A 62 year-old male presents with hepatosplenomegaly, skin lesions, and cardiomyopathy. A representative field of his blood smear is shown here.
Which of the following is the most likely diagnosis?
A. Multiple myeloma
B. Chronic myeloid leukemia
C. Metastatic prostate carcinoma
D. Hypereosinophilic syndrome
E. Bacterial sepsis
Â
(Scroll down for the answer)
The diagnosis in this case is hypereosinophilic syndrome, a rare myeloproliferative disorder characterized by a marked and persistent elevation in the eosinophil count. Multiple organs are usually affected, and patients may present with cardiomyopathy, skin lesions, thromboembolic disease, neuropathy, hepatosplenomegaly, or pulmonary disease. The pathophysiology behind the damage in these organs isn’t well understood, but it probably has something to do with the release of eosinophil granules in those tissues.
You need to have three things in order to make the diagnosis:
- Eosinophilia (absolute eosinophil count >1500/μL)
- No other cause for the eosinophilia
- Signs and symptoms of organ involvement
Other more common causes of eosinophilia, such as drug reactions, autoimmune disease, infection, hematologic malignancy, and familial eosinophilia, must be ruled out before making the diagnosis.
Usually, patients aren’t treated unless or until they have symptoms (because the treatment itself has its risks). These patients are monitored closely with serum troponin levels (to monitor for MI), echocardiograms and pulmonary function tests.
Some patients with hypereosinophilic syndrome have a tiny deletion in 4q12, which ends up producing a fusion transcript called FIP1LI-PDGFRA (which is also present in some cases of systemic mastocytosis). Imatinib works very well in the majority of these patients.
Patients who have symptoms (but not FIP1LI-PDGFRA) are generally treated with steroids first. If those don’t work, interferon alpha and hydroxyurea are used. If those don’t work either, then imatinib is the treatment of choice (it doesn’t work as well as it does in patients with FIP1LI-PDGFRA, but it does seem to work in at least some of these patients).
Here’s an explanation of why the other answers are incorrect:
A. In a blood smear from a patient with myeloma, the main thing you see is rouleaux (red cell stacking up on top of each other). You generally don’t see plasma cells in the blood – but you do see them in the marrow. This blood smear just has a lot of eosinophils – so it’s not consistent at all with myeloma. The clinical history also doesn’t fit. In myeloma patients usually have bone pain, signs of anemia (fatigue, palpitations), and maybe signs of renal failure. Hepatosplenomegaly, skin lesions, and cardiomyopathy aren’t generally seen in myeloma.
B. In CMLÂ (check out the photo above), you see a massive leukocytosis which is composed of neutrophils and precursors. There is a big left shift, with “bulges” (more cells) at the myelocyte and segmented neutrophil stages. Most cases also show a distinct basophilia. This smear just has a ton of eosinophils, which is not a finding you see in CML. The patient’s hepatosplenomegaly would be consistent with CML (especially the splenomegaly part) – but the skin lesions and cardiomyopathy don’t go along with that diagnosis.
C. In metastatic prostate cancer, it’s possible that you might see a rare tumor cell in the blood; you might also see a monocytosis (you occasionally can see that with solid tumors). Eosinophilia is not consistent with prostate cancer. The history, too, doesn’t fit. Patients with prostate cancer often have urinary symptoms (trouble urinating, blood in the urine), and some have signs of metastasis (bone pain, particularly in the spine).
D. The characteristic blood finding in bacterial sepsis is a neutrophilia, with or without a left shift. You may also see toxic changes in the neutrophils: toxic granulation, Dohle bodies, and/or cytoplasmic vacuolization. Bacterial infections generally don’t produce an eosinophilia – but some parasitic infections can.
If you want to test yourself with other unknown cases, here are some to try:
- Case 1: 20-year-old male who died suddenly
- Case 2: 72-year-old male with right calf mass
- Case 3: 67-year-old female with pancytopenia
- Case 4: 59-year-old male with severe headaches
- Case 5: 38-year-old female with deep venous thrombi
- Case 6: 13-year-old male with cerebellar mass
- Case 7: 45-year-old male with pulmonary emphysema
- Case 8: 38-year-old male with AIDS and headaches
- Case 9: 25-year-old male with arm mass
- Case 10: 57-year-old male with fatigue and left upper quadrant heaviness
Those look like basophils. Is the picture bad, or are there an increased number of basophils in hypereosinophilic syndrome?
No – they are eosinophils. Sometimes you can see a darker, almost bluish-purple color in eosinophil granules, particularly when they are part of a malignant process. There is a type of acute myeloid leukemia – AML-M4 Eo, with an inv(16) – in which this abnormal color is particularly prominent. If you look closely, you can see normal eosinophil granules in each of these cells. Take a look at the granules in the cell at 2 o’clock – then compare those to the granules in the rest of the cells.
Would eosinophil infiltrates be present in a biopsy of symptomatic organ involvement?
Yes – affected tissues show eosinophilic infiltrates. In the skin, they are often perivascular in location. Other findings vary by organ, but often include thrombotic microangiopathy (with necrosis, sometimes). Chronic hypereosinophilia can damage the endothelium and can activate coagulation directly.
those cells really looked like basophils to me. how do you tell the difference between a basophil and an eosinophil on peripheral film??
I know – it can be tough, particularly when the cells are abnormal. Normal eosinophils have orangeish granules and several nuclear lobes. Normal basophils have deep blue-purple granules and usually only two lobes to the nucleus (though the nucleus is usually almost totally obscured by the granules). Abnormal eosinophils can have the darker, almost bluish granules you see here. The granules are usually really big (bigger than normal basophil granules) and you can almost always find some orange ones if you look around (as in this picture).
how can we differenciate blood cells looking on.their morphology. i want to.know…
Here are a couple posts that might help: one on white blood cells in general, and one on neutrophils and precursors.