To the Editor In their systematic review of the Lariat (SentreHeart Inc) left atrial appendage (LAA) ligation system, Chatterjee et al1 question the safety classification of the device, citing adverse events recorded in the US Food and Drug Administration’s (FDA) Manufacturer and User Facility Device Experience (MAUDE) database.1
While true complication rates cannot be ascertained from the MAUDE database, more than 3000 Lariat devices have been implanted in the United States since FDA approval in 2008 (email communication with a SentreHeart Inc representative, July 22, 2015), providing some perspective for the magnitude of adverse events reported by Chatterjee et al. Furthermore, review of the MAUDE database for Lariat-related adverse events through June 2015 implicates operator error in 6 of 45 case reports (13%), including 3 LAA transections resulting from device withdrawal against resistance or before suture cutting, suggesting that inadequate operator training likely contributed to events. Surgery was required in another patient after Lariat ligation was attempted despite anatomical contraindications. This indicates that rigorous operator training protocols and improved patient selectivity would readily effect complication rates. Safety concerns of the WATCHMAN (Boston Scientific) device were addressed by implementing similar measures, accomplishing step-wise improvements in procedural safety, reflected in the PREVAIL trial of the WATCHMAN.2 The critique by Chatterjee et al of the FDA’s 510(k) process for Lariat approval relies heavily on their MAUDE database analysis. Interestingly, since FDA approval in March 2015, 5 WATCHMAN device embolizations and 3 procedural deaths have been recorded in the MAUDE database. Comparatively, in the randomized clinical PREVAIL trial of the WATCHMAN device, only 2 device embolizations and no procedural deaths were reported after 252 implants, reflecting that real-world use risks are not readily identified in clinical trials. Reviewing approximately 3000 510(k) applications annually, the FDA balances prompt integration of scientific innovations with ensuring safety of approved devices, recognizing that all off-label device uses cannot be contemplated within their framework.3 Nevertheless, system safeguards like the MAUDE database are an effective mechanism for continued device surveillance. Current MAUDE database information regarding Lariat events, however, appears inconclusive and inadequate to make new determinations regarding the device.