Reversible cerebral vasoconstriction syndrome in a female patient with systemic lupus erythematosus

Sir, Headaches are common in patients with SLE, especially those suffering from CNS involvement (CNS-SLE) with or without cerebral vasculitis [1, 2]. Reversible cerebral vasoconstriction syndrome (RCVS) is characterized by the association of severe ‘thunderclap headaches’ with or without additional neurological manifestations and a ‘string of beads’ appearance on cerebral arteries, which resolves spontaneously in 1–3 months [3]. RCVS and cerebral vasculitis have several overlapping features. We report a case of RCVS in an SLE patient that was misdiagnosed initially as lupus-associated cerebral vasculitis. To our knowledge, this is the first case of RCVS described in an SLE patient. A 40-year-old female patient was diagnosed with SLE at the age of 25 years. She presented at the age of 27 years with a severe flare-up characterized by intra-alveolar haemorrhage and CNS-SLE (headaches, psychosis and seizure). She had no aPLs. Her follow-up was characterized by poor treatment adherence. Since December 2008, she has been followed up in our department for World Health Organization classes IV–V lupus nephritis and treated according to the Euro-Lupus regimen. While clinico-biological remission was noted, she presented 2 days before the fifth cyclophosphamide pulse with an acute severe headache that disappeared spontaneously and the patient received the planned 500 mg i.v. cyclophosphamide. Her treatment included prednisone (30 mg/day), atorvastatin, omeprazole, calcium carbonate and cholecalciferol. She was readmitted to the hospital 2 days later for acute severe headache associated with nausea and vomiting. The headache was of sudden onset, recurrent, described as ‘thunderclap headache’, peaking in <10 s with maximal pain intensity graded as 9.5/10 on the visual scale. She also experienced a brief generalized seizure. Physical examination was normal. A head CT scan was normal. The cerebrospinal fluid was haematic. A cerebral magnetic resonance angiogram (MRA) demonstrated multifocal cerebral arterial stenosis (Fig. 1A). The diagnosis of lupus-associated cerebral vasculitis was considered; she was treated promptly with three daily pulses of 750 mg i.v. methylprednisolone, plasmapheresis and levetiracetam. Cerebral MRA findings. MRA showing multifocal segmental stenosis of cerebral arteries (A). Control MRA done 40 days later showing resolution of the cerebral arterial vasoconstriction (B). Arrows refer to the multifocal arterial stenoses (A) that have almost completely regressed (B). The headaches persisted leading us to reconsider the diagnosis. A careful awareness of the characteristics of headaches and the neuro-angiographic findings led us to advocate the hypothesis of RCVS; cyclophosphamide was suspected to contribute to the cerebral vasoconstriction. We then changed our strategy: i.v. nimodipine, discontinuation of plasmapheresis and replacement of cyclophosphamide by mycophenolate mofetil (3 g/day). The headaches disappeared after 5 days. The control cerebral MRA done 40 days later demonstrated an almost complete regression of the vascular abnormalities (Fig. 1B) confirming the diagnosis of RCVS. Till today, the patient has not experienced any recurrence of ‘thunderclap headache’ after 8 months of surveillance. True cerebral vasculitis is difficult to distinguish from RCVS on initial evaluation; in the Singhal experience [4], patients with vasculitis tend to have insidious-onset, dull headaches with stepwise clinical progression, rather than the acute, self-limited ‘thunderclap headaches’ characteristic of RCVS. Moreover, the angiographic abnormalities usually involve the distal cerebral arteries in vasculitis rather than the circle of Willis arteries or their proximal branches that are affected by vasoconstriction [4, 5]. In RCVS, spontaneous reversibility of the cerebral vasospasm within 1–3 months is the key diagnostic feature. Furthermore, cerebral vasculitis in SLE patients is rare and usually occurs in patients with active lupus [1], whereas our patient had an inactive lupus when severe headaches occurred. Although case reports of RCVS secondary to cyclophosphamide are very rare in the literature, several authors mentioned cyclophosphamide as potential cause [3–5]. Therefore, we decided to discontinue it in our patient. The diagnosis of RCVS in SLE patients is of practical importance, because treatment and prognosis for lupus-associated cerebral vasculitis differ strongly from those of RCVS. The clinicians must always be alerted that headaches and neurological abnormalities in patients with inactive SLE may not be CNS-SLE; awareness of the presenting clinical and neurovascular features should lead to consideration of the diagnosis of RCVS. Disclosure statement: The authors have declared no conflicts of interest.