Heart failure, often called congestive heart failure (CHF), is a condition where the heart cannot pump blood to meet the needs of metabolic tissues. To date, there has been a lot of research on methods of gene expression analysis to produce the information needed regarding the diagnosis and prognosis of various diseases, including heart failure. The 3-dimensional data analysis was finally brought to the context of computational biologies, such as gene-sample-time, patient-record-time, druggene-dose expression data, and so on. This 3-dimensional data is essential to help understand the complex biology and physical processes related to a disease's progress and response to stimulus/drug/therapy. δ-Trimax is a triclustering algorithm used to find a tricluster with an MSR smaller than the predetermined threshold. For determing the threshold, we use the K-means clustering algorithm with the Silhouette Method to find a threshold that can maximize the quality of the tricluster (minimum MSR accompanied by maximum tricluster volume). In this study, δ-Trimax was implemented with several scenarios in gene expression data of patients with and without PAH (Pulmonary Arterial Hypertension). PAH is a kind of symptom of heart failure disease. The low value of the Tricluster Quality Index (TQI) indicate best triclustering result. In this case, we found that the scenario with a threshold δ of 0.08156218 and produced a higher quality tricluster than the other scenarios.