Monday, November 27, 2006

Blood clots, eh?

ABO groups and thrombosis

or Blood clots, eh? Big grin

The title is a Canuck's mnemonic to recall that thrombosis occurs more in group A than group O individuals.

When teaching, one of the tactics that I regularly used to increase student interest was to include sex or anything unusual into the discussion. One of the topics in the latter category was whether blood groups served biological functions.

The most famous association is between the Duffy blood group system and malaria, in which the FyFy genotype confers resistance to Plasmodium vivax.

ABO was a favorite system for investigating blood group and disease associations probably because it was the first system discovered and because ABH carbohydrates are expressed on many human tissues, including red cells, platelets, vascular endothelium, and epithelium. Speculation was fueled by studies such as those which showed that group A persons were more susceptible to gastric cancer than group O persons, whereas group O individuals were more susceptible to duodenal ulcers than group A persons.

Some associations made sense, e.g., the worldwide distribution of ABO blood groups being influenced by pandemics. For example, the low frequency of A and high frequency of O blood group genes in various parts of the world could have resulted from a selective disadvantage of groups A during severe smallpox epidemics since the smallpox virus has A-like antigens. Group O individuals would have better survival rates since they have pre-existing anti-A that would destroy the virus. These and similar studies suggested that epidemics such as smallpox, plague, and cholera may have been responsible for the major differences in ABO blood group frequency observed around the globe.

But other associations were clearly without meaning, such as one showing higher intelligence in group A2 individuals. Students loved these kooky associations, which also made it possible to discuss statistical associations in general and what they actually meant.

A recent paper made me recall these earlier fun discussions:

A statistical association between ABO blood groups and risk of thrombosis has been recognized for many years. Specifically, group A, B, and AB people have demonstrated an increased incidence of thrombotic disease compared to group O individuals, presumably because ABO influences plasma levels of von Willebrand factor (vWF). vWF levels are 25% higher in non-O compared to group O individuals, but the mechanism by which ABO group determines plasma vWF levels has not been determined.

The review in Transfusion focuses on the carbohydrate structures of vWF and recent studies suggesting that differences in ABH antigen expression - which are expressed on the N-linked glycan chains of circulating plasma vWF - may have clinically significant effects on vWF proteolysis and clearance.

For more, see blood groups and disease associations (PubMed search)

New on TraQ

Monday, October 02, 2006

SmartBrief (new AABB resource)

AABB has raised its membership fees for 2007 and plans to introduce "new and modified communications vehicles."

One is AABB SmartBrief, a free daily news summary e-mailed to subscribers. There is a sample online and it appears to be available to both AABB members and non-members alike.

The sample issue includes transfusion medicine (TM) news but is fairly broad-based, e.g., there is an article on the continued search for a HIV vaccine by Don Francis, who was featured in Randy Shilt's 1988 book "And the Band Played On" and the movie of the same name.

Since the news summary occurs daily, it's easy to understand why SmartBrief needs to extend beyond TM - there's just not that much TM news happening daily.

Because the news items are not original, i.e., they link to online newspapers, they will be useful for a limited time only since web-based papers and news agency reports disappear regularly.


TraQ news is regularly reviewed for working links and links to news items are routinely culled or updated. News is available in these categories:

Monday, August 28, 2006

Agent 006001 is promiscuous? (anti-K is so common!)

The blog title is a take-off on James Bond, Agent 007.

As a long-time blood bank instructor, I was fascinated by this paper by French authors in the August 2006 issue of Transfusion:

Noizat-Pirenne F, Tournamille C, Bierling P, Roudot-Thoraval F, Le Pennec P-Y, Rouger P, Ansart-Pirenne H. Relative immunogenicity of Fya and K antigens in a Caucasian population, based on HLA class II restriction analysis. Transfusion 2006 Aug;46 (8):1328-33.

One of the many mysteries in transfusion medicine is why some red cell antigens are more immunogenic than others. Reasons for the relative immunogenicity was thought to relate to the nature of the antigen, including the number and affinity of potential derived peptides once the antigen is recognized as foreign and processed by antigen presenting cells. This papers adds one more piece to the puzzle.


The D antigen in the Rh system is known to be the most immunogenic. In the Kell blood group system the K antigen (ISBT #006001) is known to be considerably more immunogenic that antigens in the Duffy and Kidd blood group systems. The immunogenicity of an antigen is assessed by comparing the observed frequency of alloantibodies with the calculated frequency of the opportunity for alloimmunization. For example, the K antigen has been shown to be 9 times as immunogenic as Fya.

A brief review of how antibodies are produced:

We have long known how important MHC is to the immune response. Red cell antigens, like other protein antigens, elicit an immune response as follows:

  • They are processed and displayed in conjunction with HLA molecules on antigen-presenting cells (APC) such as macrophages,
  • and presented to T-cell receptors on T lymphocytes.
  • Recognition of RBC-derived peptide displayed by HLA class II molecules activates a specific CD4+ T cell (see diagram of an APC activating a CD4+ T cell)
  • The T cell divides and produces a clone of CD4+ T cells,
  • which in turn activates a specific B cell that produces a clone of antibody-producing cells.

HLA-DR molecules account for more than 90 percent of the HLA class II isotypes expressed on APCs, the HLA-DRB1* locus being highly polymorphic.

The authors found that for people who made anti-Fya (n=29) the DRB1*04 phenotypic frequency was 100%, indicating that the DRB1*04 molecule is the restriction molecule for Fya. For those who made anti-K (n=30) many DRB1* molecules were identified, demonstrating that the K antigen has a high degree of histocompatibility promiscuity.

This suggests that

  • only people with the DRB1*04 phenotype can produce anti-Fya
  • all or nearly all individuals in the Caucasian population are able to bind K-derived peptides and therefore could produce anti-K when stimulated.

Assuming that this finding is confirmed by future studies, it turns out that the immunogenicity of a given red cell antigen is related both to the nature of the antigen and to the distribution within the population of HLA-DRB1* molecules on antigen-presenting cells.

Why anti-K is so common may be mainly because the K antigen is promiscuous when it comes to which HLA-DRB1* molecules it binds with. Now is that cool or what?

New on TraQ

Monday, June 19, 2006

Finding a colleague's e-mail address

Finding a colleague's e-mail address is messy with much trial and error. Today many people want their address to be confidential due to the onslaught of spam. The best strategy is to find their phone number and then call to ask their e-mail address. <8-)

Chances for success increase if they
  • have participated in newsgroups or mailing lists
  • work for a major organization, e.g., university, health service, government, etc.
  • have authored published research papers

That said, here's a few tricks to try if you want to find "Jane Doe":

1. Search online phone books to get Jane's phone number You also need the city or town where she lives, otherwise the "hit" list of possibilities will be large.

Note that people can remove their listing from online "white page" directories.

Some examples of online directories:

Try searching for yourself (scary stuff!) or learn how to remove yourself.

2. If Jane works for an organization, find the organization on the web . If it's a university, you can almost always find the e-mail address (via a dept. listing or staff search utility). For example, search:

You can also contact Jane by phone by calling the department's main office.

3. Note how an organization's addresses are formatted by visiting its website. Once you find one address at an organization, you have the key to its address formatting. For example:

  • CBS addresses are formatted (
  • NBS address are ( )

4. Do a simple web search of the person's name ("jane doe") in Google - you can sometimes luck out.

5. If Jane has authored scientitfic papers, do an author search on PubMed in the format:

  • doe j[au]

Select several abstracts where Jane is the main author and her e-mail address will soon be discovered. Usually it takes only 1 to 5 abstracts to reveal an e-mail address. For fun, try a selection of your colleagues who have been published.

6. If Jane has been active on the Internet, try searching newsgroups at Google (Google Groups Beta - at top). For example, try "ed uthman"

7. Contact a person who may know Jane's address. For example, I get a many inquiries from colleages since I manage several mailing lists.

  • Precaution: Never give an e-mail address to someone you do not know and preferably let the person decide if they want the requester to have their address.

8. Send a general inquiry to e-mail address possibilities:

"I am looking for a colleague Jane Doe who works at Jane, If it's you, hi from Pat! Please drop me a line when you can. If it's not Jane, my apologies - please delete message."

Generic possibilities for hypothetical Canadian institution "":


If the first and last name together are long (e.g., Jane Smithsonian), try truncating at 8 letters (a common e-mail protocol):


Invalid addresses will bounce. The worst that can happen is a stranger gets your message and deletes it.

Hope this is useful! Here are some other search goodies.

Cheers, Pat

What's New on TraQ

Friday, May 12, 2006

Canadian Blood Services statistics (collections and staffing)

Below are some interesting statistics from a Canadian Blood Services' (CBS) presentation regarding provincial legislation that could impact CBS in an extreme emergency*.

What initially surprised me was that Ontario, albeit our most populous province, donates about 50% of the blood in Canada. The key statistics:

  • Annual whole blood collections: ~870,000 units
  • Annual collections in Ontario: 50% of national total
  • Permanent collection sites: 42
  • Blood donor clinics:14,000
  • Hospitals served: ~ 750

National Employees (Ontario in brackets)

  • Physicians: 62 (22)
  • Registered nurses: 597 (304)
  • Medical laboratory technologists:490 (158)

Canada's total population is almost 33 million and Ontario's is about 12.5 million (~37.9% of the total population). Quebec is missing from the CBS statistics, since that province's blood system is operated by Héma-Québec. Québec's population of 7.5 million represents about 24% of the Canadian population.

A few simple calculations: CBS serves a population of 25.5 million, of which Ontarians constitute 49%. Therefore, Ontarians donating 50% of the national total is about right.

Makes me wonder if all Canadian provinces and territories are pulling their weight, donation-wise, as well.

Another surprising statistic is that CBS employs about 20% more nurses than technologists. This could make sense if the cited figures include many more part-time nurses than part-time technologists. Otherwise, it seems odd, given the myriad of technologists who test, supervise component preparation, perform quality assurance, and liase with hospitals to distribute blood products, as well as the smaller number of CBS technologists who work in patient service laboratories performing pretransfusion and perinatal testing.

* Source (PDF): CBS presentation on Ontario's proposed Emergency Management Statute Law Amendment Act 2006 - Bill 56

What's New on TraQ

Sunday, March 26, 2006

AABB Board of Directors: non-Americans need not apply?

The Jan.-Feb. 2005 issue of AABB News featured global initiatives and international membership. Being a Canadian who is interested in international transfusion medicine and loves foreign travel, I made a note to write a blog sometime on an aspect of international membership in the AABB.

This blog focuses on the possibility of a non-American being elected to the AABB's Board of Directors. Sorry about the provocative blog title but it's meant only as an attention grabber.

Further motivation for writing on this topic came from several sources:
  • 2005 AABB elections. For the first time that I can recall (having been an AABB member for 31 years), a Canadian, Graham Sher, CEO of Canadian Blood Services, ran for elected office in the AABB.
  • In 2005, Canadian Nancy Heddle of the McMaster University Transfusion Research Program was appointed as an associate editor of Transfusion.
  • In 2008 the AABB Annual Meeting will be held in Montréal, Québec, Canada, the first time it has ever been held outside the USA.
  • And currently the AABB has its call for nominations for the 2006-2007 Board of Directors on its website.

According to the Jan.-Feb. 2005 AABB News, the key statistics on individual membership were:

  • USA: 6926 (83.7%)
  • Canada: 444 (5.4%)
  • Japan: 77 (0.93%)
  • Australia: 74
  • Germany: 66
  • UK: 60
  • Brazil: 50
  • All other countries (n=59): each less than 50
  • GRAND TOTAL: 8279

As shown, 2 countries (USA and Canada) account for almost 90% of the membership and the other 64 countries account for 10%.

Which brings me to the 2005 elections to the AABB's Board of Directors. As a Canadian I was hoping for a fellow countryman to be elected as an At-Large-Director but it did not happen.

Part of the difficulty is simply the numbers. For example, 2 At-Large directors are elected each year. Consider the scenario of 2 American candidates and 1 non-American running for the 2 positions.

Of course, no one votes strictly along nationalistic lines, but consider this scenario. If only10% (692) of American AABB members voted and their votes were split evenly among the 2 American candidates, each would receive 346 votes. If a whopping 50% of Canadian members voted, and they all voted for the Canuck, he or she would receive 222 votes and lose, assuming that votes from other nations were evenly split among the 3 candidates. Under this set of assumptions, if significantly more than 10% of Americans voted, any foreign candidate would be hard pressed to become a Board member.

Another factor in getting elected is name recognition. Studies have shown that name recognition is a major contributor to success in local, regional, and national elections. Whereas most, if not all, Canadian AABB members would recognize the name of the CEO of CBS, my guess is that, while many American leaders of the AABB would, the vast majority of rank-and-file American members would not recognize any non-American, short of someone of the stature of Karl Landsteiner.

A third element is the AABB system of USA-based district directors, some of whom run for national office after gaining experience and name exposure at the regional level. This feeder system also makes it more difficult for foreigners to win elected office.

Interestingly, the current AABB Board shows Graham Sher as 1 of 2 appointed directors, who are appointed for 1-year terms for "relevant expertise".

The AABB is undoubtedly a global leader in transfusion medicine, but this does not translate into large foreign memberships. For example, Canada's membership in the AABB is more than 5 times that of any foreign country. Consider the populations of these countries and number of AABB members:

  • Japan: 128 million (77 members)
  • UK: 60 million (60 members)
  • Canada: 32 million (444 members)

Canada's proximity to the USA appears to be one factor in explaining the relatively large number of Canadian blood bankers who join the AABB and subsequently get perks such as reduced registration fees for annual meetings. Travel to meetings in American cities is less onerous for Canadians than for overseas members. There are other reasons why so many Canadians as opposed to other international blood bankers join the AABB, and this may be the subject of a subsequent blog.

Being a member of an organization is one thing. True equality comes only when people participate in running it. So, could a non-American, particularly a Canadian, ever be elected to the AABB Board? Anything is possible but it seems unlikely without such a candidate having high recognition among American members. As to how that could happen - your guess as good as mine.

Comments on this or any other blog are most welcome (see comments link below). Readers are reminded that the views expressed in this blog are mine alone.

What's New on TraQ

Friday, February 03, 2006

Manslaughter case against UK lab worker collapses

Updated 9 Nov. 2013
As a follow-up to my earlier blog (Manslaughter charged following wrong blood fatality in UK):

The case against the biomedical scientist on trial in the UK for manslaughter has collapsed. The crown decided that, while negligence could have been proved for a civil case, the crime of manslaughter (gross negligence), was not provable in the case:
Negligence Primer

Disclaimer: I am not a lawyer. The following information has been compiled from an Internet search and is presented solely for the interest of readers. Readers are advised to consult the resources in Further Reading, which are provided as aids to identify more information on this complex topic.
By way of review, negligence is often defined as (1) not doing something which a reasonable person would do, or (2) doing something which a reasonable person would not do.

Thus negligence can involve acts of commission and acts of omission. In order to succeed in a negligence civil case, the plaintiff, or person suing, must generally satisfy the court of the following four elements:
  1. Duty of care. A person who practices in a health profession owes the patient a duty. The duty of care involves applying skill, knowledge, diligence and caution when caring for patients.

  2. Breach of standard of care. The standard of care is primarily determined by the general practice of the profession. The practitioner does not need to live up to the highest standards but rather the reasonable, accepted standards set for the profession. Standards of care are determined by consulting experts, and relevant practice guidelines and standards (such as blood safety standards). Patients also have the right to expect a reasonable standard of care from healthcare students who treat them.

  3. Injury or loss. For negligence to occur, the patient must have experienced injury or loss of some kind due to the negligent act.

  4. Causation - the breach must be the proximate cause of the harm (the causal link between the defendant's act and the injury or loss). The most common test is the "but for" test. That is, if the accident would not have occurred but for the defendant's negligence, then the conduct is the cause of the injury. There must be a clear direct connection between the negligent act and the harm caused to the patient.
In the case of the UK biomedical scientist , all 4 elements needed for a civil negligence case would seem to have been met. However, criminal negligence resulting in a charge of manslaughter require additional criteria.

Criminal Negligence
To meet the standard for criminal negligence, the act or omission must show a wanton or reckless disregard for the lives or safety of other persons. For example, a nurse in the USA was found guilty of criminal negligence (reckless manslaughter) for hiding an empty bag of blood after it was transfused to the wrong person. Transfusing the wrong blood by failing to perform required pretransfusion identity checks may show negligence but what made the nurse's act criminal - reckless manslaughter - was the failure to disclose the error upon its discovery.
For discussion of negligence in the context of student health professionals, see TraQ's Case A8:
Further Reading

  1. Berry DB. The physician's guide to medical malpractice. Proc (Bayl Univ Med Cent). 2001 Jan; 14(1): 109–12. (Free full text)

What's New on TraQ

Thursday, February 02, 2006

Manslaughter charged following wrong blood fatality in UK

Updated 9 Nov. 2013
There have been only a few court cases involving medical laboratory technologists/scientists but there is currently one happening in the UK:
Biomedical scientist on trial in the UK for gross negligence manslaughter

The case involves a group O patient who was mistyped as AB, received group A red cells, and died from multi-organ failure, presumably secondary to a severe hemolytic transfuion reaction caused by multiple errors. Not only was the patient mistyped, but she was apparently crossmatched with outdated gel cards. An in-house protocol to have a second person verify the results at the blood grouping stage was not followed. And a second person was also mistyped as group AB, fortunately not requiring transfusion.

The defendant has admitted his mistakes, claimed "mental aberration", and argued that there were contributing deficiencies at the hospital which had nothing to do with him, such as staff shortages, and that the transfusion had been unnecessary.

Anyone who has performed ABO blood typing knows that mistyping a group O person as group AB means that the grouping was read "bassackwards". In a way it's like reading the number 2006 as 6002. Regardless, it's a relatively common error made by newbee students but an incredible error for an experienced technologist to make.

Breaking in-house protocols designed to increase safety, such as requiring a second person to verify ABO blood grouping, is unacceptable but does happen, particularly under the stress of urgent calls for blood. In this case ignoring the protocol appears to be a systemic error, since the lab was cited for it during an earlier audit.

Using outdated reagents such as the gel cards used for antibody screening and crossmatching is another major error. The newspaper report of the court case suggests that the laboratory may have been trying to save money by using outdated cards that it had in excess quantity due to poor inventory management. Even if quality control of the outdated reagents was acceptable, using outdated reagents for something as critical as antibody detection and compatibility testing breaks regulatory blood safety standards.

My experience with using outdated reagents in the student lab shows that most reagents work well past their expiry date. And many transfusion services use outdated reagents under special circumstances, e.g., use rare reagent antisera for antigen phenotyping; use outdated rare red cells as positive controls for antigen phenotyping.

It would be interesting to investigate if outdated gel cards function acceptably past their shelf life and for how, even if they cannot be used for patient testing, since this may indicate another technical error by the accused in this case. Regardless of any suspending media and reagents, the crossmatch gel cards would have contained patient group O plasma and donor group A red cells.

In summary, from a regulatory perspective, this case involves multiple errors:

  • technical (human) error in misinterpreting the ABO blood group
  • failure to follow SOPs (no second person verified the ABO blood)
  • failure to use in-date reagents (gel cards)
  • possible technical (human) error in not detecting the ABO incompatible donor units in the crossmatch.
Some of the errors seem system-wide but others seem specific to the individual's actions. All of which serves to illustrate that you cannot entirely prevent human error. But much progress can be made and efforts continue. See, for example:
More details of manslaughter case:
What's New on TraQ

Wednesday, January 04, 2006

Cost of an obituary in Transfusion

Here's some food for thought: How much do you think it would cost to distribute a copy of a published obituary of a transfusion medicine pioneer in an e-mail or on a website? Curious? Then read on....

A transfusion medicine giant, JJ (Joghem) van Loghem of the Netherlands, died in 2005 and his obituary was featured in Transfusion Nov. 2005; 45(11):1823. Upon reading the obituary, I wrote aTraQ blog in Dr. van Loghem's memory (see below).

The same Nov. issue of Transfusion had an editorial on the movement towards open (free) access to published scientific literature and the competing reality that publishing quality journals is costly.The editorial focused on the NIH policy requesting recipients of NIH funding to deposit on PubMed Central (free access to all) the author's version of an accepted manuscript produced with NIH support within 12 months of publication and what that meant for authors submitting papers to both Transfusion and PubMed Central.

The authors end by noting:

AABB, Blackwell Publishing, and TRANSFUSION editors have been discussing open access, and the Journal may introduce options in the future to make authors' work available in an additional open archive. Although it is clearly attractive to gain wider exposure for articles via public access, we must carefully weigh the pros and cons of such exposure to ensure that any negative effects on the Journal are minimized. We urge TRANSFUSION authors and readers to remain aware of these evolving developments and to participate in the lively dialogue that is likely to continue in the coming years.

In writing the TraQ blog about Dr. van Loghem, I had hoped to include the Transfusion obituary, so investigated what it would cost. On the publisher's website (Blackwell Publishing) you can click on "Order permissions", which brings up the Copyright Clearance Center from which you can choose to distribute an article in several ways and get a "quick price". A few examples:

  • post the obituary on a website (personal site, educational site, etc.): $306 US
  • republish it in a newsletter (individual, educational institution): $109 US
  • send it in an e-mail to one person: $31 US
  • send it in an e-mail to 100 people: $3100 US

For comparison, the cost of my 2006 AABB membership, which includes a personal subscription to Transfusion, was $98 US or ~ $114 CDN. And with the paper copy of Transfusion I can leave it in the staff coffee room and library for all to enjoy and benefit.

What I really wanted to do was include the obituary (giving full credit to Transfusion) on a website for transfusion professionals (TraQ) , so that would cost $306 US, about $355 CDN. The cost for the obituary was the same as for any scientific paper.

Personally I think there is something wrong with paying $306 US to give a wider distribution to an obituary that pays tribute to a great transfusion medicine pioneer. It's not as though reproducing the obituary elsewhere would deprive the authors of their livlihood or impact the publisher's revenues. Such a high fee for something written as a public service is one of the many reasons that open access is so appealing.

More information on open access:

What's New on TraQ