Plasma CCN2/connective tissue growth factor is associated with right ventricular dysfunction in patients with neuroendocrine tumors
1 Section of Gastroenterology, Department of Medicine, Oslo University Hospital, Rikshospitalet, Sognsvannsveien 20, 0027 Oslo, Norway
2 Department of Cardiology, Oslo University Hospital, Rikshospitalet, Sognsvannsveien 20, 0027 Oslo, Norway
3 Institute for Surgical Research, Oslo University Hospital, Rikshospitalet, Sognsvannsveien 20, 0027 Oslo, Norway
4 Section of Immunology and Infectious Diseases, Department of Medicine, Oslo University Hospital Rikshospitalet, Sognsvannsveien 20, 0027 Oslo, Norway
5 Research Institute for Internal Medicine, Oslo University Hospital, Rikshospitalet, Sognsvannsveien 20, 0027 Oslo, Norway
6 Faculty of Medicine, University of Oslo, P.O. Box 1078 Blindern, 0316 Oslo, Norway
BMC Cancer 2010, 10:6 doi:10.1186/1471-2407-10-6Published: 6 January 2010
Carcinoid heart disease, a known complication of neuroendocrine tumors, is characterized by right heart fibrotic lesions. Carcinoid heart disease has traditionally been defined by the degree of valvular involvement. Right ventricular (RV) dysfunction due to mural involvement may also be a manifestation. Connective tissue growth factor (CCN2) is elevated in many fibrotic disorders. Its role in carcinoid heart disease is unknown. We sought to investigate the relationship between plasma CCN2 and valvular and mural involvement in carcinoid heart disease.
Echocardiography was performed in 69 patients with neuroendocrine tumors. RV function was assessed using tissue Doppler analysis of myocardial systolic strain. Plasma CCN2 was analyzed using an enzyme-linked immunosorbent assay. Mann-Whitney U, Kruskal-Wallis, Chi-squared and Fisher's exact tests were used to compare groups where appropriate. Linear regression was used to evaluate correlation.
Mean strain was -21% ± 5. Thirty-three patients had reduced RV function (strain > -20%, mean -16% ± 3). Of these, 8 had no or minimal tricuspid and/or pulmonary regurgitation (TR/PR). Thirty-six patients had normal or mildly reduced RV function (strain ≤ -20%, mean -25% ± 3). There was a significant inverse correlation between RV function and plasma CCN2 levels (r = 0.47, p < 0.001). Patients with reduced RV function had higher plasma CCN2 levels than those with normal or mildly reduced RV function (p < 0.001). Plasma CCN2 ≥ 77 μg/L was an independent predictor of reduced RV function (odds ratio 15.36 [95% CI 4.15;56.86]) and had 88% sensitivity and 69% specificity for its detection (p < 0.001). Plasma CCN2 was elevated in patients with mild or greater TR/PR compared to those with no or minimal TR/PR (p = 0.008), with the highest levels seen in moderate to severe TR/PR (p = 0.03).
Elevated plasma CCN2 levels are associated with RV dysfunction and valvular regurgitation in NET patients. CCN2 may play a role in neuroendocrine tumor-related cardiac fibrosis and may serve as a marker of its earliest stages.