Approach to Patient with a Bladder Tumor with Severe Hematuria and a Hemoglobin Level of Eight with Associated Dyspnea

I would like to re-state the concept that hemorrhage occurs within Virchow's triad. To manage bleeding, one must identify and rank the pathophysiological factors affecting vascular integrity, blood flow, and hemostasis.

1. The most important cause of painless gross hematuria in anyone, even when it is self-limited, is vascular injury from bladder cancer. In a patient without known transitional cell carcinoma with painless gross hematuria, one must direct all interventions at getting tissue, and this usually begins with cystoscopy. Between 10 and 20 percent of these patients will be diagnosed with bladder cancer.¹ 

2. Local hyperfibrinolysis contributes to hematuria in patients with urothelial tumors, but systemic hyperfibrinolysis is unlikely in the absence of disseminated cancer. Recent data from liver transplant recipients indicate that thromboelastometry is two to four times more sensitive than thromboelastography (TEG) for identifying a systemic fibrinolytic state. TEG was 99 percent specific but only 23 percent sensitive, so it cannot be used to rule out systemic hyperfibrinolysis.² 

3. Intractable hematuria from almost any cause – with or without systemic hyperfibrinolysis – should be treated with systemic antifibrinolytic therapy plus local measures, which could include intravesical instillation with formalin, alum, or prostaglandin PGF2α (best for low flow bleeding); arterial embolization; or hydrostatic pressure or urinary diversion (as a last resort). For patients with bladder cancer, hypofractionated radiotherapy or chemoperfusion with mitoxantrone often controls bleeding.³ 

4. Today, more than five and a half years since this case was first reviewed, there are no new data that recombinant activated factor VII (rFVIIa) should be used for massive hematuria from any condition other than acquired hemophilia or inherited hemophilia with an inhibitor. The therapeutic index of rFVIIa as a nonspecific hemostatic agent is unfavorable.⁴ 

(Editor’s Note: The original question was submitted to Dr. Kroll through Consult a Colleague. He expanded his answer for print.)

A 78-year-old male with a bladder tumor had severe hematuria and a hemoglobin level of 8 with associated dyspnea. Urologists were unable to evaluate secondary to severe bleeding in the bladder. Coagulation testing revealed a normal prothrombin time (PT) of 10 seconds, but his activated partial thromboplastin time (aPTT) was elevated at 108 seconds, lowering to 56 seconds when mixed with normal plasma. Levels of coagulation factors VIII, IX, X, XI, and XII were normal, although the report noted the possibility of an inhibitor. Von Willebrand factor (VWF) studies were normal, as were assays for anticardiolipin antibodies (ACA) and lupus anticoagulant (LA). The patient continues to bleed and his hemoglobin continues to drop. Should any other work-up be done? Given that he had no benefit from fresh frozen plasma infusions, should I give factor VII complex?

Bleeding is impaired hemostasis and therefore can be evaluated from the same perspective that Virchow organized for thrombosis — as dysfunction within the triad of blood (mainly soluble coagulation factors, fibrinolytic apparatus, or platelets), blood vessel (mainly its structural integrity or permeability), and/or blood flow (mainly the effects of ischemia or hyperperfusion). In most cases of bleeding, a simple, single explanation emerges. But in more complicated cases, like the one in question, I try to identify and rank the hemostatic elements that have broken down.

When I see bleeding from the bladder, I focus first on tumor-associated hemorrhage. This is because one must immediately direct therapy toward the cancer, if the fundamental pathophysiological element is tumor-erosive hemorrhage. But how does one establish the diagnosis if hematuria is so severe as to prevent visualization and biopsy of the bladder mucosa? I would retreat slightly and look for disturbed and perhaps related hemostatic elements that could be reversed and thereby permit proper evaluation of the root cause of the hematuria.

When there is severe hematuria, perhaps related to transitional cell carcinoma, I ask the question: Is there hyperfibrinolysis? The malignant urothelium is rich in urokinase-type plasminogen activator,¹  and local and sometimes systemic hyperfibrinolysis develops and contributes to bleeding in patients with bladder cancer. Systemic hyperfibrinolysis is not apparent by examining the PT and aPTT, but always results in elevated fibrin split products and d-dimers. It can be identified by thromboelastography (TEG) or by measuring the euglobulin lysis time. Neither test is available in every clinical laboratory, but in our institution we perform TEG. When there is hyperfibrinolysis, or anything else that de-stabilizes polymerized fibrin, the TEG tracing over 30 to 60 minutes shows a decline in the clot tensile strength — the plateau begins to decay or decline.²  When hyperfibrinolysis is identified, one can try to control it with an antifibrinolytic drug such as aminocaproic acid or tranexamic acid.³  The main risk of these treatments in someone with hematuria is ureteral obstruction if blood refluxes and clots in a ureter, so one must monitor urine output and serum creatinine concentration closely.

In this patient, there may be a weak coagulation factor VIII inhibitor. The isolated elevated aPTT must reflect a deficiency of, or an inhibitor to, either factor XII, XI, IX, or VIII and the incomplete aPTT correction with mixing (from 108 to 56 seconds) indicates that there is an inhibitor. The most commonly seen inhibitor is a lupus anticoagulant, which is an antiphospholipid antibody that prolongs an in vitro clotting assay, which, in this case, would be the aPTT. But lupus anticoagulants are not associated with bleeding — except in the rare case where they have antiprothrombin activity⁴  — and this patient’s anticardiolipin antibody assay and lupus anticoagulant screen were normal. So the most parsimonious explanation for this case of bleeding and abnormal coagulation studies is the presence of the most common acquired inhibitor encountered in a bleeding patient: an acquired factor VIII inhibitor. It can emerge in patients with solid tumors,⁵  including bladder cancer.⁶  It almost always causes clinically significant bleeding that can be controlled with a pharmacological intervention such as recombinant activated factor VII (rFVIIa).⁷  And the inhibitor itself (an auto-antibody to factor VIII that causes rapid clearance) is often reduced or eradicated with immunosuppressive therapy.⁸  To evaluate for this possibility, I suggest repeating the factor VIII assay and measuring the inhibitor titer in Bethesda units, which is the inverse of the dilution of the patient’s plasma that decreases factor VIII activity in control plasma by 50 percent. My speculation is that when this patient’s factor VIII level is repeated, it will be borderline low or low and that he will have an inhibitor level < 5 units. If an inhibitor is identified, I would administer rFVIIa, although some would first try human factor VIII in cases of low titer acquired factor VIII inhibitors.⁸ 

Should one use rFVIIa under any circumstances? Assuming that the PT and platelet count remain satisfactory, I would first try bladder irrigation. If this or other urological manuevers don’t work and everyone remains mystified by the pathophysiology, I would begin aminocaproic acid at an infusion rate of 1 gm/h. If aminocaproic acid doesn’t work, I would then add rFVIIa, but only at 20 to 30 mcg/kg and beginning at four- or even six-hour intervals.⁹  This dose is prohemostatic but probably causes fewer thromboses than the conventional dose of 90 mcg/kg every two hours.¹⁰