P125 Miliary tuberculosis triggering haemophagocytic lymphohistiocytosis in a patient with new-onset systemic lupus erythematosus: a diagnostic challenge in acute respiratory failure

Abstract Introduction Severe systemic lupus erythematosus (SLE) can present with multi-organ involvement, including respiratory manifestations. SLE can both mimic and trigger hyperinflammatory states, including haemophagocytic lymphohistiocytosis (HLH). This case describes a patient with a new diagnosis of SLE and indeterminant IGRA with severe organ involvement requiring high-dose immunosuppression, culminating in respiratory failure and secondary HLH triggered by reactivation of latent miliary TB. It highlights diagnostic and management challenges in rheumatology respiratory practice, particularly in differentiating infectious triggers of HLH in patients with a known rheumatological diagnosis. Case description A 57-year-old Filipino woman with no past medical history presented with two months of confusion, 10 kg weight loss, generalised abdominal pain, facial rash, and proximal muscle weakness. She was afebrile but delirious, with a malar rash, peripheral vasculitic lesions, and proximal myopathy. Initial investigations showed bicytopenia (Hb-85g/L, platelets-108 × 109/L), acute kidney injury (creatinine-166µmol/L), severe hypoalbuminaemia (16g/L), and proteinuria (urine PCR-52). Early differentials included malignancy and systemic autoimmune disease. Ferritin was elevated (7300µg/L), with hypertriglyceridemia, low fibrinogen, and bicytopenia, raising suspicion for impending HLH (HScore-156). However, there was no fever and a trial of anakinra showed no improvement in clinical parameters and was therefore stopped. Further investigations revealed high-titre ANA (1:1280), anti-dsDNA>379, positive Mi2, Ku, Ro, Scl-70, low complements, and raised troponin, CK, and BNP. Cross-sectional imaging revealed small bilateral pleural effusions with diffuse features of fluid overload and mildly avid mediastinal nodes. Skin biopsy showed “full-house” immunofluorescence. She was diagnosed with new-onset SLE with overlap myositis and multi-organ involvement (cardiac, renal, cutaneous, central nervous system). Treatment included IV methylprednisolone, IVIG and cyclophosphamide. She received antimicrobial prophylaxis and entecavir for chronic inactive hepatitis B. Despite clinical and serological improvement from an SLE perspective, she developed fevers and acute type 1 respiratory failure requiring intubation and ITU admission. Further investigations revealed a rising ferritin (19,000µg/L) and worsening pancytopenia and a diagnosis of secondary HLH was made (HScore-198). Echocardiogram showed an acute drop in left ventricular ejection fraction (LVEF) to 35% and CT pulmonary angiogram revealed diffuse micro-nodularity; Mycobacterium tuberculosis was identified two days later in endotracheal aspirate culture. She was treated with anti-TB therapy and re-introduction of anakinra with corticosteroids. HLH blood markers normalised, and cardiac function recovered. She was successfully extubated. She completed six doses of cyclophosphamide infusions and remains on steroids, hydroxychloroquine, mycophenolate and TB therapy. Total admission: 118-days. Discussion This case highlights the diagnostic and management challenges of severe multisystem SLE presenting with acute respiratory failure, and the difficulty in distinguishing between refractory autoimmune disease and an evolving infectious process. The differential diagnoses for her deterioration were broad. Initially, she was treated with intravenous diuretics for acute pulmonary oedema, presumed secondary to lupus myocarditis and exacerbated by profound hypoalbuminaemia. However, by this time she had received cyclophosphamide and was showing signs of lupus improvement, including falling anti-dsDNA titres, rising complement levels, and clinical improvement of neurological and cutaneous manifestations, making active SLE less likely. Broad-spectrum antibiotics were also commenced for suspected hospital-acquired pneumonia. Miliary tuberculosis was diagnosed two days after ITU admission, based on CT findings of diffuse micronodularity and sputum culture for Mycobacterium tuberculosis. Her Filipino background and recent high-dose immunosuppression placed her at significant risk of TB reactivation and may have warranted consideration of prophylactic anti-TB therapy. At initial presentation, her IGRA result was indeterminate, likely influenced by high dose steroid administration - a finding more common in immunosuppressed individuals. Notably, this does not exclude latent or active TB. At the time of clinical deterioration, she also fulfilled diagnostic criteria for secondary HLH, presenting with fever, falling blood counts, and rising ferritin - commonly referred to as the “3 Fs” of HLH. Her HScore exceeded the diagnostic threshold of 169. TB is a well-recognised infectious trigger of HLH and was likely the driver in this case. Her acute cardiac dysfunction was likely multifactorial - attributable to lupus myocarditis, TB, and systemic inflammation from HLH. The patient’s marked and rapid improvement, including normalisation of LVEF, following initiation of HLH-directed immunosuppression underscores the critical importance of promptly recognising and treating hyperinflammation. Simultaneous treatment of the underlying trigger - in this case, tuberculosis, is also essential for optimal outcomes. Key learning points This case highlights the common clinical challenge we face in rheumatology – the need to balance immune suppression with infection risk. While high-dose steroids and cyclophosphamide were necessary to manage SLE-related organ involvement, and evolving HLH, these treatments almost certainly led to latent TB reactivation. The case also reinforces the importance of antimicrobial prophylaxis and careful infection risk assessment when initiating immunosuppressive therapy in high-risk populations. The patient’s Filipino origin indeterminate IGRA and exposure to high-dose immunosuppressive therapy should, in retrospect, have prompted earlier consideration of latent tuberculosis reactivation. This case has heightened our vigilance in assessing reactivation risk prior to initiating immunosuppression. In selected high-risk individuals, empirical TB prophylaxis may warrant earlier consideration. This case provided valuable insight into the recognition and management of respiratory deterioration in patients with severe autoimmune diseases such as SLE. It reinforced the importance of maintaining a broad differential diagnosis, particularly in patients with evolving HLH despite improving underlying rheumatic disease, and the need to remain vigilant for opportunistic infections. A major learning point has been the need for adaptive, multidisciplinary (MDT) and multispecialty decision-making when managing diagnostic uncertainty and treatment escalation in complex autoimmune cases. Throughout her admission, close collaboration between rheumatology, infectious diseases, respiratory medicine, and intensive care was enabled by discussion in the HLH MDT and vital in guiding clinical care and improving outcomes. This raises important questions for clinical practice: • Should patients from TB-endemic regions receive empirical TB prophylaxis before cyclophosphamide/immunosuppression? • Can we better distinguish autoimmune inflammation from infection-triggered HLH? • Should miliary TB be more aggressively excluded in deteriorating patients with indeterminate IGRA? We hope that by presenting this case at the conference, we can facilitate reflection on the challenges of managing respiratory deterioration in autoimmune disease, and stimulate discussion around individualised infection risk assessment and immunosuppression in rheumatology practice.