Objective: To study the role of alternative complement pathway overactivation in malignant nephrosclerosis. Methods: (1) Fifty patients with confirmed malignant nephrosclerosis by renal needle biopsy were enrolled. Meanwhile, twenty-five cases of time-zero renal needle biopsy were enrolled as control subjects. Enzyme linked immunosorbent assay (ELISA) was used to detect alternative complement pathway of the complement initiation factor B, positive regulation factor P, negative regulation factor H, and the complement end products C3a and C5a in the plasma and urine. (2) Immunohistochemistry was used to detect the deposition of the complement end product C5b-9, C4d and mannan binding lectin (MBL) of lectin pathway in the renal biopsies. Double immunofluorescence labeling method was used to assay the deposition of C5b-9 and CD34 (endothelial cell marker) in the arteriolar endothelium and glomerular capillary endothelium. Results: (1) The plasma and urine levels of complement factor B, factor P, C3a and C5a in malignant nephrosclerosis patients were significantly higher than those in control subjects (all P<0.05), while the plasma and urine levels of complement factor H in malignant nephrosclerosis patients were lower than those in control subjects (all P<0.05). (2) The plasma level of factor P was positively correlated with 24 h urine protein (r_s=0.465, P=0.001). Urinary factor B/urinary creatinine, urinary factor P/urinary creatinine and urinary C3a/urinary creatinine were positively correlated with serum creatinine in malignant nephrosclerosis patients (r_s=0.483, P<0.001; r_s=0.352, P=0.012; r_s=0.319, P=0.024), while urinary factor H/urinary creatinine was negatively correlated with serum creatinine and 24 h urine protein (r_s=-0.299, P=0.035; r_s=-0.342, P=0.015). Urinary C5a/urinary creatinine was positively correlated with serum creatinine and 24 h urine protein (r_s=0.525, P<0.001; r_s=0.496, P<0.001). (3) Immunohistochemical results showed that there were C5b-9 deposited in the arterioles and glomerular capillary wall in malignant nephrosclerosis patients, and no deposition in control renal tissues. Meanwhile, the semi-quantitative scores showed that C5b-9 deposition intensity was positively correlated with serum creatinine and 24 h urine protein (r_s=0.791, P<0.001; r_s=0.345, P=0.014). The double immunofluorescence labeling analysis showed that the C5b-9 and CD34 deposited in the arteriolar endothelium and glomerular capillary endothelium. (4) Plasma level of factor B in malignant nephrosclerosis patients was positively correlated with plasma C3a level (r=0.331, P=0.022). Plasma level of factor P was positively correlated with C5b-9 score (r_s=0.300, P=0.034). Urinary B was positively correlated with urinary C3a, C5a and C5b-9 score (r_s=0.311, P=0.028; r_s=0.465, P=0.001; r_s=0.428, P=0.002). Urinary factor P was also positively correlated with urinary C3a and C5a (r_s=0.307, P=0.030; r_s=0.442, P=0.001). Immunohistochemical result showed that there were C4d deposited in the arterioles and glomerular, and no deposition of MBL. Conclusion Complement activation via the alternative pathway may be involved in malignant nephrosclerosis and related to the severity of the disease.