Advanced technologies in bioinstrumentation allows easy monitoring of biometric signals such as electrocardiogram (ECG) and respiration. In order to improve unreliable monitoring due to missing RR intervals (RRIs), this paper proposes a missing RRI complement method based on respiratory features. The proposed method first selects respiratory features from the measured data based on Granger causality, and then complements the missing RRIs based on a dynamic linear model (DLM) for RRIs with selected features. The performance of the proposed method was evaluated by comparison with standard spline interpolation, standard regression, and a vector autoregressive (VAR) model. The results are discussed in terms of the effectiveness of respiratory feature selection and utilization of the DLM to capture temporal fluctuations.