Polina Zmijewski, MD
Background
Subnormal Hemoglobin and Hematocrit levels in the ICU are remarkably common, and care should be taken to establish the cause of anemia prior to intervention. Critical illness anemia is a broad term for the multifactorial causes of subnormal hemoglobins in critically ill patients. A primary cause is decreased erythrocyte production in the acute inflammatory state. Inflammation causes the upregulation of hepcidin, which causes iron sequestration in the liver and macrophages, which diminishes erythropoiesis by the bone marrow. Critically ill patients can also exhibit renal insufficiency, with decreased serum levels of erythropoietin. In addition to patient pathophysiologic effects, critical illness anemia is often caused by frequent phlebotomy, with blood draws in some patients occurring as frequently as every 4 to 6 hours. Lastly, resuscitation with crystalloid can artificially dilute hemoglobin, creating a subnormal level.
Sepsis results in a decline in Hb of 1.5 g/dL over the first three days after presentation in a review of nonbleeding ICU patients. These patients do not need to be transfused.
Evaluation
It is important to stress that critical illness anemia is a diagnosis of exclusion, and that worsening hemodynamics or steadily decreasing levels of hemoglobin in post surgical patients are signs of hemorrhage until proven otherwise. Alternative sources of bleeding, such as hemorrhage from trauma, procedures, GI loss, etc, should be routinely ruled out with physical examination, lab testing, and occasionally imaging.
Postoperative patients that are hemodynamically unstable and thought to be bleeding need to return to the OR. There are also a number of subspecialty cases that come through the SICU including major spine, plastics, ENT, and urology patients. If a spine patient is unstable and the Hemovac drain is of high output the most likely source of the hypotension is bleeding from the operative site. Notify the SICU attending and the spine service immediately in this situation suggesting return to the OR.
Transfusion principles
In situations of acute bleeding events and hemorrhage, it is important to remember principles of balanced resuscitation, shown to increase survival and decrease ICU length of stay. DCR should be activated in cases of massive transfusion (>10 units PRBCs in 6 hours). Coagulopathy should be reversed as soon as possible. Consider cryoprecipitate in addition to other blood products if serum fibrinogen levels fall below 150 or after 10 units of PRBCs. Thromboelastogram studies are useful in situations where there is concern for a coagulopathy or in cases of massive transfusion requirement to ensure balanced resuscitation of blood products. Consider administration of DDAVP in patients in acute or chronic kidney failure to correct platelet dysfunction.
In cases of active hemorrhage, blood product should be administered irrespective of hemoglobin level. The question of “when to transfuse” in a case of critical illness anemia with a slowly drifting hemoglobin is more challenging, and should be highly individualized per patient. Current practice guidelines have established a hemoglobin of 7 g/dL as our “transfusion threshold”. Most trials have shown better outcomes and less resource use in patients who were transfused at a level of 7 vs a more liberal transfusion trigger of 10. The exception to this transfusion trigger of 7 are in those patients with acute myocardial infarction or unstable angina. It is important to note that no clinical trials exist regarding transfusion at a threshold of < 7, and that this number was chosen arbitrarily based on expert consensus, believing 7 to be the minimal amount of hemoglobin needed to adequately oxygenate the tissues. It is important to note that banked blood has low levels of 2,3 DGP, which increases hemoglobin affinity to oxygen, making it less likely that transfused hemoglobin will appropriately offload oxygen to the patient’s tissues. Transfusion in patients with critical illness anemia should be therefore carefully considered, as deleterious effects are not uncommon and patients should be spared unnecessary risk.
The following blog entry sums up anemia and transfusion targets:
Key trials looking at transfusion triggers(all RCTs)
-Hebert TRICC NEJM 1999: 833 ICU patients with Hb < 9. Goal 7-9 vs 10-12 g/dL. Trend towards reduced mortality in 7-9 group.
-FOCUS NEJM 2011: 2016 high risk hip surgery patients. Target > 8 or > 10. No difference in 60 day mortality or ability to walk.
-Villanueva NEJM 2013: 921 patients with UGIB. Target 7-9 vs. 9-11. Better 45 day mortality in 7-9 group(5% vs 9%)
-TRISS NEJM 2014: 921 patients with sepsis. Target >7 vs. >9. No difference in 90 day mortality.
TRICS3 NEJM 2018: 5243 cardiac surgery patients. Target >7.5 vs. >9.5. No difference in composite of death, CVA, MI, AKI.
Consider reviewing these trials in more detail as there are issues with use of composite outcomes and whether the groups are generalizable to our patient population.