Tense bullae in an older woman with CKD and diabetes
An 81-year-old woman presents with a 4-week history of intensely pruritic tense bullae on the trunk and limbs, arising on urticarial plaques. Oral and genital mucosae are spared. She takes furosemide, started 6 weeks ago for ankle oedema, sitagliptin for type 2 diabetes, and atorvastatin. She has stage 3b chronic kidney disease (eGFR 38 mL/min/1.73 m²) and HbA1c 8.4%. Eosinophils are 1.4 × 10⁹/L; haemoglobin 10.8 g/dL. A skin biopsy shows a subepidermal blister with eosinophils. Direct immunofluorescence shows linear IgG and C3 along the basement membrane zone. Salt-split skin shows linear IgG on the epidermal roof. She is reluctant to take steroids because of her diabetes and asks whether the new diuretic is responsible.
1. Interpret the immunofluorescence and salt-split findings. Which diagnoses remain, and which is most likely?
Read the worked answer
The combination of a subepidermal blister with eosinophils and linear IgG and C3 along the basement membrane zone places the disease in the pemphigoid group. Salt-split skin is the key discriminator: linear IgG on the epidermal roof indicates binding to antigens on the hemidesmosome, chiefly BP180 and BP230, and is characteristic of bullous pemphigoid. Epidermolysis bullosa acquisa would show linear IgG on the dermal floor, because its target is type VII collagen below the lamina densa. Linear IgA disease would usually show linear IgA, not IgG, although mixed patterns occur.
The most likely diagnosis is bullous pemphigoid. Drug-induced pemphigoid remains possible and can be immunologically indistinguishable; furosemide and DPP-4 inhibitors such as sitagliptin are recognised associations. The temporal relationship with furosemide is suggestive but not proof, because bullous pemphigoid is common in this age group. The sparing of mucosae and the urticarial plaques also fit pemphigoid better than mucous membrane pemphigoid or EBA.
2. How do you assess drug causality, and what would you do about furosemide and sitagliptin?
Read the worked answer
I would assess causality using the temporal relationship, biological plausibility, and the effect of withdrawal. Furosemide started 6 weeks before the eruption, which fits a drug-triggered pemphigoid latency; sitagliptin has also been associated with bullous pemphigoid, although she has taken it for longer. Causality remains probabilistic: many older patients develop pemphigoid without a drug trigger, and the disease can persist after withdrawal.
I would not stop furosemide unilaterally, because it may be controlling heart failure or oedema. I would discuss with her cardiologist whether an alternative diuretic is safe. Sitagliptin is easier to replace with another glucose-lowering agent, so I would discuss this with her diabetes team. If a drug trigger is suspected, withdrawal may reduce disease activity over weeks to months, but it does not replace treatment of active bullous pemphigoid. The patient should be told that the diuretic may be contributory but that proof is difficult and that stopping it without review could be harmful.
3. Design a treatment plan that respects her CKD and diabetes.
Read the worked answer
For widespread symptomatic bullous pemphigoid, I would use prednisolone at a modest dose, approximately 0.3–0.5 mg/kg/day, with a planned taper, plus a steroid-sparing agent. Doxycycline 100 mg twice daily can be used as a steroid-sparing option and is not nephrotoxic, although it requires monitoring for gastrointestinal effects and photosensitivity. Ultrapotent topical clobetasol can be used for limited disease but is impractical for widespread bullae. If the disease is severe or refractory, rituximab is an option; dupilumab is emerging but off-label.
In CKD, I would avoid NSAIDs and minimise nephrotoxic drugs, monitor renal function, and arrange bone protection with calcium, vitamin D, and a bisphosphonate where appropriate. In diabetes, I would intensify glucose monitoring during steroids and expect hyperglycaemia. I would also check for superinfection, provide wound care, and review at 2–4 weeks. The plan should be explicitly shared with the patient: steroids are not contraindicated, but they require close monitoring and a steroid-sparing strategy.