|Year : 2019 | Volume
| Issue : 1 | Page : 33-38
Surrogate markers and their correlation to bacterial contamination and other quality parameters in random-donor platelets by platelet-rich plasma method
Priyatesh Dwivedi, Abhishekh Basavarajegowda, Apurba Sankar Sastry
Departments of Transfusion Medicine and Microbiology, Jawaharlal Institute of Postgraduate Medical Education and Research, Puducherry, India
|Date of Web Publication||22-Apr-2019|
Dr. Abhishekh Basavarajegowda
Departments of Transfusion Medicine and Microbiology, Jawaharlal Institute of Postgraduate Medical Education and Research, Puducherry
Source of Support: None, Conflict of Interest: None
Introduction: Bacterial contamination in platelet concentrates (PCs) occurs more frequently than other blood components because of several factors such as storage in oxygen permeable blood bags at 20°C–24°C with continuous agitation which facilitates bacterial growth compared to other blood components which are kept frozen or refrigerated which inhibits bacterial proliferation in them. The purpose of the study was to assess the incidence of bacterial contamination of random-donor PCs and factors associated with its contamination and see how well the surrogate markers such as pH and swirling correlate with the same. Methodology: This was a cross-sectional study which included randomly chosen 500 random-donor platelets (RDPs) in blood bank prepared by platelet-rich plasma method. The samples chosen for the study were from the RDPs on the 5th day of storage after their preparation. pH, platelet count, and swirling in platelets, which act as surrogate markers for bacterial contamination, were checked on the RDP units. About 1–3 ml of PCs was inoculated from the RDP units into labeled culture bottles (BD Bactec Peds Plus/F). Results: Among a total of 499 random-donor PCs that were cultured in the automated BACTEC system for the study, none of them were culture positive. Thirty RDP units in the study were visibly lipaemic whereas 93 RDP units were visibly reddish in appearance. PCs having volumes <40 ml or >70 ml did not affect the swirling, pH, and platelet counts. There was a statistically significant difference between mean pH with RDP units having swirling Grade 2 and 3 and platelet counts with RDP units having swirling Grade 1 and 2. Conclusion: Bacterial contamination though poses a significant risk is a very rare event in a meticulously prepared and stored PCs. Surrogate markers though useful in resource-constrained settings does not correlate optimally with the quality indicators.
Keywords: Bacterial contamination, random-donor platelets, swirling
|How to cite this article:|
Dwivedi P, Basavarajegowda A, Sastry AS. Surrogate markers and their correlation to bacterial contamination and other quality parameters in random-donor platelets by platelet-rich plasma method. Glob J Transfus Med 2019;4:33-8
|How to cite this URL:|
Dwivedi P, Basavarajegowda A, Sastry AS. Surrogate markers and their correlation to bacterial contamination and other quality parameters in random-donor platelets by platelet-rich plasma method. Glob J Transfus Med [serial online] 2019 [cited 2019 Sep 16];4:33-8. Available from: http://www.gjtmonline.com/text.asp?2019/4/1/33/256763
| Introduction|| |
Bacterial contamination of platelet concentrates (PCs) is a long-standing problem and is considered the most common cause of transfusion-related sepsis. The incidence of viral infections after blood transfusion have declined over the past two decades because of stringent donor selection criteria and effective screening test methods, but the risk of bacterial infections after transfusion of blood components has not decreased due to unavailability of standard bacterial screening tests and methods and hence remains a challenge in blood bank and transfusion service. Bacterial contamination in PCs occurs more frequently than other blood components because of several factors such as storage in oxygen permeable blood bags at 20°C–24°C with continuous agitation which facilitates bacterial growth compared to other blood components which are kept frozen or refrigerated which inhibits bacterial proliferation in them.
Analysis of data from the hemovigilance system from the UK, France, and the USA indicates that risk of mortality due to sepsis after transfusion of bacterially contaminated PCs is 1 in every 7500–100,000 platelet units. The rate of transfusion-related bacteremia was found to be 9.98 for single-donor platelets and 10.64 for pooled PCs per millions of units transfused. However, the actual risk of sepsis related to transfusions has not been accurately estimated because they are largely unrecognized and/or not reported.
The mechanisms implicated in bacterial contamination of PCs may include asymptomatic bacteremia in donors, inadequate disinfection of the phlebotomy site on donor arm, contamination during blood collection, blood collection in contaminated bags, and contamination during the component separation procedures.
Bacterial detection methods vary in complexity and sensitivity. Platelets are now generally tested with culture-based systems such as BACTEC, BacT/ALERT, and Pall eBDS. Molecular-based bacterial detection has high sensitivity and specificity and makes them appealing for detection of bacteria in platelets but is not readily available and costly.
The purpose of the study was to assess the incidence of bacterial contamination of random-donor PCs and factors associated with its contamination and see how well the surrogate markers such as pH and swirling correlate with the same
| Methodology|| |
This was a cross-sectional study which included randomly chosen 500 random-donor platelets (RDPs) in blood bank prepared by platelet-rich plasma (PRP) method after collecting whole blood from healthy donors who fulfill the criteria to make donation in 350 ml bags with citrate-phosphate-dextrose-adenine as anticoagulant and tested negative for transfusion-transmitted infections, namely HIV, HBV, HCV, syphilis, and malaria. These were stored in platelet agitators at 22°C–24°C. The samples chosen for the study were from the RDPs on the 5th day of storage after their preparation. pH and swirling in platelets, which act as surrogate markers for bacterial contamination, were checked on the RDP units. The grade of swirling was checked by visual inspection and pH of PCs using digital pH meter. Swirling was scored as 0 for homogenous turbidity in all parts of bag with no change with pressure, 1 for swirling only in some part of the bag and is not clear, 2 for clear homogenous swirling in most parts of the bag, and 3 for very clear homogenous swirling in all parts of the bag. pH and swirling were tested for their predictive value and correlation with the culture reports. About 1–3 ml of PCs was inoculated from the RDP units into labeled culture bottles (BD Bactec Peds Plus/F, Becton Dickinson and Company, MD21152) under aseptic conditions on the 5th day of their storage. Then, the culture bottles were taken to the bacteriology department for incubation into the automated BACTEC culture system for 7 days. The results obtained were categorized into confirmed positives when RDPs with positive culture along with identification of the bacteria on subculture or negatives when RDPs did not exhibit a positive signal over 7 day's incubation in BACTEC culture system. Data regarding RDP unit and culture bottle numbers, pH, swirling grade, the volume of PCs, platelet count in RDP units, confirmed positives, and identification of bacteria were recorded in a predesigned pro forma entered into Microsoft Excel and analyzed using IBM SPSS Statistics for Windows, Version 21.0., 2013 (IBM Corp., Armonk, NY, USA)
| Results|| |
A total of 499 random-donor PCs were cultured in the automated BACTEC system for the study in the period between January 2017 and June 2018. Among these, none of them were culture positive. On visual inspection 30 RDP units in the study were visibly lipemic whereas 93 RDP units were visibly reddish in appearance. Among the visibly reddish RDP units, only 11 units (2.2%) had RBC contamination of more than 0.5 ml. The lipemia and RBC contamination in the PCs did not affect the pH and platelet counts in the studied RDP units.
[Table 1] shows the percentage of RDP units with adequate volumes, pH and the ones beyond the acceptable range, and distribution of the units with regard to grades of swirling and in the study. PCs having volumes <40 ml or >70 ml did not affect the swirling, pH, and platelet counts. All of the RDP units having volumes <40 ml or >70 ml had swirling, and pH and platelet counts met the standard quality control criteria.
|Table 1: Quality characteristics of the random-donor platelets units studied (n=499)|
Click here to view
There was a statistically significant difference between mean pH with RDP units having swirling Grade 2 and 3. There was also a statistically significant difference between platelet counts with RDP units having swirling Grade 1 and 2. No significant difference was found between volumes of RDP units with any swirling grade as shown in [Table 2].
|Table 2: Post hoc analysis between swirling and other variables (multiple comparisons)|
Click here to view
Pearson product-moment correlation coefficient was computed to assess the relationship of pH with platelet count and volume of RDP units in [Table 3]. There was a statistically significant weak correlation only between pH and platelet count with r = −0.12426 (P = 0.005).
|Table 3: Pearson product-moment correlation coefficient to assess the relationship between the quality variables of platelets|
Click here to view
| Discussion|| |
This cross-sectional study was conducted with an aim to assess the incidence of bacterial contamination of random-donor PCs, to identify the factors associated with bacterial contamination, and to assess the effectiveness of surrogate markers such as pH and swirling in detecting bacterial contamination in our center.
Bacterial contamination of PCs is a long-standing problem in transfusion practice and is considered to be the most common cause of transfusion-associated morbidity and mortality globally. There were no culture-positive RDP units in the study, and all the quality control parameters for random-donor PCs were met in >84.5% of the RDP units. Thus, this study reflects that strict adherence to the standard operating procedures for component separation and preparation had been followed in the blood bank of this institute. Comparison of the results of our study with other studies has been depicted in [Table 4].
|Table 4: Comparison of Results of culture of platelet concentrates in various studies|
Click here to view
The results of our study are similar to the study done by Raturi et al. In their study, none of the units showed any bacterial growth and swirling was present in all the units studied. In studied units, volume (90%), platelet yield (94.35), and RBC contamination (87%) showed compliance to the Indian standards whereas all the PCs had pH well within the recommended norms.
A study done by Das et al. with a sample size similar to our study had slightly different results from ours. A total of 410 whole blood-derived PCs were included in that study and a total of 2 (0.48%) PCs were bacterially contaminated among which coagulase-negative staphylococci (CONS) and diphtheroids both of which are skin commensals were isolated. Nine (2.19%) were considered probable contamination among which 3 (0.73%) were due to instrument error and 6 (1.46%) were inoculation contamination as repeat culture did not yield any growth. CONS was isolated on day 3 of collection and diphtheroids on day 1, but there were no isolations from PCs sampled on days 5 and 7. The study by Das et al. revealed a high rate of positive culture findings within 24–48 h of incubation in the BacT/Alert-automated culture system. Similarly, other different studies have also shown positive culture results. This is because growth characteristic might differ between clinical isolates of bacteria and laboratory bacterial strains that had been used in spiking the PCs in all those studies. Hence, performance decisions on the use of bacterial detection systems for PCs should be based on routine screening of a large number of PCs rather than artificial situations produced with laboratory bacterial strains for inoculation.
The Food and Drug Administration has issued specific guidelines under which the bacterial detection culture systems could be validated for released testing – in postmarketing surveillance studies, a large sample size of >50,000 units needs to be studied to achieve a high degree of statistical confidence (>95% confidence level).
In our study, BD BACTEC Peds Plus/F culture vials (used for aerobic blood cultures) were used for inoculation of samples from PCs. The volume of inoculation was <3 ml for these culture vials. Quality control certificates were present with each carton of media. The range of time-to-detection in hours was <72 h for each of the organisms listed on the quality control certificate for the culture vials: Streptococcus pyogenes, Escherichia More Details coli, Streptococcus pneumoniae, Pseudomonas aeruginosa, Candida albicans, Neisseria More Details meningitidis, Alcaligenes faecalis, Haemophilus influenzae, and Staphylococcus aureus. There was no information about the three most common bacteria of the skin normal flora and the Gram-positive bacteria such as Staphylococcus epidermidis, Bacillus cereus, and Propionibacterium acnes on the quality control certificate for the culture vials. Furthermore, standard internal quality control for the culture vials was not performed for any cartons before use.
Riedel et al. in their comparison study between BACTEC 9240 Plus Aerobic/F and BacT/Alert blood culture system for the detection of bacterial contamination of PCs inoculated 113 PCs with low levels of different bacteria. Growth was detected first in the BACTEC system in 93 instances, and only in 12 instances, growth was detected first in the BacT/Alert system. Growth was detected, on average, 1.7 h sooner with BACTEC system among all comparisons, and this difference in length of time to detection was statistically significant.
Similarly, Dunne et al. in their in-house validation of the automated BACTEC 9240 blood culture system for the detection of bacterial contamination in PCs found that BACTEC system gave a detection sensitivity of <10 CFUs/ml for S. aureus, S. epidermidis, E. coli, Serratia marcescens, Klebsiella pneumoniae, B. cereus, Enterobacter cloacae, and Pseudomonas aeruginosa, except Streptococcus mitis for which the limit of detection was 61 CFUs/ml. Detection time ranged from 6.5 to 17.6 h and yielded 2 true-positive samples from 3879 apheresis PCs over a period of 9 months.
In our study, 69 RDP (13.8%) units had volume <40 ml, 422 units (84.5%) had volume 40–70 ml, and only 12 units (2.4%) had volume >70 ml. PCs having volumes <40 ml or >70 ml did not affect the swirling, pH, and platelet counts. All of the units having volumes <40 ml or >70 ml had swirling, and there was no correlation between the volume of RDP units with swirling, pH, and platelet counts.
As quality control criteria, all of the RDP units prepared from PRP method should have a volume of 40–70 ml, but 81 RDP units in this study had volumes outside the recommended range in which 12 had >70 ml. The causes for variation in volumes of the RDP units can be multifactorial including manual expression of plasma with plasma expressor in PRP method of PC preparation leading to under/overexpression of plasma leading to less and excess volumes in RDP units inconsistent training to technicians about component separation and preparation. This suggests that further standardization is required for preparation of RDP units by PRP method in the blood bank of our institute.
Sufficient plasma volume in a platelet unit is required to maintain pH in PCs. Insufficient plasma volume results in lower pH at the end of the allowable storage period. Although sufficient volume is required to buffer pH, large volumes of Group O plasma in the platelet product may cause significant hemolysis of recipient red cells particularly when the donor has high titer anti-A and anti-B isoagglutinins. Most published reports of hemolysis following platelet transfusion have documented that the donor platelets were invariably Group O resulting in a minor incompatibility-associated hemolytic transfusion reaction., A balance must, therefore, be struck between sufficient volume to buffer pH and lesser volumes to minimize adverse effects. Most blood banks have a policy of using ABO-incompatible PCs: however, most have not included a method to minimize transfusion of anti-A or anti-B. Such methods should include performing isoagglutinin titers for Group O PC. If compatible platelets are unavailable, the plasma may be removed by centrifugation and replaced by saline or albumin, especially in cases where a donor antibody is directed against an antigen present in the recipient. Alternatively, the platelets may be suspended in group AB plasma which lacks anti-AB isoagglutinins. This also prevents immunoglobulin A (IgA) present in donor plasma from precipitating anaphylactic reactions in the IgA-deficient recipient.
In our study, swirling was present in all of the RDP units. One hundred and five RDP units had Grade 1 swirling (21%), 170 RDPs had Grade 2 swirling (34%), and 224 RDPs had Grade 3 swirling (45%). This is in contrast to a multicenter study (13 participating centers) done by Bertolini and Murphy where swirling was reported to be absent in 0%–18% of PCs. This variation in swirling results between our study and other studies maybe because in our study swirling was recorded by a single investigator which leads to minimal variation in grading of swirling results, whereas in multicenter studies, the swirling grade results are recorded by different investigators which lead to interpersonal and inter-institute variation.
All of the PCs in our study had a pH of more than or equal to 6.0 which meets the quality control criteria for PCs according to the Directorate General of Health Services guidelines. pH is the simplest parameter indicator of the platelet storage lesion and probably the most important quality parameter that gives an indication of viability and potential recovery of PCs at the end of the storage period. pH has been identified as the parameter having the highest correlation with recovery and survival of platelets. Platelet viability is markedly affected by pH. In the absence of oxygen, stored platelets revert to glycolytic metabolism with increased generation of lactic acid and consequent fall in pH within 3 days of preparation. The final pH of PC and hence in vivo recovery and survival will depend on the type of storage container, storage conditions, and the volume of residual plasma. Therefore, platelet storage bags must allow for free gaseous exchange. In our study, all the platelet bags were of the second generation which allows free gaseous exchange and permits storage till 5 days.
Platelets must be stored in sufficient plasma, whose bicarbonate content acts as a buffer, to maintain pH at greater than 6.2. Depletion of bicarbonate by high lactic acid level typically at 20–25 mmol/l lowers pH and results in loss of platelet viability.
Ninety-three RDP units in the study were visibly reddish in appearance. Among the visibly reddish units, only 11 units (2.2%) had RBC contamination of >0.5 ml which violates the quality control criteria for random-donor PCs. The causes for red cell contamination of RDP units can be multifactorial including improper loading, and expression of whole-blood bag inside centrifuge leads to contamination of PRP during soft spin centrifugation which ultimately leads to contamination of platelets during the second hard spin centrifugation and failure to tap the red cells from the tubings/ports at the top end of bags before loading into the centrifuge.
Some contaminating red cells can be found in even the most refined PCs which can immunize an RhD-negative recipient. High cell counts including RBCs accelerate the platelet storage lesion due to increased metabolism. However, Beutler and Kuhl found that low RBC and WBC count appears to have no effect on the glucose consumption, lactate production, or fall in pH.
About 99.7% of PCs in the study had a platelet count of more than 3.5 × 1010 per unit which meets the quality control criteria for a RDP unit prepared from a 350-ml blood donation.
A positive correlation was found between pH and platelet count in the RDP units. However, there was no correlation between pH and volume and between platelet count and volume.
The results in our study show a statistically significant difference between pH and swirling grades in the RDP units. A significant difference was found between pH and swirling Grade 2 and 3 but no difference with swirling Grade 1. No significant difference was found between volume of RDP units and swirling grade.
In a multicenter study done by Bertolini and Murphy, swirling was present in 94% of PCs with a pH value in the range of 6.7–7.5. Although data from another center in this multicenter study indicated that pH in swirling-negative PCs was significantly lower than that in swirling-positive PCs and pH values differed significantly (P < 0.001), 48% of swirling-negative PCs had a pH lower than 6.4 or higher than 7.5.
There was also a statistical significance between swirling and platelet counts in the RDP units. A significant difference was found between platelet counts with swirling Grade 1 and 2 but no difference with swirling Grade 3.
These results are in contrast to the multicenter study done by Bertolini and Murphy  where no correlation was found between the presence or absence of swirling and platelet counts. However, in one center in that multicenter study, pH was found to be slightly, although not significantly, higher, and in that center, there was a significant difference in the platelet counts in the swirling-negative PCs and the swirling-positive PCs.
The strength of the study was that it was conducted over 18 months with samples at regularly spaced intervals which can compensate for the seasonal variations in platelet counts. Most of the available studies in literature had used BacT/Alert and other recent automated culture systems for sterility testing for contamination in PCs. This was one of the few studies which used the BACTEC-automated culture system for sterility testing of PCs.
The limitations of the study were the small sample size considering the general incidence of bacterial contamination in PCs. In this study, general accuracy approach was followed which could have been a potential factor for error while measuring samples with pH close to 6.2 as the accuracy of pH meter used in the study was 0.001–0.1. Automated hematology analyzers give false values when platelet counts are very high in PCs. Though we tried dilution-correction in some RDP units, inadvertent error in measurement cannot be completely compensated.
Financial support and sponsorship
This study was funded intramurally by JIPMER.
Conflicts of interest
There are no conflicts of interest.
| References|| |
Das S, Kale M, Beena PM, Kumar H. Bacterial contamination of platelet at university hospital: A prospective surveillance study. Int J Curr Microbiol App Sci 2015;4:805-12.
Nanua S, Weber C, Isgriggs L, Dunne WM Jr. Performance evaluation of the VersaTREK blood culture system for quality control testing of platelet units. J Clin Microbiol 2009;47:817-8.
Ribault S, Harper K, Grave L, Lafontaine C, Nannini P, Raimondo A, et al.
Rapid screening method for detection of bacteria in platelet concentrates. J Clin Microbiol 2004;42:1903-8.
Patel TG, Shukla RV, Gupte SC. Impact of donor arm cleaning with different aseptic solutions for prevention of contamination in blood bags. Indian J Hematol Blood Transfus 2013;29:17-20.
Bertolini F, Murphy S. A multicenter inspection of the swirling phenomenon in platelet concentrates prepared in routine practice. Biomedical excellence for safer transfusion (BEST) working party of the international society of blood transfusion. Transfusion 1996;36:128-32.
Werch JB, Mhawech P, Stager CE, Banez EI, Lichtiger B. Detecting bacteria in platelet concentrates by use of reagent strips. Transfusion 2002;42:1027-31.
Barrett BB, Andersen JW, Anderson KC. Strategies for the avoidance of bacterial contamination of blood components. Transfusion 1993;33:228-33.
Leiby DA, Kerr KL, Campos JM, Dodd RY. A retrospective analysis of microbial contaminants in outdated random-donor platelets from multiple sites. Transfusion 1997;37:259-63.
Kulkarni N. A prospective study to determine the frequency of bacterial contamination of platelets. Indian J Hematol Blood Transfus 2014;30:319-20.
Yazer MH, Stapor D, Triulzi DJ. Use of the RQI test for bacterial screening of whole blood platelets. Am J Clin Pathol 2010;133:564-8.
Raturi M, Shastry S, Raj P. Cumulative quality assessment for whole bloodderived platelets: A compliance review. Glob J Transfus Med 2017;2:38-43. [Full text]
Schmidt M, Sireis W, Seifried E. Implementation of bacterial detection methods into blood donor screening – Overview of different technologies. Transfus Med Hemother 2011;38:259-65.
Vollmer T, Knabbe C, Geilenkeuser WJ, Schmidt M, Dreier J. Bench test for the detection of bacterial contamination in platelet concentrates using rapid and cultural detection methods with a standardized proficiency panel. Transfus Med Hemother 2015;42:220-5.
Kaufman RM. Platelets: Testing, dosing and the storage lesion – Recent advances. Hematology Am Soc Hematol Educ Program 2006; p. 492-6.
Riedel S, Siwek G, Beekmann SE, Richter SS, Raife T, Doern GV, et al.
Comparison of the BACTEC 9240 and BacT/Alert blood culture systems for detection of bacterial contamination in platelet concentrates. J Clin Microbiol 2006;44:2262-4.
Dunne WM Jr., Case LK, Isgriggs L, Lublin DM. In-house validation of the BACTEC 9240 blood culture system for detection of bacterial contamination in platelet concentrates. Transfusion 2005;45:1138-42.
Fung MK, Downes KA, Shulman IA. Transfusion of platelets containing ABO-incompatible plasma: A survey of 3156 North American laboratories. Arch Pathol Lab Med 2007;131:909-16.
Valsami S, Dimitroulis D, Gialeraki A, Chimonidou M, Politou M. Current trends in platelet transfusions practice: The role of ABO-RhD and human leukocyte antigen incompatibility. Asian J Transfus Sci 2015;9:117-23.
] [Full text]
Saran RK. Quality assurance in blood transfusion. Transfusion Medicine Technical Manual. 2nd
ed. New Delhi: Directorate General of Health Services; 2003. p. 341-60.
Beutler E, Kuhl W. Platelet glycolysis in platelet storage. IV. The effect of supplemental glucose and adenine. Transfusion 1980;20:97-100.
[Table 1], [Table 2], [Table 3], [Table 4]