Prognostic nutritional index and early mortality with percutaneous endoscopic gastrostomy

Explainable Machine Learning Models for Prediction of Surgical Site Infection After Posterior Lumbar Fusion Surgery Based on Shapley Additive Explanations

OBJECTIVE

This study aims to develop machine learning (ML) models combined with an explainable method for the prediction of surgical site infection (SSI) after posterior lumbar fusion surgery.

METHODS

In this retrospective, single-center study, a total of 1016 consecutive patients who underwent posterior lumbar fusion surgery were included. A comprehensive dataset was established, encompassing demographic variables, comorbidities, preoperative evaluation, details related to diagnosed lumbar disease, preoperative laboratory tests, surgical specifics, and postoperative factors. Utilizing this dataset, 6nullML models were developed to predict the occurrence of SSI. Performance evaluation of the models on the testing set involved several metrics, including the receiver operating characteristic curve, the area under the receiver operating characteristic curve, accuracy, recall, F1 score, and precision. The Shapley Additive Explanations (SHAP) method was employed to generate interpretable predictions, enabling a comprehensive assessment of SSI risk and providing individualized interpretations of the model results.

RESULTS

Among the 1016 retrospective cases included in the study, 36 (3.54%) experienced SSI. Out of the six models examined, the Extreme Gradient Boost model demonstrated the highest discriminatory performance on the testing set, achieving the following metrics: precision (0.9000), recall (0.8182), accuracy (0.9902), F1 score (0.8571), and area under the receiver operating characteristic curve (0.9447). By utilizing the SHAP method, several important predictors of SSI were identified, including the duration of indwelling jugular vein catheter, blood urea nitrogen levels, total protein levels, sustained fever, creatinine levels, triglycerides levels, monocyte count, diabetes mellitus, drainage time, white blood cell count, cerebral infarction, estimated blood loss, prealbumin levels, Prognostic Nutritional Index, low back pain, posterior fusion score, and osteoporosis.

CONCLUSIONS

ML-based prediction tools can accurately assess the risk of SSI after posterior lumbar fusion surgery. Additionally, ML combined with SHAP could provide a clear interpretation of individualized risk prediction and give physicians an intuitive comprehension of the effects of the model's essential features.

One-Year Mortality After Percutaneous Endoscopic Gastrostomy: The Prognostic Role of Nutritional Biomarkers and Care Settings

Background/Objectives: This study aimed to evaluate the clinical outcomes, complications, and one-year mortality of patients undergoing percutaneous endoscopic gastrostomy (PEG) in different care settings (hospital, nursing home, and home). Additionally, we investigated the comparative prognostic role of the prognostic nutritional index (PNI) and the CRP-to-albumin ratio (CAR) in predicting mortality among these patients. Methods: A retrospective analysis of 236 adult patients who underwent PEG placement between January 2022 and December 2023 was performed. Demographic, clinical, and laboratory data were collected. The PNI was calculated according to the following formula: PNI = 10 × (albumin) + 0.005 × (lymphocyte count). The CAR was obtained by the ratio of the CRP level to the albumin level. Patients were categorized based on their post-PEG care settings. Results: Neurologic disorders were the most common indication for PEG (69.9%). The one-year mortality was 32.2%, with a median survival of 38 weeks (95% CI: 35-41). In the multivariable model, a lower PNI (HR = 0.93, 95% CI: 0.89-0.97, p < 0.001), as well as being followed in a hospital setting, emerged as independent predictors of mortality. Patients with timely PEG tube replacement showed a reduced mortality risk. The ROC analysis showed that the PNI had a higher AUROC (0.78 ± 0.04) compared to the CAR (0.69 ± 0.04), indicating superior prognostic accuracy for predicting one-year mortality. Conclusions: Care settings significantly influence survival outcomes, with better mortality rates observed in nursing homes and home environments. The PNI was superior to the CAR in predicting one-year mortality, emphasizing its clinical utility in risk stratification for PEG patients. Proactive tube management and individualized care strategies are critical for improving the prognosis in this population.

Complications and survival following percutaneous endoscopic gastrostomy tube placement

INTRODUCTION

percutaneous endoscopic gastrostomy (PEG) is considered as a safe and effective method for nutritional support in patients with malnutrition and swallowing impairment with an estimated survival of over two months. Some indications, such as advanced cognitive decline, contraindicate the technique.

MATERIALS AND METHODS

all patients who underwent PEG placement between January 2001 and May 2019 were included. Clinical data, indication, complications, and mortality were retrospectively analyzed.

RESULTS

a total of 648 patients (46.5 % male, mean age 70 ± 18.5 years) were included. The most common indications for PEG were advanced cognitive decline (31.5 %) and cerebrovascular disease (18.8 %). The mean follow-up was 12.07 months (IQR 3.27-34.73); 39.5 % of patients experienced complications (systemic 17.9 %, local 28.5 %). The most frequent were bronchoaspiration (9.7 %) and rupture/dysfunction (13.9 %), respectively. The presence of early complications (HR 1.63 [1.20-2.21]) and age (HR 1.02 [1.01-1.02]) was associated with shorter survival time, while female sex was a protective factor (HR 0.78 [0.66-0.94]).

CONCLUSIONS

PEG is not without complications, with 39.5 % of patients experiencing them. Patients with early complications, male sex and older age have lower survival following PEG placement.

Nutritional and prognostic significance of abdominal wall thickness measured during percutaneous endoscopic gastrostomy in older individuals with dysphagia

BACKGROUND & AIMS

The significance of abdominal wall thickness (AWT) for nutritional assessment remains unclear. This study aimed to evaluate the nutritional and prognostic significance of AWT measured during percutaneous endoscopic gastrostomy (PEG) in older patients with dysphagia.

METHODS

This single-center retrospective cohort study enrolled older adults with dysphagia who underwent PEG for enteral nutrition using the introducer technique between February 2010 and January 2019. We examined the association between AWT measured during PEG and nutritional status items, including body mass index (BMI), serum albumin (Alb), total lymphocyte count (TLC), total cholesterol (TC), hemoglobin (Hb), and C-reactive protein (CRP). The shaft length of the PEG tube inserted, which is an approximation value of the AWT, was used for statistical analysis. Patients were divided into three groups: low-AWT group (shaft length ≤2.5 cm), medium-AWT group (shaft length 3.0-3.5 cm), and high-AWT group (shaft length ≥4.0 cm). We performed the Kruskal-Wallis test and multiple linear regression analysis with multiple imputation. We performed survival analysis using the Kaplan-Meier method, log-rank test, and Cox regression analysis.

RESULTS

A total of 520 patients were identified: 115, 258, and 147 patients in the low-AWT, medium-AWT, and high-AWT groups, respectively. Higher AWT was significantly associated with higher BMI, Alb, TLC, TC, Hb, and lower CRP levels. Multiple linear regression analysis revealed that BMI and TLC were significant predictors of AWT (BMI: coefficient 1.16E-01, 95% confidence interval [CI], 9.77E-02-1.33E-01, P < 0.001; TLC: coefficient 1.18E-04, 95% CI, 2.72E-05-2.09E-04, P = 0.011). The median survival time was the longest in the high-AWT group (low-AWT, 359 days; medium-AWT, 851 days; high-AWT, 1662 days; P < 0.001). The hazard ratio for the high-AWT group relative to the medium-AWT group was 0.59 (95% CI, 0.41-0.85, P = 0.004), and that relative to the low-AWT group was 0.34 (95% CI, 0.24-0.51, P < 0.001).

CONCLUSIONS

Higher AWT was associated with better nutritional status and survival. AWT may help assess nutritional status and predict survival in older dysphagic patients with PEG.

Complications in patients with neurological impairment after gastrostomy

BACKGROUND

Neurological impairment (NI) is responsible for most conditions that require a permanent gastrostomy tube. The present study assessed the occurrence of short- and long-term complications after video-assisted gastrostomy (VAG) in patients with NI.

METHODS

The incidence of short- (<6 months) and long-term (over 2 years) complications of VAG were analyzed in a retrospective study. The differences between the incidence of the complications of VAG according to the age at surgery (≤15 years vs. ≥16 years) were also evaluated. The short- and long-term complications observed were granulation tissue formation, infection requiring antibiotic treatment, skin problems, perigastrostomy leakage, vomiting, accidental tube dislodgement, dumping syndrome, ileus, and peritonitis.

RESULTS

Eighty-two patients were evaluated for short- and long-term complications. The long-term complication rate was significantly lower than the short-term complication rate (P = 0.0026). Onodera's prognostic nutritional index before VAG in patients with long-term complications was significantly lower than in patients without such complications (P = 0.046). The incidence of long-term granulation tissue formation, infection, and vomiting were significantly lower than those of similar short-term complications. Long-term skin problems were associated with short-term skin problems (odds ratio: 18.95; 95% confidence interval: 4.53-92.98; P < 0.001). The number of patients ≥16 years old with short- and long-term skin problems was significantly higher than in patients ≤15 years old (P = 0.0014 and P = 0.0073, respectively).

CONCLUSIONS

The incidence rate of granulation tissue formation and infection after VAG were lower in the long term than in the short term. However, patients ≥16 years old presented with persistent complications.

Analysis of frequency and risk factors for complications of enteral nutrition in children in Poland after percutaneous endoscopic gastrostomy placement

OBJECTIVES

The aim of the study was to assess the complication rate and identify whether age, nutritional status, and history of respiratory aspiration prior to percutaneous endoscopic gastrostomy (PEG) are risk factors for post-PEG placement complications in Polish children. In addition, the safety of two enteral feeding methods (3 h vs. 8 h) after PEG insertion in children was compared.

METHODS

Children with clinical indications for PEG placement were recruited from six medical centers in Poland to participate in the study. The patients were centrally randomized to receive the first bolus feed via a feeding tube at 3 h (group 1) or 8 h (group 2) after PEG placement. The preprocedural preparation, postoperative care, and resumption of feeding were performed on all of patients in accordance with the study protocol. Patients were followed for 12 mo.

RESULTS

Of the 97 randomized patients, 49 were assigned to group 1 and 48 to group 2. Full feed after PEG placement was achieved within 24 to 48 h in most cases (74% vs. 82%). There were no differences between the groups regarding the number of early mild (31.3% vs. 31.3%) and serious (2.1% vs 8.3%) complications or the duration of hospitalization after PEG placement (P > 0.05). The most common serious complication after PEG placement was accidental displacement of PEG. Most reported late complications were mild. The results of the regression analysis indicate no statistically significant effect of age, body mass index standard deviation score, white blood cell count, serum albumin level, and respiratory aspiration in the medical history on the occurrence of mild and severe complications.

CONCLUSIONS

The early initiation of post-PEG feeding was not associated with an increase in the number of complications. Most complications after the PEG procedure were mild. Age, serum albumin level, white blood cells, body mass index standard deviation score, and a history of aspiration to the respiratory tract were not confirmed as a risk factor for post-PEG complications in children.