Delayed-Onset Heparin-Induced Thrombocytopenia With Cerebral Venous Sinus Thrombosis Following Total Knee Arthroplasty: Case Report

Abstract

Heparin-induced thrombocytopenia (HIT) is a prothrombotic state caused by autoantibodies against platelet factor 4 (PF4)-heparin complexes. Although HIT typically presents 5-10 days after the initiation of heparin, thrombosis and thrombocytopenia may occur up to several weeks following heparin withdrawal, so-called delayed-onset HIT. Although rare, there have been isolated reports of HIT-induced cerebral venous sinus thrombosis (CVST), which carry high rates of morbidity and mortality. There is a need to further characterize the etiology, clinical presentation, treatment paradigms, and outcomes of patients with HIT-induced CVST. Here, we present the case of a 57-year old female who presented to the emergency department with a headache and seizure 11 days following a right total knee arthroplasty for which she received 3 post-operative doses of enoxaparin. Work-up demonstrated acute intracerebral hemorrhage (ICH), CVST, and thrombocytopenia. Intravenous heparin resulted in rapidly deteriorating platelet count and subsequent serologic testing confirmed the diagnosis of HIT. Treatment with bivalirudin was initiated, the HIT resolved, and the patient was discharged home on hospital day 19 with long-term anticoagulation mediated by warfarin. At 3-month follow up, the patient had mild upper motor neuron pattern weakness and was living independently. This case depicts a rare case of delayed-onset HIT and CVST, highlights the importance of establishing a fluid treatment plan for managing HIT-induced CVST, and illustrates the importance of employing rapid anticoagulation despite acute ICH to achieve a desirable clinical outcome.

Introduction

Classical heparin-induced thrombocytopenia (HIT) results in thrombosis and thrombocytopenia due to immune system activation by heparin exposure, typically 5-10 days before the onset of symptoms. ¹ However, patients may also develop HIT several days to weeks after discontinuation of heparin, a phenomenon known as delayed-onset HIT.^2,3 Although thrombosis associated with HIT may occur anywhere in the venous and/or arterial systems, cerebral venous sinus thrombosis (CVST) is exceedingly uncommon with less than 25 cases reported in the literature over the past 30 years. The clinical presentation of CVST is variable and may include seizure, altered mental status, focal symptoms secondary to ischemic or hemorrhagic stroke, and/or signs of increased intracranial pressure. ⁴ CVST may also be associated with acute intracerebral hemorrhage (ICH), which may cause some providers to withhold anticoagulation in this setting. ² Given its rarity and high rates of morbidity and mortality, there is a need to further characterize the etiology, clinical presentation, treatment paradigms, and outcomes of patients with HIT-induced CVST. Therefore, we present a unique case of delayed-onset HIT with associated CVST following knee arthroplasty. We consider the molecular pathogenesis, demonstrate the importance of early anticoagulation despite the presence of acute ICH, and highlight the necessity of having a fluid treatment paradigm when treating HIT-induced CVST.

Case Report

A 57-year-old post-menopausal female on hormone replacement therapy (HRT) underwent an uncomplicated right total knee arthroplasty for degenerative osteoarthritis. She received 3 post-operative doses of subcutaneous enoxaparin (30 mg each), and twice daily baby aspirin starting on post-operative day 2 as prophylaxis against deep venous thrombosis. On post-operative day 11, she presented to the emergency department with altered mental status, right hemiparesis, and a generalized tonic-clonic seizure. Her platelet count was 148 × 10⁹/L, modestly reduced compared to her pre-operative platelet count of 229 × 10⁹/L. Non-contrast head CT revealed an acute intraparenchymal hemorrhage within the right temporo-parietal junction measuring 7.5 cm × 4.5 cm × 4.4 cm with a surrounding rim of vasogenic edema and extension into the posterior horn of the right lateral ventricle (Figure 1A). The CT scan also revealed diffuse hyperattenuation within the right transverse and sigmoid sinuses as well as the posterior superior sagittal sinus (Figure 1A-B), concerning for an acute cerebral venous sinus thrombosis. CT venogram subsequently confirmed absence of flow within the aforementioned sinuses (Figure 1C).

Figure 1.

Relevant clinical imaging and hematologic course for the patient presented in the representative case. A, Axial non-contrast head CT demonstrating a large intraparenchymal hematoma at the right temporoparietal junction. Note the hyperattenuation located at the confluence of sinuses (arrowhead). B, Axial non-contrast head CT again demonstrating the inferior temporal portion of the intraparenchymal hematoma. There is diffuse hyperattenuation within the right transverse sinus (arrowheads). C, Axial contrastenhanced CT venogram of the head demonstrating absence of contrast within the right transverse sinus. D, Platelet count with interventions noted. The patient continued on warfarin therapy for 6 months after the CVST.

She was admitted to the neurosciences intensive care where she was intubated due to progressive stupor and inability to protect her airway. She was also treated with hyperosmolar therapy for increased intracranial pressure but did not require decompressive craniectomy or hematoma evacuation. Preliminary hematologic work-up included prothrombin time 11.9 sec (ref 9.4-12.5 sec), activated partial thromboplastin time 29 (ref 25-37 sec), fibrinogen 202 mg/dL (ref 200-500 mg/dL), d-dimer 16,021 ng/mL (ref <500 ng/mL), and creatinine 0.76 mg/dL (ref 0.59-1.04 mg/dL). The patient was started on 18 units/kg/hr of intravenous heparin for treatment of the CVST. Approximately 14.5 hours after the intravenous heparin was initially administered, the patient’s platelet count was 32 × 10⁹/L without a clinically significant increase in the size of the hematoma. Due to concern for heparin induced thrombocytopenia, the patient was transitioned to bivalirudin and transfused with 2 units of platelets for a goal platelet count >50 × 10⁹/L. The patient’s serum was found to harbor IgG antibodies against platelet factor 4 (PF4)/polyanion complexes with OD 1.995 (ref <0.4000 OD). Serotonin release assay was also consistent with a diagnosis of HIT, with 99% and 100% of serotonin release with low dose unfractionated heparin (0.1 IU/mL and 0.5 IU/mL, respectively) and 42% release with high dose heparin (100 IU/mL).

She continued on bivalirudin infusion for 8 days, until her platelet count reached 181 × 10⁹/L, at which point she was transitioned to warfarin with an INR goal of 2-3 (Figure 1D). Her neurologic function continued to improve, and she transferred to inpatient rehabilitation on hospital day 19, where she stayed for an additional 13 days. She continued on warfarin for 6 months. Al3 months follow up, the patient had mild upper motor neuron pattern weakness in the left hemibody, with the triceps, wrist extensors, finger extensors, and iliopsoas muscles yielding only to maximal resistance. At that time, she was independent in her activities of daily living and ambulated with the assistance of a walker.

Discussion

As is typical in cases of delayed-onset HIT, our patient presented with mild thrombocytopenia that significantly and abruptly worsened after receiving therapeutic heparin. ⁵ Similar to our case report, the patient described in Kuitunen et al also received heparin after presenting with thrombosis, thereby unmasking and worsening the already initiated HIT. ² CVST in the context of delayed-onset HIT has been described in just 2 other case reports in the literature; one patient died from CVST and extensive intracranial hemorrhage, ² whereas no long-term outcome was reported for the second patient. ³ Therefore, we are the first to report a favorable clinical outcome in a patient with CVST and delayed-onset HIT.

The molecular underpinnings of this patient’s delayed-onset HIT are of particular interest. While unfractionated heparin is the most common antigen to cause HIT, our patient received only 3 post-operative doses of subcutaneous low molecular weight heparin (LMWH) over a twenty-four-hour period. A recent study demonstrated that the incidence of HIT in patients receiving subcutaneous unfractionated heparin was less than 0.1%, compared to 0.76% incidence upon IV heparin exposure. ⁶ Of these patients receiving subcutaneous heparin, average duration of therapy was 72 hours. ⁶ Moreover, LMWH has been shown to be significantly less immunogenic than unfractionated heparin. ⁷ Thus, it is highly unusual that this formulation and duration of heparin administration would lead to HIT. Further, based on the previously described studies, it seems unlikely, albeit not impossible, that this patient’s brief exposure to enoxaparin would be sufficient to result in HIT.

One intriguing possibility for our patient’s delayed-onset HIT is that another stimulus may have also primed the patient’s immune system and contributed to the development of HIT. In fact, there are instances in which patients develop HIT without heparin exposure, known as spontaneous HIT. ⁸ This subtype of HIT is thought to occur via molecular mimicry whereby an endogenous polyanionic antigen binds with PF4 to create neoepitopes resulting HIT. The prototypical example of spontaneous HIT is following knee arthroplasty and is thought to result from chondroitin sulfate in the knee cartilage binding to PF4.^8-10 While our patient does not meet the diagnostic criteria for spontaneous HIT due to her exposure to LMWH, this case is familiar to a recent case of spontaneous HIT with CVST from our institution. ⁸ In each case, both patients were post-operative day 11 status-post total knee arthroplasty and on twice daily aspirin for pharmacologic thromboprophylaxis. In addition, both patients presented with ICH secondary to hemorrhagic transformation of a CVST and were ultimately diagnosed with HIT. Of particular note, aspirin usage has been cited as a commonality in the clinical picture of spontaneous HIT,^8,9,11 thereby further highlighting the similarities of our patient to prior cases of spontaneous HIT. Despite not satisfying the criteria for spontaneous HIT, it is possible that our patient’s immune system may have been synergistically primed by her enoxaparin exposure and recent knee arthroplasty leading to the onset of HIT, which was further unmasked by the in-hospital heparin exposure.

An alternative explanation for our patient’s CVST is the usage of her HRT—known to be pro-coagulopathic—as the source of her thrombosis. In this explanation, the severe thrombocytopenia observed after a short duration of intravenous heparin unmasked an acute form of HIT from which she was clinically asymptomatic. While this may be a plausible explanation, we propose that delayed-onset HIT is the unifying diagnosis that explains her thrombosis as well as the mild thrombocytopenia that worsened immediately upon intravenous heparin and is therefore the most likely diagnosis.

we found that our patient’s signs and symptoms were related to CNS dysfunction and notably included headache and motor dysfunction. Our patient had a favorable outcome due to the close monitoring of her platelet levels, early identification of HIT, and fluid anticoagulation management. This report highlights the importance of non-heparin anticoagulant use, especially direct thrombin inhibitors (i.e. bivalirudin), to treat the underlying cause of thrombosis and thrombocytopenia despite the presence of acute intracranial blood products. Thus, early recognition of HIT aided by an eerily familiar clinical presentation to CVST resulting from spontaneous HIT ⁸ facilitated an excellent clinical outcome.

Conclusion

We have presented the third case of delayed-onset HIT with resulting CVST and are the first to report a favorable outcome. Our patient may represent an example of glycosaminoglycans exposure from prior knew surgery enhancing the antigenicity of enoxaparin to synergistically stimulate the immune system for HIT formation. When managing HIT with associated CVST, it is imperative to establish a fluid treatment plan to prevent patient morbidity and mortality. One crucial component of this treatment paradigm includes the use of non-heparin anticoagulants to address the underlying thrombosis and thrombocytopenia, despite the presence of ICH or acute blood products.