Assessment of systemic immune-inflammation index in predicting postoperative pulmonary complications in patients undergoing lung cancer resection

Multicentre validation of a machine learning model for predicting respiratory failure after noncardiac surgery

BACKGROUND

Postoperative respiratory failure is a serious complication that could benefit from early accurate identification of high-risk patients. We developed and validated a machine learning model to predict postoperative respiratory failure, defined as prolonged (>48 h) mechanical ventilation or reintubation after surgery.

METHODS

Easily extractable electronic health record (EHR) variables that do not require subjective assessment by clinicians were used. From EHR data of 307,333 noncardiac surgical cases, the model, trained with a gradient boosting algorithm, utilised a derivation cohort of 99,025 cases from Seoul National University Hospital (2013-9). External validation was performed using three separate cohorts A-C from different hospitals comprising 208,308 cases. Model performance was assessed by area under the receiver operating characteristic (AUROC) curve and area under the precision-recall curve (AUPRC), a measure of sensitivity and precision at different thresholds.

RESULTS

The model included eight variables: serum albumin, age, duration of anaesthesia, serum glucose, prothrombin time, serum creatinine, white blood cell count, and body mass index. Internally, the model achieved an AUROC of 0.912 (95% confidence interval [CI], 0.908-0.915) and AUPRC of 0.113. In external validation cohorts A, B, and C, the model achieved AUROCs of 0.879 (95% CI, 0.876-0.882), 0.872 (95% CI, 0.870-0.874), and 0.931 (95% CI, 0.925-0.936), and AUPRCs of 0.029, 0.083, and 0.124, respectively.

CONCLUSIONS

Utilising just eight easily extractable variables, this machine learning model demonstrated excellent discrimination in both internal and external validation for predicting postoperative respiratory failure. The model enables personalised risk stratification and facilitates data-driven clinical decision-making.

Relationship Between Preoperative Inflammation Ratios Derived From Preoperative Blood Cell Count and Postoperative Pulmonary Complications in Patients Undergoing Lobectomy: A Single-Center Observational Study

OBJECTIVE

Evaluation of the association of inflammatory cell ratios, especially neutrophil-to-lymphocyte ratio (NLR), based on preoperative complete blood counts, with postoperative complications in lobectomy surgery.

DESIGN

This was a retrospective monocentric cohort study.

SETTING

The study was conducted at Foch University Hospital in Suresnes, France.

PARTICIPANTS

Patients having undergone a scheduled lobectomy from January 2018 to September 2021.

INTERVENTIONS

There were no interventions.

MEASUREMENTS AND MAIN RESULTS

The authors studied 208 consecutive patients. Preoperative NLR, monocyte-to-lymphocyte ratio, platelet-to-lymphocyte ratio, systemic inflammation index, systemic inflammation response index, and aggregate inflammation systemic index were calculated. Median and (IQR) of NLR was 2.67 (1.92-3.69). No statistically significant association was observed between any index and the occurrence of at least one major postoperative complication, which occurred in 37% of the patients. Median postoperative length of stay was 7 (5-10) days. None of the ratios was associated with prolonged length of stay (LOS), defined as a LOS above the 75th percentile.

CONCLUSIONS

The results suggested that simple available inflammatory ratios are not useful for the preoperative identification of patients at risk of postoperative major complications in elective lobectomy surgery.

Association between preoperative systemic immune inflammation index and postoperative sepsis in patients with intestinal obstruction: A retrospective observational cohort study

BACKGROUND

Sepsis is a severe complication that results in increased morbidity and mortality after intestinal obstruction surgery. This study examined the role of preoperative systemic immune inflammation index (SII) for postoperative sepsis in intestinal obstruction patients.

METHODS

Data on patients who underwent intestinal obstruction surgery were collected. SII was determined and separated into two groups (≤1792.19 and >1792.19) according to the optimal cut-off value of SII for postoperative sepsis. The odds ratio (OR) is calculated for the correlation between SII and postoperative sepsis. Additional analyses were used to estimate the robustness of SII.

RESULTS

A total of 371 intestinal obstruction patients undergoing surgery were included in the final cohort, and 60 (16.17%) patients developed postoperative sepsis. Patients with an SII ＞1792.19 had a significantly higher risk for developing postoperative sepsis after multivariable adjustment [adjusted odds ratio = 2.12, 95% confidence interval: [1.02-4.40]]. The analysis of interaction showed no correlation between the preoperative SII and postoperative sepsis regarding age, hypertension, American Society of Anesthesiologists classification, blood loss, albumin, hemoglobin, creatinine, and leukocyte (all interactions p > .05). In subgroup analysis, all statistically significant subgroups showed that SII was a risk factor for postoperative sepsis (all p < .05). The analyses of subgroups and interactions revealed that the interaction effect of a preoperative SII ＞1792.19 and postoperative sepsis remained significant. A sensitivity analysis confirmed the robustness of the results.

CONCLUSIONS

A preoperative SII ＞ 1792.19 was a risk factor for postoperative sepsis in patients undergoing intestinal obstruction surgery.

A diagnostic model based on routine blood examination for serious bacterial infections in neonates-a cross-sectional study

Routine blood examination is an easy way to examine infectious diseases. This study is aimed to develop a model to diagnose serious bacterial infections (SBI) in ICU neonates based on routine blood parameters. This was a cross-sectional study, and data were extracted from the Medical Information Mart for Intensive Care III (MIMIC-III). SBI was defined as suffering from one of the following: pyelonephritis, bacteraemia, bacterial meningitis, sepsis, pneumonia, cellulitis, and osteomyelitis. Variables with statistical significance in the univariate logistic regression analysis and log systemic immune-inflammatory index (SII) were used to develop the model. The area under the curve (AUC) was calculated to assess the performance of the model. A total of 1,880 participants were finally included for analysis. Weight, haemoglobin, mean corpuscular volume, white blood cell, monocyte, premature delivery, and log SII were selected to develop the model. The developed model showed a good performance to diagnose SBI for ICU neonates, with an AUC of 0.812 (95% confidence interval (CI): 0.737-0.888). A nomogram was developed to make this model visualise. In conclusion, our model based on routine blood parameters performed well in the diagnosis of neonatal SBI, which may be helpful for clinicians to improve treatment recommendations.

Systemic immune-inflammation index within the first postoperative hour as a predictor of severe postoperative complications in upper abdominal surgery: a retrospective single-center study

Background

Systemic pro-inflammatory factors play a critical role in mediating severe postoperative complications (SPCs) in upper abdominal surgery (UAS). The systemic immune-inflammation index (SII) has been identified as a new inflammatory marker in many occasions. The present study aims to determine the association between SII and the occurrence of SPCs after UAS.

Methods

Included in this study were 310 patients with upper abdominal tumors who received UAS and subsequently were transferred to the anesthesia intensive care unit between November 2020 and November 2021 in Nanjing Drum Hospital. SPCs, including postoperative pulmonary complications (PPCs), major adverse cardiac and cardiovascular events, postoperative infections and delirium, were recorded during the hospital stay. The clinical features of the patients with and without SPCs were compared by Student’s t-test or Fisher’s exact test as appropriate. Risk factors associated with SPC occurrence were evaluated by univariable and multivariable logistic regression analyses. Receiver operating characteristic (ROC) curve analysis was used to establish a cut-off level of SII value to predict SPCs.

Results

Of the 310 patients receiving UAS, 103 patients (33.2%) presented at least one SPC, including PPCs (n = 62), adverse cardiovascular events (n = 22), postoperative infections (n = 51), and delirium (n = 5). Both preoperative SII and 1-h postoperative SII in patients with SPCs were significantly higher than those in patients without SPCs. Multivariate analysis showed that 1-h postoperative SII was an independent predictor for SPC occurrence (OR = 1.000, 95% CI 1.000–1.000, P = 0.007), together with postoperative C-reactive protein, postoperative arterial lactate, postoperative oxygenation-index and older age. The ROC curve showed that the optimal cutoff value of 1-h postoperative SII to predict SPCs was 754.6078 × 10⁹/L, with an 88.3% sensitivity and a 29% specificity. Multivariate analysis also confirmed that 1-h postoperative SII > 754.6078 × 10⁹/L was associated with increased SPC occurrence (OR = 2.656, 95% CI 1.311–5.381, P = 0.007).

Conclusion

Our findings demonstrated an association between the higher level of 1-h postoperative SII and SPCs, suggesting that 1-h postoperative SII, especially categorized 1-h postoperative SII using cutoff value, may be a useful tool for identifying patients at risk of developing SPCs.