46 Serum BNP and Clinical Outcomes Prediction in Tetralogy of Fallot: A Prospective Analysis

Background Neurohormonal activation has been demonstrated in small cross-sectional studies in congenital heart disease. Brain natriuretic peptide(BNP) levels have been shown to be elevated in repaired Tetralogy of Fallot (rTOF) patients when compared to controls. Whilst serum BNP measurement measurement has long been used in the clinical management of heart failure patients with left ventricular pathology, its relevance and longitudinal predictive value of adverse clinical outcomes in rTOF is unknown. Methods rTOF patients aged over 16 years with no contra-indications to cardiovascular magnetic resonance (CMR) were prospectively recruited at an adult congenital heart disease tertiary centre (September 2000–March 2004). Study participants had baseline observations, ECG, chest radiographs, CMR and peripheral blood sampling for neurohormone quantification in accordance with study protocol (with local ethics approval). Age- and gender-matched healthy controls were also recruited with peripheral blood neurohormone quantification and baseline observations recorded. Follow-up data was collated from hospital medical records, with mortality status obtained from the Office for National Statistics (UK). The primary endpoint was all cause mortality. Results Ninety rTOF patients (mean age 32.7 ± 11.3 years, 58 males) and 15 matched controls (mean age 30.3 ± 5.0 years, 7 males) were included. rTOF patients had elevated levels of BNP (15.6 ± 26.9 vs 5.3 ± 3.4pmol/l, p 4 pmol/l) at baseline. Symptomatic rTOF patients had significantly higher levels of neurohormones (Table 1). Outcome data was available for 83 patients over a median follow-up of 10 years (IQR 0.71–12.4 years, 4 patients emigrated, 3 patients defaulted from clinical follow-up). There were 7 deaths (2 sudden cardiac deaths, 2 peri-operative right ventricular failure, 1 concomitant severe aortic stenosis, 1 respiratory sepsis and 1 unknown). BNP was significantly related to all cause mortality on univariate Cox proportional hazards analysis (AUC 0.68 with ROC characteristics analysis, logrank p = 0.04; Figure 1), in addition to other recognised adverse markers of clinical outcomes including cardiothoracic ratio, creatinine and right atrial area (Table 2). Conclusions In this large, prospective series with long-term follow-up data, BNP was elevated even in asymptomatic rTOF patients and was predictive of mortality. The measurement of BNP in rTOF patients has clinical utility for risk stratification during lifelong care.