Introduction

Enteral tubes are placed in patients who have a functional and accessible gastrointestinal (GI) tract but are not able to consume or absorb sufficient nutrients to sustain adequate nutrition and hydration. A wide variety of enteral tubes may be placed through the nares, mouth, stomach, or small intestine to provide liquid nutrition, fluids, and medications directly into the GI tract of patients in intensive care units (ICUs; Lord, 2018). Tubes may be wide or narrow and gastric or intestinal. They may be inserted nasally or through oropharyngeal placement. Although the use of enteral feeding tubes is beneficial, as with most healthcare interventions, there are risks. Critically ill patients with enteral tubes are at a greater risk of delayed gastric emptying (DGE) because of an increase of gastric residual volume (GRV), regurgitation or vomiting, gastroesophageal reflux, aspiration, and ventilator-associated pneumonia (Chen, Tzeng, Gau, Kuo, & Chen, 2013; Nasiri, Farsi, Ahangari, & Dadgari, 2017; Reignier et al., 2013) as well as of GI changes such as abdominal distension, constipation, vomiting, diarrhea, increased abdominal circumference, and subjective discomfort (Poveda, Castilho, Nogueira, Ferretti-Rebustini, & Silva, 2018). DGE increases with medical severity and is often seen in patients with trauma, shock, respiratory failure, gastroparesis, electrolyte disorders, poorly controlled diabetes mellitus, gastric outlet obstruction, ileus, recent surgery, cytokines, trauma, sepsis, and feeding intolerance and those requiring narcotic pain medications (Chen et al., 2013; Guo, 2015; Montejo et al., 2010). Approximately 50% of adults requiring mechanical ventilation (MV) experience DGE (Nguyen et al., 2008). DGE may cause a constant feeling of fullness, nausea, and vomiting, which negatively impacts nutritional status and increases aspiration risk, gastroesophageal reflux, pneumonia, and hospital length of stay (LOS; Dutch Dieticians Oncology Group, 2017; Guo, 2015; Stewart, 2014).

Controlling GRV is important for patient well-being and medical stability (Stewart, 2014). Several methods are currently used to control GRV, including changing the method of feeding (such as feeding after the pyloric outlet), prescribing medications to increase GI motility (such as metoclopramide, cisapride, herbal medicine, and pressure medicine), using acupuncture, and changing body position (DeLegge, 2011; Heydari & Emami Zeydi, 2014). Often, patients are moved and their positions are changed every 3 hours to prevent pressure sores. The right lateral (RL) position on a bed flat has been shown to reduce regurgitation and aspiration in adults (Hussein, 2012). The semirecumbent (SR) position, the upright positioning of the head and torso, while lying on the back has been found to decrease the aspiration of gastric contents in comparison with the supine position (Gocze et al., 2013). An angle of 45[degrees] has been identified as the most effective position for accelerating gastric emptying, as the gastric contents gravitate toward the duodenum in the SR position more easily than in other positions (Gocze et al., 2013). The prone position has been used for some patients but is often not well tolerated. Therefore, technical considerations preclude its routine use, and it has been associated with an increased rate of serious complications in ventilated adults (Gattinoni et al., 2001; Guerin et al., 2004). Furthermore, GRV does not differ significantly between the prone and supine positions (van der Voort & Zandstra, 2001). Findings across research studies on gastric emptying are inconsistent in adult patients (Chen et al., 2013; Gocze et al., 2013; Sanaka et al., 2013). One study reported that the gastric emptying of saline solutions was faster when lying in the RL position compared with the left lateral (LL) position, whereas no difference was found between lying in the RL and sitting positions. On the contrary, another study showed that gastric emptying of water occurred more quickly when a patient is in the RL position compared with the sitting position (Sanaka et al., 2013). In one investigation of the rate of gastric emptying of glucose solution, no significant difference was found between lying on the RL, lying on the LL, or the sitting position (Burn-Murdoch, Fisher, & Hunt, 1980).

Understanding the most appropriate position for critically ill and ventilated adult patients may facilitate reductions in the incidence of pneumonia, reflux, hospital LOS, mortality, and morbidity. This study aimed to assess the effects of the SR and RL positions on the GRV of mechanically ventilated, critically ill adults hospitalized in the ICU. It was hypothesized that patients in the SR position should exhibit rates of gastric emptying that are higher than those in other positions, as the gastric content should gravitate toward the duodenum in the SR position more easily than in other positions.

Methods

Design

A randomized, crossover clinical trial was implemented. A repeated-measures design was used in which each participant was assigned randomly to a sequence of different interventions. This crossover trial was "balanced," whereby all participants received two interventions (RL and SR positions) and participated for the same number of periods (3-hour interval between interventions), with all receiving three GRV measurements.

Participants and Setting

The study participants included 36 mechanically ventilated, critically ill patients in two medical ICUs in Bea'sat Hospital, Tehran, Iran, who were recruited between April and June 2017. Both ICUs were similar in terms of equipment, facilities, and patients and were considered as a single setting for the purposes of this study. Criteria for inclusion included between 18 and 65 years old, nasogastric tube (NGT) feeding only, requiring noninvasive MV 48 hours before and throughout the duration of the intervention, and no limitations to altering body position. The exclusion criteria included experiencing acute GI disorders, concern for GI motility, pregnancy, medications to increase GI motility taken within the last 2 weeks or during the study, change to nutrition regimen, transfer or discharge of the patient from the ICU, and discontinuing/weaning from MV. Patients were recruited using a convenient sampling method and were randomly assigned into one of two study groups using a random-numbers table and sealed envelope technique. We used a random-numbers table to generate the random sequence. Numbers were placed in the envelopes and were opened sequentially only after each participant's name was written on the appropriate envelope. To minimize the potential for bias, the researcher assistant who enrolled patients into this study was blinded to the random allocation sequence. All direct patient care was completed by an experienced ICU nurse with a master's degree.

Sample Size

A standard, two-sequence, two-period crossover design (m = 1) was used for trials with the objective of establishing therapeutic equivalence between the two different interventions. An 80% (1 - [beta] = 0.8) power with a confidence level of 99% was adopted to establish therapeutic equivalence. On the basis of the results of a previous study, the variance was estimated as 8.03² (

Equation (Uncited)

= 8.03)². The true mean difference was 5.36 (Aslani, Hanifi, Ahmadi, & Fallah, 2014). Furthermore, we assumed that the equivalence limit was 25% ([delta] = 0.25). A minimum of 50% difference was assumed for the equivalence limit based on the standard of the sample size in the crossover studies (Chow, Shao, & Wang, 2008). Thus, a minimum of 18 participants were required in each group for this study.

Equation (Uncited)

Ethical Approval

This study followed the principles of the Declaration of Helsinki, was approved by the Ethics Committee of the AJA University of Medical Sciences under Code No. IR.AJAUMC.REC.1394.48, and was registered with the Iranian Registry of Clinical Trials (http://www.irct.ir/) under No. IRCT2015110623446N6. The legal guardians of the participants were fully informed regarding which aspects of care were related to this study, the purpose of the study, the potential benefits of the interventions to the participants, the intended health-promotion benefits of the research, and the minimal risk and minimal burden involved in participation. Then, if they agreed, the patients' legal guardians provided written informed consent on behalf of the participants. All information regarding individual participants was kept confidential, and during all phases of the study, the legal guardians of the participants could withdraw their consent to participate without affecting the quality of care provided.

Procedure and Data Collection

The medical status of participants was collected using physical examinations conducted by the resident physician, reviews of medical records, and assessments of illness severity using the Acute Physiology and Chronic Health Evaluation (APACHE) III. The APACHE III scale measures illness severity using multiple variables including age, heart rate, mean blood pressure, temperature, respiratory rate, partial pressure of oxygen in arterial blood (PaO₂), arterial pH, alveolar-arterial O₂ difference (AaDO₂), hematocrit, white blood cell count, serum creatinine, 24-hour urine output, serum blood urea nitrogen, serum sodium, serum albumin, serum bilirubin, serum glucose, comorbidities, and the Glasgow Coma Scale (Knaus et al., 1991). Blood samples were collected for the APACHE III measurement. The total APACHE III scale score was classified into three levels (0-40: mild, 41-80: moderate, and > 80: severe), with a total possible score ranging from 0 to 299. In previous research, higher APACHE scale scores have been associated with higher rates of mortality and morbidity (Knaus et al., 1991). The NGT was replaced every 7 days per hospital protocol. All of the participants received a premade complete diet through NGT. The nutritional recipe provided 100 cc of nutrition and included seven cups of a meal powder mixed with 90-cc water (1 kcal/cc). The total number of calories and the required volume per individual patient were calculated by a nutritionist and a critical care medicine fellow using the Harris-Benedict equation, which takes into account the gender, body weight, height, age, and activity level of the patient, to determine the most appropriate volume. The metric basal metabolic rate (BMR) formula was used to determine the total daily energy expenditure (calories) for each participant:

Women: BMR = 655 + (9.6 x weight in kg) + (1.8 x height in cm) - (4.7 x age in years)

Men: BMR = 66 + (13.7 x weight in kg) + (5 x height in cm) - (6.8 x age in years)

As the participants were sedentary (little or no exercise), the patients' BMR was multiplied by 1.2 to calculate the total calories (Douglas et al., 2007). All feedings were bolus feedings, with 3 hours between each NGT feeding. In both groups, the intervention began after confirmation of NGT placement just before feeding, using aspiration of gastric fluid by syringe, and testing of content acidity along with auscultation of bowel sounds. Laryngoscopy was used if the position of NGT was uncertain. The gastric fluid was returned to the stomach. The next feeding was adjusted according to the intake during the previous meal (Dietitians Association, Australia, 2015). GRV was measured at three different points in time (Stage 1: 3 hours after the first feeding, Stage 2: 3 hours after the second feeding, and Stage 3: 3 hours after the third feeding of the study). Each GRV measurement was taken in a different interventional position. Stage 1 was always in the supine position on a bed flat. Positioning during Stages 2 and 3 varied by study group. For Group A, Stage 2 was in the SR position with the head of bed at an incline of 45[degrees] and Stage 3 was in the RL position. For Group B, Stage 2 was in the RL position and Stage 3 was in the SR position. Each positional change occurred immediately before an NGT feeding.

In this study, the presence of GI symptoms (abdominal distention, vomiting, and regurgitation) was evaluated and recorded. If any negative symptoms occurred, appropriate interventions were provided. Moreover, the feedings were postponed if medically indicated. If during the intervention, patients required a diagnostic or therapeutic procedure (e.g., x-ray, endotracheal suctioning) that required a change in the patient's body position, the study was stopped and postponed until the next day.

Data Analysis

The statistical analyst was blinded to the study goals and descriptions of each study group. The data were analyzed using SPSS Software Version 18. The Kolmogorov-Smirnov test was used to examine the normal distribution of the variables. The Fisher's exact test was used to examine differences in gender between the two groups. The Mann-Whitney U test was used to examine the differences in age between the two groups. The independent sample t test was used to examine the difference in the APACHE III scale score in both groups and the difference in GRV between the SR and RL positions at the three data collection points. The repeated-measures analysis of variance was used to assess the differences between the GRV in different positions over the three data collection points. A level of p < .05 was considered significant.

Results

Participant Characteristics

Of the 52 patients admitted to the hospital during the study recruitment period, 36 met the inclusion criteria, provided informed consent, and were included in the final analysis (Figure 1). The participants in the two intervention groups (Groups A and B) were similar in terms of gender (p = .72); age (p = .39); ICU LOS (p = .54); specific medical conditions in terms of diabetes mellitus, hypertension, acute renal failure, chronic renal failure, cerebrovascular accident, leukemia, and hepatic cirrhosis (p = .35); diagnosis (p = .467); and illness severity based on APACHE III scale score (p = .37). The participants' medical and background data were combined to describe the population. Slightly more than half (54.1%) of the participants were male, the mean age was 53.64 +/- 13.53 years (range: 23-65 years), the mean APACHE III scale score was 65.58 +/- 14.08 (range: 32-109), the mean GRV was 56.69 +/- 3.01, and the mean gavage volume was 150 +/- 41.40 cc (range: 100-200 cc).

Figure 1. Study flowchart.

Interventional Positioning

No significant difference was found in the GRV between groups while in the supine position (p = .085), SR position (p = .106), or RL position (p = .059; Table 1). The mean GRV was not significantly different either during Stage 2, which was measured after the second feeding (Group A in the SR position and Group B in the RL position; mean difference = -10.27, t = -1.74, p = .091), or during Stage 3, which was measured after the third feeding (Group A in the RL position and Group B in the SR position; mean difference = -10.94, t = 1.86, p = .071).

TABLE 1. Comparison of Patient GRV Across Groups in the Three Interventional Positions

When comparing the GRV across position and group, although the main effect of position was found to be statistically significant (F = 21.89, p < .001), no significant interaction between position and group was found (F = 1.49, p = .232; Table 2).

TABLE 2. Repeated-Measures ANOVA of the Effects of Position (Supine, SR, RL) and Group (A and B) on GRV

The effect of group (A vs. B) and position (supine, SR, or RL) on GRV was statistically significant for both groups (both at p = .001; Table 1, Figure 2). For Group A, GRV was significantly lower than at the previous data collection point at each measurement point after the first. Moreover, GRV was significantly lower in the SR position compared with the supine position in both groups (p < .05), and GRV in the RL position was significantly lower than in the supine position in both groups (p < .05). Finally, GRVs in the SR and RL positions, although significantly and respectively different from the supine position, were not significantly different from each other (p > .05; Table 3).

Figure 2. Comparison of gastric residual volume (GRV) by position (supine, semirecumbent [SR], and right lateral [RL]) and by group (A and B). Groups A and B were in the supine position in Stage 1. Group A was in the SR position in Stage 2 and in the RL position in Stage 3. Group B was in the RL position in Stage 2 and in the SR position in Stage 3.

TABLE 3. The Mean Difference in GRV by Groups by Position (Supine, SR, and RL)

Discussion

The hypothesis in this study was that, while in the SR position, participants would experience faster gastric emptying than while in other positions. The results of this study showed that altering body position effectively reduces GRV, with the SR and RL positions determined to be better at gastric emptying than the supine position. These findings suggest that positioning patients in either the RL or SR position accelerates the digestion of feedings. No significant difference between the SR and RL positions was identified in this study, although an increasing trend in GRV was identified for the RL position. No difference was found in terms of the order of positioning placement or of whether the patient was placed in the SR or RL position first. Similarly, Schallom, Dykeman, Metheny, Kirby, and Pierce (2015) found that elevating the head of the bed greater than 30[degrees] reduced oral secretion volume, reflux, and aspiration without inducing the development of pressure ulcers in gastric-fed adult patients receiving MV. In contrast, van der Voort and Zandstra (2001) showed that the GRV of mechanically ventilated adult patients in, respectively, prone and supine positions did not differ significantly. In addition, Hussein (2012) reported no significant difference between infants set, respectively, in RL and SR positions with regard to their mean GRV after feeding. Furthermore, Rezae et al. (2018) found an increase in gastric emptying in the RL position versus in the LL and supine positions in adults. Furthermore, Sanaka et al. (2013) showed that patients in the RL position experienced faster gastric emptying than their peers in the SR position. The use of the C-acetate breath test in this latter study to assess gastric emptying may have led to its divergent findings.

Patients in the ICUs, although admitted for a variety of problems and injuries and receiving standard treatments, may experience adverse complications. Some complications such as DGE may cause other serious complications such as pneumonia and malnutrition. Therefore, the awareness of the healthcare providers regarding the methods available to control GRV, its prevalence, and its risk factors is necessary to improve the health and quality of patient care in the ICU.

Limitations

Several limitations affect this study. First, only the SR, RL, and supine positions were used. Future studies using additional positions such as the prone and right recumbent positions are suggested. Second, GRV was quantified as the volume aspirated via an NGT. This aspiration technique may be an unreliable method to assess GRV. Further studies using more-reliable techniques such as the C-acetate breath test are recommended to assess gastric emptying. Third, it should be assessed whether returning to the supine position makes a difference in terms of GRV or whether the movement alone decreases GRV. Fourth, only adults were assessed in this study, and previous research indicates that infants and young children may experience significantly different results than adults in terms of gastric emptying. Thus, conducting studies that include participants covering multiple age groups is recommended.

Conclusions

Overall, the findings of this study suggest that posing patients in either the RL or SR position is effective in accelerating the digestion of feedings. Understanding the best positioning for critically ill patients in ICUs may help decrease DGE and thus decrease other serious complications such as pneumonia and malnutrition. The results of this study contribute to a better understanding of the relationship among patient position, time, and GRV.

Acknowledgments

This study was adopted from an intensive care nursing master's thesis performed at Aja University of Medical Sciences, Tehran, Iran. The authors appreciate the Aja University of Medical Sciences for financial support (Grant No. 594260) as well as the valuable assistance of the participating patients and their families. We also sincerely thank the Chief of Bea'sat Hospital, Tehran, Iran, and the staff of the intensive care units at this hospital.

Author Contributions

Study conception and design: ZF, MK

Data collection: MK

Data analysis and interpretation: ZF, SB, AZ

Drafting of the article: ZF, MK, SB

Critical revision of the article: ZF

References