In this study, trans-thoracic echocardiography proved to be highly reliable as an exclusively noninvasive imaging modality in the preoperative assessment of patients with CHD undergoing surgery. These findings correlate with those from some cardiac centers where the sensitivity and specificity of echocardiographic diagnosis in the preoperative assessment of patients with CHD is established.1–3 5, 6,8, Most studies cited in the literature used both trans-thoracic and trans-oesophageal echocardiography in the preoperative assessment. This study, however, used only trans-thoracic echocardiography as an isolated imaging modality in the preoperative assessment. This is important since only trans-thoracic echocardiography is used in the few centres that have such services in the West African sub-region and this showed from the study to be sufficiently safe as a preoprerative imaging modality to indicate surgery in patients with CHD.
The diagnostic error in identifying extra-cardiac structures such as persistent left superior vena cava in patients with ventricular septal defects and tetralogy of Fallot in this study is thought to be due to the difficulty in assessing the extra-cardiac structures on echocardiography, because such structures have similar echo densities. This was also observed by Lillian et al.1 The assessment of extra-cardiac structures through the suprasternal window often used in routine trans-thoracic echocardiography is known to be difficult, mainly in small children. This limitation undoubtedly might have contributed to incomplete diagnosis in patients with persistent left superior vena cava, mainly when no dilatation of the coronary sinus is observed which may suggest anomalous systemic venous drainage, a finding corroborated by Soonswang et al7 as well as Lilian et al1
False positive and false negative results were observed in the assessment of patients with CHD with increased blood flow such as VSD and ASD in which double chambered right ventricle and pulmonary valvular stenosis were missed. A finding also reported by Lilian et al.1, 3–6 12 This is thought to be secondary to an over or underestimated assessment of the right ventricular cavity and relative valvular stenosis, in the presence of increased pulmonary blood flow. A more detailed evaluation of the right ventricular cavity and valvular morphology as well as analysis of the subvalvular and postvalvular pulmonary blood flow which will appear with different velocity is suggested.
Assessment of the pulmonary valve may be hindered in patients with TOF with a significant obstruction to right ventricular outflow tract, in which the pulmonary valve is usually hypoplastic as noted in this study in which pulmonary atresia was found at surgery, but missed on echocardiography in a patient with TOF. In that case analysis with pulsed Doppler located right after the pulmonary valve could help to differentiate functional pulmonary atresia from anatomic pulmonary atresia.
Three patients were diagnosed by echocardiography as having TOF, but were found at surgery to have DORV. This is not surprising since both are within the spectrum of CHD with right ventricular out flow tract obstruction, the difference being in the degree of aortic over ride. The literature is however, still controversial about the differentiation between the two. While some authors consider the presence of complete bilateral double infundibulum essential for the diagnosis of DORV, others consider that the criterion hinders the diagnosis of DORV with TOF and would rather choose to define as DORV those cases in which more than 50% of the circumference of the aorta relates to the right ventricle independent of the nature of the structures that support it. Differentiating these two entities is however, important for surgical repair because in TOF, a straight patch may be sufficient for repair of the VSD, while in DORV tunneling of the patch is required. A straight patch in the repair of DORV will result in left ventricular outflow tract obstruction. While it is true that there was a change in operative approach in view of the intra-operative finding, the overall outcome was unaltered in this study.
The false positive result that occurred in a patient with atrial septal defect (ASD), was that of a child whose echocardiography showed that he had an ASD, but at operation no lesion was found. This diagnostic error is an isolated case and the patient is in the category that would benefit from cardiac catheterization or CT angiography, since the absence of a lesion at surgery does not rule out the presence of a lesion. The false negative in relation to ASD included one patient with supra-mitral ring which was not seen on echocardiography but identified at surgery. A patient was also found at surgery to have unroofed coronary sinus which was not diagnosed by echocardiography. The diagnostic impression of atrial septal defect may often occur in adult patients due to the difficulty of the echocardiographic window, because the atrial septum is located very far away from the transducer, resulting in lack of definition of the tissues of the oval foramen which is thicker than the rest of the septum. One patient had autologous VSD pericardial patch dehiscence on post operative day one necessitating reoperation. We noticed pericardial adhesions at initial surgery possibly from previous endocarditis. This was thought to have contributed to this complication.
The possibility of observer error in the results of the echocardiography cannot totally be ruled out considering the fact that echocardiography is operator dependent. It is to be noted however, that none of the errors affected the outcome of surgery significantly. This is a good development for our sub-region where the practice of echocardiography is still evolving.