Bacterial persistence on blood donors' arms after phlebotomy site preparation: analysis of risk factors.

Bacterial contamination of blood products (BCBP) was one of the earliest recognized complications of blood transfusion.1 The source of contamination is rarely found, but that most frequently reported is the blood donor’s skin.2 Despite wellconducted antisepsis, high concentrations of bacteria may remain on the surface of the skin.3-6 The aim of our study was to assess determinants associated with persistence of bacteria on blood donors’ arms after preparation of the phlebotomy site. Consecutive donors were included if they had been accepted for donation after a medical selection complying with our current legal regulations and had given their informed consent to participation in this study. The phlebotomy site was the antecubital fossa and was prepared according to our current two-stage procedure with povidone-iodine 7.5% scrub (Betadine-Scrub®, Munipharma AG, Switzerland) followed by application of 1% PVP-iodine (Orsan®, B. Braun-Dexon, SA, Barcelona, Spain). We recorded demographic variables related to the donors, such as sex, age, arm of phlebotomy and number of previous donations. We also asked them about their place of work (outdoors or indoors). Other variables recorded were related to the collection site (blood bank or mobile sessions) and collection time (winter or summer). We also recorded the number of donors attending each collection session. Finally, we performed the cultures after the 1% PVP-iodine had dried (30 seconds vs. 60 seconds). After disinfection, cultures were taken using a 6×2-cm flexible tongue following the manufacturer’s instructions (Microcheck®, Biomedics, SL, Madrid, Spain). Bacteria were identified according to standard microbiological methods. Culture results were recorded on a separate form at the laboratory and transferred to the corresponding donation form. Unpaired t-tests or Mann-Whitney U analysis according to Levene, were used to compare the means of the quantitative variables between bacterial and non-bacterial persistence on blood donors’ arms after preparation of the phlebotomy site. The association between a qualitative variable and bacterial persistence on the donors’ skin was estimated by the odds ratio (OR). A p value less than 0.05 was considered statistically significant. Statistical analysis was performed with SPSS for Windows 11.0 statistical software (SPSS inc., Chicago, IL, USA). Our study comprised 120 blood donors. Persistent bacterial contamination was found in 42 cases (35%). The most commonly isolated bacterial species were coagulase negative staphylococci (36 cases; 85,7%). Other bacterial species isolated were Bacillus sp. (3 cases; 7.1%), Acinetobacter lwoffi (2 cases; 4.8%) and Micrococcus sp. (1 case, 2.4%). The analyses of quantitative and qualitative variables are shown in Tables 1 and 2, respectively. In this study, we found bacterial persistence at 42 (35%) out of 120 skin sites prepared for blood donation. These results agree with Goldman’s and McDonald’s studies.5,6 The authors reported bacteria persistence on blood donors’ arms after disinfection with all evaluated techniques. The chosen method, which was considered a validated, optimal best practice disinfection technique, produced zero counts on 70% of arms. Thus, our results extend these findings and confirm that it may be almost impossible to decontaminate human skin.7 We found three reports on factors contributing to bacterial contamination at the collection stage; the available data are scarce and contradictory. First, McDonald et al. reported no correlation between age, gender, occupation, donor appearance or amount of hair on the donor’s arm and the bacterial load preor post-disinfection.8 Second, Perez et al. revealed an association between transfusion-associated bacterial contamination and the number of previous donations in the univariate analysis and also in the final multivariate analysis.9 Third, a multivariate analysis of determinants of bacterial contamination of whole-blood donations performed by Perez et al. showed a donor age >35 years, the participant blood bank and lack of repetition of the scrubbing to be significantly assoTable 1. Univariate analysis of associations between quantitative variables and bacterial persistence on blood donors' skin after phlebotomy site preparation.