This bibliometric study explores the evolving field of multimodal sentiment analysis by analysing a comprehensive dataset extracted from the Scopus database. Using VOSviewer for bibliometric and co-occurrence network analysis, we examine key trends, influential authors, institutions, journals, and countries driving the research. A total of 2310 authors, 1398 institutions, and 333 sources were analysed, with strict thresholds for document and citation counts to ensure meaningful insights. Our findings reveal the dominance of leading researchers and highlight prominent institutions such as Nanyang Technological University. The countries that significantly impact this area include the United States, China, and India. Several journals, including neurocomputing and information fusion, also strongly contribute to this area. We also analysed the top co-occurring keywords, emphasizing the growing focus on multimodal sentiment analysis, deep learning, multimodal fusion, and attention mechanisms. This study highlights the trends and growth of multimodal sentiment analysis, enabling future researchers to better understand the key drivers of innovation and research in this area.