We analyzed blood product use by utilizing large administrative data (DPC data) provided by 73 acute care hospitals in Japan. We developed three multiple linear regression models to predict hospital-wide use of blood products and components (red blood cells, platelets, fresh frozen plasma, and albumin), and evaluated predictive accuracy of each model with R2 values. The first model had five predictive factors representing the structure of each hospital: bed size, annual volume of general anesthesia, provision of cardiac surgery, bone marrow transplantation, and plasma exchange therapy. The second model predicted blood usage by patient age and across several diseases (e.g., hematopoietic disease) and surgical procedures (e.g., cardiovascular surgery), which consume a large volume of blood products. The third model predicted hospital-wide blood product use by adjusting for the distribution of diagnostic groups at the hospital. By identifying the most important diagnostic group (disease or surgical procedures) with respect to blood usage, we found that most blood products were consumed by a small number of diagnostic groups. The third model had significant predictive accuracy, and the second model also showed relatively satisfactory accuracy. In contrast, bed size and annual volume of general anesthesia were not significantly associated with the use of blood products after adjusting for other factors. By utilizing DPC data, we are now able to easily assess risk-adjusted blood product use.

View full abstract

Breakthrough bacterial contamination during blood collection is a potential risk in predeposit-type autologous blood transfusion. A total of 2,209 blood samples were collected from July 2005 to May 2009 for autologous blood transfusion at the University Hospital of the Ryukyus. The first 5 ml of blood samples was cultured in BACTEC Plus Aerobic/F blood culture bottles to detect bacterial contamination. Bacterial growth was monitored by an automated blood culture system, BACTEC 9120. Of 2,209 blood samples, 4 (0.18%) were positive, with two strains of Staphylococcus hominis, and one each of S.warneri and S.capitis. Bacterial concentrations in blood samples were determined using linear regression between the time (hour) to detect positive and colony-forming units (cfu)/ml of the respective isolates. It was demonstrated that three positive blood samples contained 0.06 to 5.3 cfu/ml while the other, positive for S.warnerii, was 1.4×10⁴ cfu/ml. To evaluate the benefit of removing white blood cells, S.hominis isolate was spiked into an aseptic blood sample from which white blood cells were removed. This preparation revealed significant exponential bacterial growth after standing for 16 hours at room temperature. However, an aseptic blood sample containing a normal number of white blood cells did not show any significant bacterial growth within a 24-hour period. Microscopic examination confirmed that a number of white blood cells had phagocytosed the spiked staphylococcal cells. From these results, we concluded that the breakthrough bacterial contamination ratio was significantly low, but not negligible. Thus, care is necessary in the collection of blood, particularly in disinfection procedures. The benefits of removing white blood cells from the preparation for autologous blood transfusion is an issue that should be discussed further.

View full abstract

The safety and efficacy of GVHD prophylaxis using tacrolimus alone following reduced-intensity cord blood transplantation were retrospectively assessed based on the tacrolimus whole blood concentration. Among 39 patients, toxicity profiles observed were renal (n=8, 20.5%), central nervous system (n=4, 10.3%), GI tract (n=2, 5.1%), and liver (n=2, 5.1%). Mean tacrolimus whole blood concentration of the previous 10 days in those who had renal damage was higher than those who did not (20.4 ng/ml vs. 14.8 ng/ml, P<0.05). The incidence of renal damage was 60.0% and 6.9% for those who showed tacrolimus whole blood concentrations of ≥17 ng/ml and < 17 ng/ml, respectively (P<0.05). The incidence of severe acute GVHD (grade II-IV) was 27.7%, suggesting the effectiveness of single tacrolimus prophylaxis. In those who showed a tacrolimus whole blood concentration above 17 ng/ml, no acute GVHD was observed, suggesting the potential advantage of maintaining higher a whole blood concentration of tacrolimus to reduce the incidence and severity of GVHD. Thus, it is suggested that the toxicity and efficacy of tacrolimus were dose-dependent, and that close monitoring of tacrolimus whole blood concentration is of particular importance both to reducing toxicity and increasing safety.

View full abstract

Washed/replaced-platelet concentrates (W/R-PC) are effective in avoiding adverse transfusion-reactions caused by plasma-containing platelets. For washing and replacement of plasma from platelet concentrates, we use M-sol, an additive solution prepared by mixing several solutions approved for clinical use. M-sols are prepared on-demand in a bio-clean room, which is not convenient for preparing W/R-PC. We have modified the method for preparing M-sols with regard to the following two points. 1) The bacteria barrier filter-integrated bag can be effectively used to avoid the use of a bio-clean room. 2) The pH of M-sols can be stabilized by vacuum-packing M-sol bags with aluminum packages, which allows preservation for at least one year. In addition, no differences were observed between M-sol preparations for single use and multiple use. There was no change in pH during storage for one year at 30°C with vacuum-packed M-sol. No significant difference was observed in the quality of replaced-platelet concentrates (R-PC) between freshly prepared and stored M-sols. Use of M-sols transported from a distant place did not affect the quality of washed-platelet concentrates (W-PC). W/R-PC can be readily prepared using M-sols prepared by this modified method.

View full abstract

In this study, we retrospectively carried out a survey on trigger values of hemoglobin concentration for red cell concentrates (RCC) transfusion, prothrombin time for fresh frozen plasma (FFP) transfusion, and platelet counts for platelet concentrates (PC) transfusion at Matsuyama Red Cross Hospital in 2006. Trigger values were then compared with the blood usage guidelines revised in 2005 in Japan. Actual rates of trigger value assessment on the day of transfusion were 78.1% in cases with RCC transfusion, 54.8% with FFP, and 77.5% with PC. The rate of appropriate RCC usage was high in patients with hematological disease (76.7%), but low in perioperative patients (41.0%) and those with chronic hemorrhage (33.3%). Concerning FFP, the rate of trigger value assessment remained low, and the rate of appropriate usage remained at lower levels in all patients (7.7∼22.7%). The rate of appropriate PC usage was high in patients with hemorrhage or disseminated intravascular coagulation syndrome (92.6%), perioperative patients (67.4%), and in those with acute leukemia or malignant lymphoma (85.0%). On the contrary, the rate of appropriate PC usage was low in patients with aplastic anemia or myelodysplastic syndrome (37.6%), and solid tumors or miscellaneous diseases (18.5%). Concerning patients with hematological disease, the rate of appropriate PC usage increased just after the reservation system was introduced. This was because doctors recognized trigger platelet counts used for PC, and then medical technologists checked the platelet counts of patients before transfusion and confirmed the clinical decision made by the doctor. In conclusion, the checking system used by medical technologists and reconsideration of the trigger values by doctors were useful for the appropriate usage of blood products.

View full abstract

We experienced a case of a D-negative woman in whom anti-D antibody rose to a titer of 4,096 by Coombs' test, despite receiving a red blood cell transfusion from apparently D-negative donors. The RHD gene of the blood components that were transfused to her was analyzed by the PCR-SSP method, and one-point mutation G1227A was found in the gene from two blood components. This case was therefore thought to be exposed to D-antigen by blood transfusion of D_el type red blood cells, as a result of which she produced anti-D antibodies. However, no hemolytic episodes due to the anti-D antibody were found during her clinical course. It should be recognized that D_el type blood is occasionally included in D-negative blood, and that a D-negative recipient may produce anti-D antibody on blood transfusion of D_el type red blood cells.

View full abstract

Based on a survey by a joint meeting of transfusion committees in Miyagi Prefecture, we analyzed the association between red blood cell (RBC) wastage and the establishment of a transfusion management system using data submitted by 34 hospitals with high transfusion activity in Miyagi Prefecture. Hospitals with higher transfusion activity had a lower median value of RBC wastage rate than other hospitals over the year 2006 to 2007. Among hospitals categorized as having a "partially established" group for transfusion quality management and with a frequency of transfusion committee meetings six times and above yearly, RBC wastage rate was decreased in 2007 compared with that in 2006, while hospitals below this level of frequency had no improvement in RBC wastage rate over the two years. Moreover, the frequency of transfusion committee meetings six times and above yearly was found to be a significant factor in reducing RBC wastage in 2007. Based on these findings, the establishment of a transfusion management system was identified as an important factor in helping hospitals to reduce RBC wastage down to the lower targets.

View full abstract

[in Japanese]

View full abstract