Resident Readiness General Surgery 1st Ed.

A 65-year-old Man Status Post Surgery With a HCT of 24%

Alden H. Harken, MD and Brian C. George, MD

Three days ago, a 65-year-old man underwent a sigmoid resection, end colostomy, and Hartman’s for a free diverticular perforation and raging peritonitis. Perioperatively the patient was aggressively fluid-resuscitated and monitored with a CVP, A-line, and Foley. Current vitals are 100/60, HR 110, and CVP 12. He remains intubated on an FiO2 of 0.6 and ABGs: PO2 100, PCO2 35, and pH 7.35. When you arrive in the SICU, this patient’s most recent urine output was 20 mL/h. The 6:00 AM labs just returned, and his HCT is 24% and Hgb 8 g%.

1. Should this patient be transfused with packed red blood cells?

2. What else is in banked blood besides hemoglobin?

3. What is an appropriate transfusion threshold for a patient who is actively bleeding?

4. When should you order a type and screen, and when should you order a type and cross?



1. This patient’s anemia is most likely dilutional, caused by his body beginning to mobilize all the fluid he received during his perioperative resuscitation 3 days prior. Still, his hemoglobin does seem low and his oliguria most likely represents some element of renal hypoperfusion. The most rapid way to increase this patient’s systemic oxygen delivery is to transfuse him with bank blood. But will that improve his overall outcome?

Goal-directed therapy (Rivers et al; see Chapter 12) protocol mandates that, following initial fluid resuscitation and inotropic support, and if the mixed venous O2 saturation remains below 70%, then transfuse blood to an Hct of 30%. But the Transfusion Requirements in Critical Care (TRICC) trial (Hébert et al) provides persuasive evidence that permitting patients to float down to an Hbg of 7 g% is safe (if not safer than transfusing at an Hbg threshold of 9 g%).

In fact, there is no controversy. Both of these beautifully conducted trials were conducted in very different cohorts of patients:

• GDT patients were all in septic shock and were treated within 6 hours of admission.

• TRICC patients were all documented euvolemic (normal CVP) and were randomized only after 3 days in the ICU.

So, in this case you search hard for evidence of myocardial ischemia (12-lead electrocardiogram) and in the absence of ischemia, your patient is euvolemic and has been in the ICU for 3 days. So, TRICC trumps GDT—don’t give blood yet.

2. When you transfuse banked blood, you are infusing much more than oxygen-carrying hemoglobin. There is a nonzero rate of infectious contamination. Also, the white cells in transfused blood are all “primed” to release toxic oxygen metabolites and proteases. Those white cells can also attack the patient, in a process called graft-versus-host disease. Leukodepletion does cut the white count by several orders of magnitude—but not to zero. And hemolyzed RBC membranes are also toxic. In our hospital, it is OK to transfuse blood that has been sitting in the refrigerator for up to 42 days. More like gymnasts than wine, banked blood does not improve with age.

3. The GDT or TRICC trials cannot and should not be generalized to patients who are actively bleeding. In fact, even using the Hgb to determine when to transfuse is a risky proposition. As an example, let’s say your patient’s Hgb is 13. If you remove half of his blood volume in the next 5 minutes, his Hgb will still be 13—as many red cells were removed as were white cells, platelets, and serum. Eventually equilibration with interstitial fluid will result in a low hematocrit, but not before he is dead. In general, you should aim to replace any ongoing blood loss greater than about 1.5 L—and to do so early, even prophylactically. Oh, and stop the bleeding!

4. Both “type and screen” and “type and cross” result in a determination of the patient’s blood type (ie, ABO antigen and Rh[D] antigen). The difference really comes down to the “screen” and the “cross.” For a type and screen, the patient’s blood is screened for a standard panel of foreign red blood cell antibodies. It takes about 30 minutes to complete. For a type and cross, the blood bank also cross-matches the patient’s blood against a specific unit of blood to determine if there is any cross-reactivity. It takes another 30 minutes in addition to the time required for the type and screen (ie, 1 hour total for both). Unless it is an emergency and you are going to transfuse O negative blood, you must order a type and cross before any transfusion.

Once a type and cross has been performed, the blood bank will set those units of blood aside specifically for that patient. This means they are out of circulation, and can impact the utilization of what is a very precious and limited resource. In order to address this issue most blood banks will not reserve cross-matched blood for longer than 3 days.

These tests are not only used when a patient already needs a transfusion, but also sent as part of the preoperative workup for all but the most trivial elective operations. Ordering these tests will often be your responsibility. But do you order a type and screen or a type and cross? The risk of significant bleeding determines which test is ordered. For low-risk procedures (eg, inguinal hernia repair, small bowel resection), a type and screen is done. On the off chance that blood is needed intraoperatively, this will save 30 minutes. It also enables the blood bank to provide ABO and Rh(D) matched (but uncrossed) blood while the cross-match is being performed. For procedures that have a high risk of bleeding (eg, liver transplant, open aortic aneurysm repair, cardiac surgery), a specific number of units of blood are typed and crossed and set aside for the patient. In that case it is available almost immediately in the more likely event that the patient needs it.


Image Red blood cell transfusions can sometimes be more harmful than helpful.

Image The TRICC trial suggests that ICU patients who are euvolemic should be transfused only for an Hgb less than 7.

Image The Early Goal-directed Therapy trial suggests that patients who are acutely septic, have been volume resuscitated, are on appropriate inotropic support, and still have a mixed venous oxygen saturation of <70% should be transfused to a goal Hct of 30%.

Image Type and cross reserves blood specifically for that patient, but only for a limited time. It should be used for procedures with a significant risk of significant bleeding.


1. A patient in the ICU has a Hgb of 9.2. It has been slowly drifting down over the past 5 days. She is noted to be euvolemic but is on pressors. When should you transfuse her?

A. When the Hgb is <9, per the GDT trial

B. When the Hgb is <7, per the TRICC trial

C. Now

D. 2 days ago

2. A trauma patient just rolled in with a sword stuck in his RUQ. He has an Hgb of 12 but is lethargic, cold, clammy, and tachycardic. When should you transfuse him?

A. When the Hgb is <9, per the GDT trial

B. When the Hgb is <7, per the TRICC trial

C. Now

D. 10 minutes ago

3. What is the most appropriate order for a patient who is scheduled to undergo a coronary artery bypass graft in 10 days?

A. Type and screen only

B. Type and cross 2 U

C. Type and cross 10 U

D. None of the above


1. B. This is a euvolemic patient in the ICU with no evidence of bleeding. The TRICC trial applies. Hold off on transfusion.

2. D. You are already behind because the patient is in shock. The patient has probably already lost at least 30% of the blood volume, regardless of the Hgb.

3. D. This patient will need to be typed and crossed for blood given the high likelihood of bleeding, but not until closer to the operation. The blood bank will not reserve the blood for this patient for 10 days.


Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. NEngl J Med. 1999;340(6):409–417.

Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368–1377.