A Cold Van, an MCHC of 45.6 g/dL, and a Request for a "Stat" Haemoglobin
A 58-year-old woman attends the emergency department on a January morning with a temperature outside of −5°C. Blood is drawn at 08:10 into a 4 mL K2EDTA tube, transported in an unheated courier van and received at 10:05; it is analysed at 10:40. The CBC shows WBC 9.2 ×10^9/L, RBC 1.62 ×10^12/L, Hb 12.8 g/dL, Hct 28.1%, MCV 173 fL, MCHC 45.6 g/dL and platelets 180 ×10^9/L. The analyser flags possible red cell agglutination. A film made from the same tube shows irregular clumps of red cells in the monolayer, with some rouleaux-like clusters; the white cell differential is unremarkable. The tube is warmed to 37°C for 30 minutes and re-analysed: RBC 3.98 ×10^12/L, Hb 12.9 g/dL, Hct 38.4%, MCV 96.5 fL, MCHC 33.6 g/dL, platelets 176 ×10^9/L. Her haemoglobin a year ago was 13.1 g/dL. The ED physician wants the "corrected haemoglobin" immediately to decide on transfusion before a CT scan.
1. Explain the discordance among the RBC count, Hb, Hct, MCV and MCHC in the cold sample, and why haemoglobin is the most reliable parameter.
Read the worked answer
Cold agglutinins cause red cells to clump. The analyser counts each clump as a single particle, so the RBC count falls (1.62 instead of about 4.0). The clump is also measured as one large cell, so the MCV rises (173 fL). The haematocrit is derived from RBC × MCV, so it is distorted - here it fell to 28.1%. The MCHC is Hb ÷ Hct, and because the denominator is falsely low the MCHC rises to 45.6 g/dL, a value that is physiologically almost impossible and is the classic red flag for agglutination, lipaemia, icterus, haemolysis or spherocytosis.
Haemoglobin is measured by spectrophotometry after lysis of the red cells and does not depend on counting particles or on cell size. Agglutinates can cause slight turbidity and a small positive bias, but here the Hb was 12.8 g/dL before warming and 12.9 g/dL after, so it is stable and reliable.
The general rule is that an MCHC above about 36–37 g/dL should be treated as an analytical alarm until proven otherwise, and the first step is to check the sample and the film before reporting any derived index.
2. What should be reported, and what else should be done before the patient goes to CT?
Read the worked answer
The corrected, 37°C results should be reported: RBC 3.98 ×10^12/L, Hb 12.9 g/dL, Hct 38.4%, MCV 96.5 fL, MCHC 33.6 g/dL. The report should carry a comment that the initial indices were invalid because of cold agglutination and that the repeat was performed after warming.
The haemoglobin of 12.9 g/dL is only mildly reduced and, by itself, is not an indication for transfusion in a stable patient; the decision should be based on symptoms, vital signs and the reason for the CT. The physician can proceed rather than delay imaging.
Before the result is finalised, it is worth checking a retic count, bilirubin, LDH and haptoglobin, because a cold agglutinin may reflect a haemolytic process, and examining the film for spherocytes. It is also worth confirming that the platelet count is not affected by clumping, and that the WBC count and differential are consistent with the film. If there is any question about the haematocrit, a spun microhaematocrit on a warmed sample gives an independent check, although trapped plasma can slightly raise it.
3. What preanalytical instructions should be given for this patient's future samples, and what would you do if warming did not resolve the agglutination?
Read the worked answer
Future samples should be drawn into a pre-warmed tube where possible, kept at 37°C during transport (a warm water bath or incubator, not a pocket), delivered directly to the laboratory and analysed immediately, with the request marked to indicate a known cold agglutinin. The tube should never be left in a cold van or a refrigerator. If a pneumatic tube system is used, the effect of the transit time and temperature should be considered.
If warming did not correct the indices, the possibilities are a high-titre cold agglutinin that persists at 37°C, a warm autoantibody, or true rouleaux from a paraprotein. In that case the report should give the haemoglobin with a comment that the RBC count, MCV and derived Hct are unreliable, rather than issuing a calculated MCHC. Manual methods - a spun haematocrit and a manual haemoglobin - or a saline replacement technique may be needed. The film should be examined after warming, because true rouleaux disperses while agglutination does not. The underlying diagnosis should be pursued with a cold agglutinin titre and, if indicated, investigation for an underlying lymphoproliferative disorder.