Ultrasound-guided femoral vein catheterization in neonates with cardiac disease*.

OBJECTIVE: To describe a novel technique for real-time, ultrasound-guided femoral vein catheterization in neonates with cardiac disease, and to compare it to a contemporaneous cohort of neonates undergoing femoral vein central venous line placement via landmark technique. DESIGN: Retrospective cohort study of data extracted from a quality improvement database. SETTING: Pediatric cardiac intensive care unit and cardiovascular operating room in pediatric tertiary hospital. PATIENTS: One hundred fifteen neonates (mean weight, 3.07 ± 0.41 kg) with cardiac disease who underwent femoral central venous line attempts from January 2009 to September 2011. MEASUREMENTS AND MAIN RESULTS: Study populations were similar in age, weight, and Risk Adjustment for Congenital Heart Surgery-1 category, but differed in intubation status (32% vs. 100%, ultrasound vs. landmark, p < .0001). Central venous line success rate was superior in the ultrasound group: 72 of the 76 (94.7%) vs. 31 of the 39 (79.5%), p = .02. Ultrasound group also had a superior first (75% vs. 30.8 %) and second attempt success rate (90.8% vs. 51.3%), p value for both < .0001. Inadvertent arterial puncture occurred less frequently in the ultrasound group: four of the 76 (5.3%) vs. nine of the 39 (23.1%), p = .01. There was a trend toward more venous thrombosis in the landmark group, 16 of the 39 (41%) vs. 18 of the 76 (23.7%), p = .08. Among all 115 subjects, there was a very strong association between greater than two central venous line attempts and the odds of being diagnosed with a deep venous thrombosis (odds ratio, 9.3; 95% confidence interval 3.5-24.8) and the odds of suffering an inadvertent femoral arterial puncture during the central venous line event (odds ratio, 8.8; 95% confidence interval 10.6-730). CONCLUSIONS: This novel long-axis real-time ultrasound technique facilitates placement of femoral vein central venous line in critically ill neonates with cardiac disease at a higher rate of success with fewer attempts and lower occurrence of complications when compared with the landmark technique.

The hemostatic history of a 15-month-old child implanted with a Berlin heart left ventricular assist device.

We documented the hemostatic changes associated with placement of a EXCOR Berlin Heart left ventricular assist device in a 15-month-old child before heart transplantation. The development of hypercoagulability was rapid, manifested first by a plasmatic and subsequently platelet-mediated increase in coagulation kinetics and strength that persisted for weeks. The patient had no thrombotic complications for 6 wk before transplant but required extraordinary blood product administration to achieve hemostasis secondary to aggressive, multimodal anticoagulation. In summary, when proscribing anesthetic and surgical management of patients with a Berlin Heart, consideration of hypercoagulable features and anticoagulant therapy must be made to maximize patient safety.

Contact activation prolongs clot lysis time in human plasma: role of thrombin-activatable fibrinolysis inhibitor and Factor XIII.

BACKGROUND: Contact activation system proteins (e.g., Factor XII, kallikrein) have been implicated as direct or indirect activators of plasminogen. However, contact activation and Factor XI have enhanced thrombin-activatable fibrinolysis inhibitor (TAFI) activation and decreased fibrinolysis, and Factor XIII (FXIII) also delays fibrinolysis via alpha(2)-anti-plasmin deposition on fibrin polymers. Thus, the goals of this study were to define how fibrinolysis is modulated in human plasma by contact or tissue factor (TF) activation, and what role TAFI and FXIII plays in this system. METHODS: Normal, TAFI-deficient and TAFI-deficient/FXIII-supplemented plasma was exposed to tissue-type plasminogen activator and activated with either celite or TF. Clot growth/disintegration kinetics were documented with thrombelastography. RESULTS: Normal plasma activated with celite had significantly prolonged onset and duration of clot lysis compared with samples activated with TF. TAFI-deficient plasma activated with celite was noted to have a duration of clot lysis not different from samples activated with TF, but a significant difference in time to onset of lysis persisted. Celite activation of TAFI-deficient/FXIII-supplemented plasma showed significantly prolonged onset and duration of clot lysis compared with samples activated with TF. CONCLUSIONS: Primarily TAFI, and to a lesser extent FXIII, contributed to contact system protein-mediated attenuation of fibrinolysis. Clinical investigation of these phenomena is warranted in clinical settings involving contact activation (e.g., intra-aortic balloon pumps and ventricular assist devices) to determine whether these devices modulate fibrinolysis and perhaps contribute to thromboembolic morbidity.

Argatroban, bivalirudin, and lepirudin do not decrease clot propagation and strength as effectively as heparin-activated antithrombin in vitro.

BACKGROUND: Heparin-induced thrombocytopenia is a potentially limb- and life-threatening response to heparin exposure. Direct thrombin inhibitors (DTIs) have been reported to provide anti-coagulation for cardiopulmonary bypass; however, clot formation within the cardiopulmonary bypass circuit has been reported after the administration of DTIs. We present a case of thrombosis of the cardiopulmonary bypass circuit and, ultimately, death after argatroban administration. An in vitro thrombelastographic assessment of the effects of DTIs on clot kinetics was consequently performed to determine potential causes for this complication. METHODS: Normal human plasma was unmodified or exposed to heparin (1, 2, 3 U/ml), argatroban (5, 10, 50 microg/ml), bivalirudin (12, 20, 120 microg/ml), or lepirudin (3, 6, 10 microg/ml) before activation with tissue factor/kaolin in a thrombelastograph. Clot initiation (R, reaction time), propagation (MTG, maximum thrombus generation), and strength (MG, maximum elastic modulus) were determined. Analysis of variance was performed, with p < 0.05 considered significant. RESULTS: Compared with unmodified plasma, heparin significantly prolonged R and essentially reduced MTG and MG to the limits of detection in an activity-dependent fashion. In general, the DTIs tested prolonged R in a concentration-dependent fashion but did not diminish MTG or MG nearly as well as heparin. The only exception was 10 microg/ml lepirudin, which eliminated coagulation. CONCLUSIONS: DTIs demonstrated a significant prolongation of clot initiation but poor attenuation of propagation and strength. Further in vitro and clinical investigations to design a heparin-equivalent regimen to provide anti-coagulation for patients with heparin-induced thrombocytopenia are indicated.

The effect of dilution on plasma coagulation kinetics determined by thrombelastography is dependent on antithrombin activity and mode of activation.

Hemodilution-associated hypercoagulability has been the focus of several investigations because significant morbidity and mortality have been associated with perioperative thrombophilia. Because most investigations implicate imbalances in procoagulant/anticoagulant activity as the etiology of hemodilution-associated hypercoagulability, we determined the effects of dilution on coagulation kinetics and clot strength with thrombelastography (TEG(R)). Control plasma (+/-celite activation) and antithrombin (AT)-deficient (<10% activity) plasma were diluted 0%, 10%, 20%, and 30% with saline. TEG(R) variables measured included time to clot initiation (reaction time, R), speed of clot propagation (angle, alpha), and clot strength (amplitude, A; or shear elastic modulus, G). Dilution of control plasma (10%-30%) resulted in a significant (P < 0.05) 16% decrease in R values, no change in alpha values, and decrease in A and G values. AT-deficient plasma had significantly smaller R values compared with control, and dilution did not change R values in AT-deficient plasma. Celite activation eliminated dilution-associated changes in R values in control plasma but resulted in linear decreases (R(2) = 0.88-0.96, P < 0.0001) in alpha, A, and G in response to dilution. Thus, our data indirectly support the concept that decreases in AT activity cause dilution-mediated hypercoagulability in plasma. Finally, celite activation permits quantification of dilution with TEG.

The impact of factor XIII on coagulation kinetics and clot strength determined by thrombelastography.

Fibrinogen has been shown to be responsible for most protein-mediated clot strength via thrombelastography. However, factor XIII (FXIII) activity also plays a prominent role in the development of clot strength. Thus, we hypothesized that changes in FXIII activity would significantly increase clot strength. FXIII (0%, 1%, 6.25%, 12.5%, 25%, 50%, and 100% normal activity) was placed in a fixed volume of citrated FXIII-deficient plasma with 1% tissue factor and calcium chloride and underwent thrombelastography for 10 min. We measured the variables reaction time (R; a measurement of clot initiation), alpha (a measure of the rate of clot formation), amplitude (A; a measure of clot strength), and shear elastic modulus (G; a measure of clot strength). FXIII activity significantly decreased R in a pattern of exponential decay (R2 = 0.77; P < 0.001). FXIII activity significantly increased alpha, following a sigmoidal pattern (R2 = 0.88; P < 0.001). Finally, increases in FXIII activity significantly increased A and G in a sigmoidal pattern (R = 0.89; P < 0.001). We concluded that FXHI significantly affects R, alpha, A, and G. Thus, transfusion decision making with protein-mediated thrombelastographic patterns must account for the contribution of both fibrinogen and FXIII.