Case Presentation: Our patient was a 46-year-old man who presented with two days of malaise, headache, fever and productive cough. He denied chest pain, shortness of breath, exposure to sick contacts or recent travel. He has a past medical history of Celiac disease with associated dermatitis herpetiformis treated with dapsone and betamethasone cream for several years. Two weeks prior to presentation, the dapsone dose was increased due to ongoing dermatitis.Physical ExamThe patient was tachycardic with a heart rate of 110 and febrile with a temperature of 101.2. His oxygen saturation was mid to high 80s on 6-10 litres/minute of oxygen by high flow nasal canula. He was in mild distress, but able to speak in complete sentences. His cardiac exam was normal. Pulmonary exam was notable for reduced breath sounds and ronchi in the right lower lobe. Perioral and auricular cyanosis was noted. Laboratory Work and imagingHis white blood cell count was 18.4 x 10^3/uL, Hemoglobin concentration was 13.5 g/dL, and the MCV was 103.5 fL. Total bilirubin was 3 mg/dL. A right lower lobe opacification was seen on his chest XR. Blood cultures revealed no growth. Pulmonary CT angiography revealed the absence of a pulmonary embolus. Diagnosis The patient was treated for sepsis secondary to lobar pneumonia. Despite antibiotics and supplementary oxygenation via nasal canula, oxygen saturation remained 80-90%. The hypoxia was out of proportion to the pulmonary infiltrate, prompting additional investigation including evaluation of methemoglobin level which was elevated at 15.2%
Discussion: This patient was diagnosed with acquired methemoglobinemia due to dapsone. Methemoglobinemia can be a challenging diagnosis to uncover given its rare incidence in the population and nonspecific presentation. Methylene blue (MB) and ascorbic acid are the standard treatments for this disease. We elected to use ascorbic acid, as MB can induce hemolysis in patients with G6PD deficiency.
Conclusions: In this case, the discovery of methemoglobinemia was made by investigating his persistent hypoxia. Accepting the current diagnosis of pneumonia would have ended the diagnostic process. This is known as premature closure, a cognitive bias. We employed type 2 clinical reasoning, a slow and deliberate analysis, to ask ourselves why this patient was not improving despite treatment. Repetition of type 2 clinical reasoning ultimately results in better ability for physicians to use type 1 reasoning, which is quicker, more intuitive and relies on recognition of patterns.