Introduction: Pan-CT (chest, abdomen, pelvis) is used for malignancy screening in patients with AIS (acute ischemic stroke); however, its diagnostic yield in the inpatient setting has not been studied. In this retrospective analysis, we estimate yield of inpatient pan-CT for malignancy in AIS and describe its associated diagnostic pathways and effects on length-of-stay (LOS) and stroke management. This may guide patient selection and timing of pan-CT screening after AIS. We hypothesized inpatient pan-CT would increase LOS and would have low diagnostic yield. Methods: This was a single-center, retrospective, cross-sectional study (2020-2023). Patients were identified from billing data using ICD-10 code for AIS and concurrent CPT code for pan-CT performed for malignancy screen. We retrieved age, sex, LOS, imaging, and pathology results. We assessed the three-territory sign (TTS)–a proposed MRI marker for malignancy-related AIS, defined as infarcts in ≥3 vascular territories. Pan-CT results were categorized as positive, negative, or indeterminate for malignancy. Mann-Whitney test compared LOS between pan-CT categories. Chi-square test tested association between TTS and malignancy diagnosis. P<0.05 was considered significant. Results: We identified 135 patients with AIS and inpatient pan-CT (mean age 67±14 years, female 46%). Pan-CT was positive in 6 (4%), negative in 79 (59%), and indeterminate in 50 (37%). Of the indeterminates that had further inpatient testing (n=12), 1 (8%) had malignancy. With deferred outpatient testing (n=36), 19 (53%) did not complete workup, and 17 (47%) had no malignancy (Figure). Further inpatient testing was associated with increased LOS (P=0.04) (Table). Presence of TTS (19%) was not related to malignancy diagnosis (P=0.99). Malignancy diagnosis on inpatient pan-CT prompted anticoagulation treatment in 2 (1%). Excluding loss to follow-up, overall diagnostic yield was 6% (7/114). Conclusion: Inpatient pan-CT detects occult malignancy comparable to prior estimates of malignancy prevalence in AIS. It is limited by high proportion of indeterminate results, changes stroke management in few, and leads to additional tests that increase length of stay; however, deferred outpatient testing has high attrition. Further research is necessary into stroke and cancer outcomes with inpatient vs. outpatient malignancy screening in AIS. TTS may not be a reliable single predictor of pan-CT results, but analysis was limited by low sample size.
Colah et al. (Thu,) studied this question.