Horses and Zebras

Tuesday, December 8, 2015

Acute Myeloid Leukemia / Pancytopenia

Today's case centred around a classic internal medicine differential diagnosis for multilineage cytopenias. The patient is an elderly man with cirrhosis who presented with fatigue and lethargy. He was found to have a profound anemia, thrombocytopenia and leukopenia. He was admitted to the internal medicine service wherein a bone marrow aspiration and biopsy was performed revealing acute myelogenous leukemia.

The approach to the differential diagnosis in this case is very important, as there are multiple diagnostic entities that can produce this syndrome:

· Malignant processes (think of bone marrow replacement with hematologic malignancies like leukemias, scar as in myelofibrosis, or non-hematologic malignancies with a significant degree of bone marrow infiltration).

· Cellular destruction (it is unusual for this to cause trilineage cytopenias, but certainly anemia/thrombocycopenia combinations are possible in diagnoses like HUS or TTP)

· Inflammatory/Autoimmune diseases like SLE can lead to pancytopenia.

· Drug effects can be categorized into drugs that are intentionally halting cellular maturation and production like chemotherapeutic agents, and drugs that exhibit this as an idiosyncratic side effect like phenytoin or Septra.

· Substrate deficiency – the example used during morning report was B12 deficiency which can lead to megaloblastic anemia, hypersegmented neutrophils on the blood film, and neurologic complications like dorsal column dysfunction and peripheral neuropathies.

· Sequestration – Sequestration can happen in tissues like the spleen and liver in conditions that produce portal hypertension, such as cirrhosis. We usually attribute thrombocytopenia to this condition, but it can certainly be seen in other cell lines.

· Infections – viral infections like hepatitis B can lead to aplastic anemia. HIV can certainly lead to cytopenias. In immunocompromised hosts such as those with advanced HIV, disseminated Mycobacterium avium infections can replace bone marrow leading to pancytopenias.

The physical examination of a patient with these clinical entities can be very revealing:

· On vital signs – findings such as hypotension or tachycardia can be revealing of a hyperdynamic cardiac output secondary to anemia. Orthostatic changes would suggest hypovolemia. Fever may suggest an underlying malignant or infectious process.

· On general appearance – pallor or jaundice may suggest anemia and/or hemolysis. It’s also important to characterize the degree of distress the patient may have. Rashes like patecheiae or purpura may relate to the degree of thrombocytopenia. Inside the mouth, gingival hyperplasia may occur with some variants of leukemia.

· On cardiovascular examination, a flow murmur over the tricuspid valve due to ramped up cardiac output from anemia may be audible. In someone with cirrhosis, signs of pulmonary hypertension (TR murmur, high JVP or CV waves, RV heave, loud and/or palpable P2) may be related to portopulmonary hypertension.

· On abdominal examination, the liver size is important, but splenomegaly is also very important. Generally, percussion of Traube’s space, followed by Castell’s sign, followed by percussion using Nixon’s method increase in specificity for splenomegaly. Importantly, your physical examination is only around 80% sensitive for splenomegaly meaning that you cannot rule out splenomegaly on exam alone. Abdominal ultrasound would be the mainstay of diagnostic tests for splenomegaly.

Management of this condition involves treatment of the underlying process (the leukemia). This typically involves chemotherapy, the type and duration of which is a highly subspecialized field. It is unusual but possible for leukemia to be present without circulating blasts, but they are otherwise the hallmark of diagnosis.

In addition to treating the leukemia, it is important to recognize the complications of leukemia including bleeding from thrombocytopenia, disseminated intravascular coagulation from some forms of leukemia, and infections as a result of inadequate numbers and function of leukocytes. Aside from some of these complications, the pathophysiologic process leading to death from acute leukemias is the cytopenias themselves.

Image Credit: de.academic.ru

Friday, December 4, 2015

Acute Bloody Diarrhea

Today's case involved an older man with an acute history of watery, bloody diarrhea. He had a history notable for radiation to the abdomen. His stool studies later became positive for Campylobacter (and he actually had a Campylobacter bacteraemia). In addition, his stool ova and parasites were positive for Entamoeba histolytica.

There were some valuable learning points today:

-Acute diarrhea is a common presenting illness both in primary care and in our hospital settings. The differential diagnosis is substantially different for acute and chronic diarrhea (see previous blog post on chronic diarrhea). Acute diarrhea is generally caused by infections.

-The history is helpful in determining the cause of the patient’s symptoms. First and foremost, the duration of symptoms and the qualities of the stool (watery, loose, bloody, mucousy, etc.) are important. Other useful information would be a travel history, history of antibiotic use, recent sick contacts, exposure to other people with diarrheal illnesses, presence of upper gastrointestinal symptoms, recent consumption of undercooked meats/eggs/poultry or unpasteurized milk, and a past medical history of immunocompromise. Also, we mentioned several cases in which rectal sexually-transmitted diseases have presented as “diarrhea” even though the primary issue is usually rectal pain and discharge; therefore, taking a sexual history is also helpful. In younger patients, a new presentation of an inflammatory bowel disease like Crohn disease or Ulcerative colitis should be considered. Don’t forget that there is a bimodal distribution of new diagnoses of IBD, and that elderly people in their 50’s to 80’s can also newly develop IBD. In their differential diagnosis, things like microscopic colitis and ischemic colitis usually also need to be considered.

-The infectious agents that can cause an acute diarrhea can be bacterial or viral. I like to think of the invasive bacterial pathogens (those that cause bloody diarrhea or dysentery) with the acronym “SSCYE” – Salmonella, Shigella, Campylobacter, Yersinia, E. coli. Other bacteria like Vibrio are frequent causes of watery travelers’ diarrhea and not necessarily enteroinvasive. In the right clinical context (or even without it) Clostridium difficile needs to be considered. It rarely produces a bloody diarrheal illness, but can in people with inflammatory bowel diseases or those on anticoagulants. Viruses like norovirus, rotavirus, and other enteroviruses may cause an acute, self-limited, usually non-bloody diarrhea. Parasitic infections often produce a longer duration of symptoms, but these can include non-invasive pathogens like Giardia lamblia and invasive pathogens like Entamoeba histolytica. Giardia is typically transmitted in water from lakes/wells in rural areas.

-Not all proven infectious diarrheas need to be treated. Entamoeba histolytica diarrhea always needs to be treated. Bacterial infections like Campylobacter in this case are often self limited. Reasons for treatment would include prolonged or non-resolving symptoms, systemic symptoms like fever and general malaise, bacteremia as in this case, immunocompromise, or significant other medical comorbidities.

-We spoke briefly about radiation-induced colitis. Radiation forms part of the management of tethered bowel structures (essentially rectal cancer) and can cause diarrhea and enteritis. This is typically when small bowel or large bowel are in the radiation field. The symptoms may start several weeks following radiation therapy and may persist for months.

Tuesday, December 1, 2015

Altered Mentation and Hyponatremia

Today's case was of a man with a previous history of lung cancer and brain metastasis, a primary psychotic disorder, and COPD who presented with a subacute alteration in his mental status characterized by some delusional thoughts.

There were a lot of valuable learning points:

-We discussed a differential diagnosis and approach to altered mentation. This will come up continuously when seeing ward patients or admitting from the ED. Broad categories include drugs (illicits, narcotics, alcohol, CNS depressants, withdrawal), infections (meningitis is the most concerning), metabolic derangements (acidemia, hypercalcemia, B12 deficiency, thyroid disease), and structural derangements (intracranial hemorrhage, mass effect, etc). In most cases, it is a good idea to obtain some form of neuroimaging, and some bloodwork (CBC, lytes, calcium profile, liver enzymes) +/- an arterial blood gas for acidemia.

-Delirium is something that we see frequently in medicine. Delirium is an acute change in mental status characterized by a fluctuating course, disordered thought, and agitation or depressed level of consciousness. It has a very wide differential diagnosis, and can be caused by quite a number of things. The typical treatment of delirium is fixing the underlying cause which can vary quite substantially depending on what it is. Around 20% of the time, no definitive cause is found for the delirium. Delirium is independently associated with higher risk of poor outcomes from hospitalization, suggesting that it should be treated as a medical emergency on that basis. In the case of our patient, it is less clear whether there is also some component of disordered thought and delusions on the basis of his psychiatric disorder as well.

-Hyponatremia is a common problem in hospitalized patients. Think of hyponatremia as a water disorder rather than a sodium disorder. Essentially there is too much free water for the amount of sodium in the body.

-Why do we care so much about hyponatremia? There are a couple of reasons. Acutely, hyponatremia can produce alterations in mental status and seizures a result of cerebral edema (too much water outside brain cells, osmotic shift of water into brain cells, cerebral edema, raised intracranial pressure, seizure). The other component that we care about is overcorrection of hyponatremia which can lead to osmotic demyelination syndrome (previously central pontine myelinolysis). For those who have never seen this complication, it is a devastating injury to the brainstem that can result in “locked-in” syndrome (only able to move eyes). Risk factors include low muscle mass (cirrhotic, malnourished), hypokalemia, and female sex. The classic teaching recommended not increasing the sodium by more than 10mmol/L in a 24 hour period, but nephrologists are now stratifying patients by their risk for ODS, and recommending more conservative correction for those at higher risk.

-There are a lot of ‘approach to hyponatremia’ documents out there but I will try to give you a simplified one that I use. Firstly, there are only THREE mechanisms that can produce hyponatremia:

(1) – free water is unable to be excreted by the kidney leading to an elevated water:sodium ratio in the body (this is the case in extreme renal failure).

(2) – Antidiuretic hormone is turned on, and is the only mechanism that separates salt from water retention. With this on, water is preferentially reabsorbed by kidneys leading to an increased water:sodium ratio in the body.

(3) – there is insufficient solute to excrete the free water. Our kidneys can only dilute urine down to around 50 mOsm/L meaning that guzzling free water and eating things without any solute (sodium, proteins, etc.) will invariably lead to, you guessed it, an increased water:sodium ratio in the body.

-As far as an approach goes, I like to think of things this way:

(1) – is the hyponatremia real? (check the serum osmolality). Things that can cause hyperosmolar or euosmolar hyponatremia are elevated paraproteins, lipids, glucose, etc.

(2) – Now that it’s real, is it symptomatic? (i.e. is the patient seizing)

If it’s symptomatic, the treatment should be to rapidly correct the sodium by 1-2mmol/L until the patient stops seizing. This is done in ICU’s with 200cc of 3% NaCl solution rapidly infused.

(3) – Now that it’s real, and asymptomatic, is it acute or chronic? Most of the time you won’t know whether it’s acute or chronic because people don’t get their labs checked at home. That said, acute hyponatremia can be corrected acutely without risk of ODS. Chronic hyponatremia (48-72h) must be corrected slowly to avoid ODS.

(4) – Now that it’s real, asymptomatic, and chronic, what is the patient’s volume status? What volume status is the kidney seeing? Ordering the urine lytes answers the second question (if the kidney is seeing low volume, the urine sodium will be under 20). Diuretics will interfere with this. The rest of your physical exam answers the first question. Many people will order urine osmolality as well. This tells you whether ADH is acting or not (it is high if ADH is acting). An example of hyponatremia when ADH is not acting would be psychogenic polydipsia, or insufficient solute intake.

From here, it’s helpful to break things down into causes by volume. Hypovolemia causes ADH to be released to maintain plasma volume. This is not because extracellular water stays intravascular, but because ADH acts as a potent vasoconstrictor to maintain systemic vascular resistance. So patients with diarrhea, vomiting, sepsis, etc. may have appropriate ADH on board.

Hypervolemic patients like CHF, cirrhotic, or nephrotic syndrome patients are hypovolemic from a renal perspective, meaning that their kidneys behave like the first group. Treatment of their underlying problem may resolve the hyponatremia.

The euvolemic group has a lot of causes: hypothyroidism, adrenal insufficiency, medications (diuretics, antiepileptics, SSRI’s), “tea and toasters”, psychogenic polydipsia, and the ever-famous SIADH (syndrome of inappropriate antidiuretic hormone).

SIADH occurs for a number of reasons: lung pathology, tumours, CNS disease, pain, nausea, etc.

So based on the above, you can see that the treatment varies widely. If you give someone with SIADH normal saline, their sodium may go down because that potent ADH will reabsorb all the water, and allow excretion of all the sodium. On the contrary, someone with hypovolemic hyponatremia will correct their sodium rapidly because the saline restores intravascular volume, leading to a shutoff of their ADH release, and a water diuresis. That’s why we sometimes give salt cubes, or urea, to patients with low solute intake – it drags out some of the excess water and increases the capacity for free water excretion.