Prediction model for Thyrotoxic Atrial Fibrillation

Husna, Amity University Kolkata

What is Thyrotoxic Atrial Fibrillation (TAF)? 

Hyperthyroidism is a medical condition in which there is an overactivity of the thyroid gland which leads to rapid heartbeat and excessive production of thyroxine hormone in the body. The most common severe complication of hyperthyroidism in an individual is Thyrotoxic Atrial fibrillation (TAF). Atrial fibrillation is the condition of an irregular or rapid heart rate which is known as arrhythmia and when it occurs due to hyperthyroidism, it is known as Thyrotoxic Atrial fibrillation. The condition of TAF can give rise to thromboembolism in which there is a blockage of the blood vessel, and it may also provoke heart failure and increase the event of cardiovascular mortality. The incidence of Thyrotoxic Atrial Fibrillation (TAF) is majorly seen in working-age individuals. So, its prevention is very crucial. Early identification of the individuals prone to TAF would improve the management of thyrotoxic patients. However, to date, there’s no such instrument that can establish the risk factor of this condition in an individual.                         

Why does TAF predictor need to be developed? 

Although a fairly large number of Thyrotoxic Atrial Fibrillation (TAF) predictors have been identified, they consider only some variables like advanced age, associated cardiovascular diseases, and male gender. However, severe factors like prolonged duration of hyperthyroidism and increased heart rate are the less investigated predictors of TAF. Another prevalent condition associated with TAF is the Nonimmune genesis of thyrotoxicosis. However, it is considered to be caused only in old age patients. Moreover, some studies have confirmed a new set of risk factors that increase the chance of TAF like obesity, presence of chronic kidney disease, proteinuria (a condition of discharging protein in urine), increased levels of hepatic transaminases, and C-reactive protein. In some cases, the findings regarding thyroid hormones level have been controversial this is because the investigation of hyperthyroidism, doesn’t consider the association of free triiodothyronine (fT3) or free thyroxine (fT4) level with TAF frequency. Therefore, many TAF predictors are known but with insufficient and controversial information. Recently, a retrospective study published in the BMC Endocrine Disorders aims to build a model that can predict Thyrotoxic Atrial Fibrillation (TAF) and rank TAF predictors. This TAF predictor uses machine learning techniques. It is a well-known fact that machine learning can instantly improve the accuracy of prediction, so it was used to develop this model. The application of machine learning techniques has even yielded promising results in the medical field. Machine learning is a data-driven approach that can identify complex interactions between variables without the need to pre-specify these relationships. It starts with the patient-level observations then the algorithms sift through vast numbers of variables and simultaneously look for combinations that can reliably predict the outcomes. Due to these reasons, machine learning is an excellent method for constructing prediction instruments. 

Conclusion of the study:

 A machine learning model which predicts TAF with 84% accuracy has been developed. By this study, the TAF risk factors with the highest predictive ability were identified; a number of them include, Premature Atrial Contraction (PAC), premature ventricular contractions (PVC), age, heart rate during hyperthyroidism, and hyperthyroidism duration. These risk factors are considered to be the new TAF predictors. This study could serve as a basis for further research focused on the prediction of TAF and facilitate the management of thyrotoxic patients. These results could be considered in the development of TAF risk scales and can be introduced to clinical practice as it has the potential to reduce the incidence of Thyrotoxic Atrial fibrillation (TAF).

Also read Da Vinci Family Tree: An intriguing DNA project!

Source: Ponomartseva, Daria Aleksandrovna, et al. “Prediction Model for Thyrotoxic Atrial Fibrillation: A Retrospective Study.” BMC Endocrine Disorders, vol. 21, no. 1, July 2021, p. 150. BioMed Central, doi: http://10.1186/s12902-021-00809-3.

• The Corrosion Prediction from the Corrosion Product Performance
• Nitrogen Resilience in Waterlogged Soybean plants
• Cell Senescence in Type II Diabetes: Therapeutic Potential
• Transgene-Free Canker-Resistant Citrus sinensis with Cas12/RNP
• AI Literacy in Early Childhood Education: Challenges and Opportunities

About the Author: Husna is an undergraduate student of B.TECH Biotechnology at Amity University Kolkata. She is a research enthusiast in Immunology and Immunotherapy but she has a keen interest in various other Bioscience subjects as well. She is constantly focused on improving her knowledge and laboratory skills through various internships. She is a Scientific content writer who has knowledge in diverse backgrounds of Biotechnology.