This was a single-center, prospective, and observational study. We enrolled all consecutive patients who were administered with enteral nutrition via a feeding tube in the neurointensive care unit (NICU) of Xuanwu Hospital, Capital Medical University, between July 2018 and June 2021. Patients were included in the study if they received EN via a nasogastric or orogastric tube, were aged 18 years or older, and had a length of stay ≥ 24 h. Exclusion criteria included patients who received EN via a nasojejunal, jejunostomy, or gastrostomy tube; were pregnant; had contraindications to EN (e.g., occlusion, ileus, or gastrointestinal ischemia); or had contraindications to gastric probing (e.g., esophageal stenosis, surgery, recent gastroesophageal trauma, and hematemesis). They were divided into two groups: the control group (GRV 100 ml) and the study group (GRV 200 ml) according to a table of the random digit.

This study was approved by the Ethics Committee of Xuanwu Hospital, Capital Medical University, and is adhered to the Declaration of Helsinki. Written informed consent was obtained from the patients or their guardians.

A working group including physicians, nurses, and dietitians reviewed the feeding protocol following the recommendations of the Chinese Nutritional Working Group (16).

All patients received EN via a 14-Fr nasogastric tube. The intragastric position of tube feeding was radiographically confirmed before diet infusion. EN was delivered continuously via a feeding pump in both groups. The goal EN for the two groups was calculated according to the patient’s body weight, with 1 ml of formula per kg of body weight being administered if the BMI was ≤ 30 kg/m², or 1 ml of formula per kg of ideal body weight if the body mass index (BMI) was > 30 kg/m². The optimal calorie supply was assumed to be 25 Kcal/kg/day. Patients were kept in a semi-recumbent position with the head of the bed elevated to 30°. The initial infusion rate was 25 ml/h, which was then gradually increased to the optimal infusion rate.

We used the syringe aspiration method with a 50-ml syringe to measure the GRV every 4 h. If 50 mL was withdrawn into the syringe, the fluid was emptied into a calibrated container, and the procedure was repeated until no more fluid could be withdrawn. Any amount less than the threshold value was reinstilled into the feeding tube. The tube was then flushed with 30 mL of water before being reconnected to the feeding pump. If a high GRV event occurred, EN could be interrupted for at least 2 h to promote gastric rest before restarting EN. If the GRV remained above the threshold, prokinetic treatment was administered, and the EN infusion rate was reduced. Insertion of a nasojejunal tube was performed in cases of repeated events of high GRV (Figure 1). Parenteral nutrition (PN) was used to complement nutritional requirements in cases of underfeeding with EN.

Demographic data, admission diagnosis, Glasgow Coma Scale (GCS) score, acute physiologic and chronic health evaluation (APACHE II) score, and sequential organ failure assessment (SOFA) score were collected from the patients’ medical records. The NICU and hospital lengths of stay and mortality rates were also recorded. High GRV was defined as a GRV equal to or greater than 100 ml (control group) or 200 ml (study group). Tolerability, referring to gastrointestinal complications, was defined as follows: abdominal distension (abdominal changes on daily physical exam with tympany and/or absence of bowel sounds), high gastric residuals (GRV ≥ threshold value), vomiting (enteral formula ejected through the mouth), astriction (dry stool and frequency less than 3 times a week), diarrhea (five or more liquid stools in a 24-h period or an estimated stool volume equal to or greater than 2,000 ml/day), and pulmonary aspiration (feed was detected in the tracheal aspirate). Calorie intake was calculated as follows: diet volume ratio = (administered volume of diet/prescribed volume) × 100%.

All data were analyzed with SPSS, version 26.0 (SPSS Inc., Chicago, IL, United States). Categorical variables between the two groups were compared using chi-squared tests. A Pearson χ² test was used when no subgroup had an expected count below 5; otherwise, Fisher’s exact test was performed. A two-tailed t-test for normally distributed continuous variables was performed. The Mann–Whitney U test was used in cases where the variable was not normally distributed. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated. P-values of less than 0.05 were considered statistically significant.

A total of 344 patients were enrolled in the NICU, and 9 patients were excluded from the final analysis for the following reasons: five had contraindications to EN, two were pregnant, and two had contraindications to gastric probing. Of the 335 included patients, 167 were in the study group, and 168 were in the control group. The patients were aged between 18 and 78 years, with a mean age of 57.1 years (SD 17.2). The primary diagnosis at admission was cerebrovascular disease (n = 188, 56.1%), followed by central nervous system (CNS) infection (n = 65, 19.4%), CNS immune disease (n = 44, 13.1%), Guillain–Barre syndrome (GBS, n = 11, 3.3%), myasthenia gravis (MG, n = 9, 2.7%), neuromuscular disease (n = 9, 2.7%), and others (n = 9, 2.7%). A total of 197 patients (58.8%) were on mechanical ventilation support (control group: 97, study group: 100).

There were no significant differences between the two groups in terms of age, sex, BMI, GCS score, APACHE II score, SOFA score, or diagnosis at admission. The baseline characteristics of both groups are shown in Table 1.

Although there were more severe adverse events (aspiration) in the study group than in the control group, there was no significant difference (1.2 vs. 0%, χ² = 1.012, OR = 0.047, 95% CI = 0.995–1.029, p = 0.248). The overall gastrointestinal complication rates were 61.1% (102/167) in the study group and 67.9% (114/168) in the control group (χ² = 1.681, OR = 0.743, 95% CI = 0.474–1.165, P = 0.195). The difference in the frequency of high GRV was statistically significant, 22.8% (38/167) vs. 38.1% (64/168) for the study and control groups, respectively (χ² = 9.307, OR = 0.047, 95% CI = 0.297–0.772, P = 0.002). There were no significant differences between the study group (GRV 200 ml) and control group (GRV 100 ml) with respect to abdominal distension (8.9 vs. 6.5%, χ² = 0.693, OR = 1.027, 95% CI = 0.965–1.093, p = 0.405), astriction (58.7 vs. 54.8%, χ² = 0.524, OR = 1.095, 95% CI = 0.856–1.400, p = 0.469), or diarrhea (14.4 vs. 11.3%, χ² = 0.702, OR = 1.271, 95% CI = 0.724–2.230, p = 0.402) (Figure 2).

The study group had superior mean ratios to the control group at week 1 (88.6 vs. 83.4%), week 2 (89.1 vs. 84.9%), and week 3 (93.2 vs. 90.5%) without significant differences. A total of 45 patients could not complete 3 weeks of EN for the following reasons: death within 3 weeks (study group: 18 vs. control group: 25, P = 0.262) and EN transition to oral feeding (study group: 10 vs. control group: 9, P = 0.818).

The study group showed better outcomes at discharge than the control group, but these differences did not reach significance. The mortality (GRV 200 ml vs. GRV 100 ml) rate was 10.8% (18/167) vs. 14.9% (25/168) at discharge (χ² = 1.260, OR = 0.724, 95% CI = 0.411–1.277, p = 0.262). The median NICU length of stay was 19.5 days vs. 25.3 days (P > 0.05) for the study and control groups, respectively.