Article Text
Abstract
Background To identify the relevant parameters for SLE-VPO evaluation on computed tomography (CT), and try to develop a CT image based evaluation system for SLE-VPO.
Methods Patients with SLE gastrointestinal involvement from two independent retrospective cohorts were included in our study. All patients fulfilled the ACR 1997 revised classification criteria for SLE. CT scan at least once is necessary prior to glucocorticoid treatment in all enrolled patients. The primary endpoint was set to time to GI functional recovery, defined as intake 50% ideal calories POPO50. A new CT scoring system for SLE-VPO was established in the derivation cohort and validated in the validation cohort.
Results A total of 90 and 47 patients with SLE gastrointestinal involvement were enrolled in the derivation cohort and the validation cohort, respectively. The time to PO50 was significantly correlated with extent of anatomical involvementP ≤0.0015 and thickness of the intestinal wallP=0.0008 in the derivation cohort. A CT scoring system was developed with the combination of extent of anatomical involvement and thickness of the intestinal wall. CT score for VPO was positively correlated with patients time to PO50 (r=0.47, p<0.0001 in derivation cohort, r=0.44, p=0.0018 in validation cohort), and to a lesser extent correlated with the length of hospital stay (r=0.27, p=0.01 in derivation cohort, r=0.21, p=0.1577 in validation cohort). Patients with a CT score ≤2 tend to have a more rapid reversible course with an average time to PO50 of 5.72 and 7.44 days in the derivation cohort and the validation cohort, respectively; while patients with a CT score >2 have a significantly prolonged recovery with an average time to PO50 of 14.55 (p<0.0001) and 18.45 (p=0.0064) days, respectively. More TPN was prescribed in the high CT score group in both cohorts (74.47% vs 51.16%, p=0.0289 in the derivation cohort; 80% vs 40.74%, p=0.0089 in the validation cohort). And the proportions of patients underwent TPN ≥7 days were also much higher in the high CT score group in both cohorts (p<0.05).
Conclusions The terminology of VPO and a CT-based scoring system may facilitate more accurate assessment and individualized management for SLE patients with GI involvement. It also meant to be helpful in terms of future clinical trial design and in-depth mechanistic research for this unique complication of lupus.
Funding Source(s): This work was supported by the National key research and development program of China (2017YFC0909002), the National Key Research and Development Program of China(2016YFC0903900).