Why do blood types have to match in a transfusion?
― A doctor mixed his own blood with his colleagues' to find out
All blood looks like the same red liquid. But mix the blood of two different people, and the red particles can stick together and clump. Around 1900, one doctor patiently worked out which combinations clump and which don't. That was the beginning of the blood types we know today.
A poster at a blood donation centre says, "We are short of type A and type O blood." Before you have an operation in hospital, your blood type is always checked.
You may hear about blood types more often in horoscopes and talk about personality. But they were first discovered as a way to understand the body, so that transfusions would stop killing people.
So what were transfusions like before blood types were known? And what happens inside the blood when the types don't match?
Mismatched blood clumps for just two reasons
The surface of a red blood cell is covered with tiny sugar chains, like small decorations. The shape of their tips differs a little between type A and type B people. Type AB has both, and type O has neither.
The liquid part of blood holds substances (antibodies) that stick to markers the person doesn't have. They are in place from soon after birth. They grab red cells of another type and bind them into clumps.
So what matters in a transfusion is the match between "the markers on the incoming red cells" and "the hands in the recipient's blood." The person who first spotted these two things was a Viennese doctor, Karl Landsteiner.
The days of transfusing without knowing
Transfusions from person to person were already being tried in the 19th century. Some people were saved. But with exactly the same procedure, others suddenly became ill and died, so it is said.
Nobody knew why it sometimes worked and sometimes failed. Transfusion was a gamble with uncertain odds.
Around 1900, Landsteiner carried out a simple experiment. He collected a little blood from colleagues in his lab, himself included. He separated each sample into red cells and the liquid part (serum). Then, it is said, he dropped one person's red cells into another person's serum and checked, for every pairing, whether they clumped.
The results were not random. The clumping followed a neat pattern. He showed that human blood could be sorted into three groups. The following year, his students found a fourth type (today's type AB).
Why is clumping of red cells dangerous?
Look at Figure 1. At each ✕, the recipient's antibodies grab the markers on the incoming red cells. Each antibody has several arms, so it grabs many red cells at once and turns them into a lump like a bunch of grapes.
On top of that, red cells tagged by antibodies are destroyed in the blood. When the contents of the broken cells spill out all at once, it puts a heavy strain on organs such as the kidneys. Sudden fevers after transfusion, and many life-threatening symptoms, are thought to have been caused by this.
Conversely, once the rules are known, such failures can be prevented. From around 1907, doctors began a "crossmatch test" before transfusion: they mix a little of the recipient's serum with the blood to be given and see whether it clumps. That procedure continues today, in modified forms.
Normally, antibodies are made after meeting a germ. But even people who have never had a transfusion carry antibodies against A and B. Gut bacteria and other organisms have similar markers, and the leading idea is that babies make these antibodies after being exposed to them. They are tuned so as not to attack the markers on the person's own body.
Around 1940, Landsteiner and colleagues also discovered another marker, "Rh". In Japan, about 1 person in 200 is said to be Rh-negative. In transfusions today, the basic rule is to match both the ABO and the Rh systems.
Summary
A blood type is a difference in the markers on the surface of red blood cells. Blood naturally contains antibodies that grab markers you don't have. So if red cells of the wrong type are given, they clump and are destroyed. Landsteiner found this rule by mixing bloods and checking, one pairing at a time, whether they clumped. For this work he received the Nobel Prize in 1930.
What separated success from failure in transfusion was not luck, but
tiny markers on the surface of red blood cells.
For the story of how we learned that blood flows around the body, see "How did we find out that blood circulates around the body?" For the struggle to get medical common sense accepted, see "Why does washing hands prevent disease?"
- Find out your family's blood types from sources such as a maternal and child health handbook or blood donation records (leave blank anyone you don't know).
- A blood type is set by a pair of letters from "A, B, O", one from your father and one from your mother. A and B show more strongly than O. For example, AO is type A, and only OO is type O.
- Pick possible letter pairs for the parents and make a table of the types their children could have. Check that it doesn't contradict your family's actual types.
Two type A parents can have a type O child because both quietly carry an O. A few years after Landsteiner's discovery, it was also confirmed that blood types follow the rules of heredity.
For those who want more ― terms, formulas and links to textbooksEach part is labelled by level, from middle-school science to university courses
- Middle schoolCovered in middle-school science
- High schoolCovered in high-school "Biology"
- High school+Advanced high-school material, or textbook sidebars
- UniversityUniversity courses not taught in high school (immunology, biochemistry, population genetics)
- ResearchNot yet taught even at university as settled fact; researchers are studying it now
Middle schoolTerms: this phenomenon has names
- ABO blood group system: A way of sorting blood into four types, A, B, O and AB, by whether the A antigen and B antigen are present on red cell surfaces.
- Antigen and antibody: An antigen is a substance that serves as a marker; an antibody is a protein that sticks specifically to it. Type A people have anti-B antibodies, and type B people have anti-A antibodies.
- Agglutination: When antibodies grab many red cells at once and bind them into clumps. Blood typing is done by looking for this by eye.
- Crossmatch test: A test before transfusion in which a little of the recipient's and donor's blood are mixed to check that no agglutination occurs.
Middle schoolHigh schoolChecking with a formula: how often would blood match if the type were unknown?
Before blood types were known, whether a transfusion worked was a matter of chance. Using the blood type proportions of people in Japan, let's calculate the probability of a match when red cells come from a randomly chosen person.
| In symbols | P ＝ Σ（ p_recipient × q_match ） |
| In words | Probability of a match ＝ the sum, over the four types, of "the proportion of recipients with that type × the proportion of donors that match that type" |
| Where it comes from | The multiplication rule and addition rule of probability. The recipient's and donor's types are decided independently, so the probability of a combination is a product, and separate cases are combined by addition. |
| Proportion in Japan (type A) | About 0.4 (roughly 4 in 10) |
| Proportion in Japan (type O) | About 0.3 |
| Proportion in Japan (type B) | About 0.2 |
| Proportion in Japan (type AB) | About 0.1 |
| Proportion of matching donors | For type A recipients, types A and O (0.7); for type O, only type O (0.3); for type B, types B and O (0.5); for type AB, everyone (1) |
Meaning of the symbols: P is the probability of a match, p_recipient is the proportion of recipients with that type, and q_match is the proportion of donors that match that type. There are no units; each is a number from 0 to 1.
| Recipient is type A, and it matches | 0.4 × 0.7 ＝ 0.28 |
| Recipient is type O, and it matches | 0.3 × 0.3 ＝ 0.09 |
| Recipient is type B, and it matches | 0.2 × 0.5 ＝ 0.1 |
| Recipient is type AB, and it matches | 0.1 × 1 ＝ 0.1 |
| Add the A and O parts | 0.28 ＋ 0.09 ＝ 0.37 |
| Add the B part | 0.37 ＋ 0.1 ＝ 0.47 |
| Add the AB part to get the probability of a match | 0.47 ＋ 0.1 ＝ 0.57 |
| Probability of a mismatch | 1 － 0.57 ＝ 0.43 |
By this calculation, a random transfusion would be a mismatch about 4 times in 10. That fits the records of the time, in which some people were saved and others were not. Landsteiner's discovery turned a gamble that failed nearly half the time into a test governed by a rule.
High schoolHigh school+Types are set by genes, and antibodies are part of the body's defence
High schoolThe ABO blood group is determined by three versions (A, B, O) of a single gene. A and B are dominant over O, and A and B both show when present together. So AO and AA are type A, BO and BB are type B, AB is type AB, and only OO is type O. Antibodies are proteins responsible for humoral immunity, which you study in the "immunity" unit of high-school biology.
High school+The markers on red cells are chains of sugar (glycans) sticking out from the cell surface. To the base chain that type O people have, type A people add one kind of sugar and type B people add a different single sugar, each using an enzyme. The difference between type A and type B is just one sugar at the end of the chain. Antibodies can tell that tiny difference apart.
UniversityGlycosyltransferases and the proportions of types in a population
The genes that decide type A and type B are blueprints for enzymes that add sugars (glycosyltransferases). The type A enzyme attaches N-acetylgalactosamine to the base H antigen, and the type B enzyme attaches galactose. The type O gene is broken partway through and cannot make a working enzyme. The proportions of blood types in a population can be calculated from the frequencies of the three genes using the "Hardy–Weinberg principle," which makes blood types a classic teaching example in population genetics.
📖 Derivation of the formula, and further reading: ABO blood group system (Wikipedia, Japanese) ／ Hardy–Weinberg principle (Wikipedia, Japanese)
ResearchWhat is still not well understood
- Why do people have several blood types at all? Many links have been reported, such as certain types being less prone to certain infections. One idea is that a variety of types survived through a long history with pathogens, but the whole picture is still unclear.
- The detailed route by which natural antibodies form. The leading idea involves gut bacteria, but which component of which bacteria acts, and at what age, is still being studied.
- Can we make red cells usable for any type? Research is under way on using enzymes that cut off the A or B sugars to make red cells more like type O. If it becomes practical, it is hoped to ease shortages of blood caused by type.
In other words, what this article says is also "an explanation of what is known so far." Still, the principle of matching types in transfusion has been confirmed by more than 100 years of clinical practice.
Links to textbooks (by level)
| Level | Subject / unit | Where in this article |
|---|---|---|
| Middle school | Science, year 2: "Functions of blood"; maths: "Probability" | Red cells and plasma; calculating the chance of a match |
| High school | Biology: "Immunity" and "Heredity" | Antigens and antibodies; the three types A, B, O and dominance |
| High school+ | Biology (advanced) | The markers being sugar chains |
| University | Immunology, biochemistry, population genetics | Glycosyltransferases; the Hardy–Weinberg principle |
| Research | Evolutionary medicine, transfusion medicine | Why types are diverse; natural antibodies; enzymes that change the type |
| ― | Links to daily life | Blood donation, blood tests before surgery, the blood type box in the maternal and child health handbook |
- Karl Landsteiner (Wikipedia, Japanese)
- ABO blood group system (Wikipedia, Japanese)
- Landsteiner K. "Ueber Agglutinationserscheinungen normalen menschlichen Blutes." Wiener klinische Wochenschrift, 1901.
- Japanese Red Cross Society (日本赤十字社), "About blood donation" (献血について), various explanatory pages (blood type proportions, basics of transfusion)
- Japanese high-school textbook "Biology" (高校教科書『生物』), immunity and heredity units
* This article is a general-audience science explainer. The figures given are rough guides to help you understand the mechanism. Transfusions and blood typing are carried out at medical institutions; please do not use this article as a basis for your own medical decisions. No scientific evidence has been found for a link between blood type and personality.