Fungal infections in immunocompetent and immunocompromised children — branching viva

Opening question

Examiner: Take me through this child. What is your frame for managing him, and what is at the top of your differential? [14]

Candidate: My frame is persistent febrile neutropenia with a high probability of invasive fungal disease. A profoundly neutropenic child (ANC 0.1) with a central line and fever persisting beyond four to seven days on appropriate broad-spectrum antibiotics has crossed the threshold where empirical antifungal therapy is mandatory. The pleuritic chest pain points toward an invasive pulmonary mould process. At the top of my differential is invasive pulmonary aspergillosis, with invasive candidiasis (line-related) and mucormycosis also in play, alongside unresolving bacterial infection. [14] [1]

Examiner: Why is persistent febrile neutropenia the trigger for empirical antifungal therapy? [1]

Candidate: Because the probability of an invasive fungal infection rises sharply after four to seven days of persistent fever in profound neutropenia, and the neutrophil that would contain the organism is absent. The cultures and biomarkers lag the clinical deterioration, so waiting for confirmation costs the child time. An echinocandin or liposomal amphotericin B is started empirically, and the chest and sinuses are imaged. The principle is that a missed or delayed invasive fungal disease in the neutropenic child is measured in lives, not in days of unnecessary therapy. [1] [14]

Branch 1 — host-defence biology and the breach points

Examiner: Explain why the neutrophil is the key defence against invasive Candida and Aspergillus. [14]

Candidate: Neutrophils phagocytose and kill Candida yeasts and Aspergillus hyphae by oxidative and non-oxidative mechanisms. When the neutrophil count falls — in chemotherapy-induced neutropenia, in leukaemia, in marrow failure — the organism is left unchecked and invades the bloodstream and the tissue. This is why prolonged profound neutropenia is the key risk factor for invasive candidiasis and invasive aspergillosis, and why granulocyte recovery can itself provoke the hepatosplenic candidiasis that declares itself as the returning neutrophils attack the seeded tissue. [14]

Examiner: And the T-cell? What does it defend against? [9]

Candidate: T-cell-mediated immunity controls Pneumocystis jirovecii and the dimorphic fungi. This is why HIV with a low CD4 count, solid-organ transplant on calcineurin inhibitors, and chronic high-dose steroids all open the door to Pneumocystis pneumonia and to cryptococcal meningitis. Macrophages, in turn, ingest and contain the inhaled mould spores that reach the alveolus. So the host-defence tier — neutrophil, T-cell, macrophage — points directly to the likely organism and the empirical cover. [9] [14]

Branch 2 — investigations and the halo sign

Examiner: How will you investigate this child's chest? [2]

Candidate: I would obtain an urgent high-resolution chest computed tomogram rather than rely on the plain radiograph, which is insensitive early in the neutropenic patient. On CT I am looking for pulmonary nodules with the halo sign — a nodule surrounded by a ground-glass halo of haemorrhage — which is the early finding of invasive pulmonary aspergillosis. The reversed-halo sign, by contrast, favours mucormycosis. I would send a serum galactomannan and, if bronchoscopy is feasible, a bronchoalveolar-lavage galactomannan, and I would pursue tissue biopsy for histopathology if a lesion is accessible. [2] [14]

Examiner: Why is the halo sign haemorrhagic? [7]

Candidate: Because Aspergillus hyphae are angioinvasive — they invade the blood-vessel walls, producing thrombosis, infarction and surrounding haemorrhage. The halo of ground-glass on CT is the haemorrhage around the infarcted nodule, and it is exactly this angioinvasive biology that underlies the catastrophic haemoptysis of pulmonary aspergillosis when an infarcted cavity erodes a bronchial vessel. The Mucorales share this angioinvasive behaviour, which is why mucormycosis produces the necrotic, black eschar in the nose and palate — that eschar is infarcted tissue, which is why medical therapy alone cannot sterilise it and surgical debridement is mandatory. [7] [14]

Branch 3 — empirical and definitive therapy

Examiner: What empirical therapy will you start, and what is your first-line if invasive aspergillosis is confirmed? [14]

Candidate: I would start an empirical echinocandin (caspofungin or micafungin) or liposomal amphotericin B, depending on the suspected organism, the local mould epidemiology and the concurrent medications. If invasive aspergillosis is confirmed, voriconazole becomes first-line, with liposomal amphotericin B as an alternative. If candidaemia is confirmed, I would remove the central line, take daily blood cultures until clearance, and give a minimum of two weeks of therapy after the first negative culture, with fundoscopy to detect retinal seeding. The echinocandin-first stance for paediatric invasive candidiasis rests on the Fisher comparative-effectiveness data. [2] [14]

Examiner: What about mucormycosis? [7]

Candidate: Mucormycosis is treated with high-dose liposomal amphotericin B and urgent surgical debridement of necrotic tissue. Voriconazole does not cover the Mucorales — a critical point, because starting voriconazole empirically in a child who actually has mucormycosis is a fatal error. I would also reverse the metabolic defect — control diabetic ketoacidosis, reduce immunosuppression where possible, and address iron overload. Outcome is decided by the speed of diagnosis, the adequacy of debridement and the reversal of immunosuppression, because the antifungal cannot sterilise infarcted, avascular tissue alone. [7]

Branch 4 — Pneumocystis pneumonia and prophylaxis

Examiner: Move on. Suppose the patient was an HIV-exposed infant with hypoxaemia disproportionate to the chest X-ray. How would that change your frame? [9]

Candidate: That is the classic presentation of Pneumocystis jirovecii pneumonia — the AIDS-defining illness of infancy. The hypoxaemia is out of proportion to the chest radiograph because the organism fills the alveoli with a foamy exudate, and the child is tachypnoeic and hypoxaemic with a subacute cough. I would start high-dose intravenous trimethoprim-sulfamethoxazole, and I would add adjunctive corticosteroids for hypoxaemia — a PaO2 below about 70 millimetres of mercury — because the steroid dampens the inflammatory pneumonitis that flares as the organisms die, and this single adjunct improves survival. I would confirm with bronchoalveolar-lavage immunofluorescence or polymerase chain reaction where feasible, without delaying therapy. [9]

Examiner: Finally, how do you prevent fungal disease in the at-risk child? [1]

Candidate: Through targeted antifungal prophylaxis matched to the host and the risk window. Fluconazole or caspofungin during acute myeloid leukaemia induction and prolonged neutropenia — the Fisher JAMA trial showed caspofungin reduced invasive fungal disease versus fluconazole. Posaconazole in graft-versus-host disease, which adds mould cover. Trimethoprim-sulfamethoxazole for Pneumocystis, from the start of immunosuppression to immune reconstitution. And the non-pharmacological prophylaxis — meticulous line-care bundles, minimising parenteral nutrition, antibiotic stewardship, and environmental controls against aerosolised mould. Prophylaxis is the durable prevention; empirical therapy is the safeguard when prophylaxis fails or the risk window is breached. [1] [9]

Wrap

Examiner: Summarise the paediatric fungal-infection stance in one sentence. [14]

Candidate: Set the host-risk tier first — immunocompetent with superficial disease or immunocompromised with invasive disease — because it fixes the antifungal strength, the route and the urgency, from topical therapy for the well child to empirical echinocandin or liposomal amphotericin B for the persistent febrile neutropenic child, with line removal, surgical debridement and reversal of immunosuppression as the source-control partners, and targeted prophylaxis as the durable prevention. [14] [1]