In this study of 203 patients with appendicitis who underwent preoperative rsDECT, we found that PVP, VNC, and combined PVP + VNC images were generally sufficient for detecting and numbering appendicoliths, suggesting that TNC acquisition may not always be necessary. Some appendicoliths were overlooked on PVP and VNC images, often being small or lower in attenuation; the clinical significance of missing these smaller appendicoliths remains uncertain. Notably, larger appendicoliths were significantly associated with complicated appendicitis, underscoring the importance of careful interpretation. Eliminating routine TNC acquisition could substantially reduce radiation exposure, a particularly meaningful consideration in the predominantly younger appendicitis population.

Among our cohort, 71 out of 203 patients (35%) had appendicoliths, aligning with prior CT-based studies reporting rates from 25.5% to 47.1% [7, 13, 17, 18]. Some reports suggest CT underestimates prevalence compared to pathology [19,20,21,22]. While CT may not detect every appendicolith, it remains crucial in appendicitis management, highlighting the importance of accurate preoperative imaging. Our findings confirmed the association between appendicoliths and complicated appendicitis, as reported in several studies [6, 13, 16, 17, 18, 20,21,22,23,24,25,26,27,28,29,30,31,32,33], though conflicting results have also been reported [34,35,36,37].

We observed substantial-to-almost-perfect agreement between PVP, VNC, and PVP + VNC with TNC in detecting and counting appendicoliths with kappa values comparable to Kim HY et al. [38]. Prior studies, such as by Mariadason et al. [19], have shown variable inter-reader agreement (kappa 0.48–0.83), underscoring diagnostic challenges in appendicolith detection. Incorporating VNC as part of DECT may help improve agreement, as calcifications can theoretically be easier to detect on these images. Although we did not formally assess observer agreement, we note that accurate detection of appendicoliths on PVP requires recognizing their higher attenuation compared to muscle-not the inflamed appendix wall-which can be subtle (Fig. 2). While VNC did not outperform PVP in detection or numbering in our study, this was not the primary research question and our sample size was limited. Nonetheless, we consider VNC a valuable component of DECT for calcification detection, with trade-offs mainly in reader time and visual fatigue, rather than radiation, as no additional scan is required.

Appendicoliths overlooked on PVP + VNC were smaller and lower in attenuation (Fig. 3), reducing conspicuity. While VNC theoretically improves visibility of dense foci, it may reduce calcification size and attenuation, impairing detection. Cheng Y et al. [39] found urinary calculi appeared smaller on VNC than on lower-keV images. A meta-analysis reported pooled sensitivity of only 78.1% for DECT (mostly dual-source, dual-energy platform) in detecting urinary stones [10], attributing misses to thicker slices (> 2 mm). The rsDECT platform which uses two-material decomposition, is known to have lower tissue differentiation than dual-source systems [40]. Bone windows may aid detection [41], but small, low-density stones remain challenging. These overlooked stones may be clinically insignificant, as larger size was independently associated with complications in our study-consistent with prior reports [13, 18, 20, 33]. Proposed size cutoff varies widely, from 2.48 mm [20] to 10 mm [13]. In our analysis, a 5-mm cutoff offered reasonable specificity but low sensitivity. Other reported predictors such as base location [18] or obstruction [42] were not significant in our cohort. We did not evaluate mural enhancement, which has been suggested as a predictor [18].

Our study has limitations. First, it is a retrospective single-center study, selection bias may be present. Second, although TNC was used as the reference standard, pathology may detect more appendicoliths [19,20,21,22]. We did not have consistent surgical or pathological documentation, as appendicoliths are not routinely recorded in operative or pathology reports at our hospital. Therefore, we relied on radiologist consensus on TNC as the reference standard, acknowledging this is imperfect and may underestimate the true prevalence of appendicoliths. For this reason, we chose to evaluate diagnostic agreement with Cohen’s kappa statistics instead of performance analyses. Third, although two radiologists independently reviewed all examinations, discrepancies were adjudicated by a third senior radiologist to ensure consistent final interpretations. While this consensus process enhanced reliability and clinical applicability, it precluded formal inter-reader agreement analysis. Fourth, most patients still received TNC because scan protocols were selected according to individual radiologists’ preferences, and baseline familiarity with TNC was high in our department. Fifth, our findings are specific to rsDECT platform and may not generalize to other technologies. Although all three major DECT configurations can generate VNC images, the accuracy and quality of material decomposition may differ between platforms, and their performance in reliably depicting appendicoliths remains to be evaluated. Sixth, our sample size limited subgroup analyses for rare features. Seventh, this study did not aim to assess the utility of DECT over single-energy CT (SECT), as addressing that question would require a much larger sample size and different evaluation methods. Such studies would be valuable, as they could clarify whether shifting from SECT to DECT provides substantial benefits, particularly given that many current practices still rely on SECT as the primary workhorse modality. Lastly, since appendicoliths are subtler on PVP than TNC, radiologists should interpret PVP images carefully, especially in the absence of other signs of complicated appendicitis.

In conclusion, PVP and VNC from rsDECT reliably detected and numbered appendicoliths, which may obviate the need for TNC scans. Eliminating TNC could also reduce radiation by over 50%, a meaningful advantage given that appendicitis often affects younger patients. Small, low attenuation appendicoliths were sometimes overlooked; although larger appendicoliths were independently linked to complicated appendicitis, the clinical significance of missed smaller appendicoliths remains uncertain due to lack of surgical or pathological outcome data.