High-Yield Hematology for USMLE Step 3: Anemia, Coagulation, and TTP

Here's what catches most residents off guard about hematology on Step 3: the questions aren't really testing whether you memorized a list of lab values. They're testing whether you can look at a clinical picture — a peripheral smear, a coag panel, a patient who's bleeding and clotting at the same time — and recognize the pattern fast enough to act. I've watched hundreds of students prep for this exam, and the ones who struggle with hematology are almost always overthinking individual numbers instead of seeing the constellation.

Let me walk you through the material the way I teach it on rounds.

Anemia: Pattern Recognition Over Memorization

Microcytic Anemias

Most students can rattle off the iron studies for iron deficiency. Fewer can explain why each value moves the way it does. That understanding is what separates you from the person next to you on test day.

Iron deficiency anemia:

• Low MCV, low serum iron, low ferritin, HIGH TIBC
• High RDW (anisocytosis — your marrow is churning out variably sized cells because it's starving for iron)
• Elevated platelet count (reactive thrombocytosis — a detail test-writers love to throw into a stem as a distractor)
• Causes: blood loss (GI is the big one in adults — always think colonoscopy in a man over 50 with unexplained iron deficiency), malabsorption, pregnancy
• Treatment: oral ferrous sulfate; IV iron if malabsorption or oral intolerance

Beta-thalassemia trait:

• Low MCV, normal/high ferritin, NORMAL/LOW TIBC, HIGH RBC count
• Low RDW (uniform microcytosis — every cell is small, but they're all the same small)
• Mentzer index = MCV/RBC: < 13 favors thalassemia; > 13 favors iron deficiency
• Confirm with hemoglobin electrophoresis (elevated HbA2 > 3.5%)

The Mentzer index comes up on exams more than you'd expect. It's a quick mental shortcut when you see a microcytic anemia with a surprisingly high RBC count — that pattern alone should make you think thalassemia trait before you even see the iron studies.

Anemia of chronic disease:

• Often normocytic (can drift mildly microcytic)
• Low serum iron AND low TIBC — this is where students get tripped up. Both drop because inflammation suppresses erythropoiesis at multiple levels
• Normal/elevated ferritin (it's an acute phase reactant, so inflammation pushes it up even when iron stores are genuinely low)
• High hepcidin (the master regulator — locks iron inside macrophages and enterocytes)

Surprising insight: Ferritin between 30-100 in a hospitalized patient is essentially a diagnostic black hole. It's too high to confirm iron deficiency and too low to rule it out when acute phase reactants are in play. In real practice, I order a soluble transferrin receptor level or reticulocyte hemoglobin content to break the tie. For Step 3, if ferritin is in that gray zone and the stem mentions chronic inflammation, lean toward anemia of chronic disease — but know that mixed pictures exist.

Macrocytic Anemias

B12 deficiency:

• Megaloblastic anemia: hypersegmented neutrophils (>5 lobes) on peripheral smear, macro-ovalocytes
• Neurological involvement (subacute combined degeneration): posterior columns go first (proprioception, vibration) then lateral corticospinal tracts (spasticity, weakness). The combination of a wide-based gait with hyperreflexia is classic
• Causes: pernicious anemia (anti-intrinsic factor antibodies), strict vegan diet, gastric bypass, metformin use (less rare than textbooks suggest — I've seen several cases)
• Serum B12 low; methylmalonic acid (MMA) elevated and is the more sensitive marker

Folate deficiency:

• Megaloblastic anemia, hypersegmented neutrophils — looks identical to B12 on the smear
• NO neurological involvement. Full stop. If there are neuro findings, it's B12 until proven otherwise
• Homocysteine elevated; MMA normal (this is your differentiator)
• Causes: alcohol use, malnutrition, medications (methotrexate, trimethoprim, phenytoin), pregnancy

Critical pearl: Always check B12 BEFORE giving folate in macrocytic anemia. Folate supplementation will correct the hematologic picture — your CBC looks better, everyone relaxes — while B12-mediated neurological damage silently progresses. I've seen patients diagnosed with "folate deficiency" who came back months later with irreversible neuropathy. This sequence matters.

Coagulation Disorders

DIC

Disseminated intravascular coagulation is the board exam's favorite paradox: a patient who is clotting everywhere and bleeding everywhere simultaneously. The consumption is the key concept — your body burns through clotting factors and platelets forming microthrombi, leaving nothing in reserve for hemostasis.

Lab findings (expect all five on Step 3):

• Prolonged PT and PTT
• Elevated D-dimer (fibrin degradation products)
• LOW fibrinogen (consumed in all that pathological clotting)
• Thrombocytopenia (platelets consumed)
• Schistocytes on peripheral smear (RBCs physically sheared by fibrin strands in the microvasculature)

Common triggers: Sepsis (far and away the most common), obstetric emergencies (placental abruption, amniotic fluid embolism), malignancy (especially AML-M3/APL — start ATRA immediately), massive transfusion, snakebite (yes, it shows up on exams)

Treatment: Fix the underlying cause — everything else is temporizing. Replace what's depleted: FFP for prolonged PT/PTT, cryoprecipitate when fibrinogen drops below 150, platelets if thrombocytopenic with active bleeding.

Heparin-Induced Thrombocytopenia (HIT)

HIT is one of medicine's cruelest ironies: you give an anticoagulant to prevent clots, and the immune response it triggers makes your patient more thrombotic. Antibodies form against PF4-heparin complexes and activate platelets aggressively.

Diagnosis — the 4T score: Thrombocytopenia, Timing, Thrombosis, oTher cause unlikely

• Platelet drop ≥ 50% starting 5-10 days after heparin initiation (the timing window is important — Day 1 thrombocytopenia is almost never HIT)
• Confirm with ELISA for anti-PF4/heparin antibodies, then serotonin release assay if ELISA is positive

Management — four non-negotiable rules:

1. Stop ALL heparin. Every last bit. LMWH, flushes, heparin-coated catheters — all of it
2. Start a direct thrombin inhibitor (argatroban is the go-to; bivalirudin for cardiac cath)
3. Do NOT transfuse platelets — you're adding fuel to the fire. Activated platelets mean more thrombosis
4. Do NOT bridge to warfarin until platelets recover above 150,000 AND you've completed at least 5 days of DTI. Starting warfarin too early causes limb gangrene and skin necrosis (protein C drops faster than procoagulant factors)

Contrarian take: The 4T score gets a lot of love in board prep, and it is high-yield. But in real clinical practice, its positive predictive value in ICU patients is mediocre — maybe 10-15% for intermediate scores. Most thrombocytopenia in the ICU is sepsis, drugs, or dilution. On Step 3, the 4T score is gospel. At the bedside, treat an intermediate score as a reason to send the ELISA, not a diagnosis.

Thrombotic Microangiopathies

TTP vs. HUS

Both present with microangiopathic hemolytic anemia (MAHA) + thrombocytopenia + schistocytes on smear. The question is always: which one? And more importantly — what do you do right now?

Feature	TTP	HUS
Dominant organ	Brain (confusion, seizures, focal deficits)	Kidney (AKI, oliguria, hematuria)
Mechanism	ADAMTS13 deficiency (autoimmune or congenital)	Shiga toxin (E. coli O157:H7, Shigella); complement-mediated
Demographics	Adults, F > M	Children (typical/STEC-HUS); adults (atypical/drug-induced)
Fever	Often present	Variable
Definitive treatment	Plasmapheresis — emergent, do not delay	Supportive care for STEC-HUS; eculizumab for atypical HUS
Platelet transfusion	Contraindicated — worsens microthrombi	Use with extreme caution

The pearl that saves lives on CCS cases: Forget the "TTP pentad." Seriously. The classic teaching (fever, MAHA, thrombocytopenia, neuro symptoms, renal failure) leads students to wait for all five features before acting. You need exactly two — MAHA plus thrombocytopenia with no other explanation — to pull the trigger on emergent plasmapheresis. Waiting for the full pentad is how patients die.

Surprising insight: Here's something boards won't explicitly tell you but will absolutely test. In a child with bloody diarrhea and MAHA, do NOT give antibiotics for the E. coli O157:H7. Antibiotic-mediated bacterial lysis releases more Shiga toxin and worsens HUS. Supportive care only. This trips up students who reflexively want to treat the infection.

Transfusion Medicine

Thresholds Worth Memorizing

These numbers are tested repeatedly. Know them cold.

• Packed RBCs: Hgb < 7 g/dL in stable patients; Hgb < 8 g/dL for acute coronary syndrome or cardiac surgery
• Platelets: < 10,000 prophylactically (even without bleeding); < 50,000 before most invasive procedures; < 100,000 for neurosurgery or ocular surgery
• FFP: coagulopathy with active bleeding or before a high-risk procedure when INR > 1.5
• Cryoprecipitate: fibrinogen < 150 mg/dL with active bleeding; also your source for factor VIII, factor XIII, and von Willebrand factor when DDAVP isn't enough

Massive Transfusion Protocol

Hemorrhagic shock with ongoing blood loss → balanced resuscitation: pRBC:FFP:platelets in a 1:1:1 ratio. Permissive hypotension (target MAP 50-65 mmHg) until you get surgical hemorrhage control — overaggressive fluid resuscitation dilutes clotting factors and worsens coagulopathy. Tranexamic acid within 3 hours of injury reduces mortality; after 3 hours, evidence suggests it may actually increase harm. That time cutoff is very testable.

Leukemia High-Yield Pearls

• AML-M3 (APL): Presents with DIC. This is an oncologic emergency. Start ATRA (all-trans retinoic acid) + arsenic trioxide the moment you suspect it — do NOT wait for confirmatory cytogenetics. The DIC can kill your patient before the pathology report comes back
• CML: Philadelphia chromosome (BCR-ABL, t(9;22)) → imatinib is first-line and transformed this from a fatal diagnosis to a manageable chronic disease. Watch for blast crisis (transformation to acute leukemia) — that's when the story changes dramatically
• CLL: Most common leukemia in adults over 65. Smudge cells on the peripheral smear are your giveaway. Many patients are watch-and-wait for years. Richter transformation (conversion to aggressive DLBCL) is the complication boards love — sudden B symptoms, rapidly enlarging nodes, rising LDH

Frequently Asked Questions

How do I distinguish iron deficiency from thalassemia trait when both show microcytic anemia?

Look at the RBC count and the RDW. Iron deficiency gives you fewer red cells that vary in size (high RDW, low RBC count). Thalassemia trait gives you more red cells that are uniformly small (low RDW, high RBC count). The Mentzer index (MCV/RBC < 13 = thalassemia) is a quick screen. Ferritin and hemoglobin electrophoresis confirm.

What's the most commonly missed diagnosis in macrocytic anemia workups?

Hypothyroidism and alcohol use — neither causes megaloblastic changes on the smear, and both are far more common than B12 or folate deficiency. If you see macrocytosis with a normal smear (no hypersegmented neutrophils, no macro-ovalocytes), think non-megaloblastic causes first: thyroid disease, liver disease, alcohol, reticulocytosis, or myelodysplastic syndrome.

When should I suspect HIT versus other causes of thrombocytopenia in a hospitalized patient?

Timing is everything. HIT classically appears 5-10 days after heparin exposure, with a platelet drop of 50% or more from baseline. Day 1-2 thrombocytopenia is almost always something else (sepsis, dilution, drug-induced). The exception: patients with prior heparin exposure within the last 100 days can develop rapid-onset HIT within hours. If thrombocytopenia comes with a new clot, your suspicion should jump significantly.

Why are platelet transfusions contraindicated in TTP?

Because TTP is driven by ultra-large von Willebrand factor multimers that spontaneously recruit platelets into microthrombi. Adding more platelets is like throwing kindling on a fire — you're giving the disease more substrate to form clots. The treatment is plasmapheresis to remove the pathological multimers and replace ADAMTS13.

What's the single highest-yield transfusion fact for Step 3?

The restrictive transfusion threshold: Hgb < 7 g/dL in stable, non-cardiac patients. This comes from the landmark TRICC trial and gets tested constantly. The exception — Hgb < 8 for ACS and cardiac surgery — is equally testable. Don't transfuse based on a number alone; always consider the clinical picture. But if forced to pick a cutoff, 7 is your answer.

Practice Hematology Questions

Ready to put this knowledge to work? Step3Sim offers free USMLE Step 3 practice questions for hematology and all other organ systems. The best way to lock in pattern recognition is to see these scenarios in simulated exam format — where you're making decisions under time pressure, not just reading about them.