Current clinical guidelines from three medical societies suggest that the hemoglobin threshold for blood transfusions during surgery should be 7 g/dl or 8 g/dl. These guidelines also note that patients don’t need a transfusion when hemoglobin levels are above 10 g/dl. However, when hemoglobin levels fall between these thresholds, there is little consensus on the best course of action.

Although four landmark studies published over the past 5 years suggest that it’s safe to wait until hemoglobin levels fall to 7 g/dl or 8 g/dl before transfusing, wide variation and excessive use of blood transfusions have been reported. Advances in viral testing in recent years have made blood transfusion safer, but risks still exist for these patients, including lung injury, immune suppression, and viral transmission. Blood is also in scarce supply and expensive.

Wide Variation of Hemoglobin Thresholds

In an issue of Anesthesiology, my colleagues and I had a study published that sought to confirm the wide range of hemoglobin thresholds used by surgeons and anesthesiologists. Over 18 months, we collected data on more than 48,000 surgical patients at Johns Hopkins Hospital. Within the institution, there was a variation of up to 3 g/dl in hemoglobin thresholds among surgeons and anesthesiologists, when compared with their peers. Virtually all providers used thresholds above the ones recommended in guidelines, and none used thresholds below the recommended range.


Surprisingly, sicker patients—generally those under-going cardiac surgeries—had the lowest hemoglobin thresholds, whereas those undergoing surgery for pancreatic cancer, orthopedic issues, and aortic aneurysms received blood transfusions at higher thresholds. The amount of blood transfused did not correlate with how sick the patients were or with how much blood is typically lost during certain surgeries. It’s possible that cardiac anesthesiologists and surgeons are more aware of blood transfusion guidelines and recent trials that support lower thresholds, making them more tolerant of lower hemoglobin levels when compared with others.

Providing Feedback Changes Behavior

At Johns Hopkins, we created a list of blood use and thresholds for each surgeon and anesthesiologist. When this list was published, physicians wanted to know where they ranked among their peers. Those who learned that they ranked among providers who offered transfusions at the highest hemoglobin thresholds have since decreased the amount of blood they give to their patients.

Providing physicians and nurses with the type of feedback we offered in our study was helpful, but many hospitals don’t have this luxury. Some don’t have blood management programs, so data on blood utilization may be inaccurate. It’s likely that a large variation exists in the tolerance of anemia and the use of blood transfusion nationwide.

Using communications training together with a surgical checklist creates a culture where personnel can be patient advocates without fear of retribution.

Based on our findings and those of previous landmark trials, surgeons should make efforts to use a restrictive strategy for transfusion, and to lower the transfusion trigger to hemoglobin levels of 7 g/dl or 8 g/dl. Optimizing blood utilization is one of the few areas in medicine where outcomes can be improved while simultaneously reducing risk and saving healthcare utilization dollars.