Microblogging has become one of the most crucial tool for expressing and sharing the opinions and views of everyday life events. Digital channels are being used to monitor public health issues on the Internet. Twitter is a very popular source that provides tweets related to the sentiment of the public during the COVID-19 pandemic. Many researchers have used tweets to monitor the opinion of the people towards the coronavirus vaccine, mental health problems, treatment received by the doctors, impact of lockdown, etc. However, these works were mostly limited to the first and second waves of the pandemic. In this work, we aim to study the impact of the third wave of the pandemic, which started in December 2021 in India. We accomplished this by collecting tweet data set of two months, i.e., December 2021 and January 2022, discussing COVID-19 and having country code as ``IN". We employed the Latent Dirichlet Allocation (LDA) technique for topic modeling and labeled each tweet message with the topic words that best describe it. We also utilized sentiment labels for each tweet and analyzed the distribution of different topics across different sentiment labels. This helped us to analyze the perspectives and sentiments of the people with respect to different topic discussions. Our analysis discovered that the two most discussed topics were ``precautionary measures" like get well soon, stay safe, wear mask, etc., and ``vaccine" where people have discussed about its effectiveness and vaccination drive in India. We found that people mostly had neutral sentiments for the former topic while for the latter, overall sentiment polarity was negative, reflecting peoples' mistrust in the COVID-19 vaccine.