Risk factors of in-hospital mortality for isolated on-pump coronary artery bypass graft surgery in the northeast of Iran from 2007 to 2016

Nomali, Mahin; Heidari, Mohammad; Ayati, Aryan; Moghaddam, Keyvan; Mosallami, Soheil; Khosravi, Afifeh; Rafiei, Mona; Riahinokandeh, Gholamali; Yadegari, Mohammad; Nomali, Mahdis; Taheriyan, Moloud; Roshandel, Gholamreza

doi:10.1007/s11845-023-03298-6

Cited by 3 publications

(3 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several recent analyses identify CABG-associated risk from structured EHR data [ 21 – 24 ]. Two of the analyses achieve similar or superior performance to our analysis with AUCs of 0.811 [ 22 ] and 0.831 [ 23 ] for post-CABG mortality and bleeding.…”

Section: Discussionmentioning

confidence: 99%

“…Recent studies that utilize structured EHR data to predict coronary artery bypass grafting (CABG)-associated risk achieve similar [ 21 ] or superior [ 22 , 23 ] performance to the STS risk calculator and identify high-risk predictors [ 21 , 24 ]. We hypothesize that neural network models can complement these analyses by extracting information from unstructured data, such as patient notes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Prediction of coronary artery bypass graft outcomes using a single surgical note: An artificial intelligence-based prediction model study

Del Gaizo,

Sherard,

Shorbaji

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

Background Healthcare providers currently calculate risk of the composite outcome of morbidity or mortality associated with a coronary artery bypass grafting (CABG) surgery through manual input of variables into a logistic regression-based risk calculator. This study indicates that automated artificial intelligence (AI)-based techniques can instead calculate risk. Specifically, we present novel numerical embedding techniques that enable NLP (natural language processing) models to achieve higher performance than the risk calculator using a single preoperative surgical note. Methods The most recent preoperative surgical consult notes of 1,738 patients who received an isolated CABG from July 1, 2014 to November 1, 2022 at a single institution were analyzed. The primary outcome was the Society of Thoracic Surgeons defined composite outcome of morbidity or mortality (MM). We tested three numerical-embedding techniques on the widely used TextCNN classification model: 1a) Basic embedding, treat numbers as word tokens; 1b) Basic embedding with a dataloader that Replaces out-of-context (ROOC) numbers with a tag, where context is defined as within a number of tokens of specified keywords; 2) ScaleNum, an embedding technique that scales in-context numbers via a learned sigmoid-linear-log function; and 3) AttnToNum, a ScaleNum-derivative that updates the ScaleNum embeddings via multi-headed attention applied to local context. Predictive performance was measured via area under the receiver operating characteristic curve (AUC) on holdout sets from 10 random-split experiments. For eXplainable-AI (X-AI), we calculate SHapley Additive exPlanation (SHAP) values at an ngram resolution (SHAP-N). While the analyses focus on TextCNN, we execute an analogous performance pipeline with a long short-term memory (LSTM) model to test if the numerical embedding advantage is robust to model architecture. Results A total of 567 (32.6%) patients had MM following CABG. The embedding performances are as follows with the TextCNN architecture: 1a) Basic, mean AUC 0.788 [95% CI (confidence interval): 0.768–0.809]; 1b) ROOC, 0.801 [CI: 0.788–0.815]; 2) ScaleNum, 0.808 [CI: 0.785–0.821]; and 3) AttnToNum, 0.821 [CI: 0.806–0.834]. The LSTM architecture produced a similar trend. Permutation tests indicate that AttnToNum outperforms the other embedding techniques, though not statistically significant verse ScaleNum (p-value of .07). SHAP-N analyses indicate that the model learns to associate low blood urine nitrate (BUN) and creatinine values with survival. A correlation analysis of the attention-updated numerical embeddings indicates that AttnToNum learns to incorporate both number magnitude and local context to derive semantic similarities. Conclusion This research presents both quantitative and clinical novel contributions. Quantitatively, we contribute two new embedding techniques: AttnToNum and ScaleNum. Both can embed strictly positive and bounded numerical values, and both surpass basic embeddings in predictive performance. The results suggest AttnToNum outperforms ScaleNum. With regards to clinical research, we show that AI methods can predict outcomes after CABG using a single preoperative note at a performance that matches or surpasses the current risk calculator. These findings reveal the potential role of NLP in automated registry reporting and quality improvement.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Prediction of coronary artery bypass graft outcomes using a single surgical note: An artificial intelligence-based prediction model study

Del Gaizo,

Sherard,

Shorbaji

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…This article utilized the dataset 41 gathered during the study and obtained funding from the Student Research Committee of the Iranian Ministry of Health. The authors thank the Student Research Committee and the Research and Technology Deputy of Golestan University of Medical Sciences (Gorgan, Iran) for facilitating the research.…”

mentioning

confidence: 99%

Ethnic Disparities in Major Adverse Cardiac and Cerebrovascular Events (MACCEs) and Postoperative Outcomes Following Coronary Artery Bypass in Northeastern Iran (2007–2016)

Nomali,

Ayati,

Tayebi

et al. 2023

Arch Iran Med

Self Cite

View full text Add to dashboard Cite

Background: Turkmens are an ethnic group mainly living in northeastern Iran. Despite previous studies on coronary artery bypass surgery (CABG) outcomes among different ethnicities, the effect of Turkmen ethnicity on outcomes of CABG surgery is still unknown. We aimed to assess the association between Turkmen ethnicity and postoperative outcomes following CABG. Methods: We used the CABG data from two heart centers in northeastern Iran between 2007 and 2016. We included adult patients undergoing CABG surgery. The study outcomes were in-hospital major adverse cardiac and cerebrovascular events (MACCEs), consisting of myocardial infarction (MI), stroke, and cardiovascular death, and postoperative outcomes, including postoperative arrhythmia, acute atrial fibrillation (AF), major bleeding, and acute renal failure (ARF). Results: Over the course of one decade, 3632 patients, with an average age (standard deviation) of 59.0 (9.8) years, were studied. Of these, 3,331 patients were of non-Turkmen ethnicity, and 301 patients were Turkmens. According to adjusted analysis, ethnicity was not associated with MACCEs (OR: 1.15, 95 % CI: 0.61, 2.16; P=0.663), postoperative arrhythmia (OR: 1.10, 95% CI: 0.78, 1.54; P=0.588), acute AF (OR: 1.17, 95 % CI: 0.83, 1.66; P=0.359), major bleeding (OR: 1.21, 95 % CI: 0.55, 2.67; P=0.636), or ARF (OR: 2.60, 95 % CI: 0.60, 11.75, P=0.224). Conclusion: This study found that despite ethnic disparity and preoperative differences, Turkmen ethnicity was not associated with in-hospital MACCEs, AF, major bleeding, or ARF after coronary artery bypass.

show abstract

Type 2 diabetes mellitus and In-hospital Major Adverse Cardiac and Cerebrovascular Events (MACCEs) and postoperative complications among patients undergoing on-pump isolated coronary artery bypass surgery in Northeastern Iran

Nomali

Ayati

Tayebi³

et al. 2023

BMC Cardiovasc Disord

View full text Add to dashboard Cite

Background Diabetes Mellitus (DM) is a rapidly growing disorder worldwide, especially in the Middle East. A higher incidence of coronary artery diseases requiring coronary artery bypass graft (CABG) surgery has been reported in patients with diabetes. We assessed the association between type 2 diabetes mellitus (T2DM) and in-hospital major adverse cardiac and cerebrovascular events (MACCEs) and postoperative complications among patients who underwent on-pump isolated CABG. Methods In this retrospective cohort study, we used the data registered for CABG patients from two heart centers in the Golestan province (North of Iran) between 2007 and 2016. The study population included 1956 patients divided into two groups: 1062 non-diabetic patients and 894 patients with diabetes (fasting plasma glucose ≥126 mg/dl or using antidiabetic medications). The study outcome was in-hospital MACCEs, a composite outcome of myocardial infarction (MI), stroke and cardiovascular death, and postoperative complications, including postoperative arrhythmia, acute atrial fibrillation (AF), major bleeding (defined as reoperation due to bleeding), and acute kidney injury (AKI). Results During the 10-year study period, 1956 adult patients with a mean (SD) age of 59.0 (9.60) years were included. After adjustment for age, gender, ethnicity, obesity, opium consumption, and smoking, diabetes was a predictor of postoperative arrhythmia (AOR 1.30, 95% CI 1.08–1.57; P = 0.006). While it was not a predictor of in-hospital MACCEs (AOR 1.35, 95% CI 0.86, 2.11; P = 0.188), AF (AOR 0.85, 95% CI 0.60–1.19; P = 0.340), major bleeding (AOR 0.80, 95% CI 0.50, 1.30; P = 0.636) or AKI (AOR 1.29, 95% CI 0.42, 3.96; P 0.656) after CABG surgery. Conclusion Findings indicated that diabetes increased the risk of postoperative arrhythmia by 30%. However, we found similar in-hospital MACCEs, acute AF, major bleeding, and AKI following CABG surgery in both diabetic and non-diabetic patients.

show abstract

Risk factors of in-hospital mortality for isolated on-pump coronary artery bypass graft surgery in the northeast of Iran from 2007 to 2016

Cited by 3 publications

References 49 publications

Prediction of coronary artery bypass graft outcomes using a single surgical note: An artificial intelligence-based prediction model study

Prediction of coronary artery bypass graft outcomes using a single surgical note: An artificial intelligence-based prediction model study

Ethnic Disparities in Major Adverse Cardiac and Cerebrovascular Events (MACCEs) and Postoperative Outcomes Following Coronary Artery Bypass in Northeastern Iran (2007–2016)

Type 2 diabetes mellitus and In-hospital Major Adverse Cardiac and Cerebrovascular Events (MACCEs) and postoperative complications among patients undergoing on-pump isolated coronary artery bypass surgery in Northeastern Iran

Contact Info

Product

Resources

About