Purpose: We compared the long-term outcome of patients with lupus nephritis in whom normalization of complement levels (CH50) was sustained by adjustment of immunosuppressive therapy to those patients with persistently low complement levels despite similar immunosuppression in whom therapy was adjusted solely on the basis of clinical disease activity.
Patients and methods: Thirty-nine female patients with lupus nephritis recruited from 1972 to 1979 were prospectively studied (mean follow-up, 116.7 +/- 11 months). Entry criteria included initial renal biopsy, low CH50, and elevated anti-DNA antibody levels. A second biopsy was performed in 24 patients after an interval of 40.6 +/- 5 months. Treatment was started with prednisone (1 mg/kg/day). Azathioprine at a dose of 1.5 to 2.0 mg/kg/day was added if complement was not normalized by prednisone alone. Twenty-five of 39 patients had normal complement levels within six months (Group 1), and immunosuppressive therapy was tapered but continuously readjusted to the lowest dosage that preserved normal CH50 and maintained clinical remission. Eight of these 25 patients subsequently became persistently hypocomplementemic due to inadequate drug intake (Group 1B), whereas the complement levels continued to be controlled in the other 17 patients (Group 1A). Despite similar therapy, the remaining 14 patients did not achieve normalization of complement within the initial six months of therapy, and therefore future treatment decisions were based solely on clinical symptoms (Group 2). Renal pathologic lesions were classified according to World Health Organization criteria and a semi-quantitative chronicity index.
Results: During the first six months, there were no significant differences in clinical or histologic features between patients in whom complement levels were controlled and patients in whom complement levels were not controlled. After a mean observation period of 10 years, however, patients with consistent normalization of complement (Group 1A) did much better than patients with only short-term complement control (Group 1B) or persistent hypocomplementemia (Group 2). Both groups with low complement levels had a similar outcome with significantly worse kidney and patient survival. Life-table analysis demonstrated that the differences in outcome between complement-controlled and complement-uncontrolled groups became apparent only after five or more years of follow-up. Patients with a low chronicity score on initial biopsy whose complement level was controlled did uniformly well with no renal failure or death. (ABSTRACT TRUNCATED AT 400 WORDS)