Successful Management of Severe Refractory ARDS in Hemophagocytic Lymphohistiocytosis with VV-ECMO: A Case Report and Analysis

Introduction

Hemophagocytic lymphohistiocytosis (HLH) is a life-threatening hyperinflammatory syndrome marked by uncontrolled activation of macrophages and cytotoxic T cells, leading to a cytokine storm, pancytopenia, and multi-organ dysfunction [1,2]. Despite immunochemotherapy guided by the HLH-2004 protocol, mortality remains high, with reported rates of 13% to 31% in general cohorts and up to 70% in critically ill patients [3], largely due to complications such as fulminant myocarditis and acute respiratory distress syndrome (ARDS) [4]. ARDS in HLH presents a particular therapeutic dilemma: aggressive immunosuppression may curb inflammation but can exacerbate infection risk, while conventional ventilatory strategies often fail to maintain adequate gas exchange in the setting of capillary leak and diffuse alveolar damage.

Veno-venous extracorporeal membrane oxygenation (VV-ECMO) has emerged as a potential salvage therapy for severe, refractory ARDS by providing extracorporeal gas exchange and allowing lung-protective ventilation [5]. However, VV-ECMO use in HLH-associated ARDS has been reported only rarely, mainly in pediatric patients and in just 7 adults; these reports describe variable outcomes and offer little detail on immunomodulatory management during ECMO support [4,6–15]. This scarcity of evidence leaves clinicians without clear guidance on patient selection, timing of ECMO initiation, and integration of immunochemotherapy.

Here, we report a case of severe ARDS secondary to HLH triggered by chronic active Epstein-Barr virus (EBV) infection that was successfully rescued with VV-ECMO and tailored immunosuppression. By reviewing relevant literature and detailing our diagnostic and therapeutic approach, we aim to highlight the potential role of VV-ECMO in this high-risk population and to address current gaps in the management of HLH-related respiratory failure.

Case Report

A 45-year-old woman presented on January 31, 2021, with a 1-week history of unexplained fever (peak 38.5°C), fatigue, and anorexia. She had no chills, cough, or sore throat. Medical history was unremarkable for chronic illness, allergies, surgery, or immunosuppression. High-resolution chest computed tomography (CT) performed 3 days before admission showed no consolidation or other abnormalities. Initial blood test results revealed thrombocytopenia (platelets 80×109/L), with normal leukocyte and hemoglobin levels. There were no clinical signs of bacterial infection; therefore, empirical antibiotics were withheld. On admission, vital signs were temperature 38.0°C; heart rate 110 beats/min; respiratory rate 20 breaths/min; and blood pressure 90/51 mmHg. The patient appeared ill but remained alert. Lung auscultation revealed slightly coarse breath sounds, without crackles. Cardiovascular and abdominal examinations were within normal limits, and there was no peripheral edema. Laboratory findings were as follows: white blood cell count (WBC) 12.1×109/L; hemoglobin 126 g/L; platelets 30×109/L; C-reactive protein 282.6 mg/L; total protein 44.4 g/L; albumin 24.4 g/L; activated partial thromboplastin time 42.9 s; fibrinogen 6.7 g/L; erythrocyte sedimentation rate 21 mm/h; procalcitonin 5.37 ng/L; troponin I 0.05 ng/mL; creatinine 132 μmol/L; and potassium 2.76 mmol/L (Table 1). A working diagnosis of fever of unknown origin with thrombocytopenia was established. Supportive care included oxygen, fluid resuscitation, and vasopressors, under Infectious Diseases Unit monitoring. However, the fever persisted, and dyspnea worsened. High-flow nasal oxygen therapy (FiO2 100%) failed to maintain SpO2 above 78%. The patient was transferred to the Intensive Care Unit (ICU) on February 2, 2021, for endotracheal intubation, mechanical ventilation, central venous access, and bronchoscopic suction. The bedside chest radiograph (Figure 1) showed bilateral diffuse infiltrates, and the PaO2/FiO2 ratio was persistently below 300 mmHg (lowest value 70 mmHg), with no evidence of cardiogenic pulmonary edema. The patient met the criteria for severe ARDS according to the Berlin definition. However, the underlying etiology remained unidentified. The patient was on assisted/control ventilation mode, with FiO2 of 1.0 and a driving pressure of 15 cmH2O. The tidal volume was set at 5 mL/kg of ideal body weight, with a respiratory rate of 35 breaths/min, resulting in a minute ventilation of 7.2 L/min. Despite implementing a stepwise lung recruitment strategy that increased positive end-expiratory pressure to 18 cmH2O, the SpO2 remained persistently at 90%; hypotension (blood pressure 82/46 mmHg) required norepinephrine at 2.7 μg/kg/min. Transthoracic echocardiography showed preserved left and right ventricular structure and contractile function. There was no evidence of significant cardiac pathology or acute coronary syndrome. No structural cardiac disease or infectious shock was identified. Given the patient’s adequate volume status and lack of localized infection, distributive shock due to circulatory dysfunction was considered the primary cause. After more than 6 h with PaO2/FiO2 below 80 mmHg under optimized ventilation, VV-ECMO commenced on February 2, 2021. Initial settings were as follows: sweep gas 3 L/min; blood flow 3.92 L/min; and rotor speed 3410 rpm. SpO2 improved to 100%, and vasopressor requirements decreased.

Further workup revealed markedly elevated soluble CD25; splenomegaly on abdominal CT (February 10, 2021); bicytopenia; bone marrow hemophagocytosis; ferritin >500 μg/L; IL-6 greatly increased; evidence of liver injury; and elevated EBV DNA by polymerase chain reaction (Table 2). Two bone marrow biopsies confirmed hemophagocytosis (Figure 2). An extensive infectious panel for blood cultures, serologies for Varicella-zoster virus, influenza, and other viruses returned negative. Autoimmune markers, including antinuclear antibody, antineutrophil cytoplasmic antibody, rheumatoid factor, and complement levels, were negative. Based on HLH-2004 criteria from the Histiocyte Society, a diagnosis of HLH associated with chronic active EBV infection and secondary ARDS was established (Table 2). Methylprednisolone (1 g/day for 5 days), followed by prednisone maintenance (30 mg/day), led to gradual organ recovery and successful ECMO decannulation on day 8 of ICU admission (Figure 3). However, the patient continued to exhibit persistent thrombocytopenia. In accordance with the HLH-2004 guidelines, etoposide therapy was initiated on ICU day 18. Mechanical ventilation was discontinued on day 20 of ICU admission. By day 36 of ICU admission, laboratory values stabilized as follows: WBC 9.6×109/L; hemoglobin 69 g/L; and platelets 167×109/L. Vital signs remained stable, and organ function normalized (Figure 3). The patient was discharged in good condition after 45 days of hospitalization. Outpatient follow-up conducted 2.5 years after discharge revealed EBV DNA levels, both cell-free and cell-associated, below 5.0×102 copies/mL. Additionally, laboratory results showed a WBC of 5.6×109/L, hemoglobin of 117 g/L, and platelets of 157×109/L. At 4 years after discharge, at the time of manuscript submission, the patient remained asymptomatic, with no evidence of disease relapse.

Discussion

LIMITATIONS:

Although this case provides valuable insights into the successful application of VV-ECMO in HLH-associated ARDS, several limitations must be considered. First, this is a single case report, and the findings may not be generalizable to all patients with HLH and ARDS. As seen in Table 3, most case reports report survival [6,7,9,11,14,15], yet studies with relatively larger sample sizes demonstrate a high mortality rate [4,13]. Second, the patient’s response to ECMO is influenced by multiple factors, including the early recognition of HLH [19], timely initiation of ECMO [20,21], and appropriate immunosuppressive treatment [1]. Thus, the success observed in this case may not be replicable in patients with delayed diagnosis or insufficient treatment. Finally, ECMO itself carries inherent risks, including bleeding complications, infection, and mechanical failure. While these risks were appropriately managed in this case, they should be considered when applying ECMO in similar clinical situations.

Conclusions

In summary, our report reinforces the feasibility and value of VV-ECMO in HLH-associated ARDS. However, further studies are needed to define the optimal indications, timing, and management strategies for ECMO in this patient population.