Combined Red Cell and Plasma Exchange Is Associated with Reduced Mortality of Fat Embolism Syndrome in Sickle Cell Disease

Fat embolism syndrome (FES) is a devastating complication of sickle cell disease which carries a high mortality without prompt recognition and treatment. We report the largest single center experience to date involving 19 patients with sickle cell disease aged 20-60 years (11 Male; 8 Female) treated for suspected FES at Imperial College Healthcare NHS Trust, London between 2016 and 2024. Patient genotypes were as follows: HbSS (11), HbSC (5), HbS-Beta zero thalassemia (2), HbS-O Arab (1). Seven patients were receiving disease modifying therapy at the time of the FES episode (5 Hydroxycarbamide; 2 Regular red cell exchange transfusion). Typically patients had a relatively mild clinical phenotype prior to presentation with FES. Two patients had a history of recurrent acute chest syndrome. All but one of the patients presented with a severe painful vaso-occlusive crisis (VOC) followed in most cases by rapid clinical deterioration. A diagnosis of FES was established in 17 patients between days 2 and 4 after initial presentation and at day 24 and 25, in 2 patients. Intractable pain despite high doses of opioids was the most common clinical feature followed by persistent fever and neurological impairment. Neurological abnormalities occurred in 68% (13/19) of patients. Confusion and impaired consciousness were the predominant neurological features. One patient had accompanying limb weakness and aphasia and 1 had dysphonia. Type 1 or type 2 respiratory failure was noted in 79% (15/19) patients with 40% (6/15) of these patients requiring intubation and mechanical ventilation. Overall, 74% (14/19) patients required critical care admission. Prominent laboratory abnormalities included thrombocytopenia in 89.4% (17/19) patients, of whom 11 had a platelet nadir below 50x10^9/L. A fall in haemoglobin from steady state levels was seen in all patients, with 17 having a reduction of 1-3 g/dL. In 84% (16/19) patients, an inappropriately low reticulocyte count was noted compared to their baseline values. Blood film features included a leucoerythroblastic picture, features of oxidative red cell damage and suspected fat globules in 1 patient. All patients had a raised ferritin of >1000 ug/L with a value of >10000 ug/L in 84.2% (16/19) patients. All patients had an LDH of >1000 U/L (range 1343-7607U/L), an ALP of >200U/L and a CRP of >100mg/L (range 143-440mg/L). An elevated troponin was found in 73.3% (11/15) patients. All 13 patients in whom D-Dimers were assayed had elevated levels with most values being above 20000 ng/ml. Evidence of parvovirus B19 infection was seen in 4 patients with detection of IgM antibody or viral DNA. One patient without typical VOC symptoms had concurrent parvovirus B19 and SARS-CoV-2 infection. Five patients underwent bone marrow biopsy, 4 showed features of bone marrow necrosis and 1 haemophagocytosis. Neuroimaging performed in 12 patients revealed features diagnostic of cerebral fat embolism in 3 and micro-haemorrhages in 1. Chest imaging revealed bilateral ground glass changes/consolidation and pulmonary congestion in most cases. Two patients had evidence of pulmonary embolism. In this cohort the use of plasma exchange in FES was predicated on removal of emboli generated as a result of bone marrow necrosis as well as humoral inflammatory mediators that could not be achieved by automated erythrocytapheresis, the mainstay of treatment for life-threatening acute complications of sickle cell disease. Patients received red cell exchange, antibiotic therapy and supportive management prior to commencement of daily plasmapheresis. Eighteen patients received plasmapheresis within 24 hours of a suspected diagnosis of FES. Twelve patients underwent plasmapheresis for 3-7 days while 6 patients required a more prolonged course lasting 8-15 days. Overall survival was 100% in patients who received plasma exchange with 1 death in the entire cohort in a patient who deteriorated rapidly before plasma exchange could be instigated. The majority of patients made a complete recovery though residual neurological sequelae were seen in 4 patients. Based on these results, we propose plasmapheresis should be considered in conjunction with standard therapy including red cell exchange for cases of sickle cell disease which based on clinical, laboratory and imaging features meet diagnostic criteria consistent with a diagnosis of fat embolism syndrome to prevent mortality associated with this complication.