Total body PET/CT systems, which enable unprecedented image quality and ultrahigh sensitivity, are widely utilized for diagnosing and treating diseases like tumors. Unlike regular protocols, dual-time-point imaging (DTPI)– where patients undergo a dual PET/CT scan to enhance lesion contrast – exposes them to higher radiation doses due to an additional CT scan for PET attenuation correction and anatomical localization. To mitigate radiation exposure, we introduce STMDiff, a spatiotemporal matching diffusion model, which reuse CT images from first scanning time point for PET attenuation correction at second scanning time point. Spatiotemporal matching strategy implemented with contrastive learning aims to find the k-best-matched CT images, which enriches the multimodal features of STMdiff and bypasses the cross-modal registration, facilitating the generation of attenuation-corrected (AC) PET images alleviating alignment errors. Both qualitative and quantitative results illustrate that the AC PET images from STMDiff not only obtain the best quantitative scores (PSNR: \(37.72 \pm 6.85\) dB; SSIM: \(0.96 \pm 0.03\) ; RMSE: \(2.35\pm 1.03\) ), but also preserve metabolic information. Moreover, clinical assessment results show that the standardized uptake value (SUV) distribution of our method is more consistent with that of real AC PET images (Our code is available at https://github.com/LEE12365/STMDiff ).