Blog with interesting cases and/or problems related to anesthesia with discussion based on best evidence in the literature.

March 4, 2018

63 year old male for shoulder arthroscopy and polycythemia vera

Today I was informed by the orthopedic surgeon that the next patient scheduled for shoulder surgery had a Hgb of 19.4 g/dL.  He was 63 and appeared to be in poor health.  He was on several medications including:

  • ASA 81 mg
  • Buprion 300 mg qd
  • Citalopram 20 mg 
  • Fluoxetine 40 mg
  • Gaba 300 mg
  • Glipizide 5 mg
  • Lantus 50 units qd
  • NPH 30 units qd
  • Metformin 1000 mg
  • pravastatin 40 mg
  • Na+ 141
  • K+ 4.8
  • BUN 10
  • Cr 0.83
  • Glucose (finger stick) 63 mg/dL patient without symptoms of hypoglycemia.
  • Hct 57.8
  • Platelets 245,000
  • total CO2 33
In addition to untreated HTN, depression and diabetes, the patient had significant OSA.
The surgeon was eager to proceed with surgery.  Therefore, I had a conversation with the patient in an attempt to determine the etiology of his polycythemia.  He admitted to a chronic smoking history, but denied any trouble with SOB, orthopnea or lung disease.  His room air sat was 94%.  He had previously been made aware of his elevated Hgb levels, and it was recommended to him to donate blood.  However, he stated that he had never donated blood because they would not accept his blood due to his diabetes.  I explained to him the importance of immediate follow up with his doctor to treat his condition.  I made a presumptive diagnosis of polycythemia vera since I could not find any obvious source of hypoxemia in this patient. A formal diagnosis requires that several criteria be met as listed below.

I placed an interscalene catheter under US guidance.  In the meantime, as we waited for surgery, I gave the patient 2 L of LR in order to mitigate his PV to a small degree.  We also planned to extract a small amount of blood once we arrived to the OR (between 60 and 100 mL's).  The patient underwent an uneventful shoulder scope with a surgical time of 63 min with un uneventful PACU stay.

Polycythemia Vera (PV) is a myeloproliferative disease which results in excessive production of RBCs. In addition, it can also lead to luekocytosis and thrombocytosis.  Other myeloproliferative disorders include essential thrombysthenia and primary myelofibrosis. PV is considered when the Hct is greater than 48% in women and 52% in men. This is an important disease to recognize in the perioperative period because patients with this condition are at greater risk for venous and arterial thrombosis, and hemorrhage [1,2], due to hyperviscosity of the blood and dysfunctional platelets.  Diagnosis can be challenging and is not easily made at the bedside. In order to be positively diagnosed with polycythemia vera, two major criteria and one minor criterion must be present

Major criteria include:

  1. hemoglobin greater than 18.5 g/dl in men, greater than 16.5 g/dl in women
  2. presence of the JAK2 V617F or similar mutation.
Minor criteria include: 
  1. bone marrow biopsy showing myeloproliferation. 
  2. serum erythropoietin level below the normal range.
  3. endogenous erythroid colony formation in vitro [3].  
In additioon, in patients who NOT have the JAK2 mutation, but do have elevated Hgb, AND two minor criteria, a positive diagnosis of PV can be made.

The only available information from the above list in my patient was a Hbg of 19.4 md/dL.  My patient also had significant OSA.  I questioned whether this might account for his polycythemia rather than polycythemia vera.  A recent retrospective analysis of polycythemia in patients with OSA considered over 527 patients who had polysomnography indicating OSA.  The authors concluded that the incidence of polycythmia in OSA patients was indeed very rare.  They also reported that severity of OSA did not correlate with greater increases in Hgb levels.  The authors did concede that, "
In those rare polycythemic OSA patients, polycythemia is corrected by CPAP therapy in the majority."  In addition, my patient had diabetes and a finger strip glucose reading prior to surgery read 63 mg/dL.  I did not treat this as the patient stated that he had no symptoms and usually did not feel bad until his sugar reading fell below 50 mg/dL.  It should be noted, that glucose bedside monitors can read artificially low levels of glucose in patients with elevated hematocrits.

Polycythemia results in hyper viscous blood.  It should be recognized that Poiseuille's equation indicates that flow through a tube is proportional to the fourth power of the radius, the pressure differential, and inversely proportional to the length and viscosity of the fluid. Therefore, as viscosity increases with increasing hematocrit, the pressure differential must increase in the exact amount to maintain an equal blood flow.  Unfortunately, as the hematocrit increases it results in a semilogarithmic increase in the blood's viscosity.  In fact, as the hematocrit hits 45%, it can be said that the increase in viscosity "takes off". (see figure)

Therefore, cardiac work must increase to maintain forward flow, and indeed, as polycythemia progresses much beyond a hematocrit of 45% cardiac output decreases.  Clinically, this can manifest as SOB, lack of energy, tiredness, and even angina.  Therefore, a careful history is critical in these patients. It should be noted that hyperviscous blood is particularly problematic for the small capillaries which can lead to ischemia especially in patients with  poor heart function.

Polycythemia Vera can also result in paradoxical hemorrhage. This is believed to be a result of platelet dysfunction. Acquired  Von Willebrand's diesease could be another cause of increased bleeding tendency in these patients. Most authors recommend the avoidance of neuraxial anesthesia in patients with PV unless a coagulation profile is normal (i.e.thromboeslastography) due to known platelet dysfunction.

These patients may be at high risk for complication and death after surgery. In a review from 1963, patients with uncontrolled polycythemia had a 79% rate of complications and a mortality rate of 36% [4]. However, patients who were treated had their rate of complications decreased to 28% and mortality rate to 5%. Another study found that the incidence of thrombosis was 7.7% and that of major hemorrhage was 7.3% [5].

A lower hematocrit can be favorable in the perioperative period.  The increased viscosity of blood is particularly problematic during this period.  Some evidence suggests that the brain oxygenation and brain oxygen delivery is optimized when the hematocrit is decreased to below 45% [5].
In several case reports dealing with patients who have presumptive PV, the authors desribe performing acute normovolemic hemodulition.  The formula that should be used in order to arrive at the proper Hgb level preoperatively is V=EBV X (Ho - Hf)/Have.    

V=volume of blood to extract
EBV = Estimated Blood Volume (in adult males 70 mL/kg is often used)
Ho= starting Hematocrit level of patient 
Hf= final Hematocrit or minimal allowable (around 40 to 45%).
Have= the average between the start and final hematocrit.

An equivalent volume to replace the blood withdrawn should be infused during the extraction of blood.  I.e. if 500 mL of blood is extracted, then 500 mL of colloid might be infused.   In general, the goal is to reduce the viscosity of blood, and generous intravenous fluids (LR or NaCl) should be infused.

Even in patients that have normalized their hematocrit prior to surgery there is elevated risk of thrombosis.  A retrospective review of patients with either PV or ET and receiving anticoagulation therapy during their surgery estimated that the incidence of DVT after major surgery was increased 5-fold. Unfortunately, the risk of major bleeding was also increased in this retrospective review demonstrating the paradoxical risk of hemorrhage in PV. The authors recommended using LMWH for prevention of DVT during the perioperative period.

One final note on this particular patient was his list of medications.  He was taking bupropion, a norepinephrine/dopamine reuptake inhibitor for depression, citalopram a norepinephrine/serotonin reuptake inhibitor and fluoxetine a selective serotonin reuptake inhibitor.  This patient was therefore, at some risk for serotonin syndrome should. This syndrome is characterized by changes in autonomic, neurological, and cognitive behavioral functions and appears to result from over-stimulation of 5-HT1a and 5-HT2 receptors in the central grey nuclei and medulla.  Therefore, Demerol, tramadol and dextromethorphan would be relatively contraindicated in this patient. Furthermore, as mentioned, this patient had a glucose reading of 63 mg/dL preop.  There are a number of articles and case reports indicating the risk of hypoglycemia in patients taking SSRIs.  It has been shown that SSRIs can enhance insulin sensitivity in man [7-10].  Therefore, this patient had three possible causes of low blood sugar: 1) too much diabetic medication plus a missed meal, 2) polycythemia induced misreading of glucometer, and 3) SSRI induced insulin enhanced sensitivity.

In general, the approach to patients with PV with scheduled surgery needs to consider any clinical symptoms resulting from the PV, current medical condition of the patient, age, degree of polycythemia, and stress and duration of surgery.  Therefore, a patient having a relatively short procedure, with no evidence of any symptoms from the polycythemia and in otherwise decent health may tolerate surgery with acute preoperative normovolemic hemodilution.  A patient with uncontrolled PV undergoing major surgery on the other hand may need to be delayed to allow for a hematologist consult for preoperative treatment.

1. Berk PD, Goldberg JD, Donovan PB, Fruchtman SM, Berlin NI, Wasserman LR. Therapeutic recommendations in polycythemia vera based on Polycythemia Vera Study Group protocols. Semin Hematol. 1986;23:132–143.
2. Ruggeri M, Rodeghiero F, Tosetto A, Castaman G, Scognamiglio F, Finazzi G, et al. Postsurgery outcomes in patients with polycythemia vera and essential thrombocythemia: a retrospective survey. Blood. 2008;111:666–671.
3. Wadleigh M, Tefferi A. Classification and diagnosis of myeloproliferative neoplasms according to the 2008 World Health Organization criteria. Int J Hematol. 2010;91:174–179. 

4. Wasserman LR, Gilbert HS. Ann New York Acad Sci 1964;115:122-38.

5. Thomas DJ, Marshall J, Russell RW, Wetherley-Mein G, du Boulay GH et al.  Lancet. 1977;2(8045): 941-943

6. Ruggeri M, Rodeghiero F, Tosetto A, Castaman G, Scognamiglio F, et al.  Blood 2008; 111(2): 666-671

7. Araya V, Contreras P, Aguirre C, Depix MS, Zura ML. The effect of fluoxetine on insulin resistance in nondiabetic obese patients [in Spanish]. Rev Med Chil. 1995;123(8):943-947.

8. Potter van Loon BJ, Radder JK, Frölich M, Krans HM, Zwinderman AH, Meinders AE. Fluoxetine increases insulin action in obese nondiabetic and in obese non-insulin-dependent diabetic individuals. Int J Obes Relat Metab Disord. 1992;16(2):79-85.
9.  Lustman PJ, Clouse RE, Nix BD, et al. Sertraline for prevention of depression recurrence in diabetes mellitus: a randomized, double-blind, placebo-controlled trial. Arch Gen Psych. 2006;63(5):521-529.
10.  Paile-Hyvärinen M, Wahlbeck K, Eriksson JG. Quality of life and metabolic status in mildly depressed women with type 2 diabetes treated with paroxetine: a single-blind randomised placebo controlled trial. BMC Fam Pract. 2003;4:7.

No comments: