Early Predictive nomogram of Ventricular Fibrillation after Percutaneous Coronary Intervention in Acute Myocardial Infarction using LASSO-Logistic regression

doi:10.21203/rs.3.rs-4867330/v1

Objective

Ventricular fibrillation (VF) is a serious complication of acute myocardial infarction (AMI) that can occur after percutaneous coronary intervention (PCI). This study aimed to identify early predictors of VF and develop a nomogram to predict VF risk in this patient population.

Methods

This single-center retrospective cohort study included 162 AMI patients who underwent PCI at Beijing Friendship Hospital between 2018 and present. Patients were divided into VF and non-VF groups (1:3 matched) based on the occurrence of VF after PCI. Baseline demographics, medical history, laboratory results, medications, and in-hospital outcomes were compared between groups. LASSO regression analysis, followed by univariable and multivariable logistic regression, identified predictors of VF. A nomogram incorporating significant predictors was constructed and validated using ROC curve analysis, calibration curves, and the Hosmer-Lemeshow goodness-of-fit test.

Results

Significant differences in baseline characteristics (Table 1A), clinical data including coagulation parameters and complications (Table 1B), and in-hospital prognoses (Table 1C) were observed between VF and non-VF groups. LASSO regression analysis identified key predictors, and multivariable analysis confirmed diabetes, metabolic acidosis, anemia, hypokalemia, and ventricular premature beats as independent predictors of VF (Table 2). The nomogram incorporating these factors showed excellent discrimination (AUC = 0.942, 95% CI: 0.882-1.000) and good calibration (Hosmer-Lemeshow test p = 0.898).

Conclusion

This study identified several readily available clinical parameters as independent predictors of VF in AMI patients undergoing PCI. The developed nomogram based on these predictors showed promising predictive accuracy and calibration, potentially assisting clinicians in early risk stratification and guiding proactive management strategies to improve outcomes in this high-risk population.

Nomogram

Ventricular Fibrillation

Percutaneous Coronary Intervention (PCI)

Acute Myocardial Infarction (AMI)

Nomogram

Acute myocardial infarction (AMI) remains a major public health concern worldwide, associated with significant morbidity and mortality[1, 2]. One of the most serious complications of AMI is the development of ventricular tachyarrhythmias, particularly ventricular fibrillation (VF), which can lead to sudden cardiac death[3, 4]. Despite advances in revascularization strategies such as primary percutaneous coronary intervention (PCI), the incidence of VF complicating AMI remains high, occurring in up to 6% of patients[5, 6].

The occurrence of VF in the setting of AMI is a strong predictor of poor in-hospital and long-term outcomes[7, 8] .Patients who develop VF prior to or during PCI have significantly higher rates of mortality compared to those without VF .Identifying early predictors of VF in AMI patients undergoing PCI is therefore crucial, as it may allow for targeted prevention and management strategies to improve clinical outcomes.

Several factors have been associated with an increased risk of VF in the setting of AMI, including the extent of myocardial injury, presence of cardiogenic shock, impaired coronary artery flow, and underlying comorbidities such as chronic kidney disease[9]. However, the specific clinical, electrocardiographic, and laboratory parameters that reliably predict VF development remain incompletely characterized.

The aim of this study was to evaluate the clinical, electrocardiographic, and laboratory predictors of VF in AMI patients undergoing primary PCI at a large tertiary care center. By identifying high-risk patients early, clinicians may be able to implement prophylactic measures and closely monitor these individuals for the development of life-threatening arrhythmias. This knowledge could ultimately lead to improved outcomes in this high-risk population.

Study Design and Population

This was a retrospective, single-center cohort study conducted at Beijing Friendship Hospital from 2015 to the present. The study included adult patients (≥18 years old) admitted with a diagnosis of AMI who underwent primary PCI. Patients were excluded if they had a history of ventricular arrhythmias prior to the index AMI, significant comorbidities (e.g. end-stage renal disease, malignancy), or incomplete medical records. The study protocol was approved by the Ethics Committee of Beijing Friendship Hospital (Approval No. 2018-P2-030-01)

Inclusion Criteria:

1. Age: Patients aged 18 years and older.

2. Procedure: Patients who have undergone PCI.

3. Diagnosis: Patients diagnosed with coronary artery disease requiring PCI.

4. Consent: Patients who have provided informed consent to participate in the study.

5. Data Availability: Patients with complete medical records and follow-up data available for analysis.

Exclusion Criteria:

1. Alternative Diagnoses: Patients with non-coronary artery disease as the primary indication for PCI.

2. Lack of Consent: Patients who did not provide informed consent for participation.

3. Incomplete Data: Patients with incomplete medical records or missing follow-up data.

4. Comorbid Conditions: Patients with severe comorbid conditions that could independently influence the risk of ventricular fibrillation, such as advanced heart failure or terminal illnesses.

5. Previous VF Episodes: Patients with a history of ventricular fibrillation prior to the PCI procedure.

Grouping strategy

Patients were divided into two groups based on the occurrence of VF after PCI: the VF group and the non-VF group. The VF group included patients who experienced VF within 48 hours of the PCI procedure, while the non-VF group included patients who did not develop VF during the same time period. A 1:3 matched control design was used, with the non-VF group matched to the VF group by age, sex, and timing of PCI.

Data Collection

Baseline demographic, clinical, laboratory, and procedural characteristics were collected from the patients' electronic medical records. Specifically, data was gathered on age, sex, cardiovascular risk factors (hypertension, diabetes, dyslipidemia, smoking), prior history of coronary artery disease, Killip class on admission, time from symptom onset to PCI, culprit vessel, procedural details, and in-hospital complications. Laboratory values including troponin, creatine kinase-MB, and electrolytes were also recorded.

Statistical Analysis

Continuous variables were reported as mean ± standard deviation or median (interquartile range), as appropriate. Categorical variables were summarized as frequencies and percentages. Comparisons between the VF and non-VF groups were performed using Student's t-test, Mann-Whitney U test, chi-square test, or Fisher's exact test, as indicated.

The LASSO (least absolute shrinkage and selection operator) regression model was used for variable selection, followed by univariable and multivariable logistic regression analyses to identify independent predictors of post-PCI VF. Variables with a p-value <0.1 in the univariable analysis were entered into the multivariable model. Odds ratios (OR) and 95% confidence intervals (CI) were calculated.

A nomogram was constructed based on the results of the multivariable analysis to provide a visual tool for predicting the individual risk of post-PCI VF. The performance of the nomogram was evaluated using receiver operating characteristic (ROC) curve analysis, calibration plots, and decision curve analysis (DCA).

All statistical analyses were performed using R software (version 4.2.3). A two-sided p-value <0.05 was considered statistically significant.

The study protocol was approved by the Institutional Review Board of Beijing Friendship Hospital, and the requirement for informed consent was waived due to the retrospective nature of the study.

Table 1A investigated the potential risk factors for the occurrence of VF dafter PCI in patients with AMI. A total of 162 patients were enrolled and divided into two groups based on the occurrence of VF during PCI. Baseline characteristics, including demographics, medical history, laboratory results, and medication use, were compared between the two groups. The study found significant differences in systolic blood pressure (SBP), blood urea nitrogen (BUN), total cholesterol, HDL-C, and the use of aspirin, clopidogrel or ticagrelor, statins, and beta-blockers between the VF and non-VF groups. These findings suggest that these factors may be associated with the risk of VF during PCI in AMI patients and could potentially guide early risk stratification and preventive strategies.

Table 1A. Baseline characters for enrolled subjects

Table 1B investigated clinical data, including medical history, vital signs, laboratory results, and medications, from 162 AMI patients undergoing PCI. It revealed that elevated hs-CRP, NLR, INR, APTT, FDP, fibrinogen, potassium levels, and a lower antithrombin III activity were significantly associated with an increased risk of VF. Furthermore, patients experiencing VF exhibited a higher prevalence of complications such as acute renal insufficiency, abnormal liver function, metabolic acidosis, pleural effusion, hypokalemia, various arrhythmias, and thyroid function abnormalities (higher T4 and lower T3). These findings underscore the importance of monitoring these parameters as potential early warning signs for VF risk stratification and proactive management in AMI patients undergoing PCI.

Table 1B. Dignosis, medication and coagulation function for enrolled subjects

Table 1C analyzed various clinical parameters from 162 AMI patients. Table 1C highlights that VF during PCI was significantly associated with lower ejection fraction, higher left coronary artery occlusion, higher Gensini score (a measure of coronary artery disease severity), and elevated levels of cTnI and NT-proBNP. Furthermore, the VF group experienced a higher incidence of in-hospital complications such as MACE, acute in-stent thrombosis, malignant arrhythmia, longer hospital stays, and increased need for interventions like IABP, endotracheal intubation, ventilation, and defibrillation. These findings emphasize the importance of considering these factors for early risk assessment and tailored management strategies to mitigate VF risk in this high-risk patient population.

Table 1C. Inhosptial prognosis for enrolled subjects

In the LASSO regression analysis aimed at identifying key predictors for the occurrence of VF post-PCI in AMI patients, we set the optimal lambda value to 0.0285 and successfully selected 19 significant variables from 122 variables. These variables and their corresponding coefficients are as follows: Hypertension (0.3973), Diabetes (0.5159), Respiratory Failure (1.4288), Metabolic Acidosis (0.7986), Pleural Effusion (0.1239), Anemia (1.3912), Thrombosis (2.4352), Hypokalemia (0.4699), Ventricular Premature Beats (1.7330), Atrial Premature Beats (0.9645), Atrioventricular Block (0.8775), Statins (-0.9900), Anion Gap (0.0172), Thyroid Uptake (0.0164), Stroke Volume (-0.0071), Right Ventricle (0.0837), Pulmonary Hypertension (0.9817), IABP (0.2387), and Right Coronary Artery Lesion (-0.4177). These variables were identified as significant predictors and were used to construct a subsequent nomogram for early prediction of VF in post-PCI AMI patients, with positive coefficients indicating a direct relationship with the likelihood of VF occurrence and negative coefficients indicating an inverse relationship.

Figure 2.LASSO regression analysis was used to selecte important characters.

LASSO Path Plot. The LASSO Path Plot, also known as the coefficient path plot, displays the trajectories of the regression coefficients as the regularization parameter lambda varies. On the x-axis, we have the log of the lambda values, and on the y-axis, the coefficients of the variables. Each line represents a variable, showing how its coefficient changes with different levels of penalization. As lambda increases, more coefficients shrink towards zero, effectively performing variable selection. This plot is useful for understanding which variables remain significant predictors at different levels of regularization.

Cross-Validation Plot. The Cross-Validation Plot is used to determine the optimal lambda value that minimizes prediction error. This plot typically shows the mean squared error (MSE) or another error metric on the y-axis and the log of the lambda values on the x-axis. The plot displays the error for each fold of the cross-validation process, and the optimal lambda is usually indicated by the point with the lowest error. This plot helps in selecting the best model that balances bias and variance, ensuring good predictive performance on unseen data.

Table 2 presents the logistic regression analysis for predictors of VF, highlighting both univariable and multivariable odds ratios (OR) with 95% confidence intervals (CI) and p-values. Significant predictors in the univariable analysis include hypertension, diabetes, respiratory failure, metabolic acidosis, pleural effusion, anemia, thrombosis, hypokalemia, ventricular premature beats, atrial premature beats, atrioventricular block, statins, anion gap, thyroid uptake, stroke volume, right ventricle, pulmonary hypertension, and IABP. In the multivariable analysis, diabetes (OR: 1.891, p=0.006), metabolic acidosis (OR: 2.537, p=0.004), anemia (OR: 3.326, p=0.001), hypokalemia (OR: 1.540, p=0.014), and ventricular premature beats (OR: 6.156, p=0.000) remained significant predictors of VF. These results underscore the importance of these factors in the prediction of VF in patients.

Table 2 Logistic regression for predicters of ventricular fibrillation

This image depicts a nomogram designed to predict the probability of VF following PCI. The nomogram includes several significant predictors identified from the logistic regression analysis, such as diabetes, ventricular premature beats, hypokalemia, anemia, and metabolic acidosis. Each variable is assigned a specific point value based on the patient's condition, which is then summed to calculate the total points. This total is used to determine the linear predictor and subsequently the predicted probability of VF. The nomogram provides a visual and quantitative tool for clinicians to assess the risk of VF post-PCI, incorporating key predictors to aid in decision-making and risk stratification.

Figure 3.Nomogram was constructed to facilitate the prediction of VF risk in AMI patients.

To validate the predictive capability of the nomogram for post-PCI VF, we utilized Receiver Operating Characteristic (ROC) curve analysis. The Area Under the Curve (AUC) was found to be 0.942, with a 95% Confidence Interval (CI) ranging from 0.882 to 1, indicating excellent predictive ability. The optimal cutoff point was determined to be 89.906, yielding a specificity of 90.44% and a sensitivity of 94.12%. Additionally, the Positive Predictive Value (PPV) was 71.11% and the Negative Predictive Value (NPV) was 98.4%, demonstrating the nomogram's effectiveness in correctly identifying patients who will not develop VF. The Youden Index was calculated to be 0.846, further confirming the strong diagnostic performance of the nomogram. In summary, the ROC curve analysis demonstrates that the nomogram is a robust tool for predicting the risk of ventricular fibrillation following PCI, with high sensitivity, specificity, and overall predictive accuracy.

Figure 4.ROC Curve was employed to validate the Nomogram.

To validate the predictive capability of the nomogram for post-PCI VF using a calibration curve, we employed the Hosmer-Lemeshow goodness-of-fit test. The Hosmer-Lemeshow statistic was calculated to be 3.512, with 8 degrees of freedom, resulting in a p-value of 0.898. This high p-value indicates no significant difference between the observed and predicted probabilities, suggesting excellent calibration. In summary, the calibration curve analysis, supported by the Hosmer-Lemeshow test, demonstrates that the nomogram provides accurate and reliable predictions for the risk of ventricular fibrillation following PCI, with no significant discrepancies between predicted and observed outcomes.

Figure 5.Calibration Curve was employed to validate the Nomogram.

This retrospective study identified several independent predictors of VF following PCI in patients with AMI. Patients who experience VF following PCI typically present with several clinical characteristics. These patients often have a history of significant coronary artery disease, necessitating the PCI procedure. Common clinical features include advanced age, the presence of multiple cardiovascular risk factors such as hypertension, diabetes mellitus, and hyperlipidemia, and previous cardiac events like myocardial infarction. Additionally, these patients may exhibit signs of left ventricular dysfunction and have complex coronary lesions. The predictive nomogram validated in this study incorporates these clinical factors to accurately assess the risk of VF, demonstrating its utility in identifying high-risk individuals and guiding clinical decision-making.The key findings were that diabetes, metabolic acidosis, anemia, hypokalemia, and ventricular premature beats were associated with an increased risk of post-PCI VF. These risk factors were incorporated into a nomogram that demonstrated excellent predictive performance.

The association between diabetes and post-PCI VF is well-established in the literature. Diabetes induces structural and electrical remodeling of the myocardium, leading to increased susceptibility to ventricular arrhythmias[10]. Metabolic derangements, such as acidosis, can also directly alter cardiac electrophysiology and promote arrhythmogenesis[11]. Anemia has been linked to myocardial ischemia and infarct expansion, which may create a substrate for re-entrant circuits[12, 13].Hypokalemia is a known trigger for triggered activity and can precipitate VF, particularly in the setting of acute ischemia[14]. Ventricular premature beats are a marker of electrical instability and have been shown to precede the onset of VF in AMI patients[15].

Interestingly, the analysis did not identify some previously reported risk factors, such as advanced age, larger infarct size, and lower left ventricular ejection fraction[16, 17]. This may be due to differences in patient populations, treatment strategies, or statistical modeling approaches. Additionally, the use of statins was associated with a lower risk of post-PCI VF in the LASSO regression, consistent with the antiarrhythmic properties of these medications[18].

The development of a nomogram incorporating these key predictors provides a practical clinical tool to identify high-risk AMI patients who may benefit from closer monitoring and prophylactic antiarrhythmic strategies after PCI. Early recognition of VF risk can guide the implementation of measures such as continuous telemetry, electrolyte optimization, and the judicious use of antiarrhythmic drugs or implantable cardioverter-defibrillators[19, 20].

The strengths of this study include comprehensive data collection, and the use of robust statistical techniques, such as LASSO regression and nomogram development. However, several limitations should be acknowledged. First, the retrospective, single-center design may limit the generalizability of the findings. Second, the study did not assess long-term outcomes or the impact of post-PCI VF on mortality. Future prospective, multicenter studies are warranted to validate the predictive model and evaluate its clinical utility in guiding risk-stratified management strategies.

In conclusion, this study identified several readily available clinical and laboratory parameters that can help predict the occurrence of VF following PCI in AMI patients. The developed nomogram may assist clinicians in early identification of high-risk individuals and guide the implementation of appropriate preventive measures to mitigate the risk of this life-threatening complication.

Ethics approval and consent to participate

Not Applicable. This article does not contain any studies with human participants or animals performed by any of the authors.

Consent for publication

Not Applicable. This article does not contain any studies with human participants or animals performed by any of the authors.

Conflicts of interest

There are no conflicts of interest.

Funding

This study was supported by the National Natural Science Foundation of China (Grant No. 81600276).

Author Contribution

Ruifeng Liu guaranteed the integrity of the entire study; Gang Wang designed the study and literature research; GW defned the intellectual content; Jianghong Liu performed experiment; Jihong Fan collected the data; Huina Sun analyzed the data; Ruifeng Liu wrote the main manuscript and prepared fgures. All authors reviewed the manuscript.

Acknowledgements

We would like to thank the anonymous reviewers who have helped to

Data Availability:

Patients with complete medical records and follow-up data available for analysis.

Go AS, Mozaffarian D, Roger VL, et al. Heart disease and stroke statistics–2014 update: a report from the American Heart Association. Circulation. 2014;129(3):e28–292. 10.1161/01.cir.0000441139.02102.80.
Bougouin W, Marijon E, Puymirat E, et al. Incidence of sudden cardiac death after ventricular fibrillation complicating acute myocardial infarction: a 5-year cause-of-death analysis of the FAST-MI 2005 registry. Eur Heart J. 2014;35(2):116–22. 10.1093/eurheartj/eht453.
Mehta RH, Starr AZ, Lopes RD, et al. Incidence of and outcomes associated with ventricular tachycardia or fibrillation in patients undergoing primary percutaneous coronary intervention. JAMA. 2009;301(17):1779–89. 10.1001/jama.2009.600.
Jabbari R, Risgaard B, Fosbøl EL, et al. Factors associated with and outcomes after ventricular fibrillation before and during primary angioplasty in patients with ST-segment elevation myocardial infarction. Am J Cardiol. 2015;116(5):678–85. 10.1016/j.amjcard.2015.05.037.
Steinbeck G, Andresen D, Seidl K, et al. Defibrillator implantation early after myocardial infarction. N Engl J Med. 2009;361(15):1427–36. 10.1056/NEJMoa0901889.
Kosmidou I, Embacher M, McAndrew T, et al. Early ventricular tachycardia or fibrillation in patients with ST elevation myocardial infarction undergoing primary percutaneous coronary intervention and impact on mortality and stent thrombosis (from the harmonizing outcomes with revascularization and stents in acute myocardial infarction trial). Am J Cardiol. 2017;120(11):1755–60. 10.1016/j.amjcard.2017.07.080.
Piccini JP, White JA, Mehta RH, et al. Sustained ventricular tachycardia and ventricular fibrillation complicating non-ST-segment-elevation acute coronary syndromes. Circulation. 2012;126(1):41–9. 10.1161/CIRCULATIONAHA.111.071860.
Demidova MM, Carlson J, Erlinge D, Platonov PG. Predictors of ventricular fibrillation at reperfusion in patients with acute ST-elevation myocardial infarction treated by primary percutaneous coronary intervention. Am J Cardiol. 2015;115(4):417–22. 10.1016/j.amjcard.2014.11.025.
Luo J, Li H, Qin X, et al. Increased risk of ischemic stroke associated with new-onset atrial fibrillation complicating acute coronary syndrome: a systematic review and meta-analysis. Int J Cardiol. 2018;265:125–31. 10.1016/j.ijcard.2018.04.096.
Jabbari R, Risgaard B, Fosbøl EL, et al. Factors associated with and outcomes after ventricular fibrillation before and during primary angioplasty in patients with st-segment elevation myocardial infarction. Am J Cardiol. 2015;116(5):678–85. 10.1016/j.amjcard.2015.05.041.
Weiss JN, Qu Z, Shivkumar K. Electrophysiology of hypokalemia and hyperkalemia. Circ Arrhythm Electrophysiol. 2017;10(3):e004667. 10.1161/CIRCEP.116.004667.
Nikolaou NI, Arntz HR, Bellou A, et al. European Resuscitation Council Guidelines for Resuscitation 2015 Section 8. Initial management of acute coronary syndromes. Resuscitation. 2015;95:264–77. 10.1016/j.resuscitation.2015.07.030.
Sabatine MS, Morrow DA, Giugliano RP, et al. Association of hemoglobin levels with clinical outcomes in acute coronary syndromes. Circulation. 2005;111(16):2042–9. 10.1161/01.CIR.0000162477.70955.5F.
Antzelevitch C, Burashnikov A. Overview of basic mechanisms of cardiac arrhythmia. Card Electrophysiol Clin. 2011;3(1):23–45. 10.1016/j.ccep.2010.10.012.
Hreybe H, Saba S. Location of acute myocardial infarction and associated arrhythmias and outcome. Clin Cardiol. 2009;32(5):274–7. 10.1002/clc.20532.
Piccini JP, Berger JS, Brown DL. Early sustained ventricular arrhythmias complicating acute myocardial infarction. Am J Med. 2008;121(9):797–804. 10.1016/j.amjmed.2008.04.024.
Mehta RH, Yu J, Piccini JP, et al. Prognostic significance of the development of new Q waves during clopidogrel treatment in patients with ST-elevation myocardial infarction undergoing primary percutaneous coronary intervention. Eur Heart J. 2012;33(4):478–85. 10.1093/eurheartj/ehr400.
Gao F, Zhou YJ, Wang ZJ, et al. Meta-analysis of the effect of statins on the incidence of atrial fibrillation after transcatheter aortic valve implantation. Am J Cardiol. 2016;118(4):569–75. 10.1016/j.amjcard.2016.05.049.
Priori SG, Blomström-Lundqvist C, Mazzanti A, et al. 2015 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: The Task Force for the Management of Patients with Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death of the European Society of Cardiology (ESC). Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC). Eur Heart J. 2015;36(41):2793–867. 10.1093/eurheartj/ehv316.
Zipes DP, Camm AJ, Borggrefe M, American College of Cardiology/American Heart Association Task Force and the European Society of Cardiology Committee for Practice Guidelines. ACC/AHA/ESC 2006 guidelines for management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: a report of the (Writing Committee to Develop Guidelines for Management of Patients with Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death). J Am Coll Cardiol. 2006;48(5):e247-e346. 10.1016/j.jacc.2006.07.010

Table 1A. Baseline characters for enrolled subjects

Characteristic	Toral	non-VF group, n= 128	VF group, n = 34	P value
Age（year）	65.13±12.01	64.01±12.02	69.26±11.19	0.501
Sex（male, n, %）	128(75.29%)	105(82.03%)	31(73.81%)	0.27
Hypertension（mmHg）	119 (70.00%)	89 (65.44%)	30 (88.24%)	1.000
Systolic Blood Pressure（mmHg）	126.50 (112.00, 140.00)	130.00 (114.00, 142.00)	116.50 (100.25, 128.00)	0.001
Diastolic Blood Pressure（mmHg）	73.21 ± 12.26	73.68 ± 11.66	71.32 ± 14.44	0.318
Pulse (bpm)	75.00 (65.00, 86.00)	74.00 (64.00, 84.00)	78.50 (70.00, 90.00)	0.081
Diabetes (n, %)	80 (47.06%)	51 (37.50%)	29 (85.29%)	0.604
Dyslipidemia (n, %)	6.08 (5.37, 8.24)	5.96 (5.20, 7.87)	6.88 (5.54, 12.50)	0.035
Smoking (n, %)	86 (50.59%)	75 (55.15%)	11 (32.35%)	0.309
Alcohol consumption (n, %)	86 (50.59%)	67 (49.26%)	19 (55.88%)	1.000
History of Coronary Artery Bypass Grafting (n, %)	55 (32.35%)	38 (27.94%)	17 (50.00%)	1.000
Chronic Heart Failure (n, %)	4 (2.35%)	4 (2.94%)	0 (0.00%)	1.000
Ventricular Fibrillation (n, %)	3 (1.76%)	3 (2.21%)	0 (0.00%)	1.000
Creatinine (mmol/L)	86.70 (77.10, 100.10)	85.30 (77.60, 97.50)	92.20 (70.40, 116.77)	0.366
Blood Urea Nitrogen ()	5.44 (4.35, 7.59)	5.17 (4.22, 7.12)	9.12 (5.73, 12.07)	0.000
Glucose (mmol/L)	79.00 (67.00, 93.00)	79.00 (68.00, 91.50)	79.50 (64.50, 105.50)	0.846
ALT (U/L)	51.50 (24.77, 155.60)	45.90 (24.18, 138.22)	79.95 (28.00, 321.48)	0.087
AST (U/L)	63.30 (59.30, 66.30)	63.10 (59.20, 66.20)	64.25 (61.52, 69.10)	0.069
Total Cholesterol (mg/dL)	1.53 (1.03, 2.17)	1.65 (1.08, 2.33)	1.15 (0.96, 1.40)	0.001
Triglycerides (mg/dL)	2.70 (2.07, 3.16)	2.70 (2.10, 3.16)	2.65 (1.87, 3.30)	0.721
LDL-C (mg/dL)	0.99 (0.88, 1.15)	0.98 (0.89, 1.14)	1.01 (0.84, 1.23)	0.599
HDL-C (mg/dL)	34 (20.00%)	0 (0.00%)	34 (100.00%)	0.000
Aspirin (n, %)	143 (84.12%)	131 (96.32%)	12 (35.29%)	0.000
Clopidogrel or Ticagrelor (n, %)	131 (77.06%)	120 (88.24%)	11 (32.35%)	0.087
Anticoagulation (n, %)	2 (1.18%)	0 (0.00%)	2 (5.88%)	0.248
Statins (n, %)	134 (78.82%)	124 (91.18%)	10 (29.41%)	0.000
Beta Blockers (n, %)	110 (64.71%)	101 (74.26%)	9 (26.47%)	0.001
CCB (n, %)	34 (20.00%)	33 (24.26%)	1 (2.94%)	0.222
Diuretics (n, %)	25 (14.71%)	21 (15.44%)	4 (11.76%)	1.000
ACEI/ARB/ARNI (n, %)	94 (55.29%)	88 (64.71%)	6 (17.65%)	0.183
Nitrates (n, %)	44 (25.88%)	38 (27.94%)	6 (17.65%)	1.000

ALT, Alanine Aminotransferase; AST, Aspartate Aminotransferase; LDL-C, Low Density Lipoprotein Cholesterol; HDL-C, High Density Lipoprotein Cholesterol; CCB,Calcium Channel Blockers; ACEI, Angiotensin-Converting Enzyme Inhibitors; ARB: Angiotensin II Receptor Blockers; ARNI: Angiotensin Receptor-Neprilysin Inhibitors.

Table 1B. Dignosis, medication and coagulation function for enrolled subjects

Characteristic	Toral	non-VF group, n= 128	VF group, n = 34	P value
hs-CRP(mg/L)	8.03 (2.41, 20.22)	6.65 (1.97, 17.19)	20.80 (12.66, 32.57)	0.000
NLR	4.06 (2.55, 6.53)	3.68 (2.55, 5.65)	5.45 (2.83, 9.14)	0.028
Respiratory Failure(n, %)	15 (8.82%)	3 (2.21%)	12 (35.29%)	0.478
Acute Renal Insufficiency(n, %)	8 (4.71%)	0 (0.00%)	8 (23.53%)	1.000
Abnormal Liver Function(n, %)	33 (19.41%)	19 (13.97%)	14 (41.18%)	0.716
Metabolic Acidosis(n, %)	20 (11.76%)	3 (2.21%)	17 (50.00%)	0.019
Pleural Effusion(n, %)	15 (8.82%)	2 (1.47%)	13 (38.24%)	1.000
Infection(n, %)	37 (21.76%)	20 (14.71%)	17 (50.00%)	0.798
DIC(n, %)	9 (5.29%)	0 (0.00%)	9 (26.47%)	0.966
Anemia(n, %)	24 (14.12%)	7 (5.15%)	17 (50.00%)	0.934
Thrombosis(n, %)	6 (3.53%)	1 (0.74%)	5 (14.71%)	0.795
Hypokalemia(n, %)	30 (17.65%)	10 (7.35%)	20 (58.82%)	0.019
VPB(n, %)	22 (12.94%)	6 (4.41%)	16 (47.06%)	0.842
VT(n, %)	18 (10.59%)	3 (2.21%)	15 (44.12%)	1.000
AF(n, %)	26 (15.29%)	10 (7.35%)	16 (47.06%)	0.379
APB(n, %)	11 (6.47%)	3 (2.21%)	8 (23.53%)	0.258
AT(n, %)	2 (1.18%)	1 (0.74%)	1 (2.94%)	0.248
AVB(n, %)	10 (5.88%)	3 (2.21%)	7 (20.59%)	0.000
INR	1.04 (0.98, 1.10)	1.03 (0.98, 1.08)	1.12 (1.02, 1.19)	0.000
APTT(s)	25.40 (23.60, 28.90)	25.30 (23.45, 27.70)	28.20 (24.90, 33.35)	0.011
Antithrombin III(%)	88.40 ± 13.08	90.11 ± 12.41	80.71 ± 13.44	0.000
PT(s)	12.00 (11.40, 12.70)	11.90 (11.35, 12.50)	12.70 (11.55, 13.38)	0.019
Prothrombin Time Activity(%)	94.00 (82.30, 102.70)	95.90 (84.45, 104.35)	80.90 (73.45, 89.58)	0.000
FDP(mg/L)	2.50 (1.90, 3.90)	2.30 (1.90, 3.50)	3.40 (2.50, 7.10)	0.001
Fibrinogen(g/L)	3.10 (2.39, 3.77)	2.96 (2.29, 3.59)	3.57 (2.89, 4.85)	0.001
D-Dimer(ng/mL)	0.70 (0.50, 1.00)	0.70 (0.50, 0.90)	0.84 (0.53, 2.40)	0.028
TSH(mIU/L)	1.22 (0.79, 2.04)	1.22 (0.81, 2.03)	1.27 (0.76, 2.18)	0.914
Thyroid Uptake(%)	43.40 (42.30, 44.50)	43.20 (42.30, 44.30)	46.90 (43.30, 47.20)	0.000
T4(ng/dL)	86.33 ± 19.16	84.52 ± 17.96	94.29 ± 22.35	0.011
T3(ng/dL)	76.55 (68.17, 84.76)	77.24 (70.55, 84.35)	67.09 (49.02, 89.92)	0.043
FT3(pg/mL)	2.81 (2.53, 3.08)	2.81 (2.53, 3.07)	2.92 (2.44, 3.37)	0.364
FT4(pg/mL)	0.92 (0.81, 1.07)	0.91 (0.81, 1.04)	1.03 (0.85, 1.45)	0.021
K(mmol/L)	4.08 (3.88, 4.36)	4.04 (3.86, 4.27)	4.36 (4.08, 4.58)	0.001
P(mg/dL)	1.07 (0.94, 1.21)	1.05 (0.94, 1.19)	1.19 (1.07, 1.29)	0.007
Cl(mmol/L)	104.00 (101.00, 106.00)	104.00 (101.00, 106.00)	102.50 (98.25, 104.75)	0.015
Na(mmol/L)	139.00 (136.80, 140.60)	139.10 (137.15, 140.70)	137.70 (134.62, 139.97)	0.037
Ca(mg/dL)	2.16 ± 0.11	2.16 ± 0.10	2.16 ± 0.13	0.880
PaCO2(mmHg)	23.70 (21.70, 25.20)	23.80 (21.80, 25.20)	23.05 (20.35, 25.27)	0.263
Lactic Acid(mmol/L)	2.60 (2.08, 3.19)	2.59 (2.06, 3.13)	2.81 (2.26, 3.26)	0.367
Osmolality(mOsm/kg)	290.90 (287.10, 295.00)	290.80 (287.30, 294.95)	291.80 (285.23, 296.30)	0.593
Anion Gap(mmol/L)	13.90 (12.17, 16.60)	13.40 (11.90, 15.70)	16.50 (14.90, 18.60)	0.000

hs-CRP, High Sensitivity C Reactive Protein; NLR, NLR (Neutrophil-to-lymphocyte ratio); DIC, Disseminated Intravascular Coagulation (DIC); VPB, Ventricular Premature Beats; VT, Ventricular Tachycardia; AF, Atrial Fibrillation; APB, Atrial Premature Beats; AT, Atrial Tachycardia; AVB, Atrioventricular Block; INR, International Normalized Ratio; APTT, Activated Partial Thromboplastin Time; PT, Prothrombin Time; FDP, Fibrinogen Degradation Products; TSH, Thyroid Stimulating Hormone; T4, Thyroxine (T4); T3, Triiodothyronine (T3); FT3, Free T3; FT4, Free T4; K, Potassium; P, Phosphorus; Cl, Chlorine; Na, Sodium; Ca, Calcium; PaCO2, Carbon Dioxide.

Table 1C. Inhosptial prognosis for enrolled subjects

Characteristic	Toral	non-VF group, n= 128	VF group, n = 34	P value
EF(%)	32.00（25.00,37.00)	33.00 (27.00, 37.00)	23.00 (20.00, 26.00)	0.000
E Wave Peak(m/s)	0.88 (0.71, 1.24)	0.84 (0.70, 1.18)	1.13 (0.76, 2.00)	0.056
A Wave Peak(m/s)	302.50 (234.50, 360.50)	297.00 (233.00, 353.00)	333.00 (268.00, 389.00)	0.130
EA Ratio	122.50 (106.25, 143.00)	122.50 (107.00, 143.00)	120.00 (101.50, 142.75)	0.587
LM(n, %)	123 (72.35%)	93 (68.38%)	30 (88.24%)	1.000
LCx(n, %)	3 (1.76%)	0 (0.00%)	3 (8.82%)	1.000
RCA(n, %)	53 (31.18%)	32 (23.53%)	21 (61.76%)	0.331
LMO(n, %)	20 (11.76%)	15 (11.03%)	5 (14.71%)	1.000
LADo(n, %)	31 (18.24%)	25 (18.38%)	6 (17.65%)	0.157
LCxo(n, %)	87.50 (65.50, 124.88)	83.00 (64.25, 118.62)	112.75 (83.25, 140.62)	0.003
RCAo(n, %)	141 (82.94%)	112 (82.35%)	29 (85.29%)	1.000
Gensini Score	9.78 (1.48, 28.50)	4.95 (0.90, 19.88)	31.28 (17.00, 42.39)	0.000
Stent(n, %)	53.45 (12.85, 201.00)	53.55 (12.67, 213.00)	52.80 (13.60, 168.10)	0.573
cTnI(ng/mL)	2422.00 (861.00, 9301.00)	1801.50 (795.75, 5809.75)	9380.00 (2267.00, 20130.00)	0.000
CK-MB(ng/mL)	20 (11.76%)	15 (11.03%)	5 (14.71%)	0.181
NT-proBNP(pg/mL)	43 (25.29%)	9 (6.62%)	34 (100.00%)	0.000
Thrombus Aspiration(n, %)	26 (15.29%)	2 (1.47%)	24 (70.59%)	0.000
MACE(n, %)	21 (12.35%)	4 (2.94%)	17 (50.00%)	0.025
Cardiogenic Death(n, %)	3 (1.76%)	0 (0.00%)	3 (8.82%)	0.423
Cardiogenic Shock(n, %)	8 (4.71%)	5 (3.68%)	3 (8.82%)	0.964
Acute In Stent Thrombosis(n, %)	39 (22.94%)	5 (3.68%)	34 (100.00%)	0.000
Recurrent Myocardial Infarction(n, %)	1 (0.59%)	1 (0.74%)	0 (0.00%)	1.000
Malignant Arrhythmia(n, %)	8.00 (6.00, 11.00)	7.00 (6.00, 10.00)	9.50 (7.25, 16.00)	0.003
Gastrointestinal Bleeding(n, %)	15 (8.82%)	4 (2.94%)	11 (32.35%)	1.000
Length of Stay(day)	18 (10.59%)	0 (0.00%)	18 (52.94%)	0.127
IABP(n, %)	23 (13.53%)	2 (1.47%)	21 (61.76%)	0.041
Endotracheal Intubation(n, %)	25 (14.71%)	0 (0.00%)	25 (73.53%)	0.062
Ventilator(n, %)	23 (13.53%)	0 (0.00%)	23 (67.65%)	0.041
Defibrillation(n, %)	23 (13.53%)	2 (1.47%)	21 (61.76%)	0.000

EF, Ejection Fraction; LM, Left Main Artery Lesion; LCx, Left Circumflex Artery Lesion; RCA, Right Coronary Artery Lesion; LMO, Left Main Artery Occlusion; LADo, Left Anterior Descending Artery Occlusion; LCxo, Left Circumflex Artery Occlusion; RCAo, Right Coronary Artery Occlusion; cTnI, Cardiac Troponin I; CK-MB, Creatine Kinase-MB Mass; NT-proBNP, NT-proBNP Peak; MACE, Major Adverse Cardiovascular Events; IABP, Intraaortic Balloon Pump.

Table 2 Logistic regression for predicters of ventricular fibrillation

	OR_univariable		OR_multivariable
	OR and 95%CI	P	OR and 95%CI	P
Hypertension (n, %)	3.961(1.316-11.916)	0.014
Diabetes (n, %)	9.666(3.519-26.555)	0.000	1.891(9.237-45.108)	0.006
Respiratory Failure (n, %)	24.182(6.312-92.644)	0.000
Metabolic Acidosis (n, %)	44.331(11.758-167.142)	0.000	2.537(17.807-125.008)	0.004
Pleural Effusion (n, %)	41.475(8.732-197.001)	0.000
Anemia (n, %)	18.429(6.678-50.859)	0.000	3.326(16.380-80.668)	0.001
Thrombosis (n, %)	23.275(2.620-206.755)	0.005
Hypokalemia (n, %)	18(7.041-46.019)	0.000	1.540(8.613-48.171)	0.014
Ventricular Premature Beats (n, %)	19.26(6.674-55.582)	0.000	6.156(33.226-179.336)	0.000
Atrial Premature Beats (n, %)	13.637(3.391-54.850)	0.000
Atrioventricular Block (n, %)	11.489(2.793-47.259)	0.001
Statins (n, %)	0.04(0.016-0.104)	0.000
Anion Gap (mEq/L)	1.242(1.102-1.399)	0.000
Thyroid Uptake (%)	1.477(1.209-1.804)	0.000
Stroke Volume (mL)	0.945(0.913-0.978)	0.001
Right Ventricle (mm)	11.953(3.280-43.555)	0.000
Pulmonary Hypertension (mmHg)	59.553(12.639-280.601)	0.000
IABP (n, %)	15.783(4.627-53.838)	0.000
Right Coronary Artery Lesion (n, %)	0.015(0.000-1.646)	0.914

IABP: Intra-aortic Balloon Pump.

No competing interests reported.

Early Predictive nomogram of Ventricular Fibrillation after Percutaneous Coronary Intervention in Acute Myocardial Infarction using LASSO-Logistic regression

Status:

Version 1

Abstract

Objective

Methods

Results

Conclusion

Figures

Introduction

Method

Results

Discussion

Declarations

Funding

Author Contribution

Acknowledgements

Data Availability:

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1