Priority Topic: Anemia

Key Feature 1: Assess the risk of decompensation of anemic patients (ex: volume status, the presence of congestive heart failure [CHF], angina, or other disease states) to decide if prompt transfusion or volume replacement is necessary. 
Skill: Clinical Reasoning, Selectivity
Phase: Diagnosis, Treatment

Anemia is common problem. And sometimes it can take a bit of time to figure out what the exact cause is. Most of the time, at least in primary care, anemia is mild and we have time to figure out the etiology to try to deal with the underlying reason for the problem. But in the hospital setting, it becomes not uncommon to manage patients with significant anemia and who may or may not warrant a blood transfusion. 

According to the UpToDate article, Use of blood products in the critically ill (2017), indications for the transfusion of packed red blood cells include:

1. Hemorrhagic shock
2. Acute hemorrhage with inadequate oxygen delivery
3. A hemoglobin concentration of <70 g/L

The article states that, "High risk populations, such as those patients with acute myocardial infarction may have higher thresholds."

In practice, as a person who generally is not working in critical care settings, I tend to rely on the <70 threshold for deciding on whether or not to order a blood transfusion. But it is important to keep in mind that some patients will have less of a reserve to cope with a decreased supply of oxygen, and these patients may warrant a blood transfusion at a higher threshold than the <70 g/L rule of thumb. In the patient with angina for example, they are exhibiting a sign that they have inadequate oxygen delivery to their heart muscle, and if there is a full on acute MI happening and associated anemia, they certainly warrant a blood transfusion as their heart is already starving of oxygen. 

It is also important to consider volume status*, and whether the patient is hypovolemic or hypervolemic. In a patient who is volume deplete but not anemic (essentially dehydrated), we would normally just replace their fluid needs with a crystalloid solution such as normal saline. But in a patient who is volume deplete and more than mildly anemic, they need a blood transfusion; infusion of a crystalloid solution may help keep their body tissue perfused, but it would not increase the oxygen carrying concentration of their blood (and in fact would contribute to decreasing it), and so they really need red blood cells to rectify the problem. On the other hand, in a patient who is hypervolemic and anemic, they may appear more anemic than they otherwise would be if their blood volume wasn't so expanded, and they are also at risk of transfusion-associated circulatory overload (TACO). If a transfusion is necessary in such a patient, I suppose it would be prudent to both reduce circulatory volume and replace red blood cells, but this would be far beyond my level of comfort. I would certainly consult a critical care specialist or a physician who is trained in the management of decompensation of the disease process responsible for the volume overload, such as a cardiologist in the setting of congestive heart failure and/or a nephrologist if a patient has comorbid kidney failure.

One other point that I think is important to mention here is that it is always stressed that patients need to provide consent when it comes to receiving blood products (unless there is implied consent as in an emergency situation). It is interesting to me that this is stressed, as really, we should be gathering consent any time we are about to start an intervention. But I suppose this is particularly stressed because there is a small risk of some very serious consequences. The general risks to outline to a patient regarding blood transfusion when obtaining informed consent include:

1. Immune reactions
   1. Acute hemolytic reactions (extremely rare [<0.02%] but can be fatal [0.003%], occurring from an error in matching blood types - we have strong safeguards in place so as to prevent this from happening)
   2. Delayed hemolytic reactions (also rare [0.025%] and are often mild [jaundice and hemoglobinuria] or go entirely undetected, but can be more pronounced in patients who have received previous transfusions - mostly significant because if it happens it means the body has created new antibodies and so their blood will need to be screened for new antibodies before receiving another transfusion)
   3. Febrile nonhemolytic reactions (common [7%] from antibodies in the patient's blood reacting to antigens on white blood cells and cytokines in the donor blood - self-limited and don't even require stopping the transfusion as antipyretics can be given)
   4. Allergic reactions (severity can range from mild urticaria to anaphylaxis - severe reactions will mean the blood transfusion will need to be stopped and the patient may need to receive epinephrine)
2. Infection (serious but rare, and would be treated like any blood infection - the risk of HIV transmission is somewhere along the lines of 1 in 1.5-2.0 million units of red blood cells)
3. Volume overload (risk is that it could cause pulmonary edema, more relevant in a patient with heart or kidney failure, with the biggest acute risk being pulmonary edema)
4. Hypothermia (if cold blood is transfused - risk of this is reduced by having protocols in place to warm the blood prior to transfusion)
5. Coagulopathy (only a significant concern in a massive blood transfusion and not significant in a transfusion of 1-2 units of packed red blood cells - it would probably not be me having this conversation if I needed to transfuse more than that at once)
6. Citrate toxicity (this is a consequence of being transfused with too much citrate, a component used in the storage of blood products that prevents coagulation, and again, it would only really be a significant concern in the setting of a massive blood transfusion, and so likely not to bear significant weight in the conversation I have to obtain consent with my patients)
7. Acute lung injury (sort of like it sounds, and rare but serious, and can last hours to days - the patient will be in respiratory distress and needing ICU management if this happens)
8. Posttransfusion purpura (extremely rare, this would happen 7-10 days post-transfusion, and they would be given IgG to try to mitigate the consequences)

Okay, considering all of the above rare but possibly real consequences, it is definitely important to obtain proper consent, and to know that while usually all ends well, transfusing blood products is not without risks, and that the indication should warrant the risk. It's important to reassure the patient that there will be a physician and other health care professionals present during the transfusion who will monitor and address concerns should they arise. 

*I perform a volume assessment as part of my cardiovascular exam (I like to think of it in two parts, the first half being the precordial exam, and the second being my assessment of volume status). Importantly, each sign by itself may be an indicator of more than volume status (ex: tachycardia in the setting of significant pain), but it is the composite picture that matters. The signs I look for to assess volume status are as follows:

1. Signs of dehydration
   1. Late signs: 
      1. Decreased level of consciousness (the differential of causes for this is of course much broader than hypovolemia alone, but if there are other signs that suggest significant hypovolemia, this is an ominous sign of possibly severe hypovolemia)
      2. Hypotension (systolic blood pressure <90 or mean arterial pressure <65**)
      3. Oliguria (<20 mL/h or <500 mL/day)
   2. Early signs:
      1. Tachycardia (if >150 or irregular then cardiovert as per ACLS)
      2. Orthostatic tachycardia/hypotension
      3. Tachypnea 
      4. Dry mucous membranes
      5. Cool extremities
      6. Distal pulses not palpable
      7. Cap refill > 5s
      8. Poor skin turgor
2. Signs of fluid overload 
   1. Hypertension with a wide pulse pressure
   2. S3 heart sound
   3. Elevated jugular venous pressure (JVP) (patient should be positioned semirecumbent @ 45 degree angle) (If elevated, I perform the abdominojugular test to rule out elevation from right-sided heart failure, unless I already am aware the patient has right-sided heart failure and so don't even bother assessing the JVP at all)
   4. Bilateral lung rales or wheeze
   5. Pitting edema

** To calculate the mean arterial pressure (MAP) = diastolic BP added to (0.3 x the pulse pressure, where the pulse pressure = the systolic BP - the diastolic BP)