Preoperative Cognitive Impairment As a Predictor of Postoperative Outcomes in a Collaborative Care Model

A simple machine learning model for the prediction of acute kidney injury following noncardiac surgery in geriatric patients: a prospective cohort study

BACKGROUND

Surgery in geriatric patients often poses risk of major postoperative complications. Acute kidney injury (AKI) is a common complication following noncardiac surgery and is associated with increased mortality. Early identification of geriatric patients at high risk of AKI could facilitate preventive measures and improve patient prognosis. This study used machine learning methods to identify important features and predict AKI following noncardiac surgery in geriatric patients.

METHODS

The data for this study were obtained from a prospective cohort. Patients aged ≥ 65 years who received noncardiac surgery from June 2019 to December 2021 were enrolled. Data were split into training set (from June 2019 to March 2021) and internal validation set (from April 2021 to December 2021) by time. The least absolute shrinkage and selection operator (LASSO) regularization algorithm and the random forest recursive feature elimination algorithm (RF-RFE) were used to screen important predictors. Models were trained through extreme gradient boosting (XGBoost), random forest, and LASSO. The SHapley Additive exPlanations (SHAP) package was used to interpret the machine learning model.

RESULTS

The training set included 6753 geriatric patients. Of these, 250 (3.70%) patients developed AKI. The XGBoost model with RF-RFE selected features outperformed other models with an area under the precision-recall curve (AUPRC) of 0.505 (95% confidence interval [CI]: 0.369-0.626) and an area under the receiver operating characteristic curve (AUROC) of 0.806 (95%CI: 0.733-0.875). The model incorporated ten predictors, including operation site and hypertension. The internal validation set included 3808 geriatric patients, and 96 (2.52%) patients developed AKI. The model maintained good predictive performance with an AUPRC of 0.431 (95%CI: 0.331-0.524) and an AUROC of 0.845 (95%CI: 0.796-0.888) in the internal validation.

CONCLUSIONS

This study developed a simple machine learning model and a web calculator for predicting AKI following noncardiac surgery in geriatric patients. This model may be a valuable tool for guiding preventive measures and improving patient prognosis.

TRIAL REGISTRATION

The protocol of this study was approved by the Committee of Ethics from West China Hospital of Sichuan University (2019-473) with a waiver of informed consent and registered at www.chictr.org.cn (ChiCTR1900025160, 15/08/2019).

Predictors of early postoperative cognitive dysfunction in middle-aged patients undergoing cardiac surgery: retrospective observational study

PURPOSE

This study aimed to identify the incidence and risk factors of early post-operative cognitive dysfunction (POCD) in middle-aged patients undergoing cardiac surgery.

METHODS

Data were examined retrospectively from 71 patients aged 46-64 years who underwent elective cardiac surgery. Magnetic resonance imaging (MRI) and MR angiography were obtained preoperatively to assess prior cerebral infarctions, carotid artery stenosis, and intracranial arterial stenosis. Patients also completed six neuropsychological tests of memory, attention, and executive function before and after surgery. Mild cognitive impairment (MCI) was defined as performance 1.5 standard deviations (SD) below the population means on any neurocognitive battery, whereas POCD was defined as a decrease of 1 SD population means on at least two in the test battery. Patient characteristics were analyzed using univariate analysis, and independent predictors were analyzed using multivariate logistic regression analysis.

RESULTS

After surgery, 25 patients (35%) were assessed with POCD. Patients with POCD had significantly higher rates of preoperative MCI and cerebral infarcts on MRI. Multivariate logistic regression analysis identified preoperative MCI and cerebral infarctions detected by MRI as a predictor of POCD.

CONCLUSION

More than one-third of middle-aged patients undergoing cardiac surgery developed POCD. Our findings suggested preoperative MCI and infarcts detected by MRI were risk factors for POCD in these middle-aged patients.

Machine learning prediction of postoperative major adverse cardiovascular events in geriatric patients: a prospective cohort study

Background

Postoperative major adverse cardiovascular events (MACEs) account for more than one-third of perioperative deaths. Geriatric patients are more vulnerable to postoperative MACEs than younger patients. Identifying high-risk patients in advance can help with clinical decision making and improve prognosis. This study aimed to develop a machine learning model for the preoperative prediction of postoperative MACEs in geriatric patients.

Methods

We collected patients’ clinical data and laboratory tests prospectively. All patients over 65 years who underwent surgeries in West China Hospital of Sichuan University from June 25, 2019 to June 29, 2020 were included. Models based on extreme gradient boosting (XGB), gradient boosting machine, random forest, support vector machine, and Elastic Net logistic regression were trained. The models’ performance was compared according to area under the precision-recall curve (AUPRC), area under the receiver operating characteristic curve (AUROC) and Brier score. To minimize the influence of clinical intervention, we trained the model based on undersampling set. Variables with little contribution were excluded to simplify the model for ensuring the ease of use in clinical settings.

Results

We enrolled 5705 geriatric patients into the final dataset. Of those patients, 171 (3.0%) developed postoperative MACEs within 30 days after surgery. The XGB model outperformed other machine learning models with AUPRC of 0.404(95% confidence interval [CI]: 0.219–0.589), AUROC of 0.870(95%CI: 0.786–0.938) and Brier score of 0.024(95% CI: 0.016–0.032). Model trained on undersampling set showed improved performance with AUPRC of 0.511(95% CI: 0.344–0.667, p < 0.001), AUROC of 0.912(95% CI: 0.847–0.962, p < 0.001) and Brier score of 0.020 (95% CI: 0.013–0.028, p < 0.001). After removing variables with little contribution, the undersampling model showed comparable predictive accuracy with AUPRC of 0.507(95% CI: 0.338–0.669, p = 0.36), AUROC of 0.896(95%CI: 0.826–0.953, p < 0.001) and Brier score of 0.020(95% CI: 0.013–0.028, p = 0.20).

Conclusions

In this prospective study, we developed machine learning models for preoperative prediction of postoperative MACEs in geriatric patients. The XGB model showed the best performance. Undersampling method achieved further improvement of model performance.

Trial registration

The protocol of this study was registered at www.chictr.org.cn (15/08/2019, ChiCTR1900025160)

Supplementary Information

The online version contains supplementary material available at 10.1186/s12871-022-01827-x.

Efficacy of geriatric multidisciplinary oncology clinic in the surgical treatment decision-making process for frail elderly patients with colorectal cancer

Purpose

Multidisciplinary care has become a cornerstone of colorectal cancer management. To evaluate the clinical efficacy of a geriatric multidisciplinary oncology clinic (GMOC), we analyzed the surgical treatment decision-making process and outcomes.

Methods

This retrospective single-center study reviewed the data of patients aged ≥65 years who participated in the GMOC at a tertiary referral hospital between 2015 and 2021. The clinical adherence rate, comprehensive geriatric assessment, and a multidimensional frailty score (MFS) were obtained. The groups that were recommended and not recommended for surgery were compared, analyzing the factors impacting the decision and 1-year survival outcomes. Furthermore, the postoperative complications of patients who underwent surgery were evaluated.

Results

A total of 165 patients visited the GMOC, and 74 had colorectal cancer (mean age, 85.5 years [range, 81.2–89.0 years]). Among patients with systemic disease (n = 31), 7 were recommended for surgery, and 5 underwent surgery. Among patients with locoregional disease (n = 43), 18 were recommended for surgery, and 12 underwent surgery. Patients recommended and not recommended for surgery had significantly different activities of daily living (ADL) (P = 0.024), instrumental ADL (P = 0.001), Mini-Mental State Examination (P = 0.014), delirium risk (P = 0.039), and MFS (P = 0.001). There was no difference in the 1-year overall survival between the 2 groups (P = 0.980). Of the 17 patients who underwent surgery, the median (interquartile range) of operation time was 165.0 minutes (120.0–270.0 minutes); hospital stay, 7.0 days (6.0–8.0 days); and 3 patients had wound complications.

Conclusion

Proper counseling of patients through the GMOC could lead to appropriate management and favorable outcomes.

A multicenter prospective study on postoperative pulmonary complications prediction in geriatric patients with deep neural network model

Aim

Postoperative pulmonary complications (PPCs) can increase the risk of postoperative mortality, and the geriatric population has high incidence of PPCs. Early identification of high-risk geriatric patients is of great value for clinical decision making and prognosis improvement. Existing prediction models are based purely on structured data, and they lack predictive accuracy in geriatric patients. We aimed to develop and validate a deep neural network model based on combined natural language data and structured data for improving the prediction of PPCs in geriatric patients.

Methods

We consecutively enrolled patients aged ≥65 years who underwent surgery under general anesthesia at seven hospitals in China. Data from the West China Hospital of Sichuan University were used as the derivation dataset, and a deep neural network model was developed based on combined natural language data and structured data. Data from the six other hospitals were combined for external validation.

Results

The derivation dataset included 12,240 geriatric patients, and 1949(15.9%) patients developed PPCs. Our deep neural network model outperformed other machine learning models with an area under the precision-recall curve (AUPRC) of 0.657(95% confidence interval [CI], 0.655–0.658) and an area under the receiver operating characteristic curve (AUROC) of 0.884(95% CI, 0.883–0.885). The external dataset included 7579 patients, and 776(10.2%) patients developed PPCs. In external validation, the AUPRC was 0.632(95%CI, 0.632–0.633) and the AUROC was 0.889(95%CI, 0.888–0.889).

Conclusions

This study indicated that the deep neural network model based on combined natural language data and structured data could improve the prediction of PPCs in geriatric patients.

Geriatric Preoperative Optimization: A Review

This review summarizes best practices for the perioperative care of older adults as recommended by the American Geriatrics Society, American Society of Anesthesiologists, and American College of Surgeons, with practical implementation strategies that can be readily implemented in busy preoperative or primary care clinics. In addition to traditional cardiopulmonary screening, older patients should undergo a comprehensive geriatric assessment. Rapid screening tools such as the Mini-Cog, Patient Health Questionnaire-2, and Frail Non-Disabled Survey and Clinical Frailty Scale, can be performed by multiple provider types and allow for quick, accurate assessments of cognition, functional status, and frailty screening. To assess polypharmacy, online resources can help providers identify and safely taper high-risk medications. Based on preoperative assessment findings, providers can recommend targeted prehabilitation, rehabilitation, medication management, care coordination, and/or delirium prevention interventions to improve postoperative outcomes for older surgical patients. Structured goals of care discussions utilizing the question-prompt list ensures that older patients have a realistic understanding of their surgery, risks, and recovery. This preoperative workup, combined with engaging with family members and interdisciplinary teams, can improve postoperative outcomes.

Association between preoperative serum homocysteine and delayed neurocognitive recovery after non-cardiac surgery in elderly patients: a prospective observational study

BACKGROUND

Homocysteine, folate, and vitamin B₁₂ involved in 1-carbon metabolism are associated with cognitive disorders. We sought to investigate the relationships between these factors and delayed neurocognitive recovery (dNCR) after non-cardiac surgery.

METHODS

This was a prospective observational study of patients (n = 175) who were ≥ 60 years of age undergoing non-cardiac surgery. Patients were evaluated preoperatively and for 1 week postoperatively by using neuropsychological tests and were divided into dNCR or non-dNCR groups according to a Z-score ≤ - 1.96 on at least two of the tests. The relationship between the occurrence of dNCR and preoperative levels of homocysteine, folate, and vitamin B₁₂ was analyzed. Univariate and multivariable logistic regression analyses were conducted to identify factors associated with dNCR.

RESULTS

Delayed neurocognitive recovery was observed in 36 of 175 patients (20.6%; 95% confidence interval [CI], 14.5-26.6%) 1 week postoperatively. Patients who developed dNCR had significantly higher median [interquartile range (IQR)] homocysteine concentrations (12.8 [10.9,14.4] μmol/L vs 10.6 [8.6,14.7] μmol/L; P = 0.02) and lower folate concentrations (5.3 [4.2,7.3] ng/mL vs 6.9 [5.3,9.5] ng/mL; P = 0.01) than those without dNCR. Compared to the lowest tertile, the highest homocysteine tertile predicted dNCR onset (odds ratio [OR], 3.9; 95% CI, 1. 3 to 11.6; P = 0.02), even after adjusting for age, sex, education, and baseline Mini Mental State Examination.

CONCLUSIONS

Elderly patients with high homocysteine levels who underwent general anesthesia for non-cardiac surgery have an increased risk of dNCR. This knowledge could potentially assist in the development of preventative and/or therapeutic measures.

TRIAL REGISTRATION

NCT03084393 ( https://www.clinicaltrials.gov ).

Age-Related Risk Factors in Ventral Hernia Repairs: A Review and Call to Action

BACKGROUND

As the population ages, the incidence of ventral hernias in older adults is increasing. Ventral hernia repairs (VHR) should not be considered low risk operations, particularly in older adults who are disproportionately affected by multiple age-related factors that can complicate surgery and adversely affect outcomes. Although age-related risk factors have been well established in other surgical fields, there is currently little data describing their impact on VHR.

METHODS

We performed a systematic review of the literature to identify studies that examine the effects of age-related risk factors on VHR outcomes. This was conducted using Cochrane Library, Embase, PubMed (Medline), and Google Scholar databases, all updated through June 2020. We selected relevant studies using the keywords, multimorbidity, comorbidities, polypharmacy, functional dependence, functional status, frailty, cognitive impairment, dementia, sarcopenia, and malnutrition. Primary outcomes include mortality and overall complications following VHR.

RESULTS

We summarize the evidence basis for the significance of age-related risk factors in elective surgery and discuss how these factors increase the risk of adverse outcomes following VHR. In particular, we explore the impact of the following risk factors: multimorbidity, polypharmacy, functional dependence, frailty, cognitive impairment, sarcopenia, and malnutrition. As opposed to chronological age itself, age-related risk factors are more clinically relevant in determining VHR outcomes.

CONCLUSIONS

Given the increasing complexity of VHR, addressing age-related risk factors pre-operatively has the potential to improve surgical outcomes in older adults. Preoperative risk assessment and individualized prehabilitation programs aimed at improving patient-centered outcomes may be particularly useful in this population.

Perioperative Optimization of Senior Health (POSH): A Descriptive Analysis of Cancelled Surgery

BACKGROUND

Geriatric collaborative care models improve postoperative outcomes for older adults. However, there are limited data exploring how preoperative geriatric assessment may affect surgical cancellations.

METHODS

This is a single-center retrospective cohort analysis. Patients enrolled in the Perioperative Optimization of Senior Health (POSH) program from 2011 to 2016 were included. POSH is a collaborative care model between geriatrics, surgery, and anesthesiology. Baseline demographic and medical data were collected during the POSH pre-op appointment. Patients who attended a POSH pre-op visit but did not have surgery were identified, and a chart review was performed to identify reasons for surgical cancellation. Baseline characteristics of patients who did and did not undergo surgery were compared.

RESULTS

Of 449 eligible POSH referrals within the study period, 33 (7.3%) did not proceed to surgery; cancellation rates within the POSH program were lower than institutional cancellation rates for adults over age 65 who did not participate in POSH. Patients who did not have surgery were significantly older, more likely to have functional limitations, and had higher rates of several comorbidities compared with those who proceeded to surgery (P < 0.05). Reasons for surgical cancellations included a similar number of patient- and provider-driven causes.

CONCLUSIONS

Many reasons for surgical cancellation were related to potentially modifiable factors, such as changes in goals of care or concerns about rehabilitation, emphasizing the importance of shared decision-making in elective surgery for older adults. These results highlight the important role geriatric collaborative care can offer to older adults with complex needs.

Preoperative Cognitive Assessment Recommendations for the Older Adult

PURPOSE

The purpose of this project was to identify the need for and to improve the preoperative cognitive assessment of the older adult.

DESIGN

A retrospective chart review was used to explore the incidence of postoperative delirium (PD) and characteristics associated with it.

METHODS

A retrospective chart review was used to identify the incidence of PD in a community hospital. The data were analyzed using descriptive statistics for trends in demographic and physiological characteristics of older adults undergoing elective hip or knee surgery.

FINDINGS

The incidence of PD was found to be 11%. Older adults with PD had an increased mean age and comorbid conditions. PD was associated with a mean increase in hospital stay, postoperative complications, and 30-day readmission.

CONCLUSIONS

Preoperative cognitive assessment can identify high-risk patients, stratify care, medically optimize the older adult before surgery, and improve perioperative outcomes.

The importance of the STOP- BANG questionnaire as a preoperative assessment tool for the elderly population

Undiagnosed obstructive sleep apnea (OSA) may adversely impact surgical patients and can lead to increased morbidity and mortality during the perioperative period, especially among the geriatric patient population (Chung et al. 2008, 2012, 2014; McDonald et al., 2018; Zietlow et al., 2018; Singh et al., 2012). The setting of this quality improvement project was a preoperative anesthesia and geriatric evaluation clinic housed within a 957-bed tertiary academic affiliated hospital. The sample included 45 patients who met the criteria established for surgery and OSA screening preoperatively. Nine patients (20.0%) were assessed as low risk (Stop-bang [SB] score </=2) for OSA, and 36 patients (80.0%) had a prior diagnosis from an ICD-9/10 code or a SB score >/= 3 indicative of high-risk for OSA. The retrospective utilization of a modified SB screening on charts that did not receive a clinical OSA evaluation (n = 52) detected 23 (44.2%) patients who were considered high-risk for OSA but were not identified prior to surgery. The SB questionnaire is underutilized, and patients' OSA is often unidentified prior to surgery.

Neuropsychological Functioning in Older Adults with Obesity: Implications for Bariatric Surgery

Bariatric surgery is the most effective approach to treating morbid obesity, resulting in decreased morbidity, mortality, and improved quality of life. Research on outcomes has generally been restricted to young and middle-aged adults, despite a growing epidemic of obesity in older adults. The use of bariatric surgery has been limited in older individuals, in part due to concerns that preexisting cognitive dysfunction increases the risk of poor post-surgical outcomes, including cognitive decline. The literature on the relationship between obesity and cognition in older adults is emerging, but fraught by several methodological limitations. While there is insufficient research to determine the nature of cognitive outcomes following bariatric surgery in older adults, the aim of this paper is to review the existing evidence and make the case for further study.

Functional capacity as a predictor of postoperative delirium in transurethral resection of prostate patients in Northeast Brazil

INTRODUCTION

Postoperative delirium (POD) is a common disorder and its frequency varies from 15% to 25% after major elective surgery. There are few data on the incidence of POD in Brazil. Here, we sought to assess the incidence of POD following transurethral resection of the prostate (TURP) and to examine precipitating and predisposing factors associated.

METHOD

We performed a prospective observational study of elderly male patients undergoing TURP (N=55) in Northeast Brazil. Information on demographic, medical, cognitive and functional characteristics were collected. The participants were followed until hospital discharge. POD was diagnosed by the Confusion Assessment Method.

RESULTS

A total of three participants (5.45%) were identified with POD. Episodes of delirium lasted 3±1 days. The study sample consisted of a healthy population. Patients with POD had longer hospital stay and more precipitating factors. The POD group showed statistically significant lower Barthel index score (p<0.001) and higher Pfeffer's Functional Activities Questionnaire scores (p<0.01).

CONCLUSION

Loss of functional capacity was associated with POD in a healthy population of elderly patients undergoing TURP.