Background: Knowledge about the influences of maternal postpartum anxiety disorders (PAD) on infant development is limited. Aim of this present study is to evaluate the influence of PAD on infant attachment. Methods: In a longitudinal study, self-reported anxiety symptoms of N = 70 mothers (N = 28 with PAD diagnosed according to the DSM-IV, N = 42 controls) were examined in the postpartum period and one year later. Infants' attachment was observed in the Strange Situation Test (SST) at the age of 12–24 months. Results: Results indicate a strong relationship between PAD and infant attachment: infants of mothers with PAD were significantly more likely to be classified as insecure or disorganized than infants of control mothers. Logistic regression analysis led to a significant model with 76.8 % correct classification of infant attachment dependent on the maternal fear of anxiety associated body sensations (OR = 4.848) in the postpartum period. Including maternal sensitivity and interaction behavior, only maternal intrusiveness was additionally associated with infant attachment (ρ = 0.273, p < .05; OR = 45.021, p = .153). Limitations: Participants were highly educated. Different anxiety disorders included led to a heterogenous sample. Generalization is diminished. Maternal sensitivity was measured on a global scale, and body tension was self-reported. Conclusions: PAD plays a crucial role in the development of infant attachment. Interaction-focused interventions, helping mothers to decrease intrusiveness, and body-focused interventions, helping mothers to deal with their fear of anxiety symptoms, might be promising pathways to buffer the influence of PAD on infant attachment.