Background Systemic lupus erythematosus (SLE) is a chronic autoimmune disorder with more than one organ involvement. Kidney is the foremost commonly affected one. Gelsolin is a protein that induces depolymerization of actin filaments thus preventing downstream stimulation of inflammatory reactions. The aim of this work was to detect the relation of plasma gelsolin to SLE disease activity and severity indices in order to find out if plasma gelsolin could be used as a biomarker of the disease. This study was conducted on 50 SLE female patients and 30 matched control. SLE disease activity Index (SLEDAI) and SLE damage index (SDI) were assessed. All lupus nephritis (LN) patients were subjected to an ultrasound-guided kidney biopsy. Plasma gelsolin level was measured. Results The mean age of the patients was 38.5 ± 6.3 years (26–51 years) with median disease duration of 5 (3–9.3) years. Eighteen patients had LN, 11 had cardiac manifestations and 12 had chest manifestations. The mean SLEDAI was 13.1 ± 4.5 (4–22) and the median SDI was 2 (1–3). Plasma gelsolin level was significantly lower in SLE patients (74.9 mg/l; 57.5–98.8 mg/l) compared to control (801.5 mg/l; 225–1008.3 mg/l) (p < 0.001). There were significant negative correlations of gelsolin levels with anti-ds DNA (r = − 0.63, p < 0.001), SLEDAI (r = − 0.79, p < 0.001), and SDI (r = − 0.74, p = 0.001). Plasma gelsolin level was significantly lower in SLE patients with high/very high activity grades compared to those with low and moderate (p = 0.007 and p < 0.001 respectively). A gelsolin level of ≤ 78.95 mg/l significantly predicted renal affection (p < 0.001), with a sensitivity of 100%, specificity 71.9%, and a positive predictive value 66.7%. Conclusion A decreased gelsolin level is associated with disease activity in SLE patients. Plasma gelsolin was well related to disease activity and severity with a high predictive value for renal affection comparable to anti-ds DNA titre. Plasma gelsolin is a potentially important predictive biomarker for SLE and LN.