Bolus Versus Continuous Enteral Feeding for Critically Ill Patients: A Systematic Review and Meta-Analysis

Enteral feeding is a crucial aspect of nutritional support for critically ill patients. However, the optimal feeding approach, whether bolus or continuous, remains a subject of debate. This systematic review and meta-analysis aimed to compare the outcomes of bolus feeding and continuous enteral feeding in critically ill patients. A systemic search was carried out in PubMed, Cumulative Index to Nursing and Allied Health Literature (CINAHL) Ultimate, Web of Science, Scopus, and Google Scholar to identify relevant studies. To ensure that we obtain the latest evidence on the topic, the search was limited to the last five years. Risk of bias assessments and meta-analyses were performed for relevant clinical outcomes. A total of nine randomized controlled trials (RCTs) were included, involving a total of 863 patients. All the studies were published between 2020 and 2023. High-risk performance bias was observed in seven studies, with unclear risk in two studies. In terms of clinical outcomes, no statistically significant differences were found between bolus and continuous enteral feeding in terms of diarrhea (odds ratio {OR} 0.60, 95% CI 0.27 to 1.30, p=0.20), constipation (OR 1.52, 95% CI 0.91 to 2.53, p=0.11), vomiting (OR 0.74, 95% CI 0.36 to 1.49, p=0.39), distention (OR 0.70, 95% CI 0.14 to 3.58, p=0.66), aspiration (OR 0.61, 95% CI 0.16 to 2.73, p=0.48), and gastric residual volume (GRV) (OR 0.80, 95% CI 0.30 to 2.15, p=0.66). Furthermore, no significant differences between bolus and continuous feeding were observed in terms of intensive care unit (ICU) mortality (OR 0.66, 95% CI 0.42 to 1.04, p=0.07), hospital mortality (OR 0.57, 95% CI 0.31 to 1.03, p=0.06), ICU length of stay (OR 0.70, 95% CI 0.50 to 1.90, p=0.25), and hospital length of stay (OR -0.86, 95% CI -3.04 to 1.33, p=0.44). This systematic review and meta-analysis suggest that bolus and continuous enteral feeding methods exhibit comparable outcomes in critically ill patients. However, both ICU mortality and hospital mortality outcomes were close to achieving statistical significance, which favored the continuous feeding approach.


Introduction And Background
The intensive care unit (ICU) is a fundamental section of any medical facility where critically ill patients receive treatments.In such patients, the provision of nutritional support is a foremost necessity.Satisfactory nutritional therapy can enhance the medical outcomes, nutritional status, and immune defense mainly associated with malnutrition [1].Prior studies have demonstrated that adequate nutritional provision can improve patient outcomes and survival [2,3].Some estimates suggest that approximately 10-60% of patients admitted in hospital are malnourished which can have serious implications for such patients [4].Furthermore, critical illness has associations with catabolic stress which enhances the multiorgan dysfunction risk, prolonged interval of stay in the medical center, and increased mortality and morbidity [5].Early <48 hours of progressive nutrition supplement initiation in critically ill persons with suitable protein provision might reduce the catabolism, and improve integrity of the gastrointestinal tract [6].Enteral nutrition (EN) is also a preferred nutritional source for infants in neonatal intensive care units [7].
Parental and EN have been widely described as methods of nutritional support; however, EN has several advantages over parental nutritional support.EN is more cost-effective and safer and it does not require central venous line access.EN aims to deliver nutrients to enhance the functioning of the immune system and prevent bacterial dislocation by maintaining gastrointestinal tract integrity, reducing muscle catabolism, and optimizing mucosal host defense [8].On the other hand, critically ill patients are also affected by gastrointestinal impairment due to some factors including gastric stasis, postoperative ileus, gut hypofusion, and administration of some sedatives and certain antibiotics [9].Despite the familiarity and widespread use of EN, the ideal dosing method is still controversial.Therefore, there is a need to further search for the optimal delivery method for EN.There are two main approaches to EN administration, intermittent/bolus and continuous EN.Intermittent bolus feeding involves administering EN multiple times a day, typically between four and six sessions, using a syringe, gravity pump, or feeding pump, with a dose of 250-750 mL.In contrast, continuous feeding employs a steady, hourly delivery over a 24-hour period through an electric feeding pump [10].
In the ICU, EN is commonly provided at a continuous degree that is consistent with the recent guidelines [11].It is thought that continuous feeding might be linked with a lower delivery of nutrition as compared to intermittent boluses where nutritional administration needs termination to asses for extubation or to facilitate the investigation.It can also confine patient movement and can alter the hormonal secretion of intestines leading to long-term metabolic complexities including insulin resistance and hyperglycemia [8].Intermittent EN is considered a more physiological approach as it enhances protein synthesis and mobility, and aids in maintaining the digestion and secretion of gastrointestinal hormones [10].On the other hand, clinical studies have indicated that intermittent EN has enhanced the risk of feeding intolerance in ICU patients [9,11].Other concerns related to intermittent EN administration are increased diarrhea and the possible risks of aspiration [9].In a recent study, McNelly et al. described that intermittent nutrition was less disordered by diarrhea and vomiting and it provided superior nutritional delivery in contrast to continuous feeding in patients with serious illness [12].
The findings of the studies mentioned above present a complex and conflicting picture for medical practitioners, leaving uncertainty about the preferable EN strategy for critically ill patients.Moreover, the safety of the two EN strategies is still controversial.Taken together both continuous and intermittent administrations might be favored depending on the clinical situation.Previously, some systemic reviews and meta-analyses tried to analyze bolus versus continuous enteral feeding; however, they did not include recent data [8,10].Therefore, we decided to perform a meta-analysis of randomized controlled trials (RCTs) published in the last five years.

Review Methodology
The present meta-analysis was conducted according to the published guidelines of the Cochrane Handbook for Systematic Reviews of Interventions [13].The predetermined protocol was duly registered in the International Prospective Register of Systematic Reviews (PROSPERO) with the registration number CRD42024497668.Furthermore, this systemic review and meta-analysis also adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [14].This meta-analysis only focused on RCTs, published in the last five years.As all the included studies already received ethical approval before publication, there was no need for ethical approval for this meta-analysis.The PICO framework for this systemic review and meta-analysis was as follows: patients (P) critically ill patients; intervention (I) bolus versus continuous enteral feeding; control (C) bolus versus continuous enteral feeding; and outcomes (O) incidence of diarrhea, constipation, vomiting, distention, aspiration, gastric residual volume (GRV), ICU mortality, hospital mortality, ICU length of stay, and hospital length of stay.

Search Strategy and Data Sources
Several key databases were explored to identify the relevant studies on bolus versus continuous enteral feeding for critically ill patients.The searched databases included PubMed, Cumulative Index to Nursing and Allied Health Literature (CINAHL) Ultimate, Web of Science, and Scopus.The search strategy included the use of keywords that were joined with Boolean operators "OR" and "AND."The keywords used included "intensive care units," "critical care," "continuous," "bolus," "feeding," and "enteral feeding."The detailed search strategy used for this meta-analysis is presented in the appendix.To further expand the literature search, Google Scholar was also explored.The inclusion criteria defined for this meta-analysis included - (1) studies that compared patient outcomes in bolus versus continuous enteral feeding, (2) studies that focused on critically ill patients, and (3) studies that were published in the English language.The exclusion criteria included studies that did not compare bolus and continuous feeding regimens and did not focus on critically ill patients.To ensure that the latest evidence was summarized in this meta-analysis, the search was restricted to the last five years only.RCTs published before that were not considered.

Data Extraction
After retrieving results from the database search, the files were transferred to a reference manager for combining all the search results.The reference manager used in this meta-analysis was EndNote.At this stage, any duplicate records were removed.After the removal of duplicates, a combined research file was then uploaded to Rayyan (Cambridge, MA: Rayyan Systems, Inc.), a software specifically built for conducting systematic reviews [15].At the crucial stage of study selection, two independent reviewers were actively involved in this process.For unbiased decision-making, reviewers had activated "blind" in Rayyan so they would not see each other's selections.Firstly, judgments were made based on record titles and abstracts.In the second stage, the blind was removed and inclusion or exclusion decisions were compared until a consensus resulted.In cases of conflict, a third reviewer was approached to give their input and help in the final decision-making process.Finally, all relevant information regarding the studies was extracted in an Excel file including demographic properties and study outcomes.

Risk of Bias Assessment
For the risk of bias assessment, two authors were involved.The risk of bias was detected by Review Manager (RevMan) version 5.3 software (London, UK: Cochrane Library).The evaluation considered various crucial factors, including selection bias, concealment of allocation, performance bias, detection bias, attrition bias, reporting bias, and other biases.In case of any conflicts, a third author was involved.

Outcomes Measures
The primary outcomes were ICU mortality, hospital mortality, ICU length of stay, and hospital length of stay.Secondary outcomes included incidence of diarrhea, constipation, vomiting, distention, aspiration, and gastric residual volume (GRV).

Data Analysis
The present meta-analysis employed the DerSimonian and Laird method, incorporating a random-effects model to address potential heterogeneity among studies [16].This method yields a consolidated estimate of the effect size, represented by the odds ratio (OR) for binary outcomes, accompanied by a 95% confidence interval (CI).The heterogeneity among the studies was evaluated using the I² statistic, which measures the proportion of total variation attributable to heterogeneity rather than random chance.A value of I² greater than 50% suggests considerable heterogeneity.The findings of the meta-analysis are depicted through forest plots, displaying combined effect sizes along with their associated 95% CIs.Statistical significance was determined by p-values, with a cutoff set at 0.05.

Included Studies
The systemic search conducted in various databases revealed 2993 studies (PubMed=815, CINAHL Ultimate=406, Web of Science=779, and Scopus=993).Additionally, 236 studies were identified from Google Scholar.After removing duplicated studies, 2608 were finalized for screening.Of the screened records, 2169 articles did not meet the inclusion criteria based on title and abstract.Finally, after removing all nonrelevant articles, nine RCTs were finalized for meta-analysis.Figure 1 shows the PRISMA flow diagram.

Characteristics of Included Studies
A total of 863 participants were included in nine RCTs.The ratio of bolus and continuous feeding arm was 1:1, with 430 patients included in the continuous feeding group and 433 in the bolus feeding group.The descriptive details of all the included studies are presented in Table 1.

Risk of Bias
Figure 2 and Figure 3 show the risk of bias assessment of all the RCTs.A minimal risk of bias in random sequence creation was identified in eight out of nine papers (89%).All studies (100%) demonstrated a minimal risk of selection bias, attrition bias, reporting bias, and other biases.Performance bias was deemed high in seven studies, while two studies had an unclear risk of bias in this aspect.Only one study exhibited a high risk of bias in blinding of outcome assessment [4,12,[17][18][19][20][21][22][23].

Discussion
Currently, there are only a limited number of clinical trials published to assess the difference between intermittent and continuous enteral feeding methods.This systematic review and meta-analysis aimed to compare the outcomes of bolus versus continuous enteral feeding in critically ill patients.In the present systemic review and meta-analysis, there was no significant difference in any outcome in both bolus and continuous enteral feeding groups.Despite this fact, both ICU mortality and hospital mortality were close to achieving statistical significance (p=0.07 and p=0.06, respectively).Both these outcomes favored continuous feeding approaches.However, more studies are required to prove this association.Furthermore, the total number of studies and patients is small, therefore, it is not possible to draw specific conclusions.The included RCTs encompassed a diverse population of critically ill patients, including adults and children, with varying conditions such as trauma, mechanical ventilation, and hemorrhagic strokes.
Our findings are supported by a previous meta-analysis that revealed that there was no difference in continuous and intermittent feeding.However, they reported that continuous enteral feeding reduced the diarrhea (risk ratio=0.42)[11].Similarly, a recent meta-analysis by Heffernan et al. reported that there was an increased risk of constipation in the continuous feeding group compared to the bolus feeding approach.However, they also found an insignificant difference in other outcomes [8].This pattern can be explained by the fact that gut motility primarily depends on stimulation from intraluminal contents.This is evident that intermittent feeding includes delivering larger intraluminal contents compared to continuous feeding where smaller amounts are given at a slow rate [24].Furthermore, continuous feeding often requires prolonged bed rest to ensure non-interruptible enteral nutrition.Prolonged bed rest can serve as a contributor to constipation as well [25].Previous research has shown that both continuous and bolus-feeding practice has their pros and cons; however, they are both effective and practical approaches [26].There was no statistically significant difference in metabolite patterns between bolus and continuous feeding in the present meta-analysis.Wilkinson et al., in their RCT, investigated 594 samples and 87 metabolites.Their findings showed no significant differences between feeding arms or over time and no correlation was found with quadriceps muscle mass changes [17].Similarly, McNelly et al. reported no significant difference in muscle loss between bolus and continuous feeding, indicating that intermittent nutritional feeding may be as effective as continuous feeding in preserving muscle mass.However, they identified that intermittently fed patients achieved higher protein (p<0.001) and energy (p=0.001)targets [12].
Our findings are not consistent with the meta-analysis of Ma et al. that included 14 clinical trials comprising 1025 critically ill patients [27].They found that intermittent feeding was associated with an increased risk of feeding intolerance, GRV, and aspiration compared to continuous feeding.However, most of the included studies in their meta-analysis had a high risk of bias.Generally, critically ill patients are more likely to develop feeding intolerance.However, no study included in this meta-analysis found significant differences in these outcomes.Veldscholte et al. focused on critically ill children and concluded that condensed feeding periods during the day did not lead to increased feeding intolerance [18].However, intermittent feeding with an overnight fast resulted in lower calorie intake.Lee et al. reported that continuous feeding achieved targeted nutrition more significantly than bolus feeding, although clinical outcomes, including ICU length of stay, gastrointestinal intolerance, mortality, and organ support, did not differ significantly [19].According to the European Society for Clinical Nutrition and Metabolism (ESPEN) guidelines, it is recommended that critically ill patients requiring nutritional support should be offered enteral feeding within 48 hours [11].These guidelines recommended continuous feeding rather than intermittent; however, we were unable to find any significant difference between continuous and intermittent feeding approaches.The quality of evidence included in our meta-analysis was high.
There are some limitations of the meta-analysis as well which should be kept in mind while interpreting the findings.First, the majority of the studies had a small sample size.Second, not all studies reported all the outcomes assessed in this meta-analysis.Therefore, the findings might not be supported by all studies.Third, in the majority of the studies, physicians or participants were not blinded, leading to highperformance bias.Fourth, we limited the duration of the search to five years which could impact the generalizability of the findings.The addition of prior studies might have impacted the overall findings of the meta-analysis.Apart from these, we only included studies that were published in English.There is also no report on the type of patients included in the studies.

Conclusions
In conclusion, the systematic review and meta-analysis indicate that bolus and continuous enteral feeding methods have comparable outcomes in critically ill patients.The choice between these methods may depend on patient-specific factors, preferences, and clinical context.The findings emphasize the importance of tailoring enteral feeding strategies to individual patient needs and highlight the need for further research in this area.The high-quality assessment of the included studies adds credibility to the overall findings of this systematic review.

FIGURE 1 :
FIGURE 1: PRISMA flow diagram of the systemic review and metaanalysis.CINAHL: Cumulative Index to Nursing and Allied Health Literature; PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses

FIGURE 7 :
FIGURE 7: Forest plot of hospital length of stay.SD: standard deviation; CI: confidence interval

TABLE 1 : A brief description of the included trials.
ICU: intensive care unit; LOS: length of stay; GRV: gastric residual volume; N/A: not attempted

TABLE 2 : Terms and strategy used for literature search.
MeSH: Medical Subject Headings; Ti/Ab: title/abstract; ICU: intensive care unit