a much lower intervention threshold (150 mL), reflecting the idea that frequent measurements may be unnecessary and that small-volume elevations could still indicate dysfunction if persistent. Despite these guidelines, our review found that most clinical protocols still mandate GRV checks every 4–6 h.5,13,15–17,19–23,25,30–33 This suggests a persistent reliance on GRV in practical settings, particularly in high-risk patients, such as those receiving MV, under deep sedation, or with severe neurological conditions. 29 Therefore, the frequency of GRV monitoring should also be individualized, based on clinical risk level, hemodynamic stability, and previously observed feeding tolerance.
An emerging perspective in this review is that elevated GRV may reflect the severity of the underlying illness rather than being a direct cause of gastrointestinal complications. Faramarzi et al. 7 and Hsu et al. 23 reported that high GRV levels were associated with higher SOFA and APACHE II scores, prolonged MV, and increased mortality. These findings underscore that impaired gastric emptying is often a systemic manifestation of critical illness, so clinicians should interpret GRV values cautiously and within the broader clinical context rather than as an isolated indicator of feeding intolerance. SCCM/ASPEN also suggests that GRV not be used to assess tolerance in critical care patients receiving EN. 4 Instead, it is suggested to monitor other markers of intolerance, including emesis and/or abdominal distention.4,14,16,22,24,29 Recently, GRV has not been listed as a core outcome set for daily monitoring of gastrointestinal function in ICU patients, according to Round 2 of the Delphi Stage 1 COSMOGI consensus. 3 However, high GRV was used to define gastroparesis, which was recognized as a core outcome of EFI. Additionally, the authors also concluded that GRV is quantifiable, but cut-offs may vary depending on study protocol, and interpretation of signs and symptoms may be subjective. This finding also supports our main aims and scopes to review GRV cut-off values. GRV could be used as a marker in different outcomes of EN in the ICUs, such as EFI and the indication of post-pyloric feeding. 3
However, Montejo et al. found that allowing GRVs up to 500 mL (vs 200 mL) did not increase ventilator-associated pneumonia or other complications, although it modestly improved early feeding. 27 Reignier et al. 1 similarly demonstrated that omitting routine GRV measurements did not raise VAP rates and allowed better caloric intake. The lack of consensus on the threshold at which GRV is considered “elevated” poses a significant challenge in clinical practice. While many centers adopt a conservative cutoff of 200–250 mL, the latest SCCM/ASPEN nutrition guidelines explicitly discourage holding feeds for GRVs below 500 mL in the absence of other signs of intolerance. 4 Lower thresholds (200–250 mL) are often applied in patients on MV, under sedation, or with neurological conditions due to their higher risk of aspiration and decreased gastric motility.16,18 ICU patients with high nutritional risk (according to mNUTRIC score were also set the threshold at 250 mL to evaluate feeding tolerance, as the mNUTRIC includes APACHE II and SOFA, which are the prognosis of outcomes and mortality in critically ill patients.13,15,38 Conversely, in patients without specific risk factors or in general ICU settings, as in the study by Lakenman et al. 14 – higher GRV thresholds (300–500 mL) have been applied to avoid unnecessary interruptions in enteral feeding and to ensure adequate energy delivery. However, this “lowering the GRV threshold to intervene” approach may inadvertently increase the risk of undernutrition by prompting unnecessary interruptions in enteral feeding. Consequently, there is a growing trend toward combining GRV assessment with clinical signs such as abdominal distension, abdominal pain, vomiting/regurgitation, diarrhea, or elevated intra-abdominal pressure to more comprehensively evaluate feeding tolerance. 3 Thus, although GRV retains a certain role in clinical practice, caution should be exercised in using it as the sole determinant for withholding or adjusting EN. An individualized approach that integrates both clinical and laboratory parameters will help optimize nutritional support without increasing the risk of complications.
Management strategies for high GRV were closely linked to the predefined cutoff values and were predominantly protocol-driven. When measured GRV remained below the threshold, enteral feeding was generally continued and advanced toward target rates.7,13,15,16,21–25,28,29,31–33 Re-infusion of aspirated gastric contents was commonly reported when GRV was below the intervention threshold, although this practice was not uniformly described across studies.7,17,20,33 When GRV exceeded the cutoff, management approaches varied but followed a stepwise escalation pattern in most protocols. For GRV levels exceeding the initial threshold but remaining below an upper limit (commonly < 500 mL), strategies included maintaining the current rate or a 50% reduction, followed by reassessment at the next scheduled GRV check.7,13,15,16,19,24,25,31,33 These interventions were typically implemented without immediate cessation of enteral feeding unless GRV remained persistently elevated. For marked elevations (commonly ⩾ 400–500 mL) or persistent elevations despite initial measures, protocols more frequently described temporary interruption of enteral feeding, gastric decompression, and consideration of post-pyloric feeding (e.g. nasojejunal tube placement). Pharmacological interventions, particularly the use of prokinetic agents such as metoclopramide or erythromycin, were frequently reported as first-line responses to elevated GRV.5,7,13,39 This aligns with SCCM/ASPEN and ESPEN 2019 guidelines, which advise withholding EN only when GRV exceeds 500 mL.4,40
Interestingly, non-pharmacologic aids have also been explored. Acupuncture at certain points has shown promise in augmenting gastric emptying. A small pilot study in neuro-ICU patients found that adding electroacupuncture to standard metoclopramide therapy resulted in greater reductions in GRV and maintenance of low residual volumes (< 200 mL) after 4 days, whereas metoclopramide alone saw GRVs rebound on days 5–6. Although this is preliminary data (n = 16 patients), it points to potential adjunctive therapies for refractory gastroparesis in the ICU. 17
The management of EN in patients requiring PP represents a significant area of controversy. Early observational evidence, most notably the 2004 study by Reignier et al., suggested that PP significantly exacerbates feeding intolerance. 41 The mechanisms cited include elevated intra-abdominal pressure and gastric emptying dynamics, which can further impair the gastrointestinal tract’s ability to tolerate enteral feeds. 42 However, recent evidence challenges this concern, suggesting that EN is safe in the PP when supported by specialized protocols. A prospective study by Saez de la Fuente et al. 19 found no significant difference in the mean GRV or frequency of vomiting when comparing patients in prone and supine positions. Successful clinical practice in PP often integrates maintaining the head of the bed in a reverse Trendelenburg position and the proactive use of prokinetic agents.
The scoping review has several limitations. Significant heterogeneity was observed across studies in terms of design, population characteristics, GRV cutoff definitions, and outcome measures. Many studies had small sample sizes or were conducted at single centers, limiting generalizability. Finally, the scope of this review was restricted to adult ICU populations and studies from the past 15 years, which may limit applicability to pediatric or older evidence bases.
Conclusion
This scoping review identified broad heterogeneity in GRV thresholds and enteral feeding practices in critically ill adults. While several guidelines support thresholds up to 500 mL or discourage routine GRV monitoring, many protocols still apply GRV with low cutoffs, commonly 200–250 mL. Where ICUs used only one cut-off value of GRV or dual GRV thresholds, stepwise management using prokinetics, and feed rate adjustment were widely reported. The most common practice of frequency to check GRV was every 4–6 h. GRV may not be an adequate criterion of EFI. Besides GRV thresholds and management, we should consider other signs and symptoms of EFI, or the trend in GRV, to decide whether we should hold EN or not. GRV may not be perfect, but it could still indicate possible feeding intolerance in the ICUs. Future research should focus on validating alternative markers of EFI and evaluating the clinical impact of GRV-guided interventions across diverse ICU populations.
Footnotes
Acknowledgements
The authors of this article thank all the doctors and scientists who participated in the study.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.