A 6 year old boy is found to have a fixed split second heart sound and a systolic murmur heard best at the left upper sternal border on routine physical examination. The presence of a secundum atrial septal defect is confirmed by echocardiography. The attending cardiologist recommends closure via transcatheter device placement. The studious paediatric resident remembers reading in the most recent edition of the Nelson textbook of pediatrics that open heart surgery is also an acceptable option. Knowing that the consultants in her department think that EBM stands only for expressed breast milk, the resident decides to search the recent medical literature.

Structured clinical question

In children with atrial septal defects [patient] is transcatheter device closure [intervention] as good as surgery [comparison] at closing the defect, with fewer side effects [outcomes]?

Search strategy and outcome

Search engine—PubMed: “heart septal defects, atrial” and “heart catheterization” and “surgery” (MeSH-terms) limit to clinical trial, all child: 0–18 years.

Search results—11 studies found, one relevant. See table 4.

Commentary

While open heart surgery is the traditional means of treating atrial septal defects, transcatheter device occlusion is quickly becoming the treatment of choice in major cardiovascular centres. With this new technique, patients do not require thoracotomies or cardiopulmonary bypass, and can often be discharged from hospital on the same day as the procedure. Unfortunately, not all types of defects can be closed in this fashion, and sometimes the suitability for device closure cannot be determined until after a cardiac catheterisation has begun.

There are no randomised, controlled trials of catheter device versus surgical closure of atrial septal defects in adults or children. One non-randomised study including both adults and children compared the two methods of treatment; unfortunately, there was no subset analysis of the paediatric group.

Du et al show that closure is very successful in both patients treated with the Amplatzer septal occluder (ASO) (96%) and those undergoing open heart surgery (100%) with p = 0.002. More importantly, Du et al show a decrease in major complications (p = 0.03, NNH = 28) and length of hospital stay (p < 0.001) in patients treated with ASO. Regrettably, the eligibility criteria was different between the two groups. For example, patients in the ASO group needed to have smaller, single defects and those in the surgery group could have larger, multiple defects. The right atrial and right ventricular sizes were not different between the two groups, however, indicating similar haemodynamic implications. Group allocation was decided by the patient in consultation with his cardiologist. No attempt was made to adjust for the confounders.

The evidence to date does not adequately address our clinical question. There are no trials specifically performed in children, and none with children analysed as a subset. Future clinical trials should be randomised, if possible.

CLINICAL BOTTOM LINE

• Amplatzer septal occlusion and open heart surgery both have excellent success rates in the closure of atrial septal defects.

• Patients undergoing transcatheter ASD occlusion have shorter hospital stays with fewer major complications compared to those having surgical closure.