The application of perventricular device closure has brought selected patients with doubly committed subarterial VSD to a minimally invasive era with reasonable outcomes [6-7]. Furthermore, we have introduced the so called percutaneous-perventricular device closure technique in our previous publication combining percutaneous transcatheter device closure with perventricular device closure, which has dramatically reduced the surgical site from a small incision to a pinhole . Based on a relatively small series, the success rate was high with only one case failing due to pericardial effusion in the operating room. In the follow-up period, no other major problems were found. However, only 16 patients were included in that series, and the follow-up period was only one year. In order to investigate the effectiveness and safety of this novel technique, we have updated the sample size and continued focusing on outcomes for a longer period of time.
Not surprisingly, taking the complete and successful rate of VSD closure into consideration, the effectiveness of this technique remained. All the VSD devices have been successfully placed without any residual shunting or device dislocation. These devices did not cause any new valve or rhythm problem even with a longer follow-up period. This was quite consistent with those studies using perventricular technique alone [5-7, 10]. Acute pericardial effusion was found in one case due to dislodgement of the sheath from the RV in the previous report , and this event was explained as an “accident” by us then. In our old technique protocol, after releasing the VSD device, the tip of sheath should be kept right in the RV outflow tract without compromising the first device or dislodging out of the heart. This is actually difficult and very demanding due to the small distance between the device and the RV free wall. Technical improvements have been made in the update to avoid this pitfall. By parking the sheath in the pulmonary artery temporarily before deploying the second hemostatic device, a reasonable length of sheath could be kept safely in the heart, Thus, no more dislodgement was found in any of the 38 patients in the updated study.
At the same time, pericardial effusion was noticed in two additional patients in the Intensive Care Unit after uneventful surgeries. The effusion was drained and caused no further adverse results, but still caught our attention. These two patients were aged 7 and 8 months respectively, so we assumed that their RV free wall in the puncture region was thin. The hemostatic device (Shanghai Shape Memory Alloy Material Co. Ltd, Shanghai, China) has a 5 mm high waist, which might be greater than the myocardium. Delayed bleeding or oozing could occur from the less packed “sandwich”. This complication implied that careful monitoring for the pericardial effusion might still be needed after an uncomplicated surgery, especially in younger patients less than 1 year old. Additionally, an internal mammary artery injury occurred in a patient aged one in the updated series. This might be due to a variant course of the artery, where the artery was closer to the edge of the sternum. Despite this, we still believe that the puncture point should be located adjacent to the left border of the sternum.
As compared to valve problems or arrhythmia, an obstruction of the RV outflow tract might be more technique specific and of interest in the follow-up. We have expanded the study population and observed every patient for three years. No cases presented with RV outflow tract obstruction in the whole follow-up period, not even in infants. No obvious flow acceleration was noticed in any echocardiographic exam. We did not measure geometric numbers of the RV outflow tract in any patient before the surgery. However, it is understandable that patients with VSD, especially a doubly committed subarterial one, would have a wider or larger infundibular region. This characteristic might allow a more crowded outflow tract without flow acceleration.