++
PATIENT
Ms. W is a 56-year-old woman who comes to the office complaining of poor appetite for several weeks and black, tarry stools with generalized weakness for 1 day.
She has no prior history of bleeding, and her 3 prior obstetric deliveries were uncomplicated. Her past history is notable for cirrhosis due to chronic hepatitis C. Her medications include spironolactone and metoprolol; additionally, she has been taking ibuprofen for back pain.
On examination, she is pale. Her blood pressure is 110/80 mm Hg, pulse is 112 bpm, RR is 16 breaths per minute, temperature is 37.1°C. Her conjunctivae are pale, mucous membranes moist, lungs clear, heart regular rhythm with a systolic flow murmur at the left sternal border, liver minimally enlarged with a nodular edge, spleen palpable 3 cm below the left costal margin in the anterior axillary line, and she has no edema. Digital rectal examination discloses black stool that is Hemoccult-positive.
At this point, what is the leading hypothesis, what are the active alternatives, and is there a must not miss diagnosis? Given this differential diagnosis, what tests should be ordered?
++
Ms. W’s presentation suggests that she is having an upper gastrointestinal (GI) bleed. In addition to the specific GI causes of upper GI bleeding discussed in Chapter 19, GI Bleeding, it is important to consider whether patients who are bleeding have an underlying platelet or coagulation disorder contributing to the bleeding. Ms. W does have cirrhosis with splenomegaly that could lead to thrombocytopenia due to splenic sequestration; however, the large volume of the bleeding may suggest a coagulation factor disorder.
++
The prothrombin time (PT) measures what is commonly called the extrinsic clotting pathway (Figure 8-2), wherein tissue factor from an injury activates factor VII, followed by activation of the coagulation cascade through the "common pathway" factors (factors V, X, II [prothrombin] and I [fibrinogen]). Because the source of tissue factor reagents used in the laboratory to trigger the cascade vary, the PT will vary among different laboratories when testing the same sample. To overcome this problem of PT results not being comparable from one lab to another, the international normalized ratio (INR) was developed, to standardize PT results based on a constant associated with each laboratory reagent. The INR, which is routinely reported along with the PT, allows the clinician to feel confident that the data from different laboratories are comparable.
++
++
The activated partial thromboplastin time (aPTT) measures what is commonly called the intrinsic clotting pathway, starting with factor XII and working through factors XI, IX and VIII before entering the "common pathway."
++
In the evaluation of a prolonged clotting time, either PT or aPTT, one considers whether only 1 test is prolonged, and which factors contribute to each test. For example, an isolated prolonged PT suggests a deficiency of factor VII, since that is the only factor unique to the PT assay. An isolated prolonged aPTT raises concern about the 4 factors that are unique to the aPTT—factors, XII, XI, IX, and VIII. Prolongation of both the PT and aPTT raises concern either about the factors in the common pathway—I, II, V, and X—or a defect in multiple factors. Table 8-3 summarizes commonly seen patterns of factor deficiencies.
++
++
In clinical practice, prolongation of clotting times is most commonly acquired, either due to acquired deficiencies (eg, from malnutrition or liver disease) or acquired factor inhibitors. (While congenital factor deficiencies such as hemophilia certainly cause prolonged clotting times, these are far less commonly seen, and patients are well aware of them, making complex diagnostic evaluations unnecessary.) In order to distinguish between a factor deficiency and an inhibitor, it is often helpful to perform a mixing study, wherein one mixes 1:1 the patient’s plasma and normal plasma, to see if the clotting time corrects. If it does correct, the normal plasma has provided the missing factor to the patient’s plasma, indicating the abnormality is due to a factor deficiency. If it does not correct, the implication is that an inhibitor in the patient’s plasma is inactivating the clotting factor(s) from the normal plasma. Such inhibitors may be exogenous, such as inadvertent heparin in the mixture; or endogenous, such as an acquired factor inhibitory antibody.
++
Based on the data we have so far, Ms. W’s GI bleeding is most likely from the upper GI tract, probably induced by use of the NSAID ibuprofen. The severity of the bleeding may be exacerbated by a coagulopathy related to her cirrhosis. The history of poor appetite for a few weeks raises the consideration of vitamin K deficiency, and the presence of splenomegaly on examination suggests that thrombocytopenia due to splenic sequestration may also be contributing.
++
Table 8-4 lists the differential diagnosis.
++
++
A CBC shows WBC 9400/mcL, Hgb 7.8 g/dL, platelet count 76,000/mcL. A chemistry profile shows mild elevation of the transaminases but is otherwise normal. Her CBC 6 months ago showed an Hgb of 11.7 g/dL and platelet count of 80,000/mcL. Coagulation studies include a PT of 22 seconds (normal range 11–13 seconds), with an INR of 1.8 (normal 0.9–1.2). The aPTT is 39 seconds (normal 24–34 seconds).
Is the clinical information sufficient to make a diagnosis? If not, what other information do you need?
++
Ms. W has the stable, moderate thrombocytopenia generally seen in patients with portal hypertension and hypersplenism. Moderate thrombocytopenia such as this does not substantially increase the risk of bleeding, especially not the large volume GI bleeding she is experiencing. The coagulation abnormalities she has can certainly contribute to large volume bleeding.
+++
Leading Hypothesis: Liver Disease–Induced Coagulopathy
+++
Textbook Presentation
++
The classic presentation of liver disease–induced coagulopathy is variable. Patients may be asymptomatic, only discovered to have a coagulopathy incidentally on coagulation laboratory studies. Spontaneous bleeding is uncommon, but anything that stresses the patient (such as an injury, an operative procedure, or perhaps drug-induced gastritis) may lead to more bleeding than one might normally anticipate with that event in someone without liver disease.
++
Patients with liver disease–induced coagulopathy typically have a disproportionately longer PT (and therefore higher INR) than aPTT.
The coagulopathy is caused by impaired production of clotting factors by the diseased liver; the clotting factor with the shortest half-life, namely factor VII, would be expected to be most prominently affected. Since the PT/INR is so sensitive to factor VII levels, that test is more notably abnormal.
Coagulopathy is seen primarily in patients with severe liver disease. The liver has considerable reserve, and only when the impairment is severe does one find significant coagulopathy.
+++
Evidence-Based Diagnosis
++
In a patient with liver disease who is bleeding or in whom an invasive procedure is planned, the PT/INR and aPTT should be checked in order to screen for coagulation factor deficiencies.
If the screening tests are prolonged, it may be worth checking the levels of factor VII, factor V, factor II, factor IX, and factor X as well as fibrinogen to help determine which replacement therapy is most appropriate.
If factor VII is low but factor V normal, it suggests that vitamin K deficiency may be playing a role, whereas in severe liver impairment, both factors V and VII are reduced.
Because all the clotting factors except factor VIII are produced in the hepatocytes, all of them except factor VIII may be low in severe liver disease. Factor VIII is typically normal or even elevated in liver disease, a finding that may distinguish liver disease from DIC, in which factor VIII is low.
Another finding that may contribute to bleeding risk in severe liver disease is excessive fibrinolysis, the cause of which is a complex interplay between the production of and hepatic clearance of fibrinolytic activators and inhibitors.
While it may seem paradoxical, there may also be increased risk of thrombosis in liver disease. Several findings may account for this: reduction of the vitamin K-dependent anticoagulant proteins, protein C and protein S; and increases in factor VIII and sometimes von Willebrand factor.
++
Correct the coagulopathy using fresh frozen plasma to replete clotting factors. If the plasma fibrinogen level is particularly low (eg, < 100 mg/dL), infusion of cryoprecipitate may be helpful.
In severe cases, administration of recombinant activated factor VIIa may help stop the bleeding associated with liver disease; it is extremely expensive, however, and carries some risk of inducing thrombosis.