A transfusion medicine physician at a large blood collection center in Texas is curious as to what crossmatch to transfusion (C/T) ratio hospitals use as their benchmark for appropriate pre-transfusion test ordering, when assessing red cell transfusion therapy for elective pediatric general surgery patients. A hospital in the inquiring physician's blood center service area routinely tracks C/T ratio by clinical department, as well as determining the overall hospital C/T ratio. The hospital's Blood Utilization Review Committee has noticed that the pediatric general surgeons routinely have C/T ratios greater than 4.0. The pediatric surgeons have been asked to address the high C/T ratio, since the patients in this practice are usually not neonates, nor are they cardiac surgery patients. The pediatric surgeons have responded that they are justified in having a higher C/T ratio than what is typically considered appropriate for adult patients, although they have provided no evidence to substantiate this point of view. Instead, the surgeons have challenged the blood bank and hospital's blood utilization review committee to provide evidence why the pediatric surgeons should meet the hospital standard. The hospital in question uses a C/T ratio of 2.0 as their benchmark goal for all departments. The overall hospital C/T ratio is currently <2.0.

The following response was received.

ADDENDA June 25, 2003

1. A colleague at a Children's Hospital in Northern California reports that at his hospital the institutional average C/T ratio for the entire hospital, for crossmatchable products, is very consistent from year to year at 1.5:1. For Pediatric Surgery, there is more variability, averaging 3.1:1 over the last 5 years. The responding colleague did not provide any explanation as to why their C/T ratio was higher for Pediatric Surgery versus the institutional average, nor if any effort was being taken to address the difference.

ADDENDA June 27, 2003

2. A colleague in Sacramento believes that under certain circumstances a C/T ratio of > 4.0 for a specific population of patients may well be entirely appropriate. He comments that in his opinion it is important to remember that the C/T ratio can be highly variable, depending upon the population of patients being assessed. This is because some patient populations for whom RBCs are crossmatched are almost always subsequently transfused (e.g., the "sure-thing", transfusion-dependent hematology/oncology or dialysis patient populations), whereas other populations are much less likely to need RBCs, but nevertheless must still have blood crossmatched for them, owing to their having a transfusion likelihood of > 10% (i.e., the standard of care followed by most practitioners in determining when to crossmatch RBCs). He adds that for example, in the former ("sure-thing") situation, the C/T ratios will very closely approach 11, since almost everyone who is crossmatched subsequently will receive their crossmatched RBCs. Whereas, in (an extreme version of) the latter situation, involving, say, only patients with an approximately 10% likelihood of needing to be transfused, the C/T ratios would be (and should be) on the order of 10:1. What a lot of people forget is that the "ideal" C/T ratio of </= 2.0 is meant to apply across the entire spectrum of patients within a facility containing a typical patient makeup. It is not, however, meant to apply to each and every subset of patients. In fact, if a hospital with a broad cross-section of patients (e.g., surgery, medical, emergency, obstetrical, etc.) were to observe C/T ratios of < 2.0 in every single subset of patients, then he believes that the hospital would inadvertantly place its practitioners in a situation of needing to order an increased number of STAT crossmatches.

ADDENDA July 7, 2003

3. A colleague reports that at an institution in Minnesota, a fair number of surgeries are done on pediatric patients. As of this writing that institution is not reporting C/T ratios for surgery, as their computer system does not have the capability of accurately capturing these numbers. However, they are considering targeting a few surgical procedures (both adult and pediatrics) and manually calculating C/T ratios for them. They would be interested to know what other institutions feel is an acceptable C/T ratio for surgery, or if institutions have attempted to decrease seemingly high ratios.

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator