3.1. Baseline characteristics and initial severity of illness
Of the 1644 patients screened for eligibility, a total of 376 patients who met the inclusion criteria and did not meet the exclusion criteria were eventually included in the study. The number of excluded patients and the reasons are as follows: 919 patients did not fulfill the inclusion criteria, 42 patients received oral feeding, 73 patients received EN prior to ICU admission, 48 patients were cross-used EN formulations, 80 patients had participated in other similar clinical studies, 17 patients did not feeding follow the nutritional protocols, 15 patients had malignant tumors, 43 patients were pregnant, 21 patients had missing clinical data, and 10 patients were not followed - up. Therefore, 376 patients were evaluated for the effect of the short-peptide and standard-polymeric formulas on the feeding outcome and clinical outcome of severely ill patients with sepsis and AGI grade I - II (Figure 1).
Baseline patient characteristics of the EEN and DEN subgroups are presented in Table 2. In the EEN subgroup, 154 patients had a mean age of 62.3 years with a mean BMI of 22.8 kg/m2, 18 patients had primary AGI, and 136 patients had secondary AGI. In the first day of ICU admission, the mean arterial pressure (MAP) was 86.7 mmHg and the average serum albumin levels in these patients were 30.2 g/dL. In the DEN subgroup, 222 patients had a mean age of 58.6 years with a mean BMI of 22.8 kg/m2, 86 patients had primary AGI, and 136 patients had secondary AGI. In the first day of ICU admission, the mean MAP was 72.4 mmHg and the average serum albumin levels were 26.9 g/dL. No significant differences were observed in baseline disease severity related scores, trophic status - related scores, mechanical ventilation ratio, EF route, and other baseline characteristics between the short-peptide and standard-polymeric formulas groups in both the EEN and DEN subgroups.
Table 2. Demographic data and clinical characteristics of patients at baseline
Characteristics
|
EEN subgroup
|
|
DEN subgroup
|
All (n = 154)
|
Short-peptide
(n = 44)
|
Standard-polymeric (n = 110)
|
P Value
|
All (n = 222)
|
Short-peptide
(n = 98)
|
Standard-polymeric (n = 124)
|
P Value
|
Sex, male (%)
|
98 (63.6)
|
32 (72.7)
|
66 (60.0)
|
0.14
|
|
136 (61.3)
|
56 (57.1)
|
80 (64.5)
|
0.26
|
Age, mean ± SD, years.
|
62.3 ± 15.5
|
65.7 ± 16.1
|
60.9 ± 15.1
|
0.08
|
|
58.6 ± 17.9
|
59.5 ± 18.4
|
57.9 ± 17.5
|
0.52
|
Actual body weight, median (IQR), kg
|
65 (60 - 70)
|
70 (60 - 75)
|
65 (60 - 70)
|
0.09
|
|
65 (60 - 70)
|
65 (60 - 70)
|
65 (60 - 70)
|
0.26
|
BMI, mean ± SD
|
22.8 ± 4.0
|
22.9 ± 3.1
|
22.8 ± 4.3
|
0.82
|
|
22.8 ± 3.6
|
22.2 ± 3.6
|
22.8 ± 3.1
|
0.15
|
Underlying conditions (%)
|
|
|
|
|
|
|
|
|
|
Diabetes mellitus
|
59 (38.3)
|
17 (38.6)
|
42 (38.2)
|
0.96
|
|
88 (39.6)
|
32 (32.7)
|
56 (45.2)
|
0.06
|
Hypertension
|
68 (44.2)
|
22 (50.0)
|
46 (41.8)
|
0.36
|
|
128 (57.7)
|
60 (61.2)
|
68 (54.8)
|
0.34
|
Chronic obstructive pulmonary disease
|
32 (20.8)
|
11 (25.0)
|
21 (19.1)
|
0.41
|
|
52 (23.4)
|
22 (22.4)
|
30 (24.2)
|
0.76
|
Coronary heart disease
|
45 (29.2)
|
13 (29.5)
|
32 (29.1)
|
0.96
|
|
56 (25.2)
|
24 (24.5)
|
32 (25.8)
|
0.82
|
Primary diagnosis (%)
|
|
|
|
|
|
|
|
|
|
Neurologic
|
38 (24.7)
|
11 (25.0)
|
27 (24.5)
|
0.95
|
|
18 (8.1)
|
6 (6.1)
|
12 (9.7)
|
0.34
|
Respiratory
|
46 (29.9)
|
15 (34.1)
|
31 (28.2)
|
0.47
|
|
22 (9.9)
|
8 (8.2)
|
14 (11.3)
|
0.44
|
Cardiovascular
|
29 (18.8)
|
8 (18.2)
|
21 (19.1)
|
0.90
|
|
38 (17.1)
|
14 (14.3)
|
24 (19.4)
|
0.32
|
Gastrointestinal
|
18 (11.7)
|
6 (13.6)
|
12 (10.9)
|
0.63
|
|
86 (38.7)
|
44 (44.9)
|
42 (33.9)
|
0.09
|
Multi trauma
|
12 (7.8)
|
2 (4.5)
|
10 (9.1)
|
0.54
|
|
40 (18.0)
|
16 (16.3)
|
24 (19.4)
|
0.56
|
Others
|
11 (7.1)
|
2 (4.5)
|
9 (8.2)
|
0.66
|
|
18 (8.1)
|
10 (10.2)
|
8 (6.5)
|
0.31
|
Disease severity at defined time b
|
|
|
|
|
|
|
|
|
|
MAP c, mean ± SD
|
82.2 ± 10.0
|
80.3 ± 7.4
|
83.0 ± 10.8
|
0.08
|
|
64.2 ± 12.2
|
63.1 ± 11.0
|
65.0 ± 13.1
|
0.25
|
APACHE II score, median (IQR)
|
15 (11 - 19)
|
15 (12 - 22)
|
15 (11 - 18)
|
0.44
|
|
15 (10 - 20)
|
16 (13 - 20)
|
14 (10 - 19)
|
0.25
|
SOFA score, median (IQR)
|
7 (4 - 9)
|
6 (4 - 9)
|
7 (5 - 9)
|
0.14
|
|
7 (5 - 10)
|
7 (5 - 10)
|
7 (5 - 10)
|
0.78
|
NRS score, median (IQR)
|
4 (3 - 4)
|
3 (3 - 4)
|
4 (3 - 4)
|
0.76
|
|
3 (3 - 4)
|
4 (3 - 4)
|
3 (3 - 4)
|
0.31
|
mNUTRIC score, median (IQR)
|
4 (3 - 5)
|
4 (3 - 6)
|
4 (3 - 5)
|
0.58
|
|
4 (3 - 5)
|
4 (3 - 5)
|
4 (2 - 5)
|
0.60
|
Mechanical ventilation (%)
|
99 (64.3)
|
32 (72.7)
|
67 (60.9)
|
0.17
|
|
176 (79.3)
|
73 (74.5)
|
103 (83.1)
|
0.12
|
AGI grade II (%)
|
71 (46.1)
|
21 (47.7)
|
50 (45.5)
|
0.80
|
|
107 (48.2)
|
52 (53.1)
|
55 (44.4)
|
0.20
|
Primary AGI a (%)
|
18 (11.7)
|
6 (13.6)
|
12 (10.9)
|
0.63
|
|
86 (38.7)
|
44 (44.9)
|
42 (33.9)
|
0.09
|
ICU course prior to defined time b
|
|
|
|
|
|
|
|
|
|
Serum albumin, mean ± SD, g/L
|
30.2 ± 6.7
|
31.4 ± 7.3
|
29.7 ± 6.5
|
0.17
|
|
26.9 ± 7.2
|
27.0 ± 7.4
|
26.9 ± 7.0
|
0.87
|
Feeding route (%)
|
|
|
|
|
|
|
|
|
|
Feeding via nasogastric tube
|
141 (91.6)
|
38 (86.4)
|
103 (93.6)
|
0.14
|
|
210 (94.6)
|
92 (93.9)
|
118 (95.2)
|
0.67
|
Feeding via nasojejunal tube
|
13 (8.4)
|
6 (13.6)
|
7 (6.4)
|
0.14
|
|
12 (5.4)
|
6 (6.1)
|
6 (4.8)
|
0.67
|
Abbreviations: ICU, intensive care unit; BMI, body mass index; MAP, mean arterial pressure; APACHE, Acute Physiology, and Chronic Health Evaluation; SOFA, Sequential Organ Failure Assessment; NRS, nutrition risk screening 2002 score; mNUTRIC, modified NUTRIC score; AGI, acute gastrointestinal injury, IQR, interquartile range.
a Primary AGI (2) is associated with primary disease or direct injury to organs of the GI system (first hit); b The defined time period was the first 24 h of ICU admission; c MAP represents the worst mean arterial pressure before application of vasoactive drugs during ICU admission.
3.2. Primary outcomes
3.2.1. Caloric adequacy
In the EEN subgroup, the caloric adequacy of EN on day 7 in the short-peptide group was higher than that of the standard-polymeric group (59.1% vs. 27.3%, p = 0.001). However, in the DEN subgroup, the caloric adequacy of EN on day 7 in the short-peptide group was lower than that of the standard-polymeric group (28.6% vs. 43.5%, p = 0.02) (Table 3, Figure 2).
Table 3. Comparison of nutritional outcomes and adverse gastrointestinal events.
Nutrition Summary
|
EEN subgroup
|
|
DEN subgroup
|
Short-peptide
(n = 44)
|
Standard-polymeric
(n = 110)
|
χ²/Z
|
P Value
|
Short-peptide
(n = 98)
|
Standard-polymeric
(n = 124)
|
χ²/Z
|
P Value
|
Timing of EN, median (IQR), hours
|
14.5 (3,21)
|
14 (2,24)
|
- 0.48
|
0.63
|
|
94 (79.8,117.3)
|
96 (72,118)
|
-0.30
|
0.77
|
Caloric adequacy of EN on the 7th day a, median (IQR), (%)
|
26 (59.1)
|
30 (27.3)
|
13.75
|
0.001**
|
|
28 (28.6)
|
54 (43.5)
|
5.27
|
0.02*
|
Calories growth from EN within 3-7d, median (IQR), g
|
624 (0, 936)
|
0 (- 360, 480)
|
- 3.55
|
0.001**
|
|
960 (480,1200)
|
1080 (720,1440)
|
-2.10
|
0.04*
|
Protein growth from EN within 3-7d, median (IQR), kcal
|
24 (0, 38)
|
0 (- 14.4, 19.2)
|
- 2.96
|
0.003**
|
|
38.4 (19.2,50.4)
|
43.2 (28.8,57.6)
|
- 2.10
|
0.04*
|
Serum albumin on the 7th day, median (IQR), (g/dL)
|
30.9 (29.1, 32.6)
|
30.0 (27.1, 32.7)
|
- 1.01
|
0.32
|
|
30.1 (28.9,31.2)
|
29.8 (27.6,32.3)
|
- 0.64
|
0.52
|
Gastric retention (%)
|
8 (18.2)
|
40 (36.4)
|
4.84
|
0.03*
|
|
14 (14.3)
|
26 (21.0)
|
1.66
|
0.20
|
Diarrhea (%)
|
4 (9.1)
|
28 (25.5)
|
5.11
|
0.02*
|
|
10 (10.2)
|
20 (16.1)
|
1.64
|
0.20
|
Abbreviations: EN, enteral nutrition; IQR, interquartile range
a In this study, we defined the caloric adequacy of EN as > 80% of the caloric target; *p< 0.05, **p< 0.01.
Tables S1 and S2 describe the results of the logistic regression analysis for the EEN and DEN subgroups, respectively. mNUTRIC and APACHE II scores were excluded from the multivariate analysis because of their high collinearity (Figures S2 and S3). The short-peptide formulas were associated with EN caloric adequacy on day 7 after controlling for covariates in the EEN subgroup (OR: 3.852, 95% CI: 1.851–8.017; p < 0.001). In addition, in the DEN subgroup, the EN caloric adequacy on day 7 associated with the standard-polymeric formulas (OR: 2.409, 95% CI: 1.325–4.378; p = 0.004), gastric retention (OR: 0.272, 95% CI: 0.115–0.643; p = 0.003), and diarrhea (OR: 0.548, 95% CI: 0.378–0.796; p = 0.002).
3.3 Secondary outcomes
3.3.1. Calories and protein growth
In the EEN subgroup, calories and protein growth from EN were higher in the short-peptide group than in the standard-polymeric group (624[0, 936] vs. 0[ - 360, 480], p = 0.001; 24[0, 38] vs. 0[ - 14.4, 19.2], p = 0.003, respectively). However, in the DEN subgroup, calories and protein growth from EN were lower in the short-peptide group than in the standard-polymeric group (960[480, 1200] vs. 1080[720, 1440], p = 0.04; 38.4[19.2, 50.4] vs. 43.2[28.8, 57.6], p = 0.04, respectively) (Table 3, Figure 2).
3.3.2. Adverse gastrointestinal events
In the EEN subgroup, the incidence of gastric retention (18.2% vs. 36.4%, p = 0.03) and diarrhea (9.1% vs. 25.5%, p = 0.02) was lower in the short-peptide group than in the standard-polymeric group. However, in the DEN subgroup, there were no statistically significant differences in the incidence of gastric retention and diarrhea between the short peptide and standard polymeric groups (p > 0.05) (Table 3, Figure 3).
3.3.3. Daily caloric EN/ (EN + PN) ratios
In the EEN subgroup, caloric EN/ (EN + PN) ratios were higher on day 6 and 7 in the short-peptide group than in the standard-polymeric group (p = 0.017; p = 0.039, respectively). In the DEN subgroup, caloric EN/ (EN + PN) ratios were higher on day 3 in the standard-polymeric group than in the short-peptide group (p = 0.044) (Table S3, Figure 4).
3.3.4. Clinical outcomes
There were no statistically significant differences were observed in ICU and 28 - day mortality, ICU length of stay, hospital length of stay, ventilator - free days and hospitalization cost in the short-peptide and standard-polymeric formulas group in both the EEN and DEN subgroups (p > 0.05) (Table S4). Kaplan–Meier survival curves plotted with the 28 - day mortality rates also did not show a statistically significant difference between the EEN (log - rank p = 0.70) and DEN subgroups (log - rank p = 0.33) (Figure 5).