Friday, January 31, 2020

Stand by me (Musings on transfusion medicine errors)

Updated: 1 Feb. 2020 (Learning pt. 5, Further Reading)
To begin the new decade, I'll muse on a TraQ case study from decades ago that discusses a topic not often discussed by educators in class or included in online cases* but is still relevant and merits airing. 

*Online transfusion cases often involve antibody identification or making a diagnosis using lab and other diagnostic tests, in other words, cases with clear evidence-based results: yes and no answers that require factual knowledge and experience but not much reflective thinking. You may disagree, that's okay too. 

This blog's case study involves a student whose error resulted in a patient dying prematurely but, given another patient, could easily have been THE cause of death. Frankly, I shudder to think what the headlines may have been if local media had gotten hold of the story and were anything like the UK's tabloid press. 

The blog's title derives from a 1961 ditty by Ben. E. King.

CASE A8: Severe Hemolytic Transfusion Reaction Involving a Student
This case involved me directly as I was the clinical instructor of the student and transfusion service (TS) involved. As the clinical instructor I taught the student in a 2-week student lab (involving wet laboratories and small group tutorials) followed by a 4-week internship in the clinical transfusion lab where students were supervised by medical laboratory technologists. All the details of the case are in Further Reading. 

The student was supervised by an experienced, respected technologist  who was also responsible for working on the bench and processing routine work. The TS laboratory performs antibody screens using gel technology followed by an electronic crossmatch for patients who lack clinically significant antibodies.

The lab's policy is to initially supervise students very closely and then, depending on performance (which is documented daily),  to gradually allow students to perform with less supervision. Regardless of the intensity of supervision, the policy requires that supervising technologists always read the results of student tests.

The student did gel antibody screens on a batch of 5 patients in the late afternoon.

One of the 5 patients had a positive (3+) antibody screen ("Patient A") but, because blood was not ordered and unlikely to be required for the type of surgery, the patient specimen was set aside for antibody identification the next day.

One of the 4 patients with a negative antibody screen ("Patient B" - an 70-year old male scheduled for liver resection for metastatic colorectal cancer) had an electronic crossmatch performed for 3 units of RBC.

Early the next morning Patient "B" was transfused with 2 units of RBC.
When the morning shift began in the laboratory, an antibody identification panel on Patient A was unexpectedly negative, as was the repeat antibody screen. Antibody screens were then repeated for all patients in the batch and one (Patient B) was positive. Patient B was later found to have an anti-Fya and anti-c.

Immediately upon discovering the mix-up and positive antibody screen on patient B, the ward was contacted to stop any transfusion in progress, to inquire about Patient B's condition, and to monitor him for signs of a possible hemolytic transfusion reaction (HTR). Subsequently, Patient B was found to be experiencing a severe HTR and the TS medical director was consulted to help manage it. 

The student and supervising technologist were questioned in private, both individually and together, by the medical director and laboratory supervisor. The student could not recall making an error and thought that patient plasma had been pipetted correctly for each patient. The supervising technologist recalled not seeing the student actually pipette patient plasma into the gel cards but did read the results of all indirect antiglobulin tests (IATS) in the batch.

By chance, staff were able to retrieve gel cards from the biohazard garbage and identify that Patient B had less volume than usual in the antibody screen tests, consistent with no plasma having been added. The conclusion was that there was a pipetting error in which Patient B's plasma was not added to Patient B's gel antibody screen cards but rather to Patient A's tests.

Patient B died the day following transfusion.

The hospital's Risk Management Office was contacted and the chief counsel discussed the case with the TS medical director and Patient B's physician.

A series of group meetings were held by the lawyer from Risk Management with the TS laboratory supervisor, student, supervising technologist, and TS clinical instructor (me) all being present.

The Chief Medical Examiner was notified and investigated the death.

The TS medical director openly disclosed and explained what had happened to the patient's family. We waited for 2 years suing limitation but the family never decided to sue. 

The supervising medical technologist left the lab and began a career in a related field.

My personal learning points from the case:
1. When medical technologists in a busy lab supervise students, they can't watch every student move. Student errors may happen and supervising technologists are devastated, as in this case. Suspect this may be true for other health professionals supervising students who make serious errors too. 

2. The group sessions with the hospital's Risk Management lawyer were enlightening to me. She stressed that anything said in our meetings couldn't be used in court because of attorney-client privilege.

3. Our TS medical director was exemplary in her transparency to disclose the error to the patient's family. Perhaps ahead of her time?

4. I suspect many lab errors are latent errors (aka systems errors as well documented by UK's SHOT) caused by flaws in the system, i.e., errors that lead to human (operator) errors and whose effects typically lie dormant in the system. (Further Reading) 

5.The student in Case 8 thought she'd added drops to the antibody screens correctly. Many times in the student lab when students get incorrect results they'd almost always say, 'But I did it right. I know I did.' Then I'd gently remind them that it was, of course, normal to think that.If they knew they were adding the wrong drops, following the wrong procedure, they'd not have done it. 

Being human we make all sometimes make mistakes.What's key is to learn from them and realize the clinical laboratory has a culture of no blame that enhances patient safety. The worst thing would be for them to be spooked and frightened of making mistakes in the clinical lab (self-fulfilling prophesy) where patient safety is at risk. 

6. Reminds me of the reality that transfusion med lab technologists / scientists are often the last barrier between a patients and possible death. The transfusion service is different from many other med lab departments (not all) where unusual results are often repeated before reporting. 

I knew this as early in my career I'd crossmatched group AB blood for a patient who died because she was group O and had a wrong group AB sample sent to the off-site transfusion service where I worked. 

As always comments are most welcome. 
Chose this song because I think all health professionals need to stand by colleagues and students when human or systematic errors occur because there, but for the grace of gawd, go I.