Background

Seventy-five percent of upper extremity deep vein thrombosis is provoked by central venous catheters (CVCs) or PICCs.¹ PICC line use for antibiotics, chemotherapy, total parenteral nutrition, and venous access has increased most likely because of the associated lower risk of insertion complications compared with CVCs. Serious mechanical complications are reported to occur in approximately 3% of patients undergoing CVC placement.² Chopra et al evaluated 966 PICC line placements and showed 33 symptomatic PICC line deep vein thromboses (DVTs).³ Bivariate analysis revealed the following factors are associated with PICC line thrombosis: recent diagnosis of cancer (last 6 months), interventional radiology placement, chemotherapy administration, number of lumens, and PICC line gauges. On the other hand, multivariate analysis identified recent cancer diagnosis and PICC line gauge with hazard ratios of 2.21 and 3.56, respectively. Chopra et al found that a cancer diagnosis in the past six months and catheter gauge were the strongest predictors of PICC-associated DVT.³ In addition 5 Fr and 6 Fr PICCs showed an earlier time to DVT, suggesting an accelerated course with large devices.

Management

The use of PICC lines in NeuroICUs has demonstrated an 8.4% cumulative incidence of symptomatic PICC line–related thrombosis, of which 15% were associated with pulmonary embolism.^4,5 With this background information, the treatment with full anticoagulation is appropriate in order to prevent pulmonary embolism and further propagation of the current thrombosis.

Patients with 5 Fr and 6 Fr catheter devices are not only at greater risk, but also develop thrombosis earlier compared with those with 4 Fr devices.^6,7 Dual lumen 4 Fr PICCs may offer the best option for venous access and therapies from a complication perspective.

A significant increase in the use of single lumen and the smaller 5 Fr triple lumen PICCs was associated with a significant decrease in PICC-associated DVT. PICC-associated DVT also increases the cost of hospitalization.

In patients with an upper extremity DVT that is associated with a central venous catheter that involves the axillary or more proximal veins and is functional, therapeutic anticoagulation with intravenous unfractionated heparin (IV UFH), low-molecular-weight heparin (LMWH), fondaparinux, or weight-based subcutaneous unfractionated heparin (SC UFH) should be initiated (Table 44-1).⁸ Anticoagulation therapy should be used for 3 months. If peri-catheter thrombosis involves the axillary or more proximal veins and the catheter is nonfunctional, the catheter should be removed and full anticoagulation maintained for 3 months. There are no adequately powered studies to address the timing of catheter removal.

Background

Stroke is the third leading cause of death in the Western world. Deep vein thrombosis (DVT) and pulmonary embolism (PE) are frequent complications of ischemic stroke.^9,10 There is a large variation in the incidence of DVT and PE among different clinical studies in ischemic stroke patients not receiving thromboprophylaxis. The range is 1% to 5.2% in DVT and 0% to 5.6% with PE.¹¹ PE is an important cause of mortality in patients after stroke; early studies indicated that PE accounted for up to one-quarter of premature deaths in the absence of prophylaxis.¹² In a more recent large registry study of 13,440 patients with ischemic stroke by Heuschmann et al, 0.4% of patients developed PE and nearly half (46.8%) of these patients died before hospital discharge.¹³

In this patient the treatment of the extensive deep vein thrombosis would be anticoagulation with IV unfractionated heparin, LMWH, fondaparinux, or subcutaneous weight-based unfractionated heparin. This approach would be followed by bridging to warfarin therapy with a target International Normalized Ratio (INR) between 2 and 3 for 3 months. Another option for management would be to use LMWH or fondaparinux as monotherapy for 3 months.

Unfractionated Heparin

Unfractionated Heparin (UFH) is dosed by a weight-based regimen with an initial bolus of 80 IU/kg administered intravenously followed by a continuous infusion of 18 IU/kg/h. Monitoring of the activated partial thromboplastin (aPTT) should be done every 6 hours for the first 24 hours until achievement of an aPTT level within the therapeutic range and then daily, with the goal of maintaining a therapeutic aPTT range. An alternative to IV UFH is the administration of subcutaneous weight-based UFH with an initial dose of 333 units/kg followed by maintenance doses of 250 units/kg every 12 hours. This regimen can be reserved for patients with poor venous access for aPTT blood draws. Warfarin is initiated concomitantly with the above UFH regimens and continued until the INR is between 2 to 3 for 2 consecutive days, at which time UFH is discontinued with a minimum overlap of 5 days with both anticoagulants. Platelet counts are monitored on days 3 and 5 to observe for the development of heparin-induced thrombocytopenia. It is imperative to achieve a therapeutic aPTT level within the first 24 hours in order to reduce the incidence of recurrent venous thromboembolism.

Low-Molecular-Weight Heparin and Fondaparinux

Multiple randomized studies support the notion that low-molecular-weight heparins (LMWHs) and fondaparinux are at least as safe and effective as IV UFH in preventing thrombus propagation and pulmonary embolism. LMWHs are administered as follows: dalteparin 200 IU/kg SC, daily; enoxaparin 1 mg/kg SC q12h or 1.5 mg/kg SC daily; and fondaparinux 5, 7.5, or 10 mg SC daily, based on the patient’s weight. Warfarin is initiated concomitantly with the above LMWH and fondaparinux regimens and continued until the INR is between 2 and 3 for 2 consecutive days, at which time the parenteral drug is discontinued with a minimum overlap of 5 days with both anticoagulants. In this case we preferred the shorter acting LMWHs as opposed to the longer acting fondaparinux (half-life of 17 to 21 h), which also lacks reversibility in the event of any central nervous system hemorrhage.

Over the past 4 years there have been a number of meta-analyses of randomized controlled trials that have confirmed the value of LMWHs compared with UFH for the treatment of DVT with or without PE in both the inpatient and outpatient settings.^14,15 The most recent of these included 17 studies in which UFH was given intravenously and three older studies in which UFH was given via the subcutaneous route.¹⁶ LMWH was associated with fewer thrombotic complications (3.6% vs 5.4%), less major bleeding (1.2% vs 2.0%), and fewer deaths (4.5% vs 6.0%). The mortality advantage with LMWH appeared to be confined to those with cancer. Therefore, these studies suggest that LMWHs are at least as safe and effective as IV UFH in the treatment of venous thromboembolism (VTE) and are expected to have a lower incidence of HIT. Similarly, fondaparinux is effective in the treatment of VTE and is not believed to carry a risk for HIT development.

Warfarin

Warfarin prevents the generation of the reduced form of vitamin K, which is a necessary cofactor for the carboxylation of glutamic acid residues for the gamma carboxylation of key coagulation proteins including factors II, VII, IX, and X and protein C and protein S. Liver disease is known to enhance warfarin’s anticoagulant effect, but renal disease does not increase the response to warfarin. After oral administration, warfarin is rapidly absorbed, reaches peak blood concentrations after 90 minutes, and has a plasma half-life of approximately 40 hours with a bioavailability of close to 100%. The recommended initial dose of warfarin is 5 mg within 24 hours of initiating UFH, LMWH, or fondaparinux. Two studies have shown that there is no difference in the mean time to reach a therapeutic INR between the 5- and 10-mg dosing schedules.¹⁷ Warfarin therapy should be continued for 3 months.

LMWH and Fondaparinux Monotherapy

Another option for this patient would be the use of LMWH or fondaparinux as monotherapy. The patient would not be started on warfarin, but maintained on the respective dose of LMWH or fondaparinux. Treatment would be 3 months as noted above in warfarin-treated patients. This approach does incur increased cost plus an injectable mode of administration, but there is no necessary laboratory monitoring or problematic food or drug interaction.

New Target-Specific Oral Anticoagulants

In this patient with an acute stroke, the new target-specific oral anticoagulants should not be used. These agents include the following that are approved by the Food and Drug Administration for the treatment of deep vein thrombosis and pulmonary embolism: apixaban, dabigatran, edoxaban, and rivaroxaban. In the atrial fibrillation trials, patients were excluded with acute stroke < 2 weeks old. We therefore would not place this patient on any of these agents for the treatment of deep vein thrombosis during the acute phase of treatment.

In the above case, IV UFH, LMWH, fondaparinux, or subcutaneous weight-based UFH are acceptable options, followed by bridging to warfarin therapy with a target INR between 2 and 3 for 3 months. Another option for management would be to use LMWH or fondaparinux as monotherapy for 3 months.

In this case the use of therapeutic anticoagulation for the treatment of pulmonary embolism and proximal deep vein thrombosis is contraindicated because of the high risk of bleeding in a patient 4 days after craniotomy and aneurysm clipping. The only option would be the placement of an inferior vena cava (IVC) filter to prevent recurrent pulmonary embolism, which may result in a high morbidity and mortality. Table 44-2 is the current recommended indications for the placement of IVC filters.¹⁷

Prior to the insertion procedure, any available cross-sectional imaging should be reviewed for IVC and access site anatomy, patency, and anomalies along with the location of the renal veins. The majority of filters are placed using fluoroscopic guidance in interventional suites. Depending on the design of the filter and diameter of the delivery sheath, filters can be inserted from the femoral, jugular, or antecubital veins. The usual target-landing zone for an IVC filter (IVCF) is the infrarenal IVC, close to the level of the renal veins. The diameter of the IVC in the target-landing zone is important, as each filter is rated for a maximum IVC diameter, above which the likelihood of embolization of the filter itself is increased. This information is provided in each device’s instructions for use.

In the PREPIC study 400 patients with proximal DVT with or without pulmonary embolism were randomly assigned to receive a permanent IVC filter or no filter.¹⁸ All patients received therapeutic anticoagulation with UFH or LMWH bridged to warfarin. After 12 days of therapy, the IVC filter group (1.1%) had a significant decrease in symptomatic and asymptomatic pulmonary embolism compared with anticoagulation (4.8%). When only symptomatic pulmonary embolism was considered the difference between the filter (1%) and no-filter group (3%) was not significant. At 2 years, symptomatic pulmonary embolism tended to be less frequent among filter recipients (3%) than those receiving anticoagulation alone (6%). IVC filters (21%) were associated with significantly more recurrent DVT than was observed with anticoagulation alone (12%). In a more recent trial, the use of a retrievable IVC filter in addition to anticoagulation did not reduce the risk of a PE compared with anticoagulation alone in patients diagnosed with PE.¹⁹ Table 44-3 list the complications associated with retrievable inferior vena cava filters.^20–22

In the initial treatment of VTE, vena cava filters may be considered in the case of a contraindication for anticoagulation or in the case of PE recurrence despite optimal anticoagulation. Periodic reassessment of contraindications for anticoagulation is recommended, and anticoagulation should be resumed when safe.