Evaluation of preoperative nutritional variables to predict postoperative complications after pancreaticoduodenectomy

Postoperative prognostic nutrition index predicts survival in patients with small bowel adenocarcinoma after surgical resection

BACKGROUND

Surgical resection (SR) is the main treatment for small bowel adenocarcinoma (SBA), but it increases metabolic demand, systemic inflammation, and digestive dysfunction, resulting in major impacts on the postoperative outcomes of patients. In this study, we aimed to investigate the role of the postoperative prognostic nutritional index (PNI), a surrogate marker of inflammation and nutrition, in patients with SBA after resection.

METHODS

From June 2014 to March 2022, 44 consecutive patients who underwent SR for SBA in Taipei Veterans General Hospital were retrospectively reviewed. Factors associated with survival including PNI were analyzed.

RESULTS

PNI decreased in patients after SR for SBA (median change: -1.82), particularly in those who underwent Whipple operation or developed postoperative pancreatic fistula. Postoperative PNI <45.2 best predicted overall survival (OS) (area under the receiver operating characteristic curve [AUROC]: 0.826, p = 0.001). Patients with lower postoperative PNI had significantly worse OS compared to those with higher postoperative values (median OS: 19.3 months vs not reached, p < 0.001). Low postoperative PNI (hazard ratio [HR]: 11.404, p = 0.002), tumoral lymphovascular invasion (HR: 8.023, p = 0.012), and adjuvant chemotherapy (HR: 0.055, p = 0.002) were independent risk factors for OS. Postoperative PNI also significantly predicted recurrence-free survival independent of lymphovascular invasion and adjuvant chemotherapy (HR: 6.705, p = 0.001).

CONCLUSION

PNI commonly decreases in patients with SBA who undergo Whipple surgery or develop postoperative pancreatic fistula. Postoperative PNI independently predicts survival and may serve as a clinical marker to optimize patient outcomes.

The impact of preoperative nutritional status on postoperative outcomes: an insight from Geriatric Nutritional Risk Index in elderly pancreaticoduodenectomy patients

BACKGROUND

Malnutrition is not uncommon among the elderly undergoing pancreatoduodenectomy (PD) and is related to increased complications. Previous studies have shown that the Geriatric Nutritional Risk Index (GNRI) predicts outcomes in various populations. Nevertheless, the research exploring the correlation between GNRI and postoperative outcomes in PD is scarce. This study aimed to investigate the preoperative malnutrition, as measured by GNRI, on outcomes in elderly patients undergoing PD.

MATERIALS AND METHODS

This retrospective analysis enrolled 144 elderly patients underwent PD for periampullary tumors from November 2016 to December 2021. Patients were stratified based on the GNRI value: high/moderate nutrition risk (GNRI ≤ 92, N = 54), low nutrition risk (92 < GNRI ≤ 98, N = 35), and no nutrition risk (GNRI > 98, N = 55). Perioperative outcomes and postoperative surgical complications were compared between these groups. Univariate and multivariate analyses were performed on major postoperative complications and prolonged postoperative length of stay (PLOS).

RESULTS

Patients in the high/moderate risk group were significantly older, with lower BMI (P = 0.012), higher mortality rate (11.1%, P = 0.024), longer PLOS (P < 0.001), and higher incidence of over grade IIIB complications (37.0%, P = 0.001), Univariate and multivariate analyses showed the high/moderate risk GNRI group (OR 3.61, P = 0.032), increased age (OR 1.11, P = 0.014) and operative time over 8 h (OR 3.04, P = 0.027) were significantly associated with increased major postoperative complications. The high/moderate risk GNRI group was also a significant predictor for prolonged PLOS (OR 3.91, P = 0.002).

CONCLUSIONS

Preoperative GNRI has the potential to be a predictive tool for identifying high-risk elderly patients and monitoring nutritional status preoperatively to improve postoperative surgical outcomes following PD.

Two Cases of Rapidly Progressive Fatty Liver Disease due to Pancreatic Exocrine Insufficiency without a History of Surgery

We herein report two cases of rapidly progressive fatty liver (FL) disease due to pancreatic exocrine insufficiency (PEI) without a surgical history. Two women, 59 and 72 years old, with no history of abdominal surgery presented to our hospital with severe anorexia and nausea persisting for one week. Examinations revealed progressive, marked FL disease with hepatomegaly and PEI, for which pancreatic enzyme replacement therapy was effective. Commonly known causes of PEI include chronic pancreatitis, abdominal surgery (e.g. pancreaticoduodenectomy), pancreatic cancer, and obstruction of the pancreatic duct, none of which were present in either of these two cases.

The Impact of Nutritional Markers and Dietary Habits on the Bioimpedance Phase Angle in Older Individuals

Low phase angle (PhA), as determined via bioelectrical impedance analysis, reflects unhealthy aging and mortality. In this study, we assessed whether nutritional status, including serum nutritional markers and dietary habits, is related to PhA in older individuals. We recruited 212 participants (aged ≥ 65 years) who underwent medical health checkups. PhA was measured using a multi-frequency impedance body composition analyzer. Habitual food and nutrient intake was evaluated using a brief, self-administered diet history questionnaire. Low PhA values were defined as ≤4.95 in males and ≤4.35 in females. Males with low PhA had poor exercise habits (p = 0.0429) and a lower body mass index (p = 0.0024). PhA was significantly correlated with serum cholinesterase levels, a nutritional status marker (r = 0.3313, p = 0.0004 in males; r = 0.3221, p = 0.0070 in females). The low-PhA group had significantly lower total energy and carbohydrate intake per ideal body weight (IBW) than the high-PhA group in males (total energy intake:30.2 ± 9.8 and 34.5 ± 9.3 kcal/kg/day, p = 0.0307; carbohydrate intake:15.2 ± 4.9 and 18.0 ± 5.8 kcal/kg/day, p = 0.0157). Total energy intake per IBW (adjusted odds ratio [95% confidence interval], 0.94 [0.89-1.00] per 1 kcal/kg/day increase) was independently associated with a low PhA in males. Our study revealed that lower total energy intake independently impacted low PhA in older males.

Physiological stress level and screening for malnutrition as preoperative predictors of postoperative complications in pancreatic surgery: a retrospective study

BACKGROUND

Assessment of 'physiological stress levels' and 'nutritional status' before surgery is important for predicting complications and indirect interventions on the pancreas. The aim of this study was to determine neutrophil-lymphocyte ratio (NLR) and nutritional risk index (NRI) indicators before surgery to predict 90-day complications and mortality in a cohort of patients with complicated chronic pancreatitis and cancer of the head of the pancreas.

METHODS

We evaluated preoperative levels of NLR and NRI among 225 subjects treated at different centres located in three countries. Short-term outcomes included length of hospital stay, postoperative complications, and mortality at 90 days and were appreciated based on NLR and NRI. The level of physiological stress was divided according by the formulas: neutrophil-lymphocyte ratio (NLR) = (neutrophil count, %)/(lymphocyte count, %). The nutritional state of the patients was divided according to the INR: NRI = (1.519 × serum albumin, g/L) + (41.7 × present weight, kg / usual weight, kg)].

RESULTS

All patients were operated. An analysis of the operations performed in three institutions demonstrated mortality in chronic pancreatitis and pancreatic pseudocysts in 1.4%, in chronic pancreatitis and the presence of an inflammatory mass mainly in the pancreatic head in 1.2%, and in cancer of the pancreatic head in 5.9%. The mean preoperative NLR was normal in 33.8% of the patients, the mild physiologic stress level was 54.7%, and the moderate was 11.5% before surgery. 10.2% of patients had a normal nutritional status, 20% had mild, 19.6% had moderate, and 50.2% had severe malnutrition. In a univariate analysis, at the cutoff of NLR ≥ 9.5 (AUC = 0.803) and the cutoff of NRI ≤ 98.5 (AUC = 0.801), increasing the risk of complications was observed (hazard ratio, 2.01; 95% CI, 1.247-3.250, p = 0.006), but at the cutoff of NRI ≤ 83.55 (AUC = 0.81), we observed a survival difference in operated patients (hazard ratio, 2.15; 95% CI, 1.334-3.477, p = 0.0025).

CONCLUSIONS

Our study demonstrated that NLR and NRI were predictors of postoperative complications, but only NRI was a predictor of 90-day mortality in patients after surgery.

Machine Learning for the Prediction of Complications in Patients After Mitral Valve Surgery

Background: This study intended to use a machine learning model to identify critical preoperative and intraoperative variables and predict the risk of several severe complications (myocardial infarction, stroke, renal failure, and hospital mortality) after cardiac valvular surgery.

Study Design and Methods: A total of 1,488 patients undergoing cardiac valvular surgery in eight large tertiary hospitals in China were examined. Fifty-four perioperative variables, such as essential demographic characteristics, concomitant disease, preoperative laboratory indicators, operation type, and intraoperative information, were collected. Machine learning models were developed and validated by 10-fold cross-validation. In each fold, Recursive Feature Elimination was used to select key variables. Ten machine learning models and logistic regression were developed. The area under the receiver operating characteristic (AUROC), accuracy (ACC), Youden index, sensitivity, specificity, F1-score, positive predictive value (PPV), and negative predictive value (NPV) were used to compare the prediction performance of different models. The SHapley Additive ex Planations package was applied to interpret the best machine learning model. Finally, a model was trained on the whole dataset with the merged key variables, and a web tool was created for clinicians to use.

Results: In this study, 14 vital variables, namely, intraoperative total input, intraoperative blood loss, intraoperative colloid bolus, Classification of New York Heart Association (NYHA) heart function, preoperative hemoglobin (Hb), preoperative platelet (PLT), age, preoperative fibrinogen (FIB), intraoperative minimum red blood cell volume (Hct), body mass index (BMI), creatinine, preoperative Hct, intraoperative minimum Hb, and intraoperative autologous blood, were finally selected. The eXtreme Gradient Boosting algorithms (XGBOOST) algorithm model presented a significantly better predictive performance (AUROC: 0.90) than the other models (ACC: 81%, Youden index: 70%, sensitivity: 89%, specificity: 81%, F1-score:0.26, PPV: 15%, and NPV: 99%).

Conclusion: A model for predicting several severe complications after cardiac valvular surgery was successfully developed using a machine learning algorithm based on 14 perioperative variables, which could guide clinical physicians to take appropriate preventive measures and diminish the complications for patients at high risk.