(WHS-L2.01) DEEP VEIN THROMBOSIS AFFECTS HEALING OUTCOMES IN PATIENTS WITH PYODERMA GANGRENOSUM: A SINGLE-CENTER PROSPECTIVE CASE-CONTROL STUDY
Thursday, May 16, 2024
10:30 AM – 11:30 AM East Coast USA Time
While inflammation and thrombosis are closely integrated processes, there has been little exploration between PG and thromboembolism. In this single-center case-control study, we assess the association between a history of DVT and healing outcomes in PG patients. This prospective case-control study was approved by the OHSU IRB and was conducted between December 2019 and July 2023. Inclusion criteria was defined as patients with classic, ulcerative PG of the lower extremity confirmed by a PARACELSUS score of ≥10 from the Pyoderma Gangrenosum Study Registry (PYGAS) at OHSU. Data regarding patient demographics, comorbidities, and healing outcomes were extracted for further analysis. Patients were then grouped based on the presence or absence of ultrasound-proven lower extremity DVT prior to PG diagnosis. Cox regression and Kaplan-Meier survival analysis were applied to detect the effect of DVT on healing status within a 1-year period from initial presentation. Chi-squared tests were used to compare the descriptive statistics of the DVT and non-DVT groups. Statistical significance was established as P ≤0.05. In total, 99 patients met inclusion criteria, with 13 (13.1%) having a history of DVT and 86 (86.9%) without. 121 total ulcers were included in the analysis—patients presenting with sequential or recurrent ulcers were recorded as separate encounters. The mean age of presentation for patients with and without DVT was 57 (SD +/-11) and 53 (SD +/-18), respectively (P=0.60). There were significantly more males presenting with DVT than females (P =0.04). There were no significant differences in race, ethnicity, or smoking history between the two groups. There were also no significant differences in confounding comorbidities such as body mass index, history of solid or hematological malignancy, or history of autoimmune or inflammatory disorder between the DVT and non-DVT groups. Evaluation of median healing time between DVT and non-DVT groups using Kaplan-Meier survival analysis confirmed a significant increase in median healing time among patients with a history of DVT. The median healing time for the DVT and non-DVT group was 224 and 145 days, respectively (p < 0.04). Univariate and multivariate Cox proportional hazard modelling was also used to assess the influence of various predictor variables on wound healing. When adjusting for obesity and smoking, the two factors most prevalent in our patient population and factors known to influence wound healing, a history of DVT remained the most significant influence on the healing time (p=0.05). Single predictor modelling also demonstrated that variables such as diabetes mellitus, solid malignancy, and multi-ulcer presentation did not significantly alter healing time. In conclusion, these findings suggest a novel association between a history of DVT and poorer healing in patients with PG, indicating the importance of identifying a history of thromboembolism when evaluating PG patients.