Effect of amino acid supplementation on short-term complications after gastrointestinal tumor surgery: the AMIGITS randomized clinical trial Authors: Junke Wang, Yuezhi Chen, Zhiwen Zhao, Guodong Lian, Lipan Peng, Xinxiu Zhou, Shengxu Sha, Wenhao Qiao, Hui Tian, Leping Li, Feng Tian, Changqing Jing Published: 2024 ABSTRACT AbstractBackground:There is a scarcity of large randomized clinical trials on the efficacy and safety of high-dose amino acid supplementation (AAS) in patients with gastrointestinal tumors undergoing surgical treatment.Methods:This pragmatic, randomized, controlled, single-center, open-label, parallel-group AMIGITS trial was performed in a tertiary care teaching hospital. Patients with gastrointestinal tumors were randomly assigned to receive either AAS or standard care (SC). Amino acid targets were 2.0 g/kg per day in the AAS group and 1.2 g/kg per day in the SC group. The AAS group received additional amino acids intravenously, while the SC group received an iso-energetic 5% glucose intravenously.Results:Overall, 407 patients (AAS group, 204; SC group, 203) were included in this study. During the intervention, the actual mean daily energy intake did not differ significantly between the AAS and SC groups (25.53 vs. 25.16 kcal/kg per day, P=0.493). However, the actual mean daily amino acid intake was significantly higher in the AAS group than that in the SC group (1.81 vs. 0.94 g/kg per day, P<0.001). The infection incidence during hospitalization and within 30 days of surgery was significantly lower in the AAS group than that in the SC group (P=0.031 and P=0.024, respectively). The 30-day postoperative incidence of amino acid treatment-related adverse events and other complications did not significantly differ between the two groups. The postoperative hospital stay was significantly different between the two groups (P<0.001).Conclusions:AAS was associated with a reduced infection incidence within 30 days of major surgery in patients with gastrointestinal tumors and can be a promising strategy.Keywords: amino acid supplementation, gastrointestinal surgery, gastrointestinal tumors, infection, nutritional support Introduction Highlights Amino acid supplementation (AAS) is associated with a reduced infection incidence 30 days following surgery. Amino acid supplementation (AAS) is associated with a reduced infection incidence 30 days following surgery. Amino acid intake levels of 2.0 g/kg per day were safe. Amino acid intake levels of 2.0 g/kg per day were safe. This study provides specific clinical insights into nutritional support therapy. This study provides specific clinical insights into nutritional support therapy. Increased muscle protein breakdown in patients with cancer increases the incidence of postoperative complications1, leading to poor prognosis2. Gastrointestinal tumor surgery causes a catabolic response that leads to inflammation, nitrogen loss3, infections, and sepsis4. Good postoperative nutritional support can reduce the incidence of complications and infections and shorten the length of hospital stay5–7. Postoperative exogenous amino acid supplementation (AAS) can promote anabolism and positive nitrogen balance8, facilitating muscle protein synthesis in patients with cancer9. Meeting concurrent protein and energy goals is more beneficial for improving outcomes than meeting energy-only goals10,11. A meta-analysis revealed that high protein or amino acid intake levels might improve patients’ clinical outcomes12. Two previous studies have reached different conclusions13 and suggested that high-dose protein may worsen outcomes for patients with acute kidney injury and high organ failure scores14; nevertheless, the current nutritional guidelines for critically ill patients recommend at least 1.2 g/kg protein per day15,16. However, the latest clinical guidelines do not clearly define postoperative amino acid intake levels in patients undergoing surgery17,18. Underfeeding is common for amino acid intake after major gastrointestinal surgery because patients’ oral intake is restricted, and maintaining fluid balance is difficult19. A multicenter study showed that most critically ill and high nutritional-risk patients fail to achieve adequate amino acid intake20. Many guideline recommendations are far from clinical practice, and nutritional therapy is not optimal. Similar conclusions were drawn in a multicenter observational study21. This supports administering higher doses of amino acids to achieve nutritional goals. Moreover, high amino acid therapy can improve renal function and grip strength in critically ill patients22,23; however, the assessment of outcome measures is lacking in areas such as infection and length of hospital stay. Furthermore, there is a lack of large randomized clinical trials (RCT) on the efficacy and safety of high-dose AAS in patients with gastrointestinal tumors undergoing surgical treatment. Therefore, this RCT aimed to determine whether patients with gastrointestinal tumors could benefit from high-dose AAS after major surgery. Methods Study design and patients AMIGITS was a 1:1 randomized, controlled, single-center, open-label, parallel-group trial. This study was conducted in a tertiary care teaching hospital between February 3 and November 3, 2022. The study protocol and the statistical analysis plan are available in Supplement 2, Supplemental Digital Content 3, http://links.lww.com/JS9/D190 (original study protocol in Supplement 1, Supplemental Digital Content 4, http://links.lww.com/JS9/D191). Written informed consent was obtained from all eligible patients before inclusion in the study. The study was conducted according to the principles of the Declaration of Helsinki24 and was approved by the Biomedical Research Ethics Committee involving humans at *** University (SWYX: NO.2022-343). We followed the Consolidated Standards of Reporting Trials (CONSORT, Supplemental Digital Content 1, http://links.lww.com/JS9/D188, Supplemental Digital Content 2, http://links.lww.com/JS9/D189) reporting guidelines25. All authors had access to the study data and reviewed and approved the final manuscript. Our inclusion criteria were non-traumatic elective major surgery for gastrointestinal tumors in patients aged 18–80 years with stable preoperative vital signs, voluntarily participated, and signed the informed consent form. The detailed inclusion and exclusion criteria are described in the study protocol. Randomization and masking Patients were randomized in a 1:1 ratio to the AAS group or the standard care (SC) group before 8 am on the first postoperative day (POD). The random allocation sequences were computer-generated. Allocation concealment was implemented by sequentially numbered, sealed, opaque envelopes. Investigators and participants were not masked to the treatment assignment; however, the follow-up assessments were performed by trained physicians and nurses who were blinded to the patient’s assignment. The statisticians were blinded to the treatment group during the data analysis. Interventions The nutritional regimen for SC was developed as recommended by the guidelines17,18 and the consensus for enhanced recovery after surgery (ERAS)26,27 with a target amino acid intake of 1.2 g/kg per day. Oral feeding was recommended as early as possible after surgery, and enteral feeding was started no later than 24 h. In addition, except for oral feeding, enteral nutrition was also administered via nasoenteric tube in patients undergoing gastric surgery (specific nutritional formula in Supplement 2, Supplemental Digital Content 3, http://links.lww.com/JS9/D190). Based on SC, the AAS group received additional amino acids (8.5% 250 ml) intravenously to achieve the target of 2.0 g/kg per day, while the SC group received an iso-energetic 5% glucose intravenously. Additional amino acids infusion was administered via the central vein in the AAS group. The intervention was administered from POD 1 to POD 4. Criteria for stopping the intervention were patient discharge or end of the intervention. Patients’ demographic and clinical characteristics Patients’ baseline data and clinical characteristics were collected, including sex, age, height, weight, BMI, Nutritional Risk Screening 2002 (NRS-2002) score (Table 9 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192), comorbidities, disease diagnosis, obstruction, neoadjuvant therapy, surgical type, surgical approach, operation duration, and postoperative tumor pathological stage. Blood was collected for laboratory parameter assessment at 6 am on POD 1 before the randomization. Outcomes The primary outcome was the infection incidence within 30 days postoperatively, classified according to the Centers for Disease Control and Prevention definition28 of intra-abdominal infection, surgical site infection (SSI), bloodstream infection, pneumonia, urinary tract infection, and other infections (Tables 1, 2 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192). Secondary outcomes included the incidence of infection during hospitalization, amino acid-related adverse events, postoperative hospital stay, other complications (including anastomotic leakage, severe gastrointestinal intolerance, ICU admission, anastomotic bleeding, intra-abdominal bleeding, and lower extremity venous thrombosis), and levels of postoperative laboratory parameters (including nutritional indicators, inflammatory biomarkers, hepatic and renal function, blood routine, immune function and metabolism-related index). Amino acid-related adverse events and hepatic and renal function were used as safety evaluation indicators. Statistical analysis Two previous systematic meta-analyses29,30 found an overall infection rate of 10–30% in patients following gastrointestinal surgery. The trial assumed a 25% incidence of infection 30 days postoperatively in patients receiving SC following major gastrointestinal surgery. We hypothesized that AAS reduces the infection incidence to 11%. With a two-tailed type I error rate of 5%, this effect was detected at a statistical power level of 80%. Considering a 5% loss to follow-up, a sample size of 120 patients per group was required. We recruited a total of 240 patients, 120 in the AAS group and 120 in the SC group. We performed a preliminary analysis of the primary outcome after the patients completed one month of clinical follow-up and found that 11 patients developed an infection within 30 days of surgery in the AAS group and 23 in the SC group. Although the incidence of infection between the two groups was significant (P = 0.041), the actual incidence of infection in the SC group differed greatly from the incidence of infection at the time of trial design (19% vs. 25%). Therefore, we recalculated the sample size according to the actual infection incidence for more rigor and reliable results. The incidence of infection was close to 19% in the SC group and 9% in the AAS group. A two-tailed type I error rate of 5% detected this effect at the 80% statistical power level. Considering a loss to follow-up rate of 5%, a sample size of 195 patients per group was required. We obtained approval from the Biomedical Research Ethics Committee to expand patient recruitment (SWYX: NO.2022-343) and completed the final trial. The analysis set was based on the intention-to-treat principle. Variables were reported as numbers (percentages), mean±standard deviation, or medians (interquartile range). We used the Shapiro–Wilk test to assess whether continuous data obeyed a normal distribution. Where appropriate, we performed χ2 test or Fisher exact tests for categorical variables and two-tailed Student’s t-tests or Mann–Whitney U tests for continuous variables. We also performed subgroup analyses of the primary outcomes, including the age (<65 vs. ≥65 years), sex (male vs. female), NRS-2002 score (<3 vs. ≥3), BMI (≤23.9 vs. >23.9 kg/m2), comorbidities (yes vs. no), obstruction (yes vs. no), neoadjuvant therapy (yes vs. no), surgical type (robotic vs. laparoscopic), and surgical site (gastric vs. colorectal). No data on the primary outcome were lost. Missing data for other variables were not imputed. Statistical significance was set at a two-sided P less than 0.05. Since we statistically tested the primary outcome twice, a Bonferroni correction for multiple testing was applied31. Therefore, P less than 0.025 was considered to reach the statistical significance for the primary outcome. All analyses were performed using SPSS software version 25.0 (IBM Corp.). Results Patients A CONSORT flow diagram of this RCT is shown in Fig. 1. We reviewed 665 patients for eligibility; 238 did not meet the enrollment criteria, 18 were enrolled in another clinical trial, and 2 declined to participate. A total of 407 patients, all of whom underwent surgical treatment for gastrointestinal tumors, were randomized: 204 were assigned to the AAS group, and 203 were assigned to the SC group. These populations were defined as the intention-to-treat population (ITTP). Three patients in the AAS group developed amino acid treatment-related adverse events, and the intervention was terminated; two presented with nausea and vomiting, and one presented with an allergic reaction. Although the intervention was discontinued, these three patients who belonged to ITTP completed the 1-month clinical follow-up, and their clinical data were analyzed. A total of 201 patients in the AAS group received the full intervention protocol, and 203 in the SC group received SC. These populations were defined as the per-protocol population (PPP). All patients completed the 1-month follow-up, and their clinical data were analyzed. Table 1 lists the patients’ demographic and clinical characteristics. Baseline characteristics were similar between groups (Table 1). Flow diagram. Table 1Patients’ demographic and clinical characteristics (intention-to-treat population).Patient characteristicsAAS (n=204)SC (n=203)Male sex132 (64.7)145 (71.4)Age, year62.5 (56.3, 68.0)63.0 (56.0, 69.0)Height, cm166.8±7.8167.6±7.5Weight, kg67.9±11.669.3±12.0BMI, kg/m2 24.4±3.0324.6±3.15NRS-2002 scorea 280 (39.2)69 (34.0) ≥3124 (60.8)134 (66.0)Diagnosis Gastric cancer82 (40.8)87 (42.9) Colorectal cancer121 (58.7)114 (56.2) Other gastrointestinal tumorsb 1 (0.5)2 (1.0)Comorbiditiesc Hypertension63 (30.9)59 (29.1) Type 2 diabetes30 (14.7)26 (12.8) Cardiovascular disorders22 (10.8)23 (11.3) Cerebrovascular disease14 (6.9)15 (7.4) Lung disease19 (9.3)29 (14.3) Hepatic cyst40 (19.6)37 (18.2) Renal cyst27 (13.2)37 (18.2)Obstruction37 (18.1)40 (19.7)Neoadjuvant therapy36 (17.6)32 (15.8)Surgical type Da Vinci robot126 (61.8)119 (58.6) Laparoscope77 (37.7)82 (40.4) Laparotomy1 (0.5)2 (1.0)Operative approach Proximal gastrectomy9 (4.4)12 (5.9) Distal gastrectomy48 (23.5)43 (21.2) Total gastrectomy26 (12.7)33 (16.3) Right hemicolectomy23 (11.3)19 (9.4) Transverse or left hemicolectomy13 (6.4)10 (4.9) Dixon75 (36.8)72 (35.5) Hartmann3 (1.5)2 (1.0) Miles7 (3.4)13 (5.9)Operation duration, min192.83±25.33191.46±28.55Pathological TNM staged Gastric cancer after neoadjuvant therapy 02 (1.0)1 (0.5) I8 (4.0)3 (1.5) II5 (2.5)6 (3.0) III2 (1.0)3 (1.5) IV01 (0.5) Gastric cancer without neoadjuvant therapy I37 (18.4)29 (14.3) II12 (6.0)19 (9.4) III16 (8.0)25 (12.3) IV00 Colorectal cancer after neoadjuvant therapy 04 (2.0)5 (2.5) I4 (2.0)5 (2.5) II5 (2.5)6 (3.0) III4 (2.0)1 (0.5) IV2 (1.0)1 (0.5) Colorectal cancer without neoadjuvant therapy I33 (32.4)29 (14.3) II32 (31.4)38 (18.7) III37 (36.3)29 (14.3) IV00Data are shown as n (%), mean±SD, or median (IQR), unless otherwise indicated. Continuous data were compared using Student’s t-test or the Mann–Whitney U test. Categorical variables were compared using the χ2 test or Fisher’s exact test.AAS, amino acid supplementation; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); NRS-2002, Nutritional Risk Screening 2002; SC, standard care; TNM, tumor-node-metastasis. a Scores on NRS-2002 range from 0 to 7, with a score ≥3 identifying patients at nutritional risk. Higher scores indicate an increased risk. b Other gastrointestinal neoplasms included gastric mucinous cystadenoma, gastrointestinal stromal tumor, or ileocecal neoplasms. c Comorbidities included type 2 diabetes mellitus, hypertension, cardiovascular disease, cerebrovascular disease, lung disease, and hepatorenal cyst. d Since there were three benign lesions in the postoperative pathological findings, only 404 patients were included in the “Pathological TNM stage.” Patients’ demographic and clinical characteristics (intention-to-treat population). Data are shown as n (%), mean±SD, or median (IQR), unless otherwise indicated. Continuous data were compared using Student’s t-test or the Mann–Whitney U test. Categorical variables were compared using the χ2 test or Fisher’s exact test. AAS, amino acid supplementation; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); NRS-2002, Nutritional Risk Screening 2002; SC, standard care; TNM, tumor-node-metastasis. Scores on NRS-2002 range from 0 to 7, with a score ≥3 identifying patients at nutritional risk. Higher scores indicate an increased risk. Other gastrointestinal neoplasms included gastric mucinous cystadenoma, gastrointestinal stromal tumor, or ileocecal neoplasms. Comorbidities included type 2 diabetes mellitus, hypertension, cardiovascular disease, cerebrovascular disease, lung disease, and hepatorenal cyst. Since there were three benign lesions in the postoperative pathological findings, only 404 patients were included in the “Pathological TNM stage.” Energy and amino acid intake During the intervention period, the actual mean daily energy intake levels were similar between the AAS and SC groups (25.53±5.27 vs. 25.16±5.36 kcal/kg, P=0.493); the actual mean daily amino acid intake in the AAS group was 1.81±0.34 g/kg, whereas that in the SC group was 0.94±0.18 g/kg (P<0.001; Fig. 2; Table 4 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192; Figs. 1, 2 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192). Mean daily energy and amino acid intake during the 5 days following surgery. AAS, amino acid supplementation, SC, standard care. Primary clinical outcome The infection incidence within 30 days postoperatively was significantly lower in the AAS than that in the SC groups. Concerning specific types of infection, including intra-abdominal infection, SSI, pneumonia, bloodstream infection, and urinary tract infection, the infection incidence did not significantly differ between the AAS and SC groups (Table 2). Table 2Clinical outcomes: postoperative complications and postoperative hospital stay for the AAS and SC groups.OutcomeAAS (n=204)SC (n=203)Absolute difference (95% CI) P Primary outcome 30-day infectious complications a 18 (8.8)33 (16.3)7.5 (1.0, 13.8)0.024c Intra-abdominal infection8 (3.9)15 (7.4)3.5 (−1.0, 7.9)0.130c Surgical site infection11 (5.4)21 (10.3)4.9 (−0.3, 10.1)0.063c Pneumonia5 (2.5)7 (3.4)0.9 (−2.3, 4.3)0.552c Bloodstream infection5 (2.5)4 (2.0)−0.5 (−3.3, 2.4)1.000d Urinary tract infection2 (1.0)3 (1.5)0.5 (−1.6, 2.6)0.996d Secondary outcomes Infectious complication in the hospital14 (6.9)27 (13.3)6.4 (0.6, 12.3)0.031c Amino acid-related adverse events3 (1.5)0−1.5 (−3.1, 0.2)0.248e Anastomotic leakage6 (2.9)8 (3.9)1.0 (−2.5, 4.5)0.580c Severe gastrointestinal intolerance b 11 (5.4)5 (2.5)−2.9 (−6.7, 0.8)0.128c Anastomotic bleeding1 (0.5)3 (1.5)1.0 (−0.9, 2.9)0.612d Intra-abdominal bleeding3 (1.5)1 (0.5)−1.0 (−2.9, 0.9)0.619d Venous thrombosis lower limb1 (0.5)2 (1.0)0.5 (−1.2, 2.16)0.997d ICU admission0 (0.0)3 (1.5)1.5 (−0.2, 3.1)0.123e Postoperative hospital stays, days5 (5–6)5 (5–6)—<0.001f Clavien–Dindo grade I6 (2.9)3 (1.5)6.0 (−1.3, 13.3)0.267e II21 (10.3)32 (15.8) III2 (1.0)3 (1.5) IV03 (1.5)Data are shown as n (%) or median (IQR), unless otherwise indicated. Percentages of patients with each outcome were compared between groups using the χ2 test or Fisher’s exact test and categorical data are reported as median (IQR) using Mann–Whitney U’s non-parametric test.AAS, amino acid supplementation; ICU, intensive care unit; IQR, interquartile range; SC, standard care.a18 patients in the AAS group and 33 in the SC group developed infectious complications within 30 days of surgery. Since one patient usually has several types of infection, the sum of the five types of infection—intra-abdominal infection, surgical site infection, pneumonia, bloodstream infection, and urinary tract infection—does not equal the total number of infections. We statistically tested the primary outcome twice, hence a Bonferroni correction for multiple testing was applied. Therefore, P<0.025 is considered to reach the statistical significance for the primary outcome.bSevere gastrointestinal intolerance, including severe vomiting, abdominal distension, diarrhea, and abdominal pain, resulting in interruption of enteral nutrition.cχ2 test.dAdjusted χ2 test.eFisher’s exact test.fMann–Whitney U test. Clinical outcomes: postoperative complications and postoperative hospital stay for the AAS and SC groups. Data are shown as n (%) or median (IQR), unless otherwise indicated. Percentages of patients with each outcome were compared between groups using the χ2 test or Fisher’s exact test and categorical data are reported as median (IQR) using Mann–Whitney U’s non-parametric test. AAS, amino acid supplementation; ICU, intensive care unit; IQR, interquartile range; SC, standard care. 18 patients in the AAS group and 33 in the SC group developed infectious complications within 30 days of surgery. Since one patient usually has several types of infection, the sum of the five types of infection—intra-abdominal infection, surgical site infection, pneumonia, bloodstream infection, and urinary tract infection—does not equal the total number of infections. We statistically tested the primary outcome twice, hence a Bonferroni correction for multiple testing was applied. Therefore, P<0.025 is considered to reach the statistical significance for the primary outcome. Severe gastrointestinal intolerance, including severe vomiting, abdominal distension, diarrhea, and abdominal pain, resulting in interruption of enteral nutrition. χ2 test. Adjusted χ2 test. Fisher’s exact test. Mann–Whitney U test. Secondary clinical outcomes The incidence of infection during hospitalization was significantly lower in the AAS group than that in the SC group. The incidence of amino acid-related adverse events did not significantly differ between the AAS and SC groups (Table 8 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192). Other complications were statistically insignificant (Table 7 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192). The postoperative hospital stay was significantly different between the two groups (P<0.001; Table 2). The postoperative laboratory parameters revealed no significant difference in serum albumin levels between the AAS and SC groups. Pre-albumin levels were significantly higher in the AAS group than those in the SC group (POD 3, 142.99±33.35 vs. 133.83±36.95 mg/l, P=0.009; POD 5, 147.72±37.57 vs. 135.32±42.90 mg/l, P=0.014). Red blood cell (RBC) and hemoglobin (HGB) levels were significantly higher in the AAS group than that in the SC group on POD 5 (RBC, P=0.004; HGB, P=0.007). Concerning renal function parameters, blood urea nitrogen (BUN) levels were significantly higher in the AAS group than those in the SC group (P<0.001). The difference between serum creatinine (Scr) levels was not statistically significant on POD 5 in both groups. However, Scr levels in the AAS group were significantly lower than those in the SC group on POD 3 (P=0.001). Estimated glomerular filtration rate (eGFR) levels were significantly higher in the AAS group than those in the SC groups on POD 3 (P=0.037). Beta 2 microglobulin (β2-MG) levels were significantly lower in the AAS groups than those in the SC groups (P=0.032). The differences between hepatic function parameters were not statistically significant on POD 5 in both groups (Table 3 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192). Subgroup analyses Subgroup analyses of infections in the full analysis sets are shown in Fig. 3. In our pre-specified subgroups, the infection incidence was significantly lower in AAS group than that in the SC group in the subgroups with gastric surgery, NRS-2002 score less than 3, comorbidities, no obstruction, and no neoadjuvant therapy. Therefore, high-dose AAS after major surgery for gastrointestinal tumors may reduce the 30-day infection rate in these populations. Risk difference of AAS vs. SC by pre-specified subgroups. The P value is for the test of interaction between the group and each subgroup factor. Error bars indicate 95% CIs. AAS, amino acid supplementation; CI, confidence intervals; NRS-2002, Nutritional Risk Screening 2002; SC, standard care. Discussion Patients with cancer experience disturbed amino acid dynamics with an increased protein turnover and muscle protein breakdown, leading to muscle wasting, skeletal muscle loss, and cachexia32–35, resulting in a poor prognosis2,36. A recent study suggested that up to 50% of patients with cancer are malnourished or at risk of malnutrition37. The ESPEN guidelines recommend for these patients, energy intake levels of 25–30 kcal/kg per day and protein intake greater than 1.0 g/kg per day (and if conditions permit, 1.5 g/kg per day)38. Patients with cancer undergoing surgery require more energy and protein; those undergoing gastrointestinal surgery face a higher nutritional risk. AMIGITS trial investigated the effect of higher doses of AAS on 30-day infection rates in patients with gastrointestinal tumors undergoing major surgeries. The intention-to-treat patients (n=407) underwent surgery for gastrointestinal tumors; only 3 were pathologically confirmed to have benign tumors, and 404 had gastric and colorectal cancers. During the intervention, energy intakes were insignificantly different between the two groups; however, amino acid intake levels were significantly higher in the AAS group than those in the SC group. Compared with the SC group, the AAS group had a significantly lower 30-day postoperative infection incidence, reduced postoperative hospital stays, and improved nutritional risk, postoperative anemia, and renal function. Previous studies have shown that high energy and amino acid intake levels in critically ill patients can improve clinical outcomes. Both high energy and high amino acid intake are crucial in these studies11,39–43. Another study investigated the permissive underfeeding44. However, in that study, both groups maintained similar protein intake and the study variable was energy. Therefore, these studies could neither exclude the confounding effect of energy nor clarify the specific role of amino acids. Few studies have assessed amino acid levels at different doses of iso-energy. An RCT on protein requirements in critically ill patients22 showed an improvement in grip strength and fatigue on day 7 in the high amino acid group, with ultrasound revealing significantly greater forearm muscle thickness than that of the low amino acid group. Energy intake was generally balanced between the two groups, avoiding the confounding effect of energy on protein balance. However, the difference in actual amino acid intake between the two groups was small (0.9 and 1.1 g/kg per day, respectively); the grip strength at discharge as a primary outcome did not significantly differ between the two groups. In the AMIGITS trial, the energy intake did not significantly differ between the groups; the AAS group had higher amino acid intake (1.81 vs. 0.94 g/kg per day). Moreover, the amino acid was administered intravenously to avoid potential bias caused by incompletely absorbed amino acids in the intestine, increasing our results’ credibility. Higher postoperative amino acid intake levels in patients with gastrointestinal tumors significantly reduced the 30-day infection incidence. This may be attributed to several reasons. First, globulin was higher in the AAS group than that in the SC group (Table3 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192). Since globulin is mainly involved in the body’s immune function, AAS may improve patients’ immune function and reduce the 30-day postoperative infection incidence. Second, AAS stimulates muscle protein anabolism45 and delays skeletal muscle reduction46, resulting in improved outcomes such as reduced respiratory complications47. Third, this trial identified an improvement in renal function. Improving organ function after AAS may help to reduce the postoperative infection incidence. The AMIGITS trial identified significantly higher BUN levels in the AAS group than in the SC group. This resulted from increased amino acid intake in the AAS group and increased amino acid utilization and metabolism rather than a consequence of decreased renal function. Conversely, β2-MG and Scr levels were significantly lower in the AAS group than in the SC group. The eGFR level in the AAS group was significantly higher than that in the SC group on POD 3. Although the Scr, β2-MG, and eGFR levels were within physiologically normal ranges in both groups, these results suggested an improvement in renal function. Our results were consistent with those of two previous RCTs. One investigated the effect of intravenous amino acids on renal function in critically ill patients23, and the other investigated the protective effect of intravenous amino acids on the kidney in patients undergoing cardiac surgery48. The AMIGITS trial results demonstrated that amino acid intake levels of 2.0 g/kg per day were safe and that increased urea levels due to increased amino acid intake did not increase the burden on renal function49 but improved eGFR. The underlying mechanism may be related to the diuretic effect of amino acids50,51. Moreover, AAS did not affect hepatic functions and postoperative blood glucose levels. This may provide some practical experience for nutritional therapy in the future. Patients receiving standard care had energy and amino acid intake levels comparable to those receiving early parenteral nutrition supplementation in a previous RCT52; however, there was a large difference in the postoperative infection incidence (16.3% vs. 8.7%). This may be due to several reasons. First, the primary outcome of the AMIGITS trial was the 30-day postoperative infection incidence, whereas that of the previous study was the incidence of infection from day 3 after surgery until patient discharge. Therefore, the AMIGITS trial spanned a longer period. Second, we treated more patients (16.7%) with neoadjuvant therapy. These populations might have had a higher infection incidence, resulting in an increased overall incidence. Third, patients with cancer accounted for a higher proportion (99.3%) of our study population; they had higher nutritional risk and metabolic stress levels and a combined hypercoagulable state. Generally, these patients have greater surgical coverage and trauma, which increases postoperative infection incidence. Overall, infectious complications were significantly lower in the AAS group than those in the SC group. However, the specific type of infection did not differ significantly between the groups. This may be because the incidence of specific infection types, such as intra-abdominal infection, pneumonia, SSI, bloodstream infection, and urinary tract infection, was low postoperatively (<10% in both groups); larger studies may be required to produce significant results. In the pre-specified subgroups of patients with NRS-2002 scores less than 3, comorbidities, no obstruction, no neoadjuvant therapy, and gastric surgery, the 30-day postoperative infection incidence was significantly lower in the AAS group than that in the SC group. Subgroups with NRS-2002 scores less than 3, no obstruction, and no neoadjuvant therapy had better preoperative nutritional status and lower preoperative inflammation and stress levels. After receiving high-dose amino acid therapy, these patients may meet the amino acid requirements for tissue repair and immune function. However, nutritional status has large energy and amino acid gaps in people at significant nutritional risk preoperatively; even high doses AAS may not compensate for the lack of amino acids in the body. This may require further exploration. In the subgroup with comorbidities, globulin was significantly higher in the AAS group than that in the SC group on POD 3 (P=0.026). High AAS levels may significantly improve immune function. Further, AAS improves organ function, such as the kidney, which may also be beneficial in reducing the postoperative infection incidence in patients with chronic comorbidities. Gastric surgery was more invasive and time-consuming (213.46 vs. 176.70 min; P<0.001) than colorectal surgery; thus, patients would require more amino acids. The postoperative nutritional risk and the lack of amino acids are greater in patients with gastric cancers. SC provides much lower levels of amino acids than the body needs, while high-dose AAS may better compensate for the absence of amino acids and produce a more optimistic outcome. Our results showed that pre-albumin, RBC, and HGB levels were significantly higher in the AAS group than those in the SC group. Nutritional risk and anemia improved with AAS. This might be due to several reasons. First, we provided an adequate postoperative energy supply. Second, the AAS group ingested more amino acids than the SC group to compensate for the consumption of amino acids by tissue healing, acute phase protein production, and gluconeogenesis processes35. Moreover, tryptophan and isoleucine in essential amino acids can promote hemoglobin production and resist anemia; AAS increases the substrate of protein synthesis and facilitates protein synthesis. In the AMIGITS trial, there was a statistically significant difference in postoperative hospital stay between the AAS and SC groups (P<0.001). Since the postoperative hospital stay did not conform to the normal distribution, the Mann–Whitney U test was performed. However, the median [interquartile range (IQR)] was the same in both groups, with a value of 5 (5, 6). Therefore, the clinical differences between the two groups were not significant. To better illustrate the difference, we generated a frequency distribution chart of postoperative hospital stay between the two groups (Fig. 3 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192). More patients in the AAS group had a postoperative hospital stay of 5 days, whereas the number of patients with prolonged hospital stays was higher in the SC group. This may have been observed because AAS reduces the postoperative infection incidence and improves immune and organ functions, thereby shortening the postoperative hospital stay. We defined the adverse reactions associated with amino acid therapy. Adverse reactions that did not occur prior to amino acid infusion, appeared 30 min following amino acid infusion, and gradually disappeared after stopping infusion were considered amino acid-related adverse reactions. In S3 Table 5, Supplemental Digital Content 5, http://links.lww.com/JS9/D192, the mean daily oral, nasoenteric, and parenteral energy intake during the intervention period in the AAS and SC groups is presented. Statistical analysis revealed no significant difference in mean oral, nasoenteric, and parenteral energy intake per day between the two groups. In this study, enteral nutritional support via nasoenteric tube feeding was exclusively provided to patients undergoing gastric resection, while those with colorectal resection did not receive nasoenteric tube feeding. Consequently, our analysis focused on nasoenteric feeding data within the population undergoing gastric surgery. No statistically significant difference was observed in mean daily nasoenteric energy intake during the intervention period for patients undergoing gastric surgery (Table 6 Supplement 3, Supplemental Digital Content 5, http://links.lww.com/JS9/D192). In conclusion, AAS is associated with a reduced infection incidence 30 days following major surgery in patients with gastrointestinal tumors. Our results provide specific clinical insights into nutritional support therapy, which may help to improve postoperative nutritional therapy. Limitations of the study Our study has some limitations. First, AMIGITS was a single-center RCT, which might have reduced the generalizability of the results. Second, we did not use indirect calorimetry, which is recommended for measuring resting energy expenditure in surgical patients according to international guidelines. Instead, we used the recommended formula. Third, guided by the ERAS concept, our patients’ average postoperative hospital stay was approximately 5–6 days; therefore, we could not perform a longer intervention, which was limited to 4 days, potentially impacting the long-term results. Fourth, we only conducted dietary guidance and education at the time of discharge and did not count energy and amino acid intake levels during the month after discharge. The two groups might have produced unbalanced confounding factors during this period, somewhat biasing our results. Fifth, although the overall incidence of infection differed significantly, the incidence of specific infection types did not. Therefore, a multicenter RCT with a larger sample may be required to confirm our conclusions. The P value is for the test of interaction between the group and each subgroup factor. Error bars indicate 95% CIs. Ethical approval The study was conducted according to the principles of the Declaration of Helsinki and was approved by the Biomedical Research Ethics Committee involving humans at Shandong Provincial Hospital, affiliated with Shandong First Medical University (SWYX: NO.2022-343). Consent Written informed consent was obtained from all eligible patients before inclusion in the study. Source of funding This study was funded by the [National Natural Science Foundation of China] under Grant [81900524]; the [Natural Science Foundation of Shandong Province] under Grant [ZR2020MH252; ZR2020MH205]; the China Postdoctoral Science Foundation under Grant [2020M672102]; the [Science and Technology Development Program of Jinan] under Grant [202134027]. Author contribution Concept and design: J.W., Y.C., Z.Z., L.L., F.T., C.J. Acquisition, analysis, or interpretation of data: J.W., X.Z., S.S., W.Q., F.T., C.J. Drafting of the manuscript: J.W., Y.C., Z.Z. Critical revision of the manuscript for important intellectual content: G.L., L.P., L.L., F.T., C.J. Statistical analysis: Z.Z., H.T. Obtained funding: Y.C., G.L., L.L., F.T., C.J. Administrative, technical, or material support: G.L., L.P., L.L., F.T., C.J. Supervision: L.L., F.T., C.J. Conflicts of interest disclosure The authors declare that they have no conflict of interest. Research registration unique identifying number (UIN) chictr.org.cn, ChiCTR2200056258. Data availability statement Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information. Provenance and peer review This paper was not invited. Thank you! Supplementary Material SUPPLEMENTARY MATERIAL Acknowledgement The authors thank all patients, their families, and all investigators, nurses, and research teams. Footnotes J.W., Y.C., Z.Z. contributed equally to this work. Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article. Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal's website, www.lww.com/international-journal-of-surgery. Published online 25 July 2024 Contributor Information Junke Wang, Email: jkwang5426@163.com. Yuezhi Chen, Email: chenyuezhi-007@163.com. Zhiwen Zhao, Email: zhaozhiwen0828@163.com. Guodong Lian, Email: guodonglian@gmail.com. Lipan Peng, Email: plp1983@126.com. Xinxiu Zhou, Email: 18866811238@163.com. Shengxu Sha, Email: sha_shengxu@163.com. Wenhao Qiao, Email: 593092889@qq.com. Hui Tian, Email: tianhui1223@163.com. Leping Li, Email: lileping@medmail.com.cn. Feng Tian, Email: tianfeng@sdu.edu.cn. Changqing Jing, Email: jingchangqing@sdfmu.edu.cn. References Associated Data Supplementary Materials SUPPLEMENTARY MATERIAL Data Availability Statement Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information.