Graph-based deep learning methods for Natural Language Processing have harnessed tremendous research interest for their ability to model textual structure and domain knowledge. One challenging problem from the clinical domain that requires the modeling of these characteristics is automated medical coding. Automated medical coding is the process of codifying clinical notes to their appropriate diagnosis and procedure codes from the standard taxonomies such as ICD (International Classification of Diseases) and CPT (Current Procedure Terminology) automatically. The manual coding process involves the identification of entities from the clinical notes followed by the querying of a commercial or non-commercial medical codes Information Retrieval (IR) system. We aim to automate this manual process by modelling it as an information completion problem. We propose GrabQC, a Graph-based Query Contextualisation method that automatically extracts queries from the clinical text, contextualises the queries using a Graph Neural Network (GNN) model and obtains the ICD Codes using the IR system. We also introduce a method for labelling the dataset for training the model. We perform experiments on two datasets of clinical text in three different setups to assert our approach's effectiveness. The experimental results show that our proposed method performs better than the compared baselines in all three settings.