Effects of perioperative probiotic supplementation on postoperative gastrointestinal recovery in elderly hip fracture patients: a randomized controlled trial

Abstract

Background

Enhanced Recovery After Surgery (ERAS) protocol aims to accelerate recovery and reduce complications in postoperative patients. Preoperative carbohydrate intake is recommended to reduce insulin resistance but is limited by safety concerns in elderly trauma patients. Probiotics may improve gastrointestinal function and barrier integrity. This study investigates whether perioperative probiotics can be a safe and effective alternative to preoperative oral nutrition in elderly hip fracture patients.

Methods

Elderly patients (≥ 65 years) with hip fractures were randomized into three groups: control (fasting), preoperative nutrition (200 mL carbohydrate-rich supplement 2 h before surgery) and perioperative probiotics (Bifidobacterium triple viable capsules twice daily from admission to discharge). The primary outcome was the time to first postoperative flatus. Secondary outcomes included perioperative levels of blood glucose, insulin, C-reactive protein (CRP) and intestinal barrier integrity biomarkers. Mean arterial pressure (MAP), heart rate (HR), postoperative pain score, delirium occurrence and length of hospital stay were also assessed.

Results

A total of 170 patients were enrolled and included in the intention-to-treat (ITT) analysis, of whom 154 were included in the per-protocol (PP) analysis for the primary outcome. Baseline characteristics and intraoperative parameters were comparable between groups. Both nutrition and probiotic groups had shorter time to first postoperative flatus than the control group, with the probiotic group showing the fastest recovery and greatest improvement in intestinal barrier markers. No significant differences were observed in postoperative delirium, pain scores or hospital stay between groups. No increase in adverse events related to probiotic or nutritional supplementation was found.

Conclusion

Both perioperative probiotics and preoperative nutritional supplementation improved postoperative gastrointestinal recovery in elderly hip fracture patients. Probiotics appeared safe and convenient, and may represent a promising ERAS strategy. Multicenter studies with longer follow-up are needed to confirm these results.

Trial registration

The trial was retrospectively registered on 14 September 2024 and the registration number is ChiCTR2400089785.

Background

ERAS refers to evidence-based perioperative care pathways designed to accelerate patient recovery and reduce postoperative complications [1]. ERAS-guided perioperative care can promote early postoperative recovery, lower post-surgical complication rates and improve patient experience [2]. While preoperative carbohydrate loading is widely recommended [3, 4], its implementation in elderly trauma patients is often hindered by safety concerns, such as aspiration risk and logistical challenges [5, 6]. Recent researches have highlighted the potential of probiotics to improve gastrointestinal function, modulate the intestinal microbiome, enhance mucosal barrier integrity and attenuate systemic inflammatory responses [7–9]. Probiotic supplementation has demonstrated benefits in elective gastrointestinal and abdominal surgeries, including faster restoration of bowel function and improved biochemical markers [10–12]. However, evidence in elderly trauma patients remains limited. Biological plausibility for probiotic use in the perioperative period is supported by their capacity to reinforce tight junctions, reduce endotoxin translocation and modulate immune responses [13]. This study therefore investigates whether perioperative probiotic supplementation can serve as a safe and effective alternative to preoperative oral carbohydrate loading, focusing on gastrointestinal recovery and barrier-related outcomes in elderly hip fracture patients.

Methods

Ethics statement

The single-center, assessor-blinded, three-arm randomized controlled trial was conducted from April 2024 to December 2024. The study received approval from the Ethics Committee of the Fourth Medical Center of the PLA General Hospital (No. 2024KY038-KS001) and was retrospectively registered in the China Clinical Trials Center (ChiCTR2400089785) on 14 September 2024. Registration was retrospective because of administrative delay in registry processing. The primary and secondary outcomes had been prespecified in the study protocol approved by the institutional ethics committee before patient enrollment, and no changes to trial methods or outcomes were made after trial commencement. Written informed consent was obtained from all enrolled participants. The study was carried out in accordance with the Declaration of Helsinki. This trial report adheres to the CONSORT 2025 reporting guideline; the completed CONSORT checklist is provided as an additional file (Additional file 1).

Patients

This study recruited elderly patients with intertrochanteric or femoral neck fractures who were willing to undergo surgery. Inclusion criteria included: (1) American Society of Anesthesiologists (ASA) class I to III; (2) age older than 65 years; (3) no communication or cognitive impairment; (4) voluntary participation and written informed consent.

Exclusion criteria included: (1) gastrointestinal dysfunction or the need for parenteral nutrition; (2) severe liver impairment (Child-Pugh class C); (3) chronic kidney disease stage 3b or higher; (4) concomitant thyroid disease; (5) preoperative psychiatric illness, such as schizophrenia, epilepsy, Alzheimer’s disease or Parkinson’s disease; (6) severe medical conditions, such as myasthenia gravis.

During the study, the patients’ participation will be terminated if the following situations occur: (1) spinal anesthesia fails and needs to be changed to epidural anesthesia or general anesthesia; (2) severe events such as shock, severe hypotension, allergic reaction, arrhythmia or cardiac arrest occur during the perioperative period; (3) postoperative admission to the intensive care unit (ICU) is required.

Randomization and intervention

Patients were randomly assigned in a 1:1:1 ratio to three groups: the control group (C group), the preoperative nutrition component group (M group) and the perioperative probiotic group (P group). The randomization sequence was generated using a computer-based random number generator in permuted blocks of six by an independent statistician not involved in patient care. Allocation was concealed in sequentially numbered, opaque, sealed envelopes opened by a research nurse after enrollment. C group fasted from both food and water after 10:00 p.m. the night before surgery. M group received 200 mL of a nutritional supplement (Mizone, Danone, Sichuan, China; per 100 mL: carbohydrate 4 g, sugar 4 g, vitamin B6 0.09 mg, vitamin C 20 mg, niacin 0.8 mg) administered two hours before surgery. P group was given Bifidobacterium triple viable capsules (administered twice daily, 420 mg per dose) from admission to discharge. Each gram of the probiotic contained Bifidobacterium longum (1.0 × 10^7 CFU), Lactobacillus acidophilus (1.0 × 10^6 CFU) and Enterococcus faecalis (1.0 × 10^6 CFU). Due to the nature of the interventions, participants and ward nurses were not blinded to group allocation. Anesthesiologists, surgeons, and outcome assessors were blinded to group assignment.

Anesthesia and surgery

Upon admission, intravenous access was established, and pulse oximetry, electrocardiograph (ECG) as well as invasive arterial pressure were routinely monitored. Spinal anesthesia was performed at L2-3/L3-4 with 0.5% hyperbaric bupivacaine (1 mL 10% glucose + 2 mL 0.75% bupivacaine). Vasoactive medications were used as needed during surgery to control blood pressure and heart rate fluctuations within 20% of baseline values. Postoperative analgesia was achieved with patient-controlled intravenous analgesia (PCIA) (sufentanil 2 µg/kg + dexamethasone 10 mg, diluted to 100 mL, maintained at 2 mL/h, with the bolus of 0.5 mL for 2 days). The patient was transferred to the post-anesthesia care unit (PACU) after surgery and discharged after achieving the Steward score of 4 or higher.

Outcomes

The primary outcome was the time to first postoperative flatus, which was used as an objective marker of gastrointestinal recovery. In addition, we recorded blood glucose at admission, admission to operating room (OR), leaving OR and postoperative day 1, as well as insulin, HOMA-IR (HOMA-IR = fasting glucose × insulin / 22.5) and endotoxin levels at admission and postoperative day 1.

In addition to evaluating gastrointestinal function, we also assessed the safety and ERAS recovery outcomes of the perioperative probiotic intervention. Other observational indicators included operation time, intraoperative blood loss, urine output, fluid infusion volume, length of hospital stay, as well as perioperative hemodynamic parameters, specifically MAP and HR measured at four time points: T1 (admission), T2 (after anesthesia), T3 (mid-surgery) and T4 (upon leaving OR). Furthermore, postoperative subjective assessments were performed, including the Self-Rating Anxiety Scale (SAS), Visual Analogue Scale (VAS) for pain, Ramsay Sedation Score and the 3D-CAM delirium assessment.

Sample size

The sample size calculation was performed using G*Power 3.1 based on previous study findings. We referenced the median time to first postoperative flatus reported in a similar study [14], which was 21.5 [16.5, 33.6] hours in the probiotic group and 29.2 [25.3, 43.3] hours in the placebo group. By converting these values, we assumed an estimated mean time of 21.5 h (SD 12.7 h) in the intervention group and 29.2 h (SD 13.3 h) in the control group. With a two-tailed α of 0.05 and a power of 0.80, the estimated sample size was 47 participants each group. To ensure the robustness of the study and account for a potential 10% dropout rate, we opted for a sample size of 52 participants each group.

Statistical analysis

The primary efficacy analysis was conducted in the per-protocol population because certain post-randomization perioperative events (e.g., conversion of anesthesia, unplanned ICU admission or other protocol-deviating conditions) precluded standardized implementation of the assigned intervention or standardized postoperative assessment. To improve transparency and assess robustness against attrition bias, we additionally performed a sensitivity analysis according to the intention-to-treat (ITT) principle based on initial randomization assignment for the primary outcome. Reasons of post-randomization exclusions were recorded and reported by randomized group in the CONSORT flow diagram.

Normality of continuous variables was assessed using the Shapiro-Wilk test. Normally distributed variables were presented as mean ± standard deviation (SD) and analyzed by one-way analysis of variance (ANOVA) with Bonferroni post hoc correction for multiple comparisons. Non-normally distributed data were presented as median and interquartile range (IQR) and compared using the Kruskal-Wallis test followed by the Mann-Whitney U test. Categorical variables were described as the number and frequency (%) and were analyzed using the chi-squared test or Fisher’s test as appropriate. Time-to-event data was analyzed using the Kaplan-Meier survival curves and inter-group differences were analyzed using the log-rank test. Cox proportional hazard regression analysis was applied to estimate hazard ratios (HR) and 95% confidence intervals (CI). Data analyses and figures were performed using R version 4.3.0 and GraphPad Prism 8.0. Statistical significance was set at P < 0.05.

Results

From April 2024 to December 2024, a total of 170 elderly patients with hip fractures were enrolled and randomly assigned to the perioperative probiotic group (P group, n = 58), the preoperative nutrition component group (M group, n = 57) or the control group (C group, n = 55). Exclusions in P group (n = 7) comprised three patients transferred to the ICU postoperatively and four requiring conversion to general anesthesia due to block failure. M group had six exclusions: three due to spinal anesthesia failure, two due to surgical cancellation and one due to postoperative ICU transfer. Besides, C group had three exclusions: two patients transferred to ICU and one requiring conversion to general anesthesia. Thus, a total of 170 patients were included in the intention-to-treat (ITT) population (58 in P group, 57 in M group and 55 in C group). After excluding 16 patients due to postoperative ICU admission or conversion of anesthesia methods, 154 patients were included in the per-protocol analysis (51 in P group, 51 in M group and 52 in C group). The study flow is presented in Fig. 1.

Fig. 1.

Flowchart of randomized controlled trial based on CONSORT guideline

Demographic and clinical characteristics of the participants

Baseline demographic and clinical characteristics, including age, sex, body mass index (BMI), Charlson Comorbidity Index (CCI), preoperative comorbidities (hypertension, diabetes, coronary heart disease and chronic obstructive pulmonary disease [COPD]), ASA classification, as well as blood glucose and insulin levels at admission, were comparable among the three groups (all P > 0.05).

Similarly, there were no significant between-group differences in operative duration, intraoperative blood loss, urine output, fluid infusion volume or perioperative hemodynamic parameters (MAP and HR at admission [T1], after anesthesia [T2], mid-surgery [T3] and upon leaving the operating room [T4]) (all P > 0.05). These findings indicate successful randomization and good baseline comparability across the three groups (Table 1).

Table 1.

Demographic and clinical characteristics of participants

	C(n = 55)	M(n = 57)	P(n = 58)	F/χ²	P
Age (y, Mean ± SD)	79.52 ± 8.21	76.53 ± 9.03	77.17 ± 7.92	3.711	0.156
Gender [n (%)]				0.782	0.676
Male	11 (20.0)	10 (17.5)	14 (24.1)
Female	44 (80.0)	47 (82.5)	44 (75.9)
BMI (kg/m², Mean ± SD)	23.3 ± 4.3	22.7 ± 3.3	24.4 ± 4.2	2.494	0.086
Preoperative comorbidities [n (%)]
Hypertension	23(41.8)	25(43.9)	26(44.8)	0.108	0.948
Diabetes	17(30.9)	15(26.3)	19(32.8)	0.6	0.74
Coronary heart disease	8(14.5)	9(15.8)	6(10.4)	0.8	0.67
COPD	7(12.8)	8(14.0)	7(12.0)	0.102	0.95
CCI [n (%)]				0.301	0.86
0–1	50 (90.9)	51 (89.5)	50 (86.2)
2–3	5 (9.1)	6 (10.5)	8 (13.8)
ASA classification [n (%)]				7.527	0.275
Ⅱ	31 (56.4)	36 (63.2)	29 (50.0)
Ⅲ	20 (36.4)	20 (35.1)	26 (44.8)
Ⅳ	4 (7.3)	1 (1.70)	3 (5.2)
Operation time (min, Mean ± SD	131.49 ± 52.15	142.71 ± 52.52	143.60 ± 45.10	2.035	0.361
Blood loss (ml, Mean ± SD)	211.4 ± 116.6	207.5 ± 124.1	207.7 ± 226.5	0.01	0.99
Urine output (ml, Mean ± SD)	316.0 ± 200.1	343.2 ± 222.9	293.3 ± 198.1	0.832	0.437
Fluid volume (ml, Mean ± SD)	1282.3 ± 444.6	1314.3 ± 528.0	1464.1 ± 620.8	1.861	0.159
HR (bpm, Mean ± SD)
T1	81.3 ± 12.6	82.6 ± 9.8	81.0 ± 11.6	0.311	0.733
T2	81.3 ± 12.6	82.4 ± 12.4	83.0 ± 9.4
T3	77.2 ± 13.1	80.1 ± 13.8	81.6 ± 12.6
T4	73.2 ± 11.7	70.6 ± 11.8	71.8 ± 11.5
MAP (mmHg, Mean ± SD)
T1	99.9 ± 11.8	98.2 ± 9.3	98.9 ± 10.5	0.386	0.68
T2	99.7 ± 11.8	99.9 ± 11.3	101.1 ± 12.7
T3	93.7 ± 11.3	87.0 ± 13.1	90.6 ± 15.0
T4	84.3 ± 11.8	86.5 ± 9.5	85.3 ± 10.5
Blood glucose upon admission (mmol/L, Mean ± SD)	7.41 ± 2.08	7.69 ± 2.56	7.30 ± 2.89	0.357	0.7
Insulin upon admission (mU/L, Mean ± SD)	13.67 ± 11.67	12.02 ± 6.69	12.08 ± 7.68	0.621	0.539

Postoperative gastrointestinal function and metabolic outcomes

Because the primary outcome (time to first postoperative flatus) and other postoperative variables did not follow the normal distribution (Shapiro–Wilk test, all P < 0.05), non-parametric methods were used for analysis.

Time to first postoperative flatus differed significantly among the three groups (overall P < 0.001, Kruskal–Wallis test; Table 2; Fig. 2). The median time to first postoperative flatus was 10.00 (7.00, 17.00) hours in C group, 5.24 (4.00, 7.00) hours in M group and 3.40 (2.00, 5.98) hours in P group. Pairwise comparisons showed that both M and P groups had significantly shorter time to first postoperative flatus compared with C group (M vs. C, P < 0.001; P vs. C, P < 0.001), and P group recovered significantly faster than the M group (P vs. M, P < 0.001). Kaplan–Meier analysis further demonstrated that patients in P group achieved the earliest gastrointestinal recovery (log-rank P < 0.001; Fig. 3). In Cox proportional hazards models, the probability of first postoperative flatus was significantly higher in both intervention groups than in the control group. Compared with C group, the hazard ratio (HR) for the first postoperative flatus was 5.08 (95% CI 3.32–7.77, P < 0.001) in P group and 3.45 (95% CI 2.24–5.32, P < 0.001) in M group, indicating the fastest recovery in P group. The per-protocol analysis yielded a similar result: compared with C group, the hazard ratio (HR) for the first postoperative flatus was 5.18 (95% CI: 3.32–8.07, P < 0.001) in P group and 3.40 (95% CI: 2.16–5.35, P < 0.001) in M group. These findings further confirm that P group experienced the fastest postoperative recovery. The consistency of results between the intention-to-treat and per-protocol analyses strengthens the reliability of the observed effect.

Table 2.

Summary of postoperative gastrointestinal function and metabolic outcomes

Outcome	P Group	M Group	C Group	P (Kruskal-Wallis)	P-C(Mann-Whitney U)	M-C(Mann-Whitney U)	P-M(Mann-Whitney U)
Time to first postoperative flatus (h, Median [IQR])	3.40 (2.00, 5.98)	5.24 (4.00, 7.00)	10.00 (7.00, 17.00)	< 0.001	< 0.001	< 0.001	< 0.001
Postoperative bacterial endotoxin (mg/L, Median [IQR])	13.78 (11.95, 16.07)	15.43 (13.14, 18.17)	18.65 (15.56, 23.83)	< 0.001	< 0.001	< 0.001	0.041
Postoperative blood glucose (mmol/L, Median [IQR])	6.53 (5.54, 7.36)	6.47 (5.74, 7.13)	7.29 (5.64, 8.86)	0.04	0.04	0.02	0.79
Postoperative HOMA2-IR (Median [IQR])	3.61 (1.82, 4.94)	3.40 (2.04, 4.69)	4.54 (2.54, 7.82)	0.006	0.007	0.005	0.887
Postoperative CRP (mg/L, Median [IQR])	33.00 (20.74, 57.90)	47.50 (35.80, 67.50)	53.60 (27.25, 73.85)	0.025	0.023	0.894	0.017

Fig. 2.

Summary of indicators of postoperative outcomes. *, **, *** Data with different symbols are significantly different between the groups (P < 0.05 or P < 0.01 or P < 0.005)

Fig. 3.

Kaplan-Meier survival curve for the probability of first postoperative flatus

Postoperative bacterial endotoxin levels were significantly lower in M and P groups than in C group (C: 18.65 [15.56, 23.83] mg/L, M: 15.43 [13.14, 18.17] mg/L, P: 13.78 [11.95, 16.07] mg/L; overall P < 0.001). Pairwise comparisons showed significant reductions in both intervention groups versus the control group (M vs. C, P < 0.001; P vs. C, P < 0.001) and a further reduction in P group compared with M group (P = 0.041).

Postoperative blood glucose levels were also lower in both intervention groups than in the control group (C: 7.29 [5.64, 8.86] mmol/L, M : 6.47 [5.74, 7.13] mmol/L, P : 6.53 [5.54, 7.36] mmol/L; overall P = 0.04). Compared with C group, both M (P = 0.02) and P (P = 0.04) groups showed significantly reduced glucose levels, whereas no difference was observed between the two intervention groups (P = 0.79).

Similarly, postoperative HOMA2-IR values were significantly lower in both intervention groups than in the control group (C: 4.54 [2.54, 7.82], M : 3.40 [2.04, 4.69], P : 3.61 [1.82, 4.94]; overall P = 0.006). Pairwise analyses showed significant differences for M vs. C (P = 0.005) and P vs. C (P = 0.007), with no significant difference between the M and P groups (P = 0.887).

Postoperative CRP levels differed significantly among the three groups (overall P = 0.025). P group had significantly lower CRP levels than C group (C: 53.60 [27.75, 73.85] mg/L, P : 33.00 [20.74, 57.90] mg/L; P = 0.025) and M group (M: 47.50 [35.80, 67.50] mg/L; P vs. M: P = 0.017), whereas CRP did not differ significantly between C and M groups (P = 0.82).

Taken together, these results indicate that both preoperative nutritional supplementation and perioperative probiotic administration were associated with earlier postoperative gastrointestinal recovery and improved metabolic as well as intestinal barrier markers, with the most pronounced and consistent benefits observed in the probiotic group (Table 2). The per-protocol analysis yielded similar results, as presented in Table 3.

Table 3.

Comparison of the primary outcome among groups using the intention-to-treat and per-protocol analyses

Outcome	Analysis type	P Group Median (IQR)	M Group Median (IQR)	C Group Median (IQR)	P (Kruskal-Wallis)	P-C (Mann-Whitney U)	M-C (Mann-Whitney U)	P-M (Mann-Whitney U)
Time to first postoperative flatus (h)	Intention-to-treat	3.40 (2.00, 5.98)	5.24 (4.00, 7.00)	10.00 (7.00, 17.00)	< 0.001	< 0.001	< 0.001	< 0.001
	Per-protocol	3.00 (2.00, 5.50)	5.00 (4.00, 7.50)	10.00 (7.00, 17.25)	< 0.001	< 0.001	< 0.001	< 0.001

Other ERAS recovery outcomes

Other ERAS-related recovery indicators included postoperative complications, pain scores and length of hospital stay. The incidence of postoperative delirium was similar across groups (C: 12.7%, M: 14.9%, P: 12.1%; P = 0.95). Median pain scores on postoperative days 1 and 2 were 2 across all groups, with no significant between-group differences (postoperative day 1 VAS, P = 0.403; postoperative day 2 VAS, P = 0.257). Length of hospital stay did not differ significantly among the three groups (all P > 0.05) (Table 4).

Table 4.

Summary of other ERAS recovery outcomes

Outcome	P Group	M Group	C Group	P (χ²/Kruskal-Wallis)
Delirium [n (%)]	7/55 (12.7%)	8/57 (14.0%)	7/58 (12.1%)	0.95
Postoperative day 1 VAS (points, Median [IQR])	2.00 (2.00,3.00)	2.00 (2.00,3.00)	2.00 (2.00,3.00)	0.403
Postoperative day 2 VAS (points, Median [IQR])	2.00 (1.00,2.00)	2.00 (1.00,2.00)	2.00 (1.00,2.00)	0.257
Length of hospital stay (days, Median [IQR])	13.0 (9.0,17.0)	12.0 (9.0,16.0)	12.0 (9.0–16.0)	0.66

Overall, perioperative probiotic and preoperative nutritional interventions did not increase postoperative pain, delirium or other complications, and appeared safe for use in elderly patients with hip fractures.

Discussion

In this randomized trial of elderly patients with hip fracture, perioperative probiotic supplementation was associated with faster postoperative gastrointestinal recovery, as reflected by earlier time to first flatus and with favorable changes in several postoperative biochemical markers compared with conventional fasting. However, these findings should be interpreted with appropriate caution. Time to first postoperative flatus is a commonly used ERAS-related recovery endpoint, but it remains an early surrogate indicator and does not necessarily translate into broader clinical benefit in all settings. In the present study, we did not observe significant between-group differences in pain scores, delirium incidence or length of hospital stay. In addition, the trial was not powered to detect modest differences in many secondary outcomes or uncommon complications. Therefore, our findings support improved early postoperative recovery signals rather than definitive improvement in major clinical outcomes.

Our results extend current ERAS concepts, which emphasize minimizing preoperative fasting and using carbohydrate loading to reduce insulin resistance and enhance patients’ comfort [15, 16]. Consistent with previous reports, preoperative carbohydrate administration was associated with improved postoperative metabolic recovery compared with standard fasting [17, 18]. At the same time, our data added to the growing body of literature suggesting that perioperative probiotic supplementation may contribute to faster restoration of gastrointestinal function and improved barrier-related biomarkers in surgical patients [19, 20]. Unlike most prior probiotic studies conducted in elective gastrointestinal or abdominal surgery populations [21], this trial focused on elderly trauma patients undergoing hip fracture surgery, a group in whom ERAS implementation is often constrained by concerns about safety and feasibility. Within this context, perioperative probiotics appeared to offer a practical alternative to time-critical preoperative carbohydrate loading, while achieving at least comparable, and in some respects more pronounced, improvements in gastrointestinal recovery and selected biomarkers.

Several plausible biological mechanisms may underlie the observed benefits. Trauma, anesthesia and surgery can disrupt the intestinal mucosal barrier, increase permeability and amplify systemic inflammatory responses, although direct causal links remain incompletely defined [22, 23]. The intestinal barrier is a composite of mechanical, chemical, microbial and immune components that together maintain mucosal integrity [24, 25]. Probiotic preparations containing bifidobacteria and lactobacilli may help restore a more favorable microbial milieu, reinforce tight junction function and limit the translocation of bacteria and endotoxin [26]. In our study, postoperative endotoxin levels were lowest in the probiotic group, which is consistent with a potential barrier-preserving effect, although endotoxin is an indirect and non-specific surrogate for barrier integrity [27]. Improvements in postoperative blood glucose and HOMA2-IR in both intervention groups further suggest attenuation of stress-related metabolic derangements, possibly mediated by reduced systemic inflammation and, as hypothesized in other settings, by modulation of incretin and glucagon-like peptide pathways [28]. The observed reductions in endotoxin, CRP, glucose and HOMA-IR in the probiotic group are consistent with possible modulation of postoperative inflammatory and metabolic stress responses. Nonetheless, these biomarkers are indirect surrogate indicators and do not provide direct evidence of intestinal barrier restoration or specific microbiome-mediated mechanisms. Because we did not perform microbiome profiling, direct intestinal permeability testing, metabolomics or histologic assessment of the intestinal mucosa, mechanistic inferences should be considered hypothesis-generating rather than confirmatory.

From a practical perspective, the intervention regimens in both active arms appeared to be implementable within routine perioperative workflows. Preoperative carbohydrate loading requires accurate timing within a narrow window before anesthesia and may be difficult to apply consistently in elderly trauma patients with variable surgical schedules or aspiration concerns [29]. By contrast, the probiotic regimen was initiated on admission and continued until discharge, without requiring precise timing relative to surgery, and did not result in hemodynamic instability, increased intraoperative blood loss or other evident safety signals. Although some reports have raised concerns about probiotic prophylaxis in specific high-risk populations, such as patients with severe acute pancreatitis [30], most studies in elective surgical cohorts have not demonstrated increased rates of serious adverse events [31, 32]. In our carefully selected elderly hip fracture population without severe gastrointestinal dysfunction or critical organ failure, perioperative probiotic supplementation appeared safe and convenient, with no observed increase in complications or mortality. Nevertheless, the modest sample size limits the ability to detect rare adverse events.

This study has several strengths, including its randomized controlled design, focus on the elderly trauma patients who were often underrepresented in ERAS research, use of an objective primary endpoint, as well as parallel assessment of metabolic, inflammatory and barrier-related biomarkers. However, some limitations must be acknowledged. First, it was a single-center study, and local perioperative care protocols may limit generalizability to other institutions. Second, participants and ward nurses were not blinded because of the nature of the interventions, which may have introduced performance bias and detection/ascertainment bias, particularly for the primary endpoint that relies on patient reporting and nursing documentation. Third, the post-randomization exclusions may introduce attrition bias despite predefined criteria. To address this concern, we reported the reasons of exclusions by group and performed the ITT analysis. Fourth, the enrolled population was relatively selected (e.g., excluding patients with severe cognitive impairment, severe gastrointestinal dysfunction and major comorbidities), which limits applicability to more severely ill hip fracture patients. Fifth, we relied on surrogate biomarkers (endotoxin, CRP, glucose and HOMA2-IR) rather than direct measures of gut microbiota composition or barrier structure. Sixth, follow-up was limited to hospitalization, and the long-term functional recovery, quality of life, nutritional outcomes as well as post-discharge events were not assessed. Finally, the trial registration was retrospective due to administrative delay, although outcomes and methods were prespecified and unchanged after trial commencement.

Future multicenter, adequately powered trials incorporating microbiome and metabolomic profiling, broader inclusion criteria and longer follow-up are warranted to confirm these preliminary findings and to clarify whether perioperative probiotic strategies can translate into meaningful improvements in long-term clinical and patient-reported outcomes.

Conclusion

In this selected elderly hip fracture population, perioperative probiotic supplementation was associated with faster postoperative gastrointestinal recovery and favorable changes in indirect inflammatory and metabolic biomarkers compared with conventional fasting. These findings support the feasibility of perioperative probiotic use and suggest short-term physiological benefit signals in this setting. However, because major clinical outcomes (including delirium and length of hospital stay) were not significantly different and mechanistic pathways were not directly assessed, larger multicenter trials with longer follow-up are needed to determine the impact on patient-centered outcomes, long-term recovery and cost-effectiveness.

Abbreviations

ERAS

Enhanced recovery after surgery

ICU

Intensive care unit

ECG

Electrocardiograph

PCIA

Patient-controlled intravenous analgesia

PACU

Post-anesthesia care unit

OR

Operating room

MAP

Mean arterial pressure

HR

Heart rate

SAS

Self-rating anxiety scale

VAS

Visual analogue scale

SD

Standard deviation

IQR

Interquartile range

HR

Hazard ratios

CI

Confidence intervals

BMI

Body mass index

CCI

Charlson Comorbidity Index

ASA

American Society of Anesthesiologists

COPD

Chronic obstructive pulmonary disease

CRP

C-reactive protein

ITT

Intention-to-treat

Authors’ contributions

LJL and MML designed the study; LJL and YS conducted the clinical study; STR collected the data; LYL and QX conducted quality control during the research process; NYT analyzed the data, wrote the manuscript, constructed the figures and tables; HZ and MML revised the manuscript. All the authors contributed substantially to the work and approved the final version of manuscript to be published.

Funding

This study was supported by the National Natural Science Foundation of China (No. 81801884).

Data availability

The data of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the Fourth Medical Center of the PLA General Hospital (No. 2024KY038-KS001) and retrospectively registered in the Chinese Clinical Trial Registry (ChiCTR2400089785). The protocol of this study adheres to CONSORT guidelines. Informed consent was obtained from all the subjects. This study was conducted in accordance with the Declaration of Helsinki. All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.