Bacterial Contamination of Blood Components

1
Bacterial Contamination of Blood Components

• Harry L. Taylor, MD
• Medical Director
• Mid-Atlantic Regional Blood Services

2
NIH Prospective Study of Post-Transfusion
Hepatitis
3
Prevention
Infected Blood
Recipient
4
The Problem
5
Bavarian Red CrossBlood Transfusion Service

• 100 in 25,171 (0.40) blood components culture
  positive
• Whole blood 0.03
• Red blood cells 0.48
• Single-donor platelets 0.42
• Ilert WE, et al Transfusion Medicine 1995
  557-61

6
American Red Cross

• Culture Positive Units
• Red Blood Cells 0 0.2
• Platelets 0 10
• Febrile Reactions 1/10,00020,000
• Deaths 1/1,000,000 6,000,000

Wagner et al, Clin Microbiol Rev 19947290
7
Sepsis following platelet transfusion (n 41
cases)
Staph. Epidermidis 25.0
Other species 35.5
Staph. Aureus 5.8
Strep. viridans 3.8
Bacillus cereus 5.8
Slmonella cholerasuis 13,5
Serratia marcescens 9.6
Wagner et al, Clin Microbiol Rev 19947290
8
Sepsis following red cell transfusion (n 51
cases)
Other species 14.3
Treponema pallidum 4.1
Pseudomonas putida 4.1
Yersinia enterocolitica 51
Pseudomonas fluorescens 26.5
Wagner et al, Clin Microbiol Rev 19947290
9
Typical platelet contaminants
Typical red cell contaminants
Pychrophiles
Skin colonizers
10
(No Transcript)
11
Potential Sources of Contamination of Blood
Components

• Donor skin at blood donation
• Unapparent donor bacteremia
• Contamination during collection or processing
• Environment
• Equipment
• Disposable supplies

12
Characteristic Growth of Bacteria in Platelets
and Red Cells
13
Signs and Symptoms of Transfusion-Associated
Bacterial Sepsis

• Signs
• Fever
• Rigors (shaking chills)
• Tachycardia
• Change in systolic blood pressure

• Symptoms
• Nausea and vomiting
• Shortness of breath
• Lumber (lower back) pain

14
Transfusion Fatalities Reported to FDA (1976
Sep 1998, 22.75 yrs, 51 cases)Bacterial
Contamination of Platelets
S. Aureus 17.3
Serratia 15.4
S. Epidermidis 9.6
Klebsiella 17.3
Streptococcus 7.7
P. mirabilis 1.9
Salmonella 7.7
Bacillus 5.8
E. coli 5.8
Pseudomonas 5.8
Enterobacter 5.8
15

• Ollgard, Abjerg, Georgen Monitoring the
  bacterial growth in platelet concentrates-one
  years experience with the BacT/AlertTM System.
  Vox Sanguinis 199874(suppl 1) Abstract 1126
  Denmark 7 days
• Laan, Tros Improved safety and extended
  shelf-life of leuco-depleted platelet
  concentrates by automated bacterial screening.
  Transfusion 19995S Supplement Netherlands - 7
  days
• Vuetic The use of BacT/AlertkTM system for
  bacterial screening in platelet concentrates. Vox
  Sanguinis 2000 P371 Supplement 1 Yugoslavia - 7
  days
• McDonald, Roy, Low, Rogers, Green, Vegoda, Erley,
  Barbara The first experience in the United
  Kingdom of the bacteriological screening of
  platelets to increase shelf life to 7 days. Vox
  Sanguinis 2000 P375 Supplement 1 United Kingdom
  7 days

16

• Cooper, Leach, Tracy, Zaura, Schwartzman,
  AuBuchon Application of a universal culturing
  approach to reduce the risk of bacterial
  contamination of apheresis platelet units.
  Transfusion 199939119S Supplement USA
• Schelstraete, Bijnens, Wuyts Prevalence of
  bacteria in leukodepleted pooled platelet
  concentrates and apheresis platelets A 3 years
  experience. Vox Sanguinis 2000 P373 Supplement 1
  - Belgium

17
BaCon Study

• Assessment of the Frequency of Blood Component
  Bacterial Contamination Associated Transfusion
  Reactions

18
BaCon Study Purpose

• Determine bacterial contamination rates of blood
  components associated with recipient transfusion
  reactions
• Identify pathogens associated with contamination
• Identify risk factors for bacterial contamination
• Identify factors associated with transfusion
  recipient morbidity and mortality

19
Criteria for Adverse Transfusion Reaction

• Fever
• Temperature gt390C or gt 1020F
• Temperature gt 20C or gt 3.50F rise
• Rigors (shaking chills)
• Tachycardia
• Heart rate gt 120/min or gt 40min/rise
• Systolic blood pressure
• Rise gt 30 mm Hg
• Drop gt 30 mm Hg

Within 90 min Of Tx
20
Criteria for Adverse Transfusion Reaction

• Supportive symptoms may include

• Nausea and vomiting
• Shortness of Breath
• Lumbar (low back) pain

21
BaCon Data '98, '99, '00
22
Reported Rates BaCon
23
What did the BaCon Study do for us?

• Sharpened the recognition skills of the hospitals
  for bacterial transfusion events
• Sharpened the investigational skills of the Red
  Cross medical and technical staff
• Pointed out the differences in the willingness to
  cooperate among blood collection and transfusion
  organizations

24
Comparison of Residual Risks
25
Requirements for an Ideal Intervention

• Easy to perform.
• Minimum of specialized training.
• Clear results and easy interpretation.
• Sensitivity
• detect units with gt104 CFU/mL.
• High specificity
• False()- dead bugs, contaminates, or clinically
  insignificant concentrations.
• Rapid- delay the issuing.
• INEXPENSIVE!
• No specialized equipment.

26
Strategies for minimizing transfusion transmitted
bacteria

• Grouped into 4 categories
• 1. Bacterial avoidance
• 2. Growth inhibition
• 3. Bacterial detection
• 4. Bacterial elimination

27
Strategy 1. Bacterial Avoidance

• Extensive donor history
• 13-50 of donors had GI symptoms 30 d before
  donation.
• Not sensitive or specific.
• Improved skin preparation
• Alcohol - tincture of iodine.
• Technician dependant.
• Skin will never be sterile
• Removal of first aliquot of collection
• diversion of first 10-40 mL of donor blood.
• used for testing.
• Skin core may also be diverted.

28
Strategy 2. Growth Inhibition

• Limiting storage time
• gt3 wks for RBCs.
• gt3 days for plt units. (1985 - FDA ? 7d to 5d
  storage).
• Decreasing the storage time, esp. for platelets,
  would result in further supply difficulties.
• Optimizing storage temperature
• Plts stored at colder temps? Function?

29
Strategy 3. Bacterial Detection

• Visual inspection (RBC) - darker color when
  compared to sterile units.
• Hemolysis, ?pO2
• GNB like Enterobacter agglomerans and Serratia
  sp.
• Sensitivity of inspection - 108 CFU/mL.
• Easy, rapid, low cost, and noninvasive.

30
Strategy 3. Continued

• Visual Inspection (Plt)
• quality control measure.
• Clumps, color change
• Plts and the swirling/streaming effect.
• Non-discoid plts do not swirl.
• Swirling decreases at bacterial levels of 107-8
  CFU/mL.
• Low pH, prolonged storage at 22C, low temps. -
  all affect swirling and plt viability.

31
Strategy 3. Continued...

• Direct staining
• Both RBCs and plts.
• Gram stain vs. Acridine orange
• Acridine orange(104-5 CFU/mL) is more sensitive
  than Gram stain. Problem?
• Would this increase technical error and delay
  issuing? Ex. A () gram stain reported on the
  wrong product.
• Some institutions have a program of
  pretransfusion Gram stains of platelet pools and
  apheresis units.

32
Direct Staining - Continued

• Pretransfusion gram staining-1-5 day old
• 80 sens./99.96 specificity.
• 4- and 5-day old
• 100 sens./ 99.93 specificity.
• Combine pretransfusion gram stain with culture of
  4-5 day old plt units.

33
Strategy 3. Continued...

• Culture of products
• Automated culture systems
• 10 CFU/mL
• Timing of the sampling- pre/post processing.
• Phlebotomist/technique dependant.
• Culture requires considerable time.
• False positive cultures
• addl time to reculture for confirmation
• sampling times of gt24 hrs for plt units,
  effectively decreasing time available for
  transfusion.
• However, this is routinely done in some hospitals
  in Europe.

34
Strategy 3. Continued...

• Metabolic Measurements
• Glucose and pH levels become abnormal at gt107-8
  CFU/mL.
• Use of multi-reagent indicator strips.
• Use with platelets.
• Glucose-? glucose ?bacteria.
• Acid production- ? pH.
• Compared to automated glucose analyzers and pH
  meters.
• Indicator strips are easy, cheap, done at the
  time of issue with rapid results.

35
Problems(challenges) with Metabolic Measurements

• Range of glucose levels in fresh, outdated plt
  units.
• Affect glucose levels
• Anticoagulants.
• Containers.
• Method of processing.
• Difficult to get high sensitivity.
• Production of neutral metabolites
• Klebsiella sp.,Bacillus subtilus, and Proteus
  mirabilis.
• No decreased pH.

36
Strategy 3. Continued...

• Endotoxin Assays Enzyme Linked Assays
• Lower limit of bacterial detection was 101-5
  CFU/mL.
• Limited to detection of endotoxin-producing bugs.
• ELISA- immunochromatographic detection of
  peptidoglycan.
• Simple to perform, inexpensive,specific, and
  sensitive up to 104 organisms.
• Has not been validated in clinical trials.

37
Strategy 3. Continued...

• DNA/RNA Techniques
• PCR Y.enterocolitica
• virF gene on pYV plasmid the ail portion of the
  Y.e. genome.
• Detects 500 org/100?L blood.
• 6 hrs to perform.
• One organism detected.
• Problems
• false positives
• high contamination rate
• not suited to blood bank use.

38
DNA/RNA Techniques Continued

• Chemoluminescence - linked universal bacterial
  rRNA probe
• Acridinium ester-labeled ssDNA probe bacterial
  rRNA regions.
• Superior when compared with Gram and acridine
  orange stains.
• Gen-Probe, developing the assay, is not pursuing
  further development.
• Labor intensive and very time consuming.

39
DNA/RNA Techniques Continued

• Microvolume fluorimetry
• Fluorescent labeled antibiotics as probes for
  bacteria(prokaryotes).
• Sample remains stationary within the volumetric
  capillary.
• Helium neon laser scans capillary length.
• Splits the signal and creates a two dimensional
  image.
• Vancomycin probe detected S.epidermidis at levels
  of 5 10² CFU/mL.
• Gentamicin probes for Serratia marcescens(unpublis
  hed).

40
Strategy 4. Bacterial Elimination

• Leukocyte filtration - reduces the level of
  bacterial contamination.
• Reduces accumulation of cytokines, breakdown
  products and possibly immunogenic fragments.
• Effective with Y.enterocolitica from RBCs.
• Haemovigilance Network, France - universal
  leukoreduction (ULR) in 1998, 66 reduction in
  rate of septic transfusions due to RBCs.
• RBC assoc.reaction rate before ULR 3.8 vs. 1.7
  after (plt.001).
• Retrospective analysis, no prospective studies
  yet.
• Not been shown to be effective in platelets.

41
Strategy 4. Continued...

• Antibiotics-Adding-? Not an acceptable solution.
• Trading drug reaction for bacterial sepsis.
• Development of drug resistant strains of
  bacteria, patient sensitivities, anaphylaxis.

42
Strategy 4. Continued...

• Pathogen Inactivation - use of photochemical
  methods to inactivate viruses, bacteria, and
  protozoa.
• Psoralen/UVA combo, riboflavin/visible light, UVB
  irradiation.
• Preliminary studies had apparent conflicting
  results.
• Spore formers may be resistant to inactivation by
  some methods.

43
Risks of Blood Transfusion
Bacteria Introduced during collection
Emerging/Unknown Viruses
Leukocytes Adverse immune responses and
transfusion reactions
Know Pathogens For which no assay is available
Window Period Limits of detection of current
assays
44
Pathogen Inactivation
Plasma and Platelets
Red Cells
Addition of inactivator
Addition of inactivator
Removal
Removal
UV Activation
pH Activation
45
Effect of Inactine on RBC Recovery In Vivo
28 Day Storage
85.0 5.0
85.9 2.5
82.3 5.4
82.2 3.5
24-Hour Recovery (mean sd)
P0.55
P0.95
Double Label
Single Label
46
Effect of Inactine on RBC Recovery In Vivo
42 Day Storage
82.9
86.3
24-Hour Recovery
Pgt0.05
47
Effect of Inactine on RBC Survival In Vivo
42 Day Storage
48
Effect of S303 on RBC Recovery and Survival
42 Day Storage
n 29 paired 11 full unit reinfusions 3
reduction of clinical efficacy
49
Practical Implications

• Medical Issues
• Decreased Recovery/Survival
• ? Increased Transfusions
• Increased Exposures
• Iron Overload
• Supply adequacy

50
Practical Implications

• Economic Issues
• ????????????
• Logistic Issues
• Equipment
• Space
• Staff
• Turnaround
• Throughput

51
Practical Implications

• Safety Implications
• Toxicity
• Mutogenesis
• Carcinogenesis
• Teratogenesis

52
Comparison of Risks
1100 11000 110,000 1100,000 11,000,000
Transmission Risk per unit
HBV
HIV
Mistransfusion Fatalities (RBCs)
HCV
84 86 88 90 92 94 96
98 00 02
53
Conclusion . . .

• Bacterial contamination is a major cause of
  transfusion-related morbidity and mortality.
• Septic reactions are more common with platelets
  units than RBCs.
• Gram negative reactions tend to be more severe
  and rapid in onset due to endotoxins.
• Organisms present at the time of donation in
  undetectable levels can reach clinically
  significant levels by time of transfusion.
• First strategies will likely be 1) diversion of
  the initial collection, and 2) detection by
  culture

54
Conclusion . . .

• Culturing plts on day 2-3 of storage extends
  shelf-life to 7 days in use in Europe.
• Visual examination, pretransfusion Gram staining,
  and glucose/pH strips are being used now.
• No single strategy is likely to solve the problem
  of bacterial contamination of products.
• It is time to implement partial solutions of
  various kinds and experiment.

55
AABB Proposed Standard

• Standards, 22 Edition ...facilities have
  method(s) to detect bacterial contamination in
  all platelet components.
• Culture method for apheresis platelets
• Performed by collection facility
• Collection hold 24 hours culture hold 24
  hours release neg. units

November, 2003
56
AABB Proposed Standard

• Standards, 22 Edition ...facilities have
  method(s) to limit and detect bacterial
  contamination in all platelet components.
• Implementation by March 1, 2004.
• Methods
• Enhance arm prep
• Use of diversion collection systems
• Culture for apheresis platelets
• Stains or dipstick for random pooled platelets

March 3, 2003
57
Last, but not least,

• Transfusion-related fatalities are reported to
  the FDA, Division of Inspections Surveillance
  within 24 hours with a written report due in 7
  days followed by a case summary and autopsy
  report.
• In the case of suspected contamination, also
  notify your blood provider. The other implicated
  components need to be secured and the cause
  investigated.