When to Culture a Product: AABB vs BEST Guidelines
- caitlinraymondmdphd
- 7 hours ago
- 4 min read
How the BEST Criteria Updated a Decade-Old AABB Approach to Septic Transfusion Reactions
One of the most uncomfortable questions in transfusion medicine is deceptively simple:
When should we culture the patient and the blood product after a transfusion reaction?
Culture too often, and you trigger false positives, unnecessary lookbacks, and wasted resources.Culture too conservatively, and you risk missing a true septic transfusion reaction — one of the most dangerous complications we manage.
For years, many institutions have relied on guidance from an AABB Association Bulletin published in 2014. But in 2019, a large multicenter study fundamentally challenged whether those criteria are sensitive enough for real-world practice.
This post walks through what changed, why it matters, and what the tradeoff actually is.
The AABB 2014 Bulletin: Safety Through Clinical Vigilance
The 2014 AABB Association Bulletin on suspected bacterial contamination of platelets was written with a clear goal:don’t miss sepsis.
Its framework is intentionally broad and clinically driven. In short, it recommends investigation when:
A patient develops fever ≥38°C with a ≥1°C rise, plus at least one associated symptom (rigors, hypotension, tachycardia, dyspnea, etc.), or
There is any clinical change that raises concern for sepsis — even without fever
Importantly, the bulletin acknowledges:
Fever may be absent in neutropenic or immunosuppressed patients
Antipyretics may blunt temperature rise
Symptoms may be delayed
This guidance reflects its era. In 2014, the dominant concern was under-recognition of septic transfusion reactions, especially with gram-positive organisms. The solution was education, vigilance, and a low threshold to act.
What the bulletin did not do was define:
Objective thresholds for hypotension or tachycardia
How to systematically account for antipyretic use
How well these criteria actually perform in practice
That gap mattered more than we realized.
The Problem: How Well Do the AABB Criteria Actually Work?
In 2019, investigators from the BEST (Biomedical Excellence for Safer Transfusion) Collaborative asked a hard question:
If we apply the AABB criteria to real-world transfusion reactions, how many culture-positive cases do we actually detect?
Using data from nearly 800,000 transfusions across 20 centers, they found that the answer was… not many.
When evaluated empirically:
The AABB criteria detected only ~40% of culture-positive reactions
The majority of reactions that ultimately yielded positive cultures never met AABB triggers
Reliance on fever and subjective symptom reporting was a major limitation
In other words, the system was doing exactly what it was designed to do — but that design was missing cases.
The BEST Criteria: Trading Specificity for Sensitivity (On Purpose)
Rather than discarding the AABB framework, the BEST investigators asked:What small, evidence-based changes would catch more cases?
They tested three modifications, all of which improved detection:
1. Isolated High Fever Counts
A temperature ≥39°C with a ≥1°C rise triggered culture even without other symptoms.
Why? Because multiple international criteria already recommended this — and AABB did not.
2. Objective Vital Sign Definitions
Instead of relying on checkbox reporting:
Hypotension required both an absolute BP threshold and a percentage drop
Tachycardia required ≥100 bpm and a significant increase from baseline
This mattered because provider-reported vital sign abnormalities were frequently inaccurate.
3. Antipyretics Matter
If a patient received antipyretics before transfusion, absence of fever could not be used to rule out sepsis when other concerning signs were present.
This was not a philosophical change — it reflected basic physiology.
Did It Work?
Yes — and predictably.
When all three modifications were combined into the BEST criteria:
Sensitivity improved to ~70–75%
Specificity decreased to ~45%
Crucially: there were no cases detected by AABB that BEST missed
In other words, BEST caught substantially more potential septic reactions — at the cost of more cultures and more false positives.
This was not an accident. It was a conscious tradeoff.
The Real Debate: False Positives vs Missed Sepsis
Critics of broader culturing thresholds often raise legitimate concerns:
Positive product cultures trigger supplier notification
Co-components may be quarantined or destroyed
Many positive cultures do not correlate with patient infection
All of that is true.
But the BEST authors make a different argument:
In a passive surveillance system, missing cases is the greater danger.
Septic transfusion reactions are rare, difficult to adjudicate, and often masked by critical illness. Fever is unreliable. Cultures are imperfect. But hypotension requiring pressors, shock, or unexplained deterioration are not benign signals, even when temperature is normal.
The BEST criteria reflect a shift from:
“Culture when sepsis is obvious” to “Culture when sepsis is plausible and high-risk.”
Where This Leaves Us
The BEST criteria don’t replace clinical judgment — they formalize what experienced clinicians already know:
Fever is not required for sepsis
Antipyretics obscure key signals
Objective thresholds matter
Sensitivity matters more than comfort when stakes are high
Institutions now face a choice:
Accept fewer cultures and higher miss rates, or
Accept more cultures to reduce the risk of missing a true septic transfusion reaction
That choice is about risk tolerance, not right vs wrong.
But it should be made using current evidence, not decade-old assumptions.
Bottom line
If you are still relying solely on fever-centric AABB criteria from 2014, you are almost certainly missing cases. The BEST criteria offer a data-driven update that reflects how septic transfusion reactions actually present — messy, masked, and dangerous.
In transfusion medicine, that tradeoff is worth naming out loud.
