William Aird Profile picture
Aug 7, 2023 9 tweets 3 min read Read on X
1/9

COLD AGGLUTININ DISEASE (CAD)

Yesterday I posted CBCs from a patient, posed three questions and asked for a diagnosis.

Kudos to @EltonWandira for coming up with the answers and diagnosis within minutes of posting!

The dx is CAD. Image
2/9

Answers to questions:

Q1. Which value at time 0 is physiologically impossible?

A1. The MCHC of 65.4 g/dL. Hb is almost supersaturated under normal conditions, and rarely increases above 40 g/dL.
3/9

Q2. What does the notation 4 refer to at time 0?

A2. Spun Hct. Note that the Hb is reported as 8.5 g/dL, the Hct as 26.5%. Recall that the MCHC = Hb/Hct, which in this case would = 32 g/dL, a far cry from the reported MCHC of 65.4 g/dL!
4/9

Q3. What has the lab technician done to the blood to obtain 4 h results?

A3: They have warmed the blood to disassemble the cold agglutinins from the RBCs, converting clumps to single cells.
5/9

Instead of simply memorizing this pattern of CBC changes in CAD, let's think our way through it.

In automated hematology analyzers ('CBC machines'), RBCs normally pass in single file as their numbers are counted and their volume measured using the Coulter principle. Image
6/9

In CAD, clumps of 3 or more RBCs cannot squeeze through the aperture used for counting/sizing cells, so fewer cells get through, leading to falsely low RBC count. Doublets can pass through, but are counted as single cells, resulting in falsely high MCV and reduced RBC count. Image
7/9

These artefactual changes can be incorporated in the equation Hct = RBC count x MCV. Because the RBC takes a double hit, it is disproportionately decreased relative to the change in MCV, resulting in a reduction in Hct. Image
8/9

Finally because the Hct is falsely low (and the Hb is one of the few parameters unaffected in CAD), the MCHC is falsely elevated. Image
9/9

We can put it all together in table format with findings and explanations. Image

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with William Aird

William Aird Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @WilliamAird4

Mar 19
1/10

Yesterday’s oral exam case (MCV 140) generated a lot of discussion.

Before I share my approach, I want to slow things down and think through two questions.
2/10

1. How do you structure the physical exam?

Do you:

– move disease by disease (B12, liver disease, hypothyroidism…), or

– move head to toe, mapping findings onto the body?

Both are acceptable.

But they are not the same.
3/10

A disease-by-disease approach follows the differential.

A head-to-toe approach reorganizes the differential into anatomy.

It forces you to translate ideas into space.
Read 10 tweets
Feb 13
1/9

I’ve long taught a simple framework for hypoproliferative normocytic anemia that I still love:

• nutritional deficiency
• inflammation
• organ dysfunction (kidney, liver, endocrine, marrow)

Clinically useful. Memorable. Teachable.

And it still holds up. Image
2/9

But conceptually, something interesting is happening.

These categories aren’t actually the same kind of explanation.

Nutritional deficiency and inflammation are mechanisms.
Organ dysfunction is different.

And even within organs, not all behave the same way.
3/9

Take kidney disease.

We often say “anemia from CKD” as if that explains it.

And in a way, it almost does.

Because kidney disease sits very close to mechanism:
↓ EPO → ↓ erythropoiesis

It’s only one step away.
Read 9 tweets
Feb 13
1/11

I posted a poll:

A woman with ferritin 10 and Hb 12.2 (baseline 14). How should this be described?

Here’s how you answered:

• non-anemic Fe deficiency: 35%
• Fe deficiency anemia: 32%
• Fe deficiency with relative anemia: 27%
• none: 6%

Really interesting spread! Image
2/11

This tells us something important: clinicians sense a mismatch between definition-based language and physiology-based thinking, even if we disagree on terminology.
3/11

By strict WHO criteria, she is not anemic.
Hb ≥12 in women = normal.

So formally the correct label is: iron deficiency without anemia.
Read 11 tweets
Feb 10
RETICS IN PV

I recently ran a poll asking whether polycythemia vera (PV) is associated with:

• ↑ absolute reticulocyte count
• ↑ % reticulocytes
• both
• neither

The most popular answer was also the best answer: ↑ absolute reticulocyte count. Image
2/10

At first glance, that can feel counterintuitive.

PV is a disease of excessive red cell production, so why wouldn’t the reticulocyte percentage also be increased?

The answer depends on what reticulocytes actually measure.
3/10

Two metrics that are often conflated:

• Absolute reticulocyte count = how many reticulocytes are circulating

• Reticulocyte percentage = reticulocytes ÷ total red cell mass

They answer different physiologic questions.
Read 11 tweets
Feb 2
1/13

I posted a poll asking:

In acute GI bleed anemia, would you give 1 g IV iron regardless of ferritin?

Results:

• 27% yes — anticipate iron debt
• 12% sometimes
• 21% only if ferritin is low
• 41% no
2/13

First, an important acknowledgment:
There is no right answer here.

There are no firm guidelines that tell us what to do in this situation. Reasonable clinicians land in different places.

This is a gray zone where physiology, timing, and judgment matter.
3/13

So rather than argue what we should do, I want to walk through the numbers and biology and explain why some clinicians anticipate iron debt even when ferritin is normal.

Let’s take a concrete example.
Read 13 tweets
Jan 23
1/9

Yesterday I posted a CBC + reticulocyte count and asked for your diagnostic thoughts. Many of you offered great reasoning. The correct diagnosis was hemoglobin C disease.

Let’s unpack why this case is such a good learning example. 👇 Image
2/9

Microcytosis often triggers a reflex binary:
iron deficiency vs thalassemia trait.

That’s a useful starting point. But it’s incomplete. Structural hemoglobin variants (like HbC and HbE) also belong on that list.
3/9

Several people calculated the Mentzer index (MCV/RBC):

75 / 4.0 ≈ 18 → “suggests iron deficiency (ID).”

Important teaching point:

The Mentzer index was designed to distinguish thal trait vs ID. It is not validated for structural hemoglobinopathies like HbC or HbE.
Read 9 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(