ABSTRACTBackgroundNLRP3 inflammasome‐related inflammation might play an important role in the pathophysiology of severe CVT. The use of steroids as anti‐inflammatory agents in improving severe CVT prognosis remains controversial.MethodsA total of 94 male Sprague–Dawley rats were used. We evaluated the dynamic and association between NLRP3 inflammasome in brain, blood, and CSF and severity in severe CVT rats and/or patients. We also explored the effect of steroids on NLRP3 activation, neurological injury, and CSF circulation disturbance after CVT in animals and/or patients.ResultsIn rats, compared with the sham group, NLRP3‐related factors rose on day 1, peaked on day 2 (NLRP3, Sham: 0.79 ± 0.22; day 2: 1.25 ± 0.08, p < 0.01; pro‐Caspase‐1, Sham: 0.58 ± 0.13, day 2: 1.20 ± 0.44, p < 0.05; GSDMD, Sham: 0.94 ± 0.22, day 2: 1.72 ± 0.46, p < 0.05; pro‐IL‐1β, Sham: 0.74 ± 0.15, day 2: 1.35 ± 0.09, p < 0.01), decreased on day 7 in rats (n = 4 per group). Thrombus (Sham: 0.00 ± 0.00, day 2: 3.44 ± 0.70, p < 0.0001), infarct size (Sham: 0.00 ± 0.00, day 2: 11.99 ± 6.26, p < 0.01) and neurological deficits appeared similar trend. In 50 patients, serum NLRP3 and IL‐6 levels correlated positively with NIHSS (r = 0.4273, p = 0.0020; r = 0.4938, p = 0.0029) and mRS (r = 0.5349, p = 0.0125; r = 0.6213, p = 0.026), while CSF IL‐6 correlated positively with mRS on admission (r = 0.5349, p = 0.0125). Compared with baseline, NLRP3 (0.36 (0.36, 0.36) vs. 0.41 (0.37, 0.84), p < 0.0001) and IL‐6 decreased (4.06 ± 1.48 vs. 12.03 ± 7.80, p < 0.05), accompanying by improvement of neurological deficits and CSF circulation (all p < 0.01) after steroids therapy in severe CVT patients at discharge and 3 months follow‐up. No significant steroid‐related adverse effects were observed.ConclusionShort‐term steroid therapy may improve prognosis of severe CVT by suppressing NLRP3 inflammasome‐related inflammation.