Page 18 - Journal of Structural Heart Disease Volume 2, Issue 5

219
Original Research Article
Table 1: Summary characteristics of patients studied
device implanted (p = 0.007 and p = 0.006, Kruskal- Wallis H Test, STATA; Figure 1). ROC curves were created for each prediction score to assess sensitivi- ty and speci city. Because of the limited number of 35-mm devices implanted in the study period (n = 2), 30- and 35-mm devices were grouped together for the ROC analysis. The area under the curve for the proposed sizing score was 0.8846 (STATA, Figure 2). The cutpoint score of ≥2 had 75% sensitivity and 92% speci city for a device diameter ≥30 mm.
Discussion
Percutaneous PFO closure has been proposed as a safe alternative or adjunct therapy to antiplatelet medication and anticoagulation in patients with cryp- togenic stroke or peripheral embolism [10, 17-19]. Observational studies have suggested a substantial bene t to PFO closure in the secondary prevention of neurologic and vascular events when compared to medical therapy [20, 21]. Three randomized controlled trials assessing PFO closure for secondary prevention of cryptogenic stroke failed to reach their primary ef- cacy endpoints [22-24]. However, per-protocol and as-treated analysis of one of the randomized trials did show a signi cant bene t of closure over medical therapy in the prevention of recurrent ischemic stroke and death [22]. At this point, percutaneous closure of PFOs in the U.S. is considered “o -label,” and there is a wide heterogeneity in device size selection.
Our study suggests that patients undergoing per- cutaneous PFO closure who display certain echocar- diographic ndings will likely require closure devices with larger disc diameters. The presence of ASA and increasing length of PFO tunnel on TEE or ICE were independently associated with larger device size. In addition, our ndings suggest that a sizing score may be used to facilitate identi cation of patients that are likely to require larger devices. A scoring system consisting of 1 point for the presence of ASA, 1 point for PFO tunnel length >10 mm, and 1 point for male sex was statistically associated with increasing de- vice size. Scores of 0 and 1 were associated with the 25 mm device. A score of 2 was associated with the 30 mm device, and a score of 3 was associated with the 35 mm device.
No. of Patients
36
Patient Characteristics Age
Male Height (cm) Weight (kg) BSA
Size of Device Implanted 25 mm Cribriform
30 mm Cribriform
35 mm Cribriform
Indication for Procedure Cryptogenic Stroke/TIA High DVT Burden Refractory Hypoxia TTE, Pre-Procedure, no. TEE, Pre-procedure, no.
ICE, Intra-procedure, no.
BSA = Body Surface Area, calculated by DuBois criteria. TIA = transient ischemic attack.
DVT = deep venous thrombosis.
TTE = trans-thoracic echocardiography.
TEE = transesophogeal echocardiography. ICE = intra-cardiac echocardiography.
Mean (STD) 56.7 (13.8) 13 (36%) 168.7 (11.1) 86.3 (25.5) 1.9 (.28)
18 (50%) 16 (44%) 2 (6%)
27 (75%) 3 (8%)
6 (17%) 26 (72%) 19 (53%) 12 (33%)
was a change from a 30- to a 35-mm diameter device. These cases were categorized by the nal device size implanted. All procedures were performed using the ASOC device in one of three disc diameters: 25 mm (18 patients, 50%), 30 mm (16 patients, 44%), or 35 mm (2 patients, 6%).
The results of univariate analysis are listed in Table 2. The presence of ASA (p = 0.027, Fisher’s exact test) and PFO tunnel length >10 mm (p = 0.038, Fish- er’s exact test) were independently associated with increased device size.
To guide sizing, a list of hypothesized sizing scores and their included variables is available in Table 3. A scoring system of 1 point for male sex, 1 point for ASA, and 1 point for PFO tunnel >10 mm in length was signi cantly associated with the size of closure
Venturini, J.M. et al.
Device Sizing for Percutaneous PFO Closure