Abstract
Sickle cell disease (SCD) has been associated with an increased incidence of pulmonary embolism (PE) but not an increased incidence of deep vein thrombosis (DVT). Because of this and other chest complications associated with SCD, these individuals undergo frequent imaging. Computerized tomographic pulmonary angiography (CTPA) is the most common diagnostic imaging test for PE but repeated CTPAs in a chronically ill, younger population with potential renal dysfunction may not be optimal. Ventilation perfusion (VQ) scintigraphy is an alternate imaging modality for PE with a generally lower radiation exposure. However, the diagnostic utility of VQ in SCD is uncertain as ventilation/perfusion mismatches from other diseases, such as acute chest syndrome, may be misdiagnosed as PE. With this in mind, we investigated the performance of CTPA and VQ for the diagnosis of PE in patients with SCD.
Material and Methods: We performed a retrospective cohort study of all adults with SCD imaged for PE with either CTPA or VQ scan at our institution, for the years 2000-2016. Our imaging algorithms recommend VQ scans for patients with normal chest radiographs (CXR) and when CTPA is contraindicated. Demographics, age at imaging and CXR results prior to PE imaging were collected. Results of index PE imaging with VQ or CTPA were reviewed and concordance with follow-up imaging within 90 days was noted. When available, venous duplex results for positive and indeterminate imaging were analyzed. Repeated imaging over the 16-year period was documented. Summary statistics, t-tests, Chi Square and Fisher exact tests were used as appropriate with significance at p<0.05.
Results: The cohort comprised 245 individual patients with SCD, 58% (141) men, mean age 33 ± 10.5 years. Index imaging was VQ in 63% (154) and CTPA in 37% (91). CXRs were performed in 96% (236/245) and were normal in 60.6% (143/236). Of those who had a CXR, Index VQ or CTPA was negative in 87.8% (207), positive in 4.2% (10) and indeterminate/chronic in 8.9% (19/2). There was no difference by imaging technique in the %positive PE diagnosed (VQ vs CTPA: 3.4% vs 5.7%, p=0.39)or in the number of indeterminate/chronic PE scans (VQ vs CTPA: 9.5% vs 8.0%, p=0.69). Of the ten positive PEs, half had duplexes; of these, 3/5 had lower extremity thrombosis; 2/10 patients had vascular access devices as a potential source. Of the 19 indeterminate exams, 2 of the 9 duplexes performed were positive.
Repeat imaging studies. 9.4% (23/245) of the study population had repeat imaging within 90 days of index exam, concordant in 54.2% (13/24). 36.7% (90/245) had at least one additional imaging study for PE after more than 90 days, during a different clinical presentation. Patients with index VQ has more second imaging exams than those with index CTPA, 52.6% vs. 36.2% (p=0.01). None of the 5 CTPA positive patients were repeated within 1-3 days but 3/5 positive VQs were repeated within 1-2 days of diagnosis. Of these, 1 was negative by CTPA. Of the 7.8% (19/245) patients with indeterminate results, 47% (9/19) underwent reimaging within 90 days. Seven indeterminate VQ scans were reimaged within 3 days by CTPA; 6 were negative and one was diagnosed as chronic PE. One CTPA was reimaged by VQ on Day 7 and was considered positive. Although additional imaging was more likely when the index examination was positive or indeterminate for both CTPA and VQ scan (p<0.01, both), 4.6% (10/216) with negative index imaging also had repeat imaging within 90 days. These results were again negative for PE in 90% (9/10) while 1 demonstrated new findings at Day 57 that did not change management.
Prevalence of PE: Over the 16-year study period 11% (27/245) of patients were diagnosed with PE at least once. Multiple exams were frequent: 47% (114/245) had at least 2, 15% at least 3 (36/245), and 7% (17/245) had 4 or more studies for PE. One patient underwent 21 scans (18 VQ, 3 CTPA).
Conclusion : In sickle cell disease, the greater index of suspicion for PE may result in a lower rate of %positive PE imaging than is generally acknowledged, for a given presentation. However, over an extended time period (16 years), 11% of patients had evidence of PE. In SCD patients with suspected PE, VQ scan and CTPA perform comparably when imaging recommendations are guided by CXR results.
Billett: Janssen Pharmaceutical: Research Funding.
