Antithrombin III Supplementation in a Premature Infant: Is There a Target Antithrombin Level?

Abstract

The optimal antithrombin (AT) activity for low-molecular-weight heparin efficacy and the benefits of antithrombin III (ATIII) supplementation in premature infants diagnosed with venous thromboembolism are unknown. Currently, there are no neonatal-specific guidelines directing the appropriate target AT activity during supplementation. This case report describes a critically ill premature infant with a progressive, occlusive inferior vena cava thrombus who received supplemental ATIII during enoxaparin treatment. The patient did not achieve therapeutic anti-Xa levels despite increasing enoxaparin dosing to 3 mg/kg every 12 hours. ATIII supplementation sufficient to attain an AT activity of >40%, in combination with an enoxaparin dosing of >2 mg/kg every 12 hours, was needed to achieve therapeutic anti-Xa levels. Future large studies are needed to determine if there is an optimal target AT activity for critically ill premature infants.

Background

Low-molecular-weight heparin (LMWH) is widely used for the treatment of venous thromboembolism (VTE) in neonates, despite lack of strong evidence to support its use.¹ LMWH binds to endogenous antithrombin (AT), accelerating AT inactivation of coagulation factors, particularly factor Xa.² Neonates require higher doses of enoxaparin to achieve therapeutic goal levels than older pediatric patients.^1,3–5 This increased dosing requirement may be secondary to the impaired action of heparin or the increased volume of distribution and clearance of heparin in neonates.³ Neonates have physiologically lower AT activity than adults. However, because neonates have lower levels of procoagulants and anticoagulants in general than adults, it is unclear whether lower AT activity alone accounts for the increased enoxaparin dosing needs.³

The optimal AT activity for LMWH efficacy and the benefits of antithrombin III (ATIII) supplementation in neonates are unknown. Currently, there are no neonatal-specific guidelines directing the appropriate target AT activity with supplementation. We report the case of a critically ill premature infant with an occlusive inferior vena cava (IVC) thrombus requiring supplemental ATIII with enoxaparin therapy after failure to achieve therapeutic anti-Xa levels with large doses of enoxaparin alone. We describe AT activities, anti-Xa levels, and doses of enoxaparin during supplementation.

Case

The neonate was born at 24 6/7 weeks' postmenstrual age, birth weight 730 g, due to maternal preterm labor. The neonate required intubation in the delivery room and received conventional ventilation. Umbilical venous and arterial catheters were placed. Both catheters were removed on day of life (DOL) 9. The neonate experienced multiple complications early in the hospitalization, including intestinal perforation and coagulase-negative Staphylococcus bacteremia (DOL 6), cutaneous mucormycosis (DOL 13), respiratory deterioration requiring high-frequency oscillatory ventilation (DOL 16), and necrotizing enterocolitis and Klebsiella pneumoniae urinary tract infection (DOL 41). Pediatric Hematology was consulted on DOL 56 for an occlusive IVC thrombus measuring 2.5 cm in length at the level of the liver, discovered on abdominal ultrasonography (US) ordered to evaluate bowel distension. Etiology for the thrombus was unclear. Possibilities included a chronic thrombus secondary to the umbilical venous catheter that was placed on DOL 1 and removed on DOL 9, or an acute thrombus secondary to the left saphenous peripherally inserted central catheter (PICC) that was present and functioning well at the time of the abdominal US. Head US was negative for intracranial hemorrhage. Protein C, free protein S, and AT activities were above or within normal reference ranges for gestational age (24%, 56%, and 24%, respectively; normal ranges, 8%–18%, 18%–40%, and 24%–55%, respectively).⁶

Enoxaparin was initiated on DOL 57 (Table). At our institution, the starting dose of enoxaparin for VTE treatment in infants <2 months of age is 1.5 mg/kg/dose subcutaneously every 12 hours. The actual dose given is determined by the pediatric hematologist after the risks and benefits of starting anticoagulation are weighed. Subsequent dose adjustments and timing of anti-Xa levels are also managed by Pediatric Hematology. Enoxaparin 100 mg/mL (Sanofi-Aventis LLC, Bridgewater, NJ) is used to prepare the doses and is measured by using 1-mL subcutaneous syringes. Syringes are demarcated in both 0.1-mL and 0.01-mL increments. Given the patient's low weight at the initiation of enoxaparin (1.3 kg) and the inability to measure volumes in between 0.01-mL increments, doses were restricted to whole milligrams (e.g., 1 mg = 0.01 mL, 2 mg = 0.02 mL). Enoxaparin 1 mg (0.8 mg/kg/dose, weight 1.3 kg) every 12 hours was initiated. A 50% reduction from 1.5 mg/kg every 12 hours was chosen because the infant was considered high risk for bleeding, and all doses were rounded to the nearest whole milligram. On DOL 60, anti-Xa levels were subtherapeutic (i.e., therapeutic anti-Xa level is defined as 0.5–1.0 units/mL). As there was no demonstrable bleed, the enoxaparin dose was increased. On DOL 62, an IVC US demonstrated progression of the thrombus to the left iliac vein. Owing to this clot progression and suspicion of an acute thrombus, the left saphenous PICC was removed. Enoxaparin was continued to treat the progressive, occlusive IVC thrombus. Despite increasing enoxaparin to 2.8 mg/kg every 12 hours, the anti-Xa level remained <0.2 units/mL (Table, DOL 68).

Table.

Enoxaparin and Antithrombin Dosing and Monitoring

Day of Life	Enoxaparin, mg	Enoxaparin, mg/kg/dose	Anti-Xa Level, units/mL§	AT Level, %	ATIII, units	ATIII, units/kg/dose
56	—	—	—	24	—	—
57	1	0.8	—	—	—	—
58	1	0.8	0.05	—	—	—
60	2	1.1	<0.05	—	—	—
61	2	1.1	0.09	—	—	—
62	3	1.7	0.20	—	—	—
66	4	2.3	0.23	—	—	—
68	5	2.8	0.16	—	—	—
70	5	2.8	0.27	18	45	24.7
71	5	2.8	0.31	28	59	32.4
72	5	2.8	0.44	32	59	32.4
73	5	2.8	0.44	32	—	—
75	6	3	0.43	23	—	—
78	6	3	0.27	26	59	29.2
79	6	3	0.37	—	45	22.3
80	6	3	0.46	30	45	22.3
81	6	3	0.05	32	60	29.7
82	6	3	0.54	43	60	29.7
83	6	3	0.61	47	60	29.7
84	6	3	—	—	60	29.7
85	6	3	—	—	60	29.7
86	6	3	0.72	51	60	29.7
87	6	3	—	—	60	29.7
88–92	6	2.8	—	—	60	27.8
93	6	2.8	0.73	61	—	—
98	6	2.4	0.44	37	—	—
100	6	2.4	0.34	—	60	23.6
101–106	6	2.4	—	—	60	23.6
107	6	2.4	0.57	49	—	—
109	6	2.4	0.44	—	—	—
152	8	2.5	0.67	63	—	—
159	8	2.4	0.98	—	—	—

AT, antithrombin; ATIII, antithrombin III

^* Dash indicates no value was available or dose given that day.

^† Shaded rows indicate days when anti-Xa levels were therapeutic and AT levels were >40%.

^‡ During DOL 113 to DOL 145, 7 of 8 anti-Xa levels were in therapeutic range. Enoxaparin was held on DOL 113–117 and DOL 133–134 for surgical procedures.

^§ Blood draws for anti-Xa levels were performed 3 to 5 hours after an enoxaparin dose.

Antithrombin III supplementation (Grifols Therapeutics LLC, Research Triangle Park, NC), managed by Pediatric Hematology, was initiated at 25 international units/kg intravenously on DOL 70 to 72 (Table). Currently, at our institution, there is no pediatric or neonatal guideline for ATIII dosing or targeted AT activity range in non–extracorporeal membrane oxygenated patients. The adult equation ([Desired AT Activity − Baseline AT Activity] × Patient's Weight in kg divided by 1.4) was used, targeting a desired AT activity of approximately 50%. On DOL 72, US demonstrated persistent occlusive thrombus in the IVC with a mild decrease in thrombus burden at the left common iliac vein. ATIII supplementation was held on DOL 73 to 77 to assess whether anti-Xa levels were dependent on ATIII supplementation. Anti-Xa levels increased on DOL 70 through DOL 73 but decreased to 0.27 units/mL on DOL 78. Supplementation was resumed on DOL 78. An attempt was made to calculate ATIII doses by using daily baseline AT activities to target a higher desired AT activity of approximately 75%. However, this proved to be impractical owing to frequent neonatal phlebotomy, which is discouraged at our institution to minimize blood loss. Additionally, the turnaround time for laboratory results led to inconsistent dose timing. As a result, starting on DOL 81, a fixed ATIII dose of 60 units was given. Therapeutic anti-Xa levels were achieved only after ATIII supplementation resulted in AT activity >40% (Table and Figure, DOL 82 to 93).

Figure.

Antithrombin activity (n=15) and corresponding Anti-Xa level.^*†

On DOL 86, US demonstrated improvement in the thrombus, which was described as partially nonocclusive in the IVC. Enoxaparin was held on DOL 95 to 96 for central venous catheter (CVC) placement and removal. An attempt was made to discontinue ATIII supplementation from DOL 93 to 99; however, the anti-Xa level decreased to subtherapeutic range when AT activity fell below 40% (Table and Figure, DOL 98). The anti-Xa level increased to therapeutic goal range when resumption of ATIII supplementation increased AT activity to > 40% (Table and Figure, DOL 107). On DOL 107, with an anti-Xa level of 0.57 units/mL and AT level of 49%, ATIII supplementation was held. On DOL 109, the anti-Xa level was subtherapeutic at 0.44 units/mL. Despite this subtherapeutic level, ATIII supplementation continued to be held to allow assessment of anti-Xa level trends without supplementation. The next anti-Xa level on DOL 113 was therapeutic (0.57 units/mL). Subsequent anti-Xa levels (DOL 119 to 159) were within therapeutic range without further ATIII supplementation, except for one subtherapeutic level on DOL 138 (0.43 units/mL). The last dose of ATIII was administered on DOL 106 (Table). Enoxaparin was held on DOL 113 to 117 (abdominal exploratory laparotomy with small-bowel resection), and DOL 133 to 134 (CVC removal). Enoxaparin was restarted when deemed appropriate by the surgical or neonatal intensive care team. On DOL 152, an AT activity (63%) was obtained in anticipation of discharge and family counseling. Serial head US was obtained during enoxaparin treatment to monitor for new intracranial hemorrhage. Platelets were transfused to maintain platelet counts >50,000/mcL during enoxaparin therapy, and >100,000/mcL for surgical procedures. There were no bleeding events noted during concomitant anticoagulation and ATIII supplementation. The infant was discharged on DOL 163 with planned outpatient follow-up.

Discussion

LMWH, particularly enoxaparin, is widely used for the treatment of VTE in neonates.^7–9 The recommended starting enoxaparin dose for VTE treatment is 1.5 mg/kg every 12 hours administered subcutaneously for infants younger than 2 months.¹ Mean maintenance dosing in full and premature infants can be greater than 1.5 mg/kg every 12 hours.³ Some studies suggest that the starting dose for infants should be increased to as high as 2 mg/kg every 12 hours³ in order to achieve therapeutic anti-factor Xa levels sooner.^4,10 Despite increases in the initial or maintenance dose, infants may continue to have difficulties achieving therapeutic anti-Xa levels.

While difficulties in achieving therapeutic anti-Xa levels during LMWH treatment may be multifactorial, potential contributing factors include diminished intrinsic AT levels and function in neonates.⁵ Like unfractionated heparin, LMWH exerts its anticoagulation effect by enhancing the inhibitory effect of AT.² A healthy term infant's AT activity is lower than that of adults.^11–16 Despite this lower activity, a well neonate's AT is functional, thereby preventing spontaneous thrombosis.¹¹ Prematurity and comorbidities such as respiratory distress syndrome, necrotizing enterocolitis, sepsis, or disseminated intravascular coagulation will decrease AT activity and functionality.^{11–15,17,18} Despite higher dosing of LMWH, the immature coagulation system of a critically ill premature infant may require ATIII supplementation, as this may increase enoxaparin efficacy to achieve therapeutic anti-Xa levels.^19,20

Antithrombin III Supplementation. A total of 25 ATIII doses ranging from 22 to 33 units/kg were administered to our patient (Table). There is conflicting data on the utility of ATIII supplementation in the pediatric population. Corder et al¹⁰ reported that exogenous ATIII supplementation during enoxaparin therapy in patients <1 year of age (mean gestational age 36 weeks; mean weight 3.8 kg) did not decrease the time needed to achieve therapeutic anti-Xa levels but increased the overall cost and the risk of bleeding events. Alternatively, Diaz et al²¹ reported that patients <18 years of age who received exogenous ATIII during heparin treatment achieved therapeutic anti-Xa levels more quickly than patients who did not receive ATIII. A recent retrospective chart review published by Logston et al²² supported ATIII supplementation in infants <1 year of age (mean gestational age 30.27 weeks) by finding a median increase in anti-Xa levels of 0.2 units/mL and a median increase of AT activity of 16.5% after ATIII supplementation.²²

The recently published American Society of Hematology pediatric VTE guidelines consider ATIII supplementation justifiable if there is thrombus progression despite adequate anticoagulation or in specific populations such as neonates.⁹ At our institution, we measure AT activity prior to initiation of anticoagulation only to determine if the patient has inherited AT deficiency. Antithrombin activity is repeated if there is difficulty achieving therapeutic anti-Xa levels or if progression of a thrombus occurs. For our patient, ATIII activity measurement was repeated after radiographic imaging suggested progression of the VTE with persistent subtherapeutic anti-Xa levels despite an enoxaparin dose of 2.8 mg/kg every 12 hours.

Currently, there are no neonatal-specific guidelines directing the appropriate target AT activity with supplementation. In adults, ATIII loading and maintenance doses are calculated by using body weight, baseline AT activity, and targeted AT activity of 80% to 120%.²³ Corder et al¹⁰ applied this calculation to patients <1 year of age, but noted that patients rarely reached the targeted AT activity range of 80% to 120%. Diaz et al²¹ and Logston et al²² used a weight-based strategy (50 units/kg rounded to the nearest vial size) in patients <18 years of age and patients <1 year of age, respectively. With this weight-based strategy, Diaz et al²¹ observed an increase of AT activity of 0.3% for each unit/kg of ATIII administered and approximately 50% decrease in activity at 18 hours after administration. This is in contrast to ATIII supplementation in asymptomatic adults with hereditary AT deficiency (increase AT activity 1.4% per unit/kg of AT concentration with 50% decrease in activity at 22 hours after administration).¹⁹ These findings suggest that ATIII supplementation pharmacokinetics in critically ill pediatric patients are altered as compared with asymptomatic adult patients. Using an adult-derived target range or weight-based formula to calculate ATIII doses may be inadequate in a critically ill premature infant. Rather, age-specific targeted ranges should be considered.

At our institution, ATIII supplementation and dosing are at the discretion of the pediatric hematologist, who considers bleeding risks, enoxaparin doses, persistence of subtherapeutic anti-Xa levels, and the limitations of frequent blood draws in a premature infant. For our patient, the initial ATIII dose of 25 units/kg was calculated by using a target AT activity of approximately 50%. The rationale for targeting a lower AT activity of 50%, instead of 80% to 120% as used in the study of Corder et al,¹⁰ was not documented in the medical records. Targeting AT activity at 50% resulted in a lower weight-based dose (25 units/kg compared with 50 units/kg in Diaz et al²¹ or Logston et al²²). One possible explanation for targeting a lower ATIII activity could be to minimize the risk of potential bleeding in the critically ill premature infant with thrombocytopenia and intestinal perforation, as the benefit of larger doses and the risks of adverse bleeding events are unclear.²¹

Our patient did not experience bleeding events. Corder et al¹⁰ reported higher bleeding events requiring holding enoxaparin doses in those with concomitant ATIII supplementation compared with those without supplementation (14.7% versus 3.8%).¹⁰ Alternatively, Logston et al²² reported no bleeding events for 17 patients treated with enoxaparin and ATIII supplementation.²² Large cohort studies are needed to evaluate this potential complication of ATIII supplementation.

Conclusions regarding the efficacy of ATIII supplementation on VTE treatment cannot be drawn from this single case study. Our patient experienced VTE progression on DOL 62, 6 days after initial diagnosis. This progression could be due to factors other than low ATIII activity, such as ongoing infection, a persistent indwelling PICC in the left lower extremity, subtherapeutic anti-Xa levels, and generally higher enoxaparin dose requirements for a critically ill premature infant compared with a well newborn.^1,3,4 Conversely, on DOL 86, the improvement in the thrombus could be the result of removal of the PICC or treated infection.

Anti-Xa Levels and AT Activities. Blood draws for 24 anti-Xa levels and 16 AT activities were performed during the infant's hospitalization. One anti-Xa level was excluded because draws for duplicate levels were performed on DOL 66. Sources of the remaining 33 anti-Xa levels included venous (27), arterial (3), and capillary (3) specimens. Twenty-four of 27 venous draws were via catheters. No AT activity was excluded; 14 were venous specimens while the infant had a catheter, and 2 were arterial specimens.

The initial anti-Xa level (Siemens BCS XP instrument, Siemens Healthcare, Germany; and STA-Liquid anti-Xa, Diagnostica Stago, NJ) is routinely drawn between the second and fifth enoxaparin doses. Measurements after the second dose are more likely to be obtained in patients who are at high risk of bleeding or for whom subtherapeutic dosing would be life-threatening. While not optimal, measurements obtained nearer to the fifth dose are reserved for patients whose venous access are extremely limited. Blood draws for anti-Xa levels are performed 3 to 5 hours after a dose is administered. Following American College of Chest Physicians guidelines, the therapeutic anti-Xa level is defined as 0.5 to 1.0 units/mL.¹ For neonatal or pediatric patients diagnosed with VTE, our institution measures AT activity (Siemens Berichrom, Siemens Healthcare) prior to initiation of anticoagulation, when anti-Xa levels are persistently subtherapeutic despite higher than typical heparin or LMWH doses, or if VTE progression occurs. Additional AT activity measurement is obtained at the discretion of Pediatric Hematology.

A limitation to this report is the challenge with obtaining frequent laboratory test results in a low-weight, critically ill premature infant. Though peripherally drawn venous specimens are favored for coagulation tests, premature infants have poor venous access. Therefore, samples for anti-Xa levels and AT activities are drawn from a heparinized umbilical venous catheter or CVC when available. When none of these routes are possible, arterial or capillary routes are used. Arterial draws are generally discouraged during anticoagulation. Specimens from capillary sites are interpreted with caution. Anti-Xa levels and ATIII activities are interpreted by the pediatric hematologist who considers confounding factors such as volume measurement inaccuracies, site accessed (venous, arterial, or capillary), or if drawn from a heparinized catheter. From the anti-Xa level trends and clinical situation, the pediatric hematologist determines if the levels should be repeated prior to dose adjustments. Owing to the retrospective nature of this case report, it is unclear which samples were truly obtained from a heparinized catheter or a peripheral venous draw. Additionally, it is unclear if blood drawn from various locations (venous, arterial, and capillary) impacts their comparability.

Enoxaparin Dose Measurement. Another limitation is ensuring accurate enoxaparin dose measurement when treating a low weight infant with LMWH. Enoxaparin 100 mg/mL was used to prepare this patient's doses. While 1-mL subcutaneous syringes demarcated in both 0.1-mL and 0.01-mL increments allow for measuring enoxaparin doses in 0.01-mL increments, precise measurements are not guaranteed, particularly when measuring the low-end doses such as 1 mg or 2 mg. Additionally, dose changes involving 0.01-mL volume are difficult to measure. A 1-mg dose change may not result in a true dose change. This can lead to difficulties in achieving targeted anti-Xa levels. For example, enoxaparin was initiated on DOL 57 at 1 mg (0.01 mL). On DOL 60, the dose was increased to the next syringe demarcation of 0.02 mL (2 mg). Despite this being a 100% dose increase, inaccurate measurements of 0.01 mL and 0.02 mL may have created a situation where no actual dose change occurred, resulting in no increase in anti-Xa levels. Dose variations can also occur at stable enoxaparin dosing. From DOL 68 to 73, the enoxaparin dose was maintained at 5 mg. ATIII supplementation was initiated on DOL 70 and administered on DOL 70, 71, and 72 for persistently subtherapeutic anti-Xa levels (Table). It is possible that dose variations that occurred when measuring 0.05 mL led to an increase in anti-Xa levels, rather than ATIII supplementation. However, the dose measurements would have to be consistently high to increase the anti-Xa levels from DOL 70 to 73. Similarly, from DOL 93 to 99, ATIII supplementation was held and the enoxaparin dose was maintained at 6 mg. Again, consistently low-dose measurements that result in a trend of decreasing anti-Xa level are unlikely. To minimize dosing inaccuracies, enoxaparin doses are measured by using the provided subcutaneous needle, delivered to the bedside nurse, and administered to the patient with the same needle, thereby limiting drug loss due to dead space. Despite this, it is important to recognize that variability in doses can occur, particularly in the lower milligram range, thereby impacting anti-Xa levels. Our institution has since changed its policies and now uses diluted enoxaparin (20 mg/mL) for low weight premature infants.

Conclusion

In summary, in this critically ill premature infant, ATIII supplementation sufficient to attain an AT activity of >40%, in combination with an enoxaparin dose of >2 mg/kg every 12 hours, was needed to achieve therapeutic anti-Xa levels. This report suggests that neonatal-specific target AT activity should be considered. Future large studies are needed to determine an optimal target AT activity for critically ill premature infants that would mitigate thrombus progression with minimal bleeding events and costs.

ABBREVIATIONS

AT

antithrombin

ATIII

antithrombin III

CVC

central venous catheter

DOL

day of life

IVC

inferior vena cava

LMWH

low-molecular-weight heparin

PICC

peripherally inserted central catheter

US

ultrasonography

VTE

venous thromboembolism