The Incidence and Characteristics of Gastrointestinal Intolerance on Mechanically Ventilated Patients with Continuous versus Intermittent Feeding Method in The Intensive Care Unit: A Randomized Controlled Trial

The study was conducted at the ICU, University Medical Center, Ho Chi Minh City, Vietnam from 3/2017 to 5/2017 using a convenience sample of patients available during the study period. Inclusion criteria: All patients aged 18 years or more who were machenically ventilated and nourished with enteral nutrition through nasogastric tube for 4 consecutive days or longer. Exclusion criteria: Patients who were connected to gastric aspiration system; with gastrointestinal tract problems such as reflux disease or cancer; who had had recent surgery on their digestive tracts; those with diabetes or hemodynamic instability, requiring inotropic medications or dialysis; patients suffering diarrhea or being prescribed with laxatives in the previous 72 hours.

After the informed consents were signatured, the nurses collecting data assigned randomly and equaly patients either to group 1 or group 2 according to a rule that the first patient would be enrolled in the continuous feeding group (group 1), the second other would be enrolled in the intermittent group (group 2). The next patients were assigned according to this rule until the end of the study period. After randomizing the patients, the nurses collecting data requested that physicians order enteral nutrition prescription with continuous or intermittent feeding. In the continuous group, feeding would be administered through a feeding pump with a speed at 20 mL/h for 4 first hours and advanced gradually by 20 mL/h after every 4 hours until the highest rate at 100 mL/h or reach the nutrition goal as the prescription. In the intermittent group, nutrition would be delivered through an enteral feeding set up used routinely in the hospital. The patients were monitored for 4 consecutive days. All patients were nourished with the same commercial feeding formula.

Data were collected by four nurses using a study tool composed by the authors. These four nurses with bachelor’s degree and 2 years of experience in this ICU were equally and well trained at the same time to ensure consistent data collection and accurate measurements. The gastrointestinal tolerance has been evaluated in terms of 4 components: the amount of gastric residual volumes (GRV), diarrhea score, and frequency of abdominal distention and tube occlusion events. The nurses aspirated and measured the amount of GRV every 4 hours in the continuous group and before each feeding in the intermittent group by a 50 mL-syringe until no more fluid could be withdrawn. The amount of GRV > 200 ml was considered an indicator of intolerance [14]. Tube occlusion’s presence was checked every shift by the data-collecting nurses. Patient’s waist circumference was measured at the beginning of the study and every 3 hours after feeding. The nurses also assessed feces’ characteristics of patients and recorded the scores for 24 hours daily for 4 consecutive days according to the guideline of King’s stool chart.

Data was analyzed in SPSS 20.0. The dependency between feeding methods and occurrence of abdominal distension or diarrhea was verified by Pearson’s Chi-squared test or Fisher exact test. The overtime fluctuation of GRV was evaluated by a repeated measure ANOVA. Student’s independent sample t-test was used to analyze the effect of two feeding methods on diarrhea scores. Median value of GRV was compared using Mann-Whitney test. P value less than 0.05 was chosen as a statistically significant threshold.

Forty-two out of one hundred-fifty-three patients meeting all the sampling criteria participated in the study. One patient from the continuous group was discharged and removed from the study on the 3^rd day. There were 41 subjects, who completed the study with 20 subjects in the continuous feeding group and 21 subjects in the intermittent feeding group.

Gastric residual volume

The amounts of GRV for both groups at the beginning of the study were not significantly different with a median of 10 ml (2.5-17.5) in the continuous feeding group and 10 ml (10-20) in the intermittent feeding group (p = 0.189) (Man-Whitney Test). During the study, the GRV measured at every 4-hour interval were consistently lower in the continuous feeding group (0.93, 95%CI = 0.63-1.83) compared to the intermittent feeding group (11.61, 95%CI = 8.53-19.16). There was a statistically significant difference when comparing the GRV between the two groups with p < 0.001 (Man-Whitney Test).

The patients in the continuous feeding group showed a significantly reduction in GRV during the study with a pooled median of 0.93 ml (0.09-1.93) compared to baseline value of 10 ml (2.5-17.5), p<0.001 by paired sample T test). In contrast, no significant change in GRV has been observed during the study in the intermittent feeding group with no significant difference in comparison with p = 0.511 (Paired sample T-test).

In the intermittent feeding group, there were measureable GRV before each feeding in all patients with one notable case experiencing a gastric emptying delay with 230ml of GRV existing on the 4^th day. In contrast, only 5/20 (25%) of patients in the continuous feeding group showed no GRV for the entire 4 days of study. The proportion of no GRV was significantly dependent on the choice of feeding methods, as confirmed by Fisher exact test (p=0.021).

The amount of GRV at baseline between the two groups was not statistically significantly different. However, the amount of GRV of patients fed continuously was significantly reduced and controlled at low values, compared to those fed intermittently over 4 consecutive days. Moreover, no interaction between time and feeding method could be expected, based on the longitudinal response graph (Figure 1).

Figure 1. Comparison the mean gastric residual volume between two groups over the 4 consecutive days Checking time: the first time at baseline; the remaining times as during the study Amount of GRV: the amount of gastric residual volume at each checking

Download Original Figure

The results demonstrated that under continuous feeding method, Gastric Residual Volume (GRV) measured every 4 hours during the study was significantly reduced compared to the baseline level, and consistently lower than that in the intermittent feeding method. In addition, there was a notable gastric emptying delay presented in one patient fed intermitently who had 230 ml of GRV at one point of checking. The volume and speed of feeding affected the gastric emptying speed and GRV of patients, especially those on mechanical ventilation. In each intermittent feeding, the patients received as much as 200ml within 30 minutes exacerbating gastric emptying. The larger amount of food patients received, the more stomach function decrease they developed. This contributed to the high GRV in these patients. Consequently, high GRV was reported in 100% of patients fed intermittently with the maximum amount of GRV of 230 ml found in one checking in one patient. Whereas, patients in the continuous feeding group recieved a continous and stable speed of feeding at 20-62.5 ml/hr that considerably improved their digestive function. In this group, GRV was diminished or eliminated.

Compared to study results by Serpa [13] with no significant differences between the two groups, the results of this study vigorously demonstrated the effectiveness of continuous feeding compared to intermittent feeding on the ability to empty the stomach and improve the absorption and tolerance of patient.

Similar to the study by Serpa [13], in this study, abdominal distention was determined when the patient’s waist circumference increased by 3cm or more compared to pre-intervention measurements. Abdominal distention is closely associated to the amount of food the patient receives in each feeding and the GRV, especially in poor digestive function. Therefore, in patients fed intermittently with high GRV, abdominal distention was likely to be more severe and happen more frequently. The study’s results illustrated that the rate of patients experiencing abdominal distention and the mean number of this occurrence were significantly lower in the continuous feeding group than those in the intermittent feeding group who experienced high GRV (p = 0.048 vs p < 0.001, respectively). Although all patients in this study were mechanically ventilated by endotracheal intubation which is one of most common causes of gas-induced bloating relating to endotracheal insertion, they were regularly evaluated by doctors and nurses for abdominal distention whenever there were any symptoms to confirm the causes of abdominal distention related to gas-induced bloating or impaired digestion. During the study period, we did not record any abdominal distention episodes caused by endotracheal intubation in any subjects. Compared to the study results by Serpa [13] with no significant differences between two groups, the results of this study clearly demonstrated the effectiveness of continuous enteral feeding compared to intermittent enteral feeding in decreasing patients’ abdominal distention.

In the continuous feeding group, the mean diarrhea scores were higher for the first two days and lower for the last two days compared to the intermittent group. This can be explained in part by the fact that when patients were admitted into the ICU, instead of being maintained on a normal physiological diet, the patients are fed with nutritional formulas that include fat and other elements. Therefore, they might experience changes in digestive physiological responses to the changes of environment as well as their food intake, resulting in changes to their fecal characteristics. The results of this study, although, were not statistically significantly different in the diarrhea scores as was also found in Ciocon’s study [8] where diarrhea was identified by the abnormal occurrence in the frequency of bowel movements. This study did show that continuous feeding had a better effect in controlling the digestive function of patients therefore limiting diarrhea. Moreover, similar to the study by MacLoed [6] and by Serpa [13], although there was no significant differences between the two groups on the diarrhea scores, the incidence of diarrhea in the continuous feeding group was lower than that in intermittent feeding group.

In the study by Ciocon [8], the rate of tube obstruction in a total of 30 subjects in each group was 50% (15/30) in the continuous group and 16.7% (5/30) in the intermittent group (p = 0.01). However, the results of this study did not find any cases of obstruction in either group over 4 days and three additional follow-up days up to the time of tube replacement per hospital policy. This could be explained by the fact that all patients with nasogastric tubes in both study groups had their tubes regularly flushed every 4 hours. In addition, none of the patients in this study was prescribed with powdered or oral soluble medications that could increase tube obstruction during the study period. This was also consistent with the general trend of most other patients at the ICU, University Medicine Center. The researchers in this study were not able to identify the type of food products used in Ciocon’s study [8]. Therefore, it could be deduced that tube obstruction might not be affected by the delivery method if it was regularly flushed before and after feeding or taking medication according to standard procedure.