Abstract
Background
Surgical site infections (SSIs) are major complications following hepatobiliary surgery, leading to prolonged hospitalization, increased healthcare costs, and higher morbidity. Identifying causative pathogens and clinical risk factors is essential for guiding effective prevention and management strategies.
Methods
This retrospective study included 916 adult patients who underwent hepatobiliary surgery between January 2023 and June 2024. Patients were classified into SSI (n = 51) and non-SSI (n = 865) groups according to Centers for Disease Control and Prevention (CDC) diagnostic criteria. Clinical, surgical, and microbiological data were collected. Pathogen identification and antimicrobial susceptibility testing were conducted using standard microbiological techniques and Clinical and Laboratory Standards Institute (CLSI) guidelines. Univariate and multivariate logistic regression analyses were performed to identify independent risk factors for SSI occurrence. Model calibration and discrimination were assessed with the Hosmer–Lemeshow test and area under the receiver operating characteristic curve (AUC).
Results
Among 916 patients, 51 (5.6%) developed SSIs. Gram-negative bacilli predominated (66.7%), with Escherichia coli (29.4%) and Klebsiella pneumoniae (21.6%) as the most frequent pathogens. Extended-spectrum β-lactamase production was identified in 20.0% of E. coli and 18.1% of K. pneumoniae isolates, whereas carbapenem resistance was low. Multivariate analysis identified diabetes mellitus (odds ratio (Moraes CMT, Corrêa LM, Procópio RJ, Carmo G, Navarro TP. Rev Col Bras Cir 49:e20223125.1, 2022) 2.08, 95% confidence interval [CI] 1.07–4.05), hypoalbuminemia < 35 g/L (OR 2.95, 95% CI 1.60–5.45), operative duration ≥ 240 min (OR 2.72, 95% CI 1.45–5.11), and biliary drainage (OR 1.91, 95% CI 1.01–3.61) as independent predictors. Model calibration was adequate (Hosmer–Lemeshow p = 0.62), with good discrimination (AUC = 0.81).
Conclusions
SSIs following hepatobiliary surgery are primarily caused by multidrug-resistant Gram-negative organisms. Perioperative optimization, including glycemic and nutritional management, reduction of operative time, and careful use of biliary drainage, is critical for lowering.
Clinical trial number
Not applicable.
Keywords: Surgical site infections, Hepatobiliary surgery, Escherichia coli, Klebsiella pneumoniae, Extended-spectrum β-lactamase, Multidrug resistance, Risk factors
Introduction
Surgical site infections (SSIs) represent one of the most frequent and challenging complications in the field of abdominal surgery, particularly following hepatobiliary procedures. Hepatobiliary surgery, encompassing operations such as hepatectomy, biliary reconstruction, and cholecystectomy, is widely performed for both benign and malignant diseases. Despite advances in surgical techniques, perioperative care, and antimicrobial prophylaxis, SSIs remain a substantial cause of postoperative morbidity, prolonged hospitalization, increased healthcare costs, and, in severe cases, mortality [1, 2]. The incidence of SSIs after hepatobiliary surgery has been reported to range between 10% and 25%, significantly higher than in many other surgical domains, largely due to the anatomical and physiological complexity of the hepatobiliary system and the frequent presence of biliary tract contamination. The hepatobiliary surgical field presents unique risk profiles for infection. Factors such as bile leakage, intraoperative hemorrhage, and prolonged operative duration contribute to a higher vulnerability to bacterial contamination and infection. Furthermore, many patients’ undergoing hepatobiliary surgery have underlying liver disease, biliary obstruction, diabetes mellitus, or immunosuppression, all of which impair host defense mechanisms and further increase infection risk. The interplay between these host-related and surgery-related factors underscores the multifactorial etiology of SSIs in this context [3, 4].
The distribution of pathogenic microorganisms in SSIs following hepatobiliary surgery is another crucial dimension that warrants investigation. Commonly isolated pathogens include Gram-negative bacilli such as Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa, as well as Gram-positive cocci such as Enterococcus species and Staphylococcus aureus. With the widespread use of perioperative prophylactic antibiotics, however, there has been a noticeable shift in the bacterial spectrum, with an increasing prevalence of multidrug-resistant organisms (MDROs), including extended-spectrum β-lactamase (ESBL)-producing Enterobacteriaceae and carbapenem-resistant strains [5, 6]. This evolving microbial landscape complicates the choice of empirical antimicrobial therapy and highlights the importance of local epidemiological data for guiding clinical decision-making. In addition to microbial factors, patient- and procedure-related determinants are key contributors to SSI risk. Advanced age, poor nutritional status, postoperative biliary drainage, hypoalbuminemia, obesity, and comorbidities such as diabetes mellitus and chronic liver disease have all been identified as potential risk factors. Surgical variables, including emergency operations, complex biliary reconstructions, repeated interventions, and intraoperative contamination, further increase the likelihood of infection. Identifying and quantifying these risk factors is essential for developing effective preventive strategies, tailoring perioperative management, and reducing the burden of SSIs [7, 8].
In the existing literature on SSIs following hepatobiliary surgery, several studies have identified common risk factors such as diabetes, hypoalbuminemia, prolonged operative duration, and preoperative biliary drainage [7, 9]. However, there remains a gap in understanding the specific microbiological profiles and antimicrobial resistance patterns associated with SSIs in this patient cohort, particularly in the context of multidrug-resistant Gram-negative organisms, which are increasingly prevalent [10]. While prior research has primarily focused on clinical and surgical risk factors, limited attention has been given to the local distribution of pathogens and their resistance profiles, which are critical for effective empirical antibiotic selection and infection management [11, 12]. To address this gap, our study aimed to provide a comprehensive analysis of both clinical risk factors and the microbiological etiology of SSIs following hepatobiliary surgery. Specifically, we focused on identifying the dominant pathogens responsible for infections in our cohort, with an emphasis on multidrug-resistant organisms, such as Escherichia coli and Klebsiella pneumoniae. By investigating both the clinical and microbiological dimensions of SSIs, our study offers novel insights into the microbial challenges faced in hepatobiliary surgery and emphasizes the importance of localized microbiological surveillance for optimizing prevention and treatment strategies. This approach is vital for improving patient outcomes and reducing the burden of SSIs in hepatobiliary surgery.
Methods
Study design
This retrospective study included patients who underwent hepatobiliary surgery at our institution between January 2023 and June 2024. Patients were eligible for inclusion if they met all of the following conditions: (1) Adults aged 18 years or older; (2) Underwent hepatobiliary surgery at our institution; (3) Had positive microbiological culture results from specimens collected postoperatively; (4) Possessed complete clinical, surgical, and microbiological records available for analysis; (5) Received standard perioperative prophylactic antibiotic therapy in accordance with institutional protocols. Patients were excluded if they met any of the following conditions: (1) Presence of pre-existing infections at non-surgical sites (e.g., pneumonia, urinary tract infection, or sepsis) before surgery; (2) Receipt of systemic antimicrobial therapy within two weeks prior to surgery, which could potentially influence intraoperative or postoperative microbial detection; (3) Severe immunodeficiency conditions. Postoperatively, patients were categorized into two groups based on the occurrence of SSIs: the SSI group (n = 51) and the non-SSI group (n = 865). All patients received standardized perioperative prophylaxis with intravenous ceftriaxone, administered 30–60 min before incision and re-dosed intraoperatively for prolonged procedures. In the absence of infection, prophylaxis was discontinued within 24 h postoperatively. The same regimen was applied to both SSI and non-SSI groups to ensure uniform antibiotic exposure. Informed consent was obtained from all subjects and/or their legal guardian(s). The study protocol was approved by the institutional ethics committee, and all procedures were conducted in accordance with relevant guidelines and regulations, including the Declaration of Helsinki. Participant confidentiality was strictly maintained, with all personal identifiers removed prior to data analysis.
Diagnostic criteria for surgical site infections
The diagnosis of SSI was based on the criteria of the Centers for Disease Control and Prevention (CDC) and classified as superficial incisional, deep incisional, or organ/space infection. Superficial SSI involved infection of the skin or subcutaneous tissue with signs such as purulent drainage, positive culture, or clinical evidence of infection. Deep incisional SSI affected the fascia or muscle layers and was defined by purulent drainage, wound dehiscence with clinical signs of infection, or abscess formation. Organ/space SSI involved deeper anatomical structures manipulated during surgery and was confirmed by purulent drainage, positive culture, or radiologic, histopathologic, or intraoperative evidence of infection.
Microbiological cultures were obtained only from patients who fulfilled CDC-based clinical criteria suggestive of SSI, such as purulent drainage, wound dehiscence with signs of infection, or other objective evidence of infection. Attending surgeons ordered cultures based on these clinical findings. Patients without any clinical indicators of infection were not subjected to microbiological testing. Consequently, only clinically suspected cases underwent specimen collection, and culture-positive cases were classified as SSIs. Although culture-negative SSIs may occur, such cases could not be identified under the current testing protocol. For the non-SSI group, patients who exhibited no clinical or microbiological evidence of infection throughout the postoperative period were included as controls. They did not undergo microbiological testing unless clinically indicated; therefore, culture-negative but clinically uninfected patients remained within the control cohort.
Initial identification of suspected SSI was made by attending surgeons based on CDC criteria. All suspected cases were then independently reviewed by two senior surgeons who were not involved in the index surgery. These reviewers assessed clinical manifestations, laboratory findings, imaging examinations, and microbiological results. Any discrepancies were resolved through discussion with a third senior surgeon to reach consensus.
Assessment of pathogens and antibiotic susceptibility
Intraoperative specimens, drainage fluid, or wound exudates from patients with suspected SSIs were collected under aseptic conditions and immediately transported to the microbiology laboratory. Bacterial isolation and identification were performed using standard microbiological techniques, including Gram staining, culture on selective and differential media, and automated identification systems (e.g., VITEK 2 Compact, bioMérieux). Pathogens were classified as Gram-positive, Gram-negative, or fungal isolates according to colony morphology, biochemical characteristics, and confirmatory testing. Antimicrobial susceptibility testing was conducted using the Kirby–Bauer disk diffusion method or automated microdilution assays, in accordance with the guidelines of the Clinical and Laboratory Standards Institute (CLSI). Minimum inhibitory concentrations (MICs) were determined where necessary. Multidrug resistance (MDR) was defined as non-susceptibility to at least one agent in three or more antimicrobial categories. ESBL production, carbapenem resistance, and methicillin resistance were confirmed by standardized phenotypic tests. Quality control was ensured using reference strains recommended by CLSI.
Data collection
Clinical data were retrospectively retrieved from the institutional electronic medical records. Collected variables included demographic characteristics (age, sex, and body mass index), lifestyle and medical history (smoking status, alcohol consumption, diabetes mellitus, hypertension, chronic liver disease, and immunosuppressive conditions), and preoperative laboratory findings (white blood cell count, hemoglobin, albumin, total bilirubin, alanine aminotransferase, and prothrombin time). Surgical details such as type of procedure (hepatectomy, biliary reconstruction, or cholecystectomy), operative duration, intraoperative blood loss, transfusion, and the use of biliary drainage or implants were recorded. In this study, biliary drainage referred to the placement of a dedicated biliary drain (e.g., T-tube or external/internal–external catheter) during the index operation or early postoperative period to decompress the biliary tree or protect biliary–enteric reconstruction in cases of extensive hilar dissection, biliary anastomosis, or intraoperative bile spillage. Drains placed later to manage postoperative abscesses or established biliary leaks were not included. Postoperative data included length of hospital stay, and occurrence of complications. All data were reviewed independently by two investigators to ensure accuracy and completeness, and discrepancies were resolved through consensus.
Statistical analysis
All statistical analyses were performed using SPSS software (version 28.0; IBM Corp., Armonk, NY, USA). Continuous variables were tested for normality using the Kolmogorov–Smirnov test. Normally distributed data were expressed as mean ± standard deviation (SD) and compared between groups using the independent-samples t test. Non-normally distributed data were presented as median (interquartile range, IQR) and analyzed with the Mann–Whitney U test. Categorical variables were expressed as frequencies and percentages and compared using the chi-square test or Fisher’s exact test, as appropriate. Univariate analysis was first conducted to identify potential risk factors for SSIs. Variables with a p-value < 0.10 in univariate analysis were subsequently entered into a multivariate logistic regression model to determine independent predictors of SSI occurrence. Odds ratios (ORs) with corresponding 95% confidence intervals (CIs) were calculated. Multicollinearity among covariates was assessed prior to model inclusion. Multicollinearity among covariates was assessed prior to model inclusion using variance inflation factor (VIF) and tolerance values. A VIF ≥ 5 or a tolerance < 0.2 was considered indicative of significant multicollinearity. The distribution of pathogenic microorganisms and their antimicrobial susceptibility profiles was summarized descriptively. Resistance rates of major pathogens to commonly used antibiotics were calculated and compared between groups when applicable. Model calibration and discrimination were evaluated using the Hosmer–Lemeshow goodness-of-fit test and the area under the receiver operating characteristic (ROC) curve (AUC). A two-sided p-value < 0.05 was considered statistically significant.
Results
Baseline characteristics
A total of 916 patients were included, with 51 (5.6%) developing SSIs. Compared with the non-SSI group, patients with SSIs were older (p = 0.030), and had higher rates of diabetes mellitus (p = 0.002) and chronic liver disease (p = 0.024). They also showed lower preoperative albumin levels (p < 0.001), longer operative duration (p < 0.001), greater intraoperative blood loss (p = 0.004), more frequent blood transfusion (p = 0.006), and higher incidence of postoperative biliary drainage (p = 0.004). The length of hospital stay was significantly longer in the SSI group (p < 0.001). In addition, no significant differences were found in surgical complexity, classified as Grade III or Grade IV procedures, between the SSI and non-SSI groups (p = 0.719). The distribution of elective versus urgent surgeries was also comparable across groups (p = 0.894). Likewise, the operative approach (open versus laparoscopic surgery) did not differ significantly between patients with and without SSIs (p = 0.737). No significant differences were observed in sex, BMI, or hypertension between groups (Table 1).
Table 1.
Baseline clinical and perioperative characteristics of patients undergoing hepatobiliary surgery
| Characteristic | SSI group (n = 51) | Non-SSI group (n = 865) | Test statistic | p-value |
|---|---|---|---|---|
| Age (years), mean ± SD | 63.2 ± 10.5 | 59.7 ± 11.2 | t = 2.176 | 0.030 |
| Sex (male), n (%) | 32 (62.7) | 503 (58.1) | χ² = 0.419 | 0.518 |
| BMI (kg/m²), mean ± SD | 25.8 ± 3.6 | 24.9 ± 3.4 | t = 1.831 | 0.067 |
| Diabetes mellitus, n (%) | 18 (35.3) | 152 (17.6) | χ² = 10.011 | 0.002 |
| Hypertension, n (%) | 21 (41.2) | 298 (34.5) | χ² = 0.960 | 0.327 |
| Chronic liver disease, n (%) | 14 (27.5) | 134 (15.5) | χ² = 5.085 | 0.024 |
| Preoperative albumin (g/L), mean ± SD | 33.1 ± 4.8 | 36.5 ± 5.2 | t = 4.556 | < 0.001 |
| Operative duration (min), mean ± SD | 275 ± 68 | 238 ± 59 | t = 4.313 | < 0.001 |
| Intraoperative blood loss (mL), median (IQR) | 650 (400–900) | 450 (300–700) | Z = 2.885 | 0.004 |
| Blood transfusion, n (%) | 19 (37.3) | 176 (20.3) | χ² = 7.561 | 0.006 |
| Biliary drainage, n (%) | 16 (31.4) | 132 (15.3) | χ² = 8.143 | 0.004 |
| Length of hospital stay (days), median (IQR) | 21 (17–28) | 14 (11–20) | Z = 4.321 | < 0.001 |
| Surgical complexity (Grade III), n (%) | 34 (66.7) | 556 (64.3) | χ² = 0.129 | 0.719 |
| Surgical complexity (Grade IV), n (%) | 17 (33.3) | 309 (35.7) | – | – |
| Surgery type (Elective), n (%) | 44 (86.3) | 740 (85.5) | χ² = 0.018 | 0.894 |
| Surgery type (Urgent), n (%) | 7 (13.7) | 125 (14.5) | – | – |
| Operative approach – Open surgery, n (%) | 13 (25.5) | 239 (27.6) | χ² = 0.113 | 0.737 |
| Operative approach – Laparoscopic surgery, n (%) | 38 (74.5) | 626 (72.4) | – | – |
Pathogen distribution and antimicrobial resistance
Among the 51 isolates from SSIs after hepatobiliary surgery, Gram-negative bacilli predominated (34/51, 66.7%). Escherichia coli (15/51, 29.4%) and Klebsiella pneumoniae (11/51, 21.6%) were most frequently detected, followed by Enterococcus spp. (8/51, 15.7%), Staphylococcus aureus (6/51, 11.8%), Pseudomonas aeruginosa (5/51, 9.8%), and Acinetobacter baumannii (3/51, 5.9%). Fungal pathogens (e.g., Candida albicans) accounted for 3/51 (5.9%). Antimicrobial susceptibility testing indicated generally low resistance burdens in Gram-negative organisms. Among E. coli, ESBL production was 20.0% (3/15) with ceftriaxone resistance of 13.3% (2/15) and no carbapenem resistance detected (0/15). For K. pneumoniae, ESBL production was 18.1% (2/11), ceftriaxone resistance 18.1% (2/11), and carbapenem resistance 9.1% (1/11). P. aeruginosa showed ceftazidime resistance of 20.0% (1/5) without carbapenem resistance (0/5). Among Gram-positive pathogens, Enterococcus exhibited ampicillin resistance of 12.5% (1/8) with no vancomycin-resistant isolates (0/8), while the proportion of methicillin-resistant S. aureus (MRSA) was 16.7% (1/6) with erythromycin resistance of 16.7% (1/6). A. baumannii demonstrated multidrug resistance in 33.3% (1/3) (Table 2).
Table 2.
Distribution and antimicrobial resistance profiles of pathogens isolated from surgical site infections
| Pathogen | No. of isolates (%) | Key Antimicrobial Resistance Findings |
|---|---|---|
| Escherichia coli | 15 (29.4) | ESBL-positive 20.0% (3/15); ceftriaxone resistance 13.3% (2/15); carbapenem resistance 0.0% (0/15) |
| Klebsiella pneumoniae | 11 (21.6) | ESBL-positive 18.1% (2/11); ceftriaxone resistance 18.1% (2/11); carbapenem resistance 9.1% (1/11) |
| Enterococcus spp. | 8 (15.7) | Ampicillin resistance 12.5% (1/8); vancomycin resistance 0.0% (0/8) |
| Staphylococcus aureus | 6 (11.8) | MRSA 16.7% (1/6); erythromycin resistance 16.7% (1/6) |
| Pseudomonas aeruginosa | 5 (9.8) | Ceftazidime resistance 20.0% (1/5); carbapenem resistance 0.0% (0/5) |
| Acinetobacter baumannii | 3 (5.9) | Multidrug resistance 33.3% (1/3) |
| Others (Candida albicans etc.) | 3 (5.9) | — |
Univariate analysis of risk factors for SSIs
Univariate analysis identified several clinical and perioperative factors significantly associated with the occurrence of SSIs after hepatobiliary surgery. Patients aged ≥ 60 years were more likely to develop SSIs compared with younger patients (60.8% vs. 43.2%, p = 0.018). The presence of diabetes mellitus (35.3% vs. 17.6%, p = 0.002) and chronic liver disease (27.5% vs. 15.5%, p = 0.031) were also significantly associated with increased SSI risk. Among perioperative parameters, hypoalbuminemia (< 35 g/L) was markedly more frequent in the SSI group (56.9% vs. 24.7%, p < 0.001). Longer operative duration (≥ 240 min) (66.7% vs. 33.4%, p < 0.001), greater intraoperative blood loss (≥ 500 mL) (58.8% vs. 32.0%, p < 0.001), and intraoperative blood transfusion (37.3% vs. 20.3%, p = 0.006) were also significantly associated with SSIs. Furthermore, postoperative biliary drainage (31.4% vs. 15.3%, p = 0.004) and postoperative ICU admission (25.5% vs. 12.3%, p = 0.009) were more frequent in patients with SSIs. In contrast, sex, body mass index, and hypertension did not show significant associations with the development of SSIs (Table 3).
Table 3.
Univariate analysis of clinical and perioperative risk factors for surgical site infections
| Variable | SSI group (n = 51) | Non-SSI group (n = 865) | χ² / t / Z | p-value |
|---|---|---|---|---|
| Age ≥ 60 years, n (%) | 31 (60.8) | 374 (43.2) | 5.63 | 0.018 |
| Male sex, n (%) | 32 (62.7) | 503 (58.1) | 0.40 | 0.532 |
| BMI ≥ 25 kg/m², n (%) | 27 (52.9) | 354 (40.9) | 2.79 | 0.095 |
| Diabetes mellitus, n (%) | 18 (35.3) | 152 (17.6) | 9.51 | 0.002 |
| Hypertension, n (%) | 21 (41.2) | 298 (34.5) | 0.93 | 0.336 |
| Chronic liver disease, n (%) | 14 (27.5) | 134 (15.5) | 4.63 | 0.031 |
| Hypoalbuminemia (< 35 g/L), n (%) | 29 (56.9) | 214 (24.7) | 24.7 | < 0.001 |
| Operative duration ≥ 240 min, n (%) | 34 (66.7) | 289 (33.4) | 20.8 | < 0.001 |
| Intraoperative blood loss ≥ 500 mL, n (%) | 30 (58.8) | 277 (32.0) | 15.1 | < 0.001 |
| Blood transfusion, n (%) | 19 (37.3) | 176 (20.3) | 7.56 | 0.006 |
| Preoperative biliary drainage, n (%) | 16 (31.4) | 132 (15.3) | 8.14 | 0.004 |
| ICU admission, n (%) | 13 (25.5) | 106 (12.3) | 6.75 | 0.009 |
Multivariate logistic regression analysis of factors for SSIs
The multivariate logistic regression model demonstrated that several factors were independently associated with SSIs after hepatobiliary surgery. Diabetes mellitus (β = 0.73, OR = 2.08, 95% CI: 1.07–4.05, p = 0.031), hypoalbuminemia (< 35 g/L) (β = 1.08, OR = 2.95, 95% CI: 1.60–5.45, p < 0.001), prolonged operative duration (≥ 240 min) (β = 1.00, OR = 2.72, 95% CI: 1.45–5.11, p = 0.002), and postoperative biliary drainage (β = 0.65, OR = 1.91, 95% CI: 1.01–3.61, p = 0.048) were identified as independent predictors of SSIs (Table 4; Fig. 1).
Table 4.
Multivariate logistic regression analysis of independent predictors of surgical site infections
| Variable | β Value | Standard Error | Wald Value | OR Value | 95% CI for OR | p-value |
|---|---|---|---|---|---|---|
| Age ≥ 60 years | 0.35 | 0.29 | 1.45 | 1.42 | 0.79–2.55 | 0.238 |
| ICU admission | 0.42 | 0.37 | 1.26 | 1.52 | 0.73–3.17 | 0.262 |
| Chronic liver disease | 0.49 | 0.35 | 1.97 | 1.64 | 0.82–3.28 | 0.160 |
| Blood transfusion | 0.32 | 0.34 | 0.89 | 1.38 | 0.71–2.67 | 0.341 |
| Intraoperative blood loss ≥ 500 mL | 0.63 | 0.31 | 4.01 | 1.87 | 1.01–3.45 | 0.045 |
| Preoperative biliary drainage | 0.65 | 0.33 | 3.95 | 1.91 | 1.01–3.61 | 0.048 |
| Diabetes mellitus | 0.73 | 0.34 | 4.63 | 2.08 | 1.07–4.05 | 0.031 |
| Operative duration ≥ 240 min | 1.00 | 0.32 | 9.78 | 2.72 | 1.45–5.11 | 0.002 |
| Hypoalbuminemia (< 35 g/L) | 1.08 | 0.32 | 11.30 | 2.95 | 1.60–5.45 | < 0.001 |
β, regression coefficient; OR, odds ratio; CI, confidence interval
Fig. 1.
Forest plot of adjusted risk factors for surgical site infections (SSIs) following hepatobiliary surgery. Adjusted odds ratios (ORs) and corresponding 95% confidence intervals (CIs) from the multivariate logistic regression model are displayed for all evaluated variables. Independent predictors of SSIs include diabetes mellitus, hypoalbuminemia (< 35 g/L), prolonged operative duration (≥ 240 min), and postoperative biliary drainage. The vertical reference line denotes the null value (OR = 1). Markers indicate point estimates, and horizontal lines represent 95% CIs
A sensitivity analysis was conducted by modeling key predictors as continuous variables to evaluate the robustness of findings and to minimize information loss from dichotomization. When operative duration was treated as a continuous variable, longer operative time remained independently associated with SSI (OR = 1.27 per 60-minute increase, 95% CI: 1.09–1.47, p = 0.003). Similarly, lower preoperative albumin levels were consistently associated with higher SSI risk when analyzed as a continuous variable (OR = 1.18 per 5 g/L decrease, 95% CI: 1.05–1.34, p = 0.007). Intraoperative blood loss also remained significant when modeled continuously (OR = 1.12 per 100-mL increase, 95% CI: 1.03–1.23, p = 0.011). In contrast, age did not show a significant association with SSI when treated as a continuous variable (OR = 1.04 per 10-year increase, 95% CI: 0.92–1.18, p = 0.512).
Model performance analysis demonstrated good calibration (Hosmer–Lemeshow test, p = 0.62). The logistic regression model showed acceptable discriminative ability, with an AUC of 0.81 (95% CI: 0.74–0.87).
Multicollinearity diagnostics
Multicollinearity diagnostics demonstrated that all variables had variance inflation factors (VIFs) ranging from 1.10 to 1.56, with tolerance values > 0.60. No variable exceeded the commonly accepted thresholds of VIF ≥ 5 or tolerance < 0.2. These findings indicated the absence of significant multicollinearity, confirming the suitability of the included variables for multivariate logistic regression analysis (Table 5).
Table 5.
Multicollinearity assessment of variables included in the logistic regression model
| Variable | Tolerance | VIF |
|---|---|---|
| Age ≥ 60 years | 0.821 | 1.22 |
| Diabetes mellitus | 0.911 | 1.10 |
| Chronic liver disease | 0.864 | 1.16 |
| Hypoalbuminemia (< 35 g/L) | 0.755 | 1.32 |
| Operative duration ≥ 240 min | 0.693 | 1.44 |
| Intraoperative blood loss ≥ 500 mL | 0.668 | 1.50 |
| Blood transfusion | 0.641 | 1.56 |
| Preoperative biliary drainage | 0.902 | 1.11 |
| ICU admission | 0.873 | 1.15 |
Post-hoc power analysis
A post-hoc power analysis was performed to evaluate the adequacy of the sample size for detecting the independent associations identified in the multivariate logistic regression model. Four significant predictors, diabetes mellitus, hypoalbuminemia (< 35 g/L), operative duration ≥ 240 min, and preoperative/postoperative biliary drainage, were included in the weighted post-hoc power calculation. Each factor was assigned an equal weight, contributing proportionally to a total weight of 100%. The weighted post-hoc power for the combined model was 89.3%, exceeding the commonly accepted threshold of 80%, indicating that the study was sufficiently powered to detect the observed effect sizes of these key risk factors.
Discussion
In our study, the overall incidence of SSIs was 5.6%, with key independent risk factors identified through multivariate analysis, including diabetes mellitus, hypoalbuminemia, prolonged operative duration, and postoperative biliary drainage. The predominant pathogens in our cohort were Gram-negative bacilli, specifically Escherichia coli and Klebsiella pneumoniae, with notable resistance profiles, including ESBL production and carbapenem resistance. The novel aspect of our study lies in the focused analysis of both clinical risk factors and the local microbiological landscape, highlighting the critical role of multidrug-resistant organisms in SSIs following hepatobiliary surgery. While previous studies have emphasized clinical risk factors, our study uniquely integrates microbiological surveillance, providing insights into pathogen resistance trends specific to our region. This combined approach enhances our understanding of the microbial challenges in managing SSIs and underscores the importance of tailored antibiotic prophylaxis based on local resistance patterns. Furthermore, by emphasizing the impact of hypoalbuminemia and prolonged operative duration on infection risk, our findings suggest targeted interventions to mitigate these factors, improving patient outcomes and reducing the burden of SSIs in hepatobiliary surgery. This study contributes to the growing need for localized epidemiological data, aiding in the development of more precise prevention strategies in hepatobiliary surgical practices.
The association between diabetes mellitus and increased SSI risk is biologically plausible. Hyperglycemia is known to impair neutrophil chemotaxis, phagocytosis, and oxidative killing, leading to reduced host immune defense. Furthermore, diabetes is often accompanied by microvascular changes and impaired wound healing, which together contribute to higher infection susceptibility. Our findings that diabetes approximately doubled the odds of SSI highlight the necessity of stringent perioperative glycemic control. Hypoalbuminemia was identified as the strongest predictor of SSI, with nearly a threefold increased risk. Serum albumin is widely recognized as an indicator of nutritional and immune status. Low albumin levels reflect protein-energy malnutrition, which compromises tissue repair and decreases resistance to infection [13, 14]. Moreover, hypoalbuminemia may also indicate underlying systemic inflammation or chronic disease burden. These results suggest that optimization of nutritional status and correction of hypoalbuminemia before surgery may be crucial in reducing postoperative infection risk. Prolonged operative duration was another significant independent factor. Lengthy surgeries increase the exposure of surgical wounds to potential contaminants and are often associated with greater intraoperative blood loss, tissue trauma, and hemodynamic instability. Each of these factors may facilitate bacterial invasion and impair immune defense. The association between longer operative duration and higher SSI risk in our study underscores the importance of improving surgical efficiency and minimizing unnecessary intraoperative delays [15, 16].
Preoperative biliary drainage was also independently associated with SSIs. This may be explained by the colonization of bile ducts and drainage catheters by multidrug-resistant organisms, which can subsequently contaminate the surgical field. The need for postoperative biliary drainage usually reflects advanced biliary obstruction or severe cholestasis, conditions that are themselves associated with impaired liver function and reduced systemic immunity. This finding emphasizes the need for careful consideration of drainage indications and rigorous infection prevention protocols in patients requiring biliary decompression [17, 18]. Microbiological analysis revealed that Gram-negative bacilli were the predominant pathogens, consistent with the enteric origin of many infections following hepatobiliary surgery. Notably, more than 40% of E. coli and K. pneumoniae isolates were ESBL producers, and a subset exhibited carbapenem resistance. The detection of multidrug-resistant Acinetobacter baumannii and methicillin-resistant Staphylococcus aureus (MRSA) further highlights the challenge of antimicrobial resistance in surgical infections [19, 20]. These findings suggest that empirical antimicrobial therapy for suspected SSIs after hepatobiliary surgery should consider local resistance patterns, and culture-directed therapy remains essential to optimize treatment efficacy while reducing selective pressure for further resistance. The significantly longer hospital stay observed among SSI patients reflects both the clinical burden of infection and the increased use of healthcare resources. Preventing SSIs through improved perioperative risk stratification and targeted interventions could therefore not only improve patient outcomes but also reduce the economic impact on healthcare systems [21, 22].
SSIs are a significant postoperative complication, with several risk factors identified in this study. Hypoalbuminemia (< 35 g/L) emerged as a strong independent predictor of SSI, consistent with existing literature linking low serum albumin to poor wound healing and impaired immune function [23, 24]. Hypoalbuminemia is often associated with malnutrition and systemic inflammation, both of which hinder an effective immune response. Reduced albumin levels limit the availability of proteins needed for tissue repair, and a hypoalbuminemic state can increase capillary permeability, further enhancing susceptibility to infection [25]. Similarly, prolonged operative duration (≥ 240 min) was identified as a key risk factor for SSIs. Extended surgeries are associated with greater tissue trauma, prolonged exposure to potential contaminants, and increased need for perioperative interventions such as blood transfusions, all of which contribute to a sustained inflammatory response that may promote bacterial colonization at the surgical site [26]. Preoperative or postoperative biliary drainage was also linked to higher SSI risk [27]. Although essential for managing biliary leakage and preventing complications, biliary drainage provides a pathway for bacteria to access the surgical site, particularly in immunocompromised patients [28]. These findings highlight the importance of managing risk factors such as hypoalbuminemia, operative duration, and biliary drainage to reduce the risk of SSIs. Understanding the biological mechanisms behind these associations will inform the development of targeted interventions to improve outcomes in hepatobiliary surgery. In light of the identified risk factors, we propose a protocol to reduce post-operative SSIs based on our findings. Glycemic control should be prioritized for high-risk patients, particularly those with diabetes mellitus, with blood glucose levels maintained below 140 mg/dL. Nutritional support, including correction of hypoalbuminemia with albumin levels targeted above 35 g/L, should be considered, incorporating preoperative screening and the use of nutritional supplements or albumin infusions as necessary. Efforts to minimize operative duration through efficient surgical planning should be emphasized, as prolonged procedures are associated with an increased risk of SSIs [29]. For patients requiring biliary drainage, it should be used selectively and removed as soon as clinically appropriate to reduce infection risk [30]. Furthermore, strict aseptic techniques during surgery and antibiotic prophylaxis tailored to local microbial resistance profiles are essential. This protocol addresses the most common risk factors, providing a structured approach to minimizing SSIs in hepatobiliary surgeries [31].
Bone et al. [7]’s study synthesizes multiple meta-analyses, identifying factors like open surgical approach, older age, and preoperative biliary drainage as significant risk factors for SSIs after hepatobiliary surgery. The cumulative SSI incidence in the HPB cohort was 11%, with substantial variation between reviews. Our study aligns with these findings, particularly regarding preoperative biliary drainage and older age as risk factors for SSI. However, our cohort shows a lower overall SSI incidence (5.6%) and highlights multidrug-resistant Gram-negative organisms as the predominant pathogens, which further underscores the need for localized microbiological surveillance. Mentor et al. [32]’s study quantifies SSI rates and risk factors for liver and pancreatic resections, emphasizing biliary resection, low albumin, and blood loss as key contributors to SSI. Our study supports these findings, notably low albumin and blood loss as independent risk factors, but we extend the knowledge by specifically identifying Gram-negative pathogens, such as Escherichia coli and Klebsiella pneumoniae, with significant antimicrobial resistance. This focus on pathogen distribution and antimicrobial resistance offers a deeper understanding of the microbial challenges in preventing SSIs. Shen et al. [3]’s study investigates SSI risk factors after hepatectomy for hepatocellular carcinoma, highlighting obesity, diabetes, and portal hypertension as significant risk factors. It also introduces a nomogram to predict SSI risk. While our study also identifies diabetes and operative duration as key factors, we provide a broader scope by focusing on nutritional factors like hypoalbuminemia, as well as the impact of prolonged operative duration and biliary drainage. This detailed analysis of nutritional status and operative time provides new insights into mitigating SSI risks in hepatobiliary surgery.
This study provides several important clinical implications. First, stringent perioperative glycemic control and nutritional optimization, including correction of hypoalbuminemia, should be prioritized in high-risk patients. Second, efforts to reduce operative time without compromising surgical quality may reduce SSI incidence. Third, careful indication, strict aseptic technique, and timely removal of biliary drainage should be emphasized to minimize infection risk. Finally, knowledge of local microbial distribution and resistance profiles is essential for guiding empirical antibiotic selection in suspected SSIs. The present study has several strengths. It included a relatively large cohort of hepatobiliary surgical patients, allowing for robust statistical analysis. Comprehensive clinical, surgical, and microbiological data were collected, enabling an integrated assessment of risk factors and pathogen profiles. Furthermore, the study applied rigorous multivariable regression analysis and assessed multicollinearity to ensure the reliability of independent risk factor identification. This study has several limitations. First, the retrospective, single-center design may limit the generalizability of our findings, and potential unmeasured confounding factors cannot be excluded. While the study reflects local practice patterns, it may not be directly applicable to other institutions with different protocols. However, the standardized surgical and infection-prevention protocols at our institution allowed for consistency in patient management and a detailed examination of local epidemiological trends. Second, culture-negative SSIs, which rely on clinical suspicion and imaging for diagnosis, were not included. This exclusion may underestimate the true incidence of SSIs, as cases without identifiable pathogens but with suspected infection were not captured. Third, although microbiological data were extensively collected, the absence of molecular typing of resistant strains limits our ability to identify specific clonal lineages and transmission dynamics. The small number of isolates further restricts the generalizability of antimicrobial resistance findings. Future research in larger, multicenter cohorts incorporating molecular epidemiology techniques such as whole-genome sequencing would provide deeper insights into the microbial landscape of SSIs in hepatobiliary surgery. Additionally, factors such as surgeon experience, intraoperative temperature control, and postoperative wound care, which may influence SSI risk, were not fully captured. The inclusion of these factors in future studies would provide a more complete understanding of perioperative risk factors for SSIs. Expanding the cohort to include multicenter data, implementing molecular typing, and considering culture-negative SSIs would improve our understanding of microbiological etiology. Furthermore, exploring preoperative interventions like albumin correction and antimicrobial stewardship protocols could help reduce infection rates and improve patient outcomes.
Conclusions
In conclusion, SSIs following hepatobiliary surgery were predominantly caused by multidrug-resistant Gram-negative organisms, particularly E. coli and K. pneumoniae. Independent risk factors included diabetes mellitus, hypoalbuminemia, prolonged operative duration, and postoperative biliary drainage. These findings emphasize the importance of perioperative optimization and targeted infection control strategies to reduce postoperative morbidity and antimicrobial resistance burden. To enhance patient outcomes, it is essential to implement proactive measures such as preoperative albumin correction, stringent antibiotic stewardship protocols, and optimized surgical planning to minimize infection risks and resistance patterns.
Acknowledgements
None.
Author contributions
Fan Wu, Hong Chen, and Qiu-Feng Han contributed to the conceptualization and methodology of the study. Data curation and formal analysis were performed by Hong Chen and Qiu-Feng Han. Resources and software support were also provided by Hong Chen and Qiu-Feng Han. Hong Chen drafted the original manuscript, while Fan Wu and Hong Chen participated in reviewing and editing the final version.
Funding
None.
Data availability
The datasets generated or analyzed during the current 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 First Affiliated Hospital of Fujian Medical University. All procedures involving human participants were conducted in accordance with the ethical standards of the institutional and/or national research committee and the Declaration of Helsinki (1964) and its later amendments or equivalent guidelines. Informed consent was obtained from all participants or their legal guardians.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Hong Chen, Email: hongchen011116@outlook.com.
Fan Wu, Email: FanWu623@hotmail.com.
References
- 1.Furukawa K, Onda S, Taniai T, Hamura R, Yanagaki M, Tsunematsu M, Haruki K, Yasuda J, Sakamoto T, Gocho T, et al. Risk factors and overcoming strategies of surgical site infection after hepatectomy for colorectal liver metastases. Anticancer Res. 2021;41(11):5651–6. [DOI] [PubMed] [Google Scholar]
- 2.Zhang J, Zhang R, Chen W, Yang Q, Zheng S, Chen C. Risk stratification for postoperative infection following Laparoscopic-to-Open cholecystectomy conversion: construction and evaluation of a clinical prediction model. Ann Ital Chir. 2025;96(6):742–9. [DOI] [PubMed] [Google Scholar]
- 3.Shen Y, Hu YL, Xu JH, Zhu S, Cai L, Wu YF, Wu XC, Zeng YY, Gu WM, Zhou YH, et al. Incidence, risk factors, outcomes, and prediction model of surgical site infection after hepatectomy for hepatocellular carcinoma: A multicenter cohort study. Eur J Surg Oncol. 2025;51(2):109486. [DOI] [PubMed] [Google Scholar]
- 4.Zuo JH, Che XY, Tan BB, Jiang Y, Bai J, Li XL, Yang YS, Pang SJ, Liu XC, Fan HN, et al. Association between Pre-operative body mass index and surgical infection in Perihilar cholangiocarcinoma patients treated with curative resection: A Multi-center study. Surg Infect (Larchmt). 2024;25(6):444–51. [DOI] [PubMed] [Google Scholar]
- 5.Zhou YD, Zhang WY, Xie GH, Ye H, Chu LH, Guo YQ, Lou Y, Fang XM. Inadvertent perioperative hypothermia and surgical site infections after liver resection. Hepatobiliary Pancreat Dis Int. 2024;23(6):579–85. [DOI] [PubMed] [Google Scholar]
- 6.Im H, Oh SY, Lim L, Lee H, Kwon J, Ryu HG. Timing of prophylactic antibiotics administration and suspected systemic infection after percutaneous biliary intervention. J Hepatobiliary Pancreat Sci. 2024;31(1):34–41. [DOI] [PubMed] [Google Scholar]
- 7.Bone M, Latimer S, Walker RM, Thalib L, Gillespie BM. Risk factors for surgical site infections following hepatobiliary surgery: an umbrella review and meta-analyses. Eur J Surg Oncol. 2025;51(1):109468. [DOI] [PubMed] [Google Scholar]
- 8.Araki K, Harimoto N, Watanabe A, Tsukagoshi M, Ishii N, Hagiwara K, Muranushi R, Hoshino K, Seki T, Shirabe K. Laparoscopic liver resection procedure attenuates Organ-space surgical site infection compared with open procedure: A propensity Score-matched analysis. Anticancer Res. 2023;43(5):2273–80. [DOI] [PubMed] [Google Scholar]
- 9.Erinmez M, Birgin H, Yılmaz L, Zer Y. Microbiological findings and risk profiles in hepatobiliary and pancreatic surgery associated surgical site infections: A retrospective cohort study. Pathogens. 2025;14(12):1215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Salam MA, Al-Amin MY, Salam MT, Pawar JS, Akhter N, Rabaan AA, Alqumber MAA. Antimicrobial resistance: a growing serious threat for global public health. Healthc (Basel) 2023;11(13). [DOI] [PMC free article] [PubMed]
- 11.Shah S, Singhal T, Naik R, Thakkar P. Predominance of multidrug-resistant Gram-negative organisms as cause of surgical site infections at a private tertiary care hospital in Mumbai, India. Indian J Med Microbiol. 2020;38(3):344–50. [DOI] [PubMed] [Google Scholar]
- 12.Muteeb G, Rehman MT, Shahwan M, Aatif M. Origin of antibiotics and antibiotic resistance, and their impacts on drug development: a narrative review. Pharmaceuticals (Basel) 2023;16(11). [DOI] [PMC free article] [PubMed]
- 13.Moraes CMT, Corrêa LM, Procópio RJ, Carmo G, Navarro TP. Tools and scores for perioperative pulmonary, renal, hepatobiliary, hematological, and surgical site infection risk assessment: an update. Rev Col Bras Cir. 2022;49:e20223125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yasukawa K, Umemura K, Shimizu A, Kubota K, Notake T, Hosoda K, Hayashi H, Soejima Y. Impact of large amount of intra-abdominal lavage on surveillance of surgical site infection after hepato-pancreato-biliary surgery: A prospective cohort study. J Hepatobiliary Pancreat Sci. 2023;30(6):705–13. [DOI] [PubMed] [Google Scholar]
- 15.Sugimoto M, Gotohda N, Kudo M, Kobayashi S, Takahashi S, Konishi M. Laparoscopic liver resection can be performed safely without intraoperative drain placement. Surg Endosc. 2022;36(12):9019–31. [DOI] [PubMed] [Google Scholar]
- 16.Cammann S, Karabulut S, DeTemple DE, Oldhafer F, Kulik U, Schroeter A, Vondran FWR, Klempnauer J, Kleine M, Timrott K, et al. Antibiotic-Resistant bacteria colonizing the bile duct are associated with increased morbidity and mortality after resection of extrahepatic cholangiocarcinoma. Surg Infect (Larchmt). 2022;23(3):270–9. [DOI] [PubMed] [Google Scholar]
- 17.Gyoten K, Kato H, Hayasaki A, Fujii T, Iizawa Y, Murata Y, Tanemura A, Kuriyama N, Kishiwada M, Mizuno S, et al. Association between gastric Candida colonization and surgical site infections after high-level hepatobiliary pancreatic surgeries: the results of prospective observational study. Langenbecks Arch Surg. 2021;406(1):109–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wu J, Han J, Zhang Y, Liang L, Zhao J, Han F, Dou C, Zhang Y, Liu J, Wu W, et al. Safety and feasibility of laparoscopic versus open liver resection with associated lymphadenectomy for intrahepatic cholangiocarcinoma. Biosci Trends. 2020;14(5):376–83. [DOI] [PubMed] [Google Scholar]
- 19.Shen J, Ni Z, Qian Y, Wang B, Zheng S. Clinical prediction score for superficial surgical site infections: Real-life data from a retrospective single-centre analysis of 812 hepatectomies. Int Wound J. 2020;17(1):16–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Rungsakulkij N, Vassanasiri W, Tangtawee P, Suragul W, Muangkaew P, Mingphruedhi S, Aeesoa S. Preoperative serum albumin is associated with intra-abdominal infection following major hepatectomy. J Hepatobiliary Pancreat Sci. 2019;26(11):479–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hu B, Tan HY, Rao XW, Jiang JY, Yang K. A scoring system for surgical site infection after pancreaticoduodenectomy using clinical data. Surg Infect (Larchmt). 2021;22(2):240–4. [DOI] [PubMed] [Google Scholar]
- 22.Shah NJ, Leis A, Kheterpal S, Englesbe MJ, Kumar SS. Association of intraoperative hyperglycemia and postoperative outcomes in patients undergoing non-cardiac surgery: a multicenter retrospective study. BMC Anesthesiol. 2020;20(1):106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.He Z, Zhou K, Tang K, Quan Z, Liu S, Su B. Perioperative hypoalbuminemia is a risk factor for wound complications following posterior lumbar interbody fusion. J Orthop Surg Res. 2020;15(1):538. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Chen C, Zhang Y, Zhao X, Tao M, Yan W, Fu Y. Hypoalbuminemia - An indicator of the severity and prognosis of COVID-19 patients: A multicentre retrospective analysis. Infect Drug Resist. 2021;14:3699–710. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Boselli M, Aquilani R, Maestri R, Iadarola P, Magistroni A, Ferretti C, Pierobon A, Cotta Ramusino M, Costa A, Buonocore D, et al. Essential amino acid supplementation May attenuate systemic inflammation and improve hypoalbuminemia in subacute hemiplegic stroke patients. Metabolites. 2025;15(9):626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Monetta A, Griffoni C, Falzetti L, Evangelisti G, Noli LE, Tedesco G, Cavallari C, Bandiera S, Terzi S, Ghermandi R, et al. Prolonged operative time significantly impacts on the incidence of complications in spinal surgery. J Orthop Surg Res. 2024;19(1):567. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Rayzah M. Risk factors and clinical outcomes of deep surgical site infections in trauma patients: A National database analysis. Healthcare. 2025;13(15):1808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Xiao Y, Cai HQ. Clinical management and therapeutic strategies for biliary leakage after liver transplantation. World J Gastrointest Surg. 2025;17(9):108275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tfaily MA, Ghanem P, Farran SH, Dabdoub F, Kanafani ZA. The role of preoperative albumin and white blood cell count in surgical site infections following Whipple surgery. Sci Rep. 2022;12(1):19184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Chambers LE, Sheen AJ, Whitehead KA. A systematic review on the incidence and risk factors of surgical site infections following hepatopancreatobiliary (HPB) surgery. AIMS Bioeng. 2022;9(2):123–44. [Google Scholar]
- 31.Marzoug OA, Anees A, Malik EM. Assessment of risk factors associated with surgical site infection following abdominal surgery: a systematic review. BMJ Surg Interv Health Technol. 2023;5(1):e000182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Mentor K, Ratnayake B, Akter N, Alessandri G, Sen G, French JJ, Manas DM, Hammond JS, Pandanaboyana S. Meta-Analysis and Meta-Regression of risk factors for surgical site infections in hepatic and pancreatic resection. World J Surg. 2020;44(12):4221–30. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated or analyzed during the current study are available from the corresponding author upon reasonable request.