Short-Term Use of Umbilical Artery Catheters May Not Be Associated With Increased Risk for Thrombosis
* Department of Pediatrics, Division of Neonatal-Perinatal Medicine, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania
Department of Pediatrics, Division of Neonatal-Perinatal Medicine, Christiana Care Health Services, Newark, Delaware
Department of Radiology, Christiana Care Health Services
|| Cardeza Foundation for Hematologic Research, Division of Hematology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective. Umbilical arterial catheters (UACs) have rare but serious complications related to thrombus formation. Two specific serum markers of thrombogenesis—prothrombin fragment (F1.2) and thrombin-antithrombin (TAT)—can be assayed and correlated with abdominal ultrasound visualization of UAC thrombosis. Levels of these markers of thrombogenesis have not been studied in infants with UACs. The objective of this study was to determine F1.2 and TAT levels longitudinally and compare the levels with platelet counts and ultrasound evidence of thrombi during the first week of life in infants with UACs.
Methods. This study was conducted as a prospective, nonblinded, observational study performed between June 2001 and January 2002 at Christiana Care Hospital, a level III neonatal intensive care unit. Infants with a UAC in place in the first 24 hours of life were studied. All received equal amounts of heparin in the UAC. F1.2, TAT, platelet counts, and abdominal aorta ultrasounds were examined every other day starting within 24 hours of life. Studies were not done when the UAC was removed within the 5-day study period. Enzyme-linked immunosorbent assay for TAT and F1.2 was performed using a commercially available kit from Enzyngost. Data were analyzed with repeated measures analysis of variance evaluating TAT, F1.2, and platelet count over time.
Results. Thirty-three patients were investigated (mean ± standard deviation; gestational age: 27.4 ± 3.5 weeks; birth weight: 1139 ± 729 g). A total of 66 measurements of TAT, F1.2, and platelet counts were obtained. Sixty-one abdominal ultrasounds were performed; only 1 study was positive for UAC thrombus. There was no significant difference between F1.2 and TAT over time during the study period. Platelet counts seemed to fall over the 5-day study period, although this decrease did not reach statistical significance.
Conclusion. Indwelling UACs in sick infants may not carry an increased risk of thrombosis during the first 5 days of use.
Key Words: catheters • umbilical • thrombosis • F1.2 • TAT
Abbreviations: NICU, neonatal intensive care unit • UAC, umbilical arterial catheter • F1.2, fibrinogen 1.2 • TAT, thrombin-antithrombin • RDS, respiratory distress syndrome • UVC, umbilical vein catheter • IVH, intraventricular hemorrhage • PRBC, packed red blood cell • SNAP, score for neonatal acute physiology
Neonatal thrombosis is a significant source of morbidity and mortality in neonatal intensive care units (NICUs).1–7 There is likely a 21% mortality rate for all patients with arterial thromboses.8
Widespread use of umbilical arterial catheters (UACs) is a common risk factor for neonatal thrombosis, because the presence of a UAC may induce local thrombus formation. The mechanism by which umbilical catheters induce thrombosis is poorly understood. It is thought that UACs mechanically damage the vascular endothelium, thereby exposing the underlying basement membrane and collagen matrix, which triggers a thrombogenic response. Neal et al, in 1971,9 using aortography immediately after the UAC was removed, showed that virtually all UACs acquire a fibrin sheath at some point. Documentation of endothelial disruption by a UAC within 24 hours of placement has also been proved in animal models. Chidi et al10 demonstrated a consistent relationship between degree of intimal injury and duration of catheterization in rabbits. The presence of the catheter, in situ, was enough to start the process of local fibrin formation.
During activation of plasmatic coagulation, 2 specific markers that closely mirror thrombin generation are produced, fibrinogen 1.2 (F1.2) and thrombin-antithrombin (TAT).11,12 These markers can be assayed by enzyme-linked immunosorbent assay. The key event in the activation of plasmatic coagulation is the conversion of prothrombin to thrombin, because only thrombin can convert fibrinogen to fibrin in vivo.11–14 As prothrombin is converted to thrombin, F1.2 is generated.12–14 F1.2 has some diagnostic potential for assessing thrombotic risk because it is a biomarker of thrombin generation during blood coagulation.12,13,15
The complex TAT is generated when thrombin and its natural inhibitor, antithrombin III, combine. Like F1.2, TAT can be a sensitive parameter for specific detection of a latent activation of the clotting pathway.11
In the early 1990s, assays to quantify plasma levels of these sensitive products of the coagulation pathway were developed.11,12 Before this time, the only laboratory tests that could indicate an active coagulation system were those that measured prolonged clotting times, depletion of platelets and coagulation factors, and elevated levels of fibrin derivatives.16,17 These methods are indirect, nonspecific, and insensitive.12
F1.2 and TAT values have been previously used in adults, to guide anticoagulation therapy for deep vein thrombosis. In infants, most studies have concentrated on determining a link between intravascular and extravascular generation of thrombin in neonatal respiratory distress syndrome (RDS).18,19 Investigators have found that although sick neonates are able to generate thrombin as well as healthy age-matched neonates, their ability to inhibit thrombin is significantly decreased, placing the sick neonate at a unique risk for thrombotic complications in the immediate postnatal period. Decreased perfusion of other organs occurs with systemic impaired hemostasis and subsequent formation of microthrombi, after activation of the clotting, fibrinolytic, and kinin-kallikrein systems.20
Direct visualization of thrombi using real-time ultrasound has been shown to detect accurately abnormal intravascular echoes and lack of pulsation when thrombi are present.21,22 Although aortography is the gold standard, its invasiveness, nonportability, and expense discourage its routine use. Doppler ultrasound has emerged as a popular means of diagnosing clot because of it ease of use, safety, and portability. Its reliability in localizing UACs has been tested and found to be uniquely suited for studying in vivo evolution of echogenic intravascular thrombi.21,22 The purpose of this study was to observe the longitudinal production of F1.2 and TAT in infants who had indwelling UACs and to determine whether thrombus formation occurred within or around the UACs over time.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The study design was a prospective, nonblinded, observational study that was performed between June 2001 and January 2002 at Christiana Care Hospital, a level III NICU. The institutional review board approved the study. All infants who were admitted to the NICU with a UAC were eligible for enrollment. Infants with major congenital anomalies were excluded from eligibility. Patients were enrolled after signed informed parental consent.
Placement of a UAC was made at the discretion of the attending neonatologist, on the basis of the infant’s clinical status. UACs were placed under sterile preparations with the tip located between the second and fourth lumbar vertebrae. The position of all UACs was confirmed radiographically. Polyvinyl, single-lumen 3.5-French Argyle catheters with end holes were used in this study. All UACs were infused with 0.2 normal saline with 1 unit of unfractionated heparin sulfate per milliliter at 1.0 mL/hour, as per standard NICU protocol. No other intravenous fluids, medications, or blood products were infused through the UAC. Blood for clinical labs was withdrawn from the UAC on an as-needed basis by the staff nurse assigned to the infant.
Serial color flow and spectral analysis Doppler ultrasound of the aorta was performed on each patient. The first ultrasound was performed during the first 24 hours after insertion of the UAC (time 1). The second and third ultrasounds were performed during hours 48 to 72 (time 2) and hours 96 to 124 (time 3) of insertion, respectively. If the UAC was removed earlier than time 2 or time 3 of the study, then the ultrasound was obtained before removal.
Color flow and spectral analysis Doppler ultrasound examinations were performed at the bedside. The ultrasound studies were performed by a certified vascular ultrasound technician and were recorded for later review by a pediatric radiologist (M.F.). The abdominal aorta and iliac arteries were visualized, and flow was confirmed. The location of the tip of the UAC was confirmed in relation to the renal vessels and analyzed for the development of thrombus. Size and extent of thrombi were recorded as well as flow within the distal aorta.
Simultaneous with each ultrasound study, blood samples were drawn for F1.2 and TAT assays. When the patient also had an umbilical vein catheter (UVC), the blood samples were drawn from the UVC. When there was no UVC, samples were drawn from the UAC. Before the blood samples were obtained, 2 mL of drawback was taken off to ensure that no heparin was contained in the sample. One milliliter of blood was removed at each sampling. Venous blood was preferred over arterial blood because the original assays for F1.2 and TAT were developed using venous samples.11,12 Nine parts of blood was mixed with 1 part of citrate buffer solution or sodium citrate solution 0.11 mmol/L. Samples were centrifuged at 1500 x g for 10 minutes, platelet-poor plasma was removed, and the samples were stored at –70 degrees until assayed. Determination of F1.2 and TAT was done using monoclonal antibody-based enzyme immunoassay (Enzyngost, Marburg, Germany). These assays were performed at the Cardeza Hematology Foundation of Thomas Jefferson University.
The following additional information was extracted from the patient’s medical record: birth date, gestational age, birth weight, gender, platelet count, occurrence and grade of intraventricular hemorrhage (IVH), occurrence and number of packed red blood cell (PRBC) and/or platelet transfusions received during the study period, and Apgar scores. A score for neonatal acute physiology (SNAP), an objective index of illness severity, was also documented for each patient and was performed by a single observer (K.L.).23,24
Data on F1.2 and TAT and platelet levels among the 3 time periods were analyzed using repeated measures analysis of variance. Mann-Whitney U test was used for comparing means between individual sample groups. Multiple linear regression was used to determine correlation between birth weight and gestational age against F1.2 and TAT values.
A power analysis was performed for the repeated measures analysis of variance design. To detect a 20% difference in TAT and F1.2 levels during the study period, with mean values as reported by Brus et al20 and Bruhn et al,25 with 80% power and the .05 level of significance, 25 patients needed to be enrolled.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A total of 33 infants were enrolled in this study. All infants had baseline (time 1) F1.2 and TAT levels done, whereas 31 of 33 infants had a baseline (time 1) ultrasound done. One infant died before the first ultrasound could be performed. The other infant’s ultrasound could not be included because it was an inadequate study as a result of excessive bowel gas preventing adequate visualization of the UAC and surrounding vessels. Twenty of the 33 infants completed 2 sets of F1.2, TAT and ultrasounds. Thirteen of the 33 infants completed 3 sets of F1.2 and TAT levels, 11 of whom had ultrasounds that were included in the study. One of the ultrasounds was not included because the technician erroneously imaged the umbilical vein, and the other ultrasound was never performed because of the parent’s withdrawal of consent.
Table 1 illustrates the pertinent information on the study patients. The mean birth weight of the study group was 1133 g, and mean gestational age was 27.4 weeks. The male-to-female ratio was 1.3:1 (see Table 1). Table 2 outlines the incidence of IVH and transfusions in the patient population (see Table 2).
|
|
Figures 1 to 3 illustrate F1.2 levels, TAT levels, and platelet counts during the study period. We examined the time course of F1.2, TAT, and platelet counts under the hypothesis that formation of a fibrin sheath on the UAC and/or thrombus formation would contribute to their generation. There was no significant difference between F1.2 (P = .26) and TAT (P = .99) over time during the study period. Platelet counts seemed to fall over the 5-day study period, although this decrease did not reach statistical significance (P = .08). There was no statistically significant correlation between birth weight and gestational age on F1.2 and TAT (multiple linear regression).
|
|
|
Only 1 thrombus and 1 instance of decreased aortic flow was detected among a total of 61 ultrasounds. Both occurred in the same patient. This patient’s F1.2 level was 5.5 nM/L, a value higher than the mean in the study population. The TAT level was 5.2 µg/L, a value consistent with the mean in the study population. The platelet count at time 1 was 141 000. The patient did not have any evidence of IVH before death but died within 48 hours of life.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The clinical diagnosis of neonatal thrombosis is made infrequently in the NICU setting.1,3–5 The medical literature contains many anecdotes describing various complications from UAC usage, particularly thrombi- and thromboembolic-related incidents, yet there remains a paucity of literature regarding neonatal thrombosis detection, diagnosis, and management.2–7 In 1995, Schmidt and Andrew8 published the results of a prospective Canadian registry (over a 3.5-year time period, 25 Canadian and 58 international institutions participated voluntarily). They reported 97 cases of neonatal thrombosis, 33 cases of arterial thrombosis, 12 aortic thrombosis, and 16 iliac and femoral artery thrombosis, yielding an overall incidence rate of
1%. In other retrospective studies, the incidence of UAC-related thrombosis is reported to be as low as 1% and as high as 60%. This variability can be partially explained by 2 variables: varying methods used for diagnosis and the timing of the study. Whether Doppler sonography, aortography, or postmortem examination is used to diagnose thrombosis affects reported incidence, as well as retrospective versus prospective analyses.8,21,26,27
Some investigators have demonstrated clot formation only after catheters have been removed, as a result of fibrin sheath formation, which forms a cast that remains in the vessel even after its removal. Boo et al28 documented a UAC thrombosis incidence of 32% when they imaged the distal aorta by ultrasound after the catheter was removed.
In our study, we did not record any symptoms that could be related to vessel thrombosis. It is important to note that during the study, the only UAC that was withdrawn as a result of discoloration of a lower extremity occurred in the infant with the confirmed thrombus. Although infants with signs of catheter-associated vascular complications have a high incidence of detectable thrombi on ultrasound, a significant number may have asymptomatic complications. Despite the likelihood that most thromboses will naturally regress over time, the long-term sequelae of either symptomatic or asymptomatic vessel obstruction are yet unknown.5–7,22
Normative values for F1.2 and TAT in either healthy or sick term or preterm newborns are not known. They also are not used as definitive markers of thrombosis in infants. Healthy adult levels of F1.2 are reported as 0.37 to 1.2 nM/L (2.5–97.5 percentile). The 95% tolerance interval for healthy adults <44 years of age is between 0.21 and 2.38 nmol/L (P = .95) and tends to increase with age after 44 years.12,15
Healthy adult values of TAT are <4.1 µg/L.14 Schmidt et al29 in a study of TAT levels and increasing severity of RDS published TAT values of 2 to 10 µg/L in infants with mild to moderate RDS.
In healthy premature newborns between 30 and 36 weeks, there is an effect of gestational age as compared with healthy term newborns for certain components of the coagulation system. In healthy neonates, thrombin expression is controlled such that neither hemorrhagic nor thrombotic complications occur.19,30 In sick neonates, the coagulation system may influence overall generation of thrombin.19,29 Our observational study supports this theory because of elevation of F1.2 and TAT compared with adults (mean ± standard deviation; F1.2: 3.5 ± 1.8 nM/L; TAT: 5.2 ± 4.1 µg/L), providing direct evidence that increased amounts of thrombin are generated. Although there was no increase in F1.2 or TAT over time, there was also lack of decline, which could reflect ongoing thrombin formation. Our results also indicated that there was no correlation between gestational age or birth weight and F1.2 and TAT levels. This finding suggests that there is no effect of gestational age for these 2 components of the coagulation system. Although this correlation was not shown, the infants in this study were ill enough to require arterial blood gas monitoring and had a mean SNAP score of 17. Richardson et al23,24 published birth weight-adjusted mortality values of 30% to 60% if SNAP scores were >19. These data raise the question of whether healthy preterm infants may have different values of F1.2 and TAT.
In our study, our mean F1.2 value was an average of 3 times higher than values reported for adults. Our mean TAT value was 25% higher than normal adult values. The percentage of TAT and/or F1.2 elevation that would clinically represent an increased likelihood of thrombosis is unknown. Larger, prospective studies need to be performed to compare levels of F1.2 and TAT in healthy versus sick premature infants and newborns and correlate them with risk of thrombosis.
The most significant risk factor to inducing local fibrin formation is the duration of catheter use.28 The probability of developing aortic thrombus in situ with a UAC in place has been reported as 16% within 1 day, 32% within 7 days, 56% within 14 days, and 80% within 21 days.28 Heparinization of the UAC fluids does not contribute to inhibition of clot formation. Four published studies all failed to document that low-dose heparin prevents thrombosis while fluids are infusing through a UAC.18,31,32 Heparin does not seem to alter levels of TAT.18 Thirty percent of the infants studied received an average of 1.5 PRBC transfusions during the course of the study. PRBCs contain small percentages of plasma proteins (10%–20%), which are slightly activated by the processing procedure.20,33 We believe that the small amounts of PRBC transfusions received by the infants did not affect levels of TAT. Head ultrasounds to evaluate IVH were routinely done on day of life 3 during the study period. Twelve percent of infants had IVH of grade 3 or higher. A direct relationship between IVH and TAT and F1.2 levels has not been studied. It is possible that infants with higher grades of IVH are predisposed to even more severe derangements of thrombin inhibition.
Our results are different compared with other reported incidences of UAC thrombosis in the literature. We would have expected to see an 25% incidence of thrombosis; instead, there was only 1 positive study in 61 studies. A major clinical impact on such low rates of UAC thrombus may be the short duration of indwelling UACs in our NICU. Of 33 infants enrolled in this study, only 13 had their UAC in for up to 5 days. In our NICU, only polyvinyl, single-lumen 3.5-French Argyle catheters with end holes are used as UACs. This type of catheter may also be less thrombogenic than earlier generations of catheters that were made from Teflon.34 We also do not infuse any fluids (other than 0.2 normal saline with a small amount of heparin), medications, or blood products through the UAC. It has been speculated that infusions that contain calcium or hyperosmolar solutions contribute significantly to UAC occlusion by thrombi.
Because the incidence of clot formation was so rare in our study, we would not expect to see any difference in the serum markers of thrombus formation over time. We acknowledge that our small sample size did not achieve adequate power and affects interpretation of our results. Although this limits us from ruling out any clinically important rate of UAC-associated thrombosis in sick infants, we did not find increased thrombosis formation in patients with indwelling UACs during the first 5 days of life.
|
FOOTNOTES |
|---|
Received for publication Jul 15, 2002; Accepted Aug 11, 2003.
Reprint requests to (M.M.C.) Neonatal Associates of Jacksonville, PA, 4205 Belfort Rd #4090, Jacksonville, FL 32216. E-mail: rilkean648{at}aol.com
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Colburn M, Gelabert H, Quinones-Baldrich W. Neonatal aortic thromboses. Surgery.1992; 111 :21 –28[Web of Science][Medline]
- Roy M, Schmidt B. Neonatal thrombosis: are we doing the right studies? Semin Thromb Hemost.1995; 21 :313 –316[Web of Science][Medline]
- Schmidt B. The etiology, diagnosis and treatment of thrombotic disorders in newborn infants: a call for international and multi-institutional studies. Semin Perinatol.1997; 21 :86 –89[CrossRef][Web of Science][Medline]
- Schmidt B, Andrew M. Neonatal thrombosis: report of a prospective Canadian and international registry.
Pediatrics.1995; 96
:939
–943
[Abstract/Free Full Text] - Schmidt B, Andrew M. Neonatal thrombotic disease: prevention, diagnosis, and treatment. J Pediatr.1988; 113 :407 –410[CrossRef][Medline]
- Smith P, Miller D, Lail S, Mehta A. Urokinase treatment of neonatal aorto-iliac thrombosis caused by umbilical artery catheterization: a case report. J Vasc Surg.1991; 14 :684 –687[CrossRef][Web of Science][Medline]
- Vailas G, Brouillette R, Scott J, et al. Neonatal aortic thrombosis: recent experience. J Pediatr.1986; 109 :101 –107[CrossRef][Web of Science][Medline]
- Schmidt B, Andrew M. Report of scientific and standardization subcommittee on neonatal hemostasis. Diagnosis and treatment of neonatal thrombosis. Thromb Haemost.1992; 67 :381 –382[Web of Science][Medline]
- Neal W, Reynolds J, Jarvis C, Williams H. Umbilical artery catheterization: demonstration of arterial thrombosis by aortography. Pediatrics.1971; 50 :6 –13[Web of Science]
- Chidi C, King D, Boles T. An ultrastructural study of the intimal injury induced by an indwelling umbilical artery catheter. J Pediatr Surg.1983; 18 :109 –115[Web of Science][Medline]
- Pelzer H, Schwarz A, Heimberger N. Determination of human thrombin-antithrombin III complex in plasma with an enzyme-linked immunoabsorbent assay. Thromb Haemost.1988; 59 :101 –106[Web of Science][Medline]
- Pelzer H, Schwarz A, Stuber W. Determination of human prothrombin activation fragment 1+2 in plasma with an antibody against a synthetic peptide. Thromb Haemost.1991; 65 :153 –159[Web of Science][Medline]
- Bauer KA, Rosenburg R. The pathophysiology of the prethrombotic state in humans: insights gained from studies using markers of hemostatic system activation.
Blood.1987; 70
:343
–350
[Abstract/Free Full Text] - Dati F, Pelzer H, Wagner C. Relevance of markers of hemostasis activation in obstetrics/gynecology and pediatrics. Semin Thromb Hemost.1988; 24 :443 –448[CrossRef]
- Hursting M, Stead A, Crout F, Horvath B, Moore B. Effects of age, race, sex, and smoking on prothrombin fragment 1.2 in a healthy population.
Clin Chem.1993; 39
:683
–686
[Abstract/Free Full Text] - Schmidt B, Vegh P, Johnston M, Andrew M, Weitz J. Do coagulation screening tests detect increased generation of thrombin and plasmin in sick newborn infants? Thromb Haemost.1993; 69 :418 –421[Web of Science][Medline]
- Yurdakok M, Yigit S, Gurakan B, Dundar S, Kirazli S. Thrombin-antithrombin and prothrombin fragment 1.2 levels in early respiratory distress syndrome. Am J Hematol.1996; 51 :247 –248[CrossRef][Web of Science][Medline]
- Rajani K, Goetzman B, Wennberg R, Turner E, Abildgaard C. Effect of heparinization of fluids through an umbilical artery catheter in catheter patency and frequency of complications.
Pediatrics.1979; 63
:552
–556
[Abstract/Free Full Text] - Shah J, Mitchell L, Paes B, et al. Thrombin inhibition is impaired in plasma of sick neonates. Pediatr Res.1992; 31 :391 –395[Web of Science][Medline]
- Brus F, van Oeveren W, Okken A, Oetomo S. Activation of the plasma clotting, fibrinolytic, and kinin-kallikrein system in preterm infants with severe idiopathic respiratory distress syndrome. Pediatr Res.1994; 36 :647 –653[Web of Science][Medline]
- Oppenheimer D, Carroll B. Ultrasonic detection of complications following umbilical arterial catheterization in the neonate.
Radiology.1982; 145
:667
–672
[Abstract/Free Full Text] - Oppenheimer D, Carroll B. Ultrasonic localization of neonatal umbilical catheters.
Radiology.1982; 142
:781
–782
[Free Full Text] - Richardson D, Gray J, McCormick M, Workman K, Goldmann D. Score for neonatal acute physiology: a physiologic severity index for neonatal intensive care.
Pediatrics.1993; 91
:617
–623
[Abstract/Free Full Text] - Richardson D, Tarnow-Mordi W, Escobar G. Neonatal risk scoring systems: can they predict mortality and morbidity? Clin Perinatol.1998; 25 :591 –610[Web of Science][Medline]
- Bruhn H, Conard J, Mannucci M, et al. Multicentric evaluation of a new assay for prothrombin fragment F1+2 determination. Thromb Haemost.1992; 68 :413 –417[Web of Science][Medline]
- O’Neill J, Neblett W, Born M. Management of major thromboembolic complications of umbilical artery catheters. J Pediatr Surg.1981; 16 :972 –978[CrossRef][Web of Science][Medline]
- Marsh J, King W, Barrett C, Fonkalsrud E. Serious complications after umbilical artery catheterization for neonatal monitoring.
Arch Surg.1975; 110
:1203
–1208
[Abstract/Free Full Text] - Boo N, Wong N, Zulkifli S, Lye M. Risk factors associated with umbilical vascular catheter-associated thrombosis in newborn infants. J Paediatr Child Health.1999; 35 :460 –465[CrossRef][Web of Science][Medline]
- Schmidt B, Vegh P, Weitz J, et al. Thrombin/antithrombin III complex formation in the neonatal respiratory distress syndrome. Am Rev Respir Dis.1992; 145 :767 –770[Web of Science][Medline]
- Andrew M, Paes B, Milner R, et al. Development of the human coagulation system in the healthy premature infant.
Blood.1988; 72
:1651
–1657
[Abstract/Free Full Text] - Bosque E, Weaver L. Continuous versus intermittent heparin infusion of umbilical artery catheters in the newborn infant. J Pediatr.1986; 108 :141 –143[CrossRef][Web of Science][Medline]
- David R, Martin D, Anderson J, et al. Prevention of umbilical artery catheter clots with heparinised infusates. Dev Pharmacol Ther.1981; 2 :117 –126[Web of Science][Medline]
- Gu Y, Obster R, de Haan J, Gallandat HR, van Oeveren W. Biocompatibility of leukocyte removal filters during bedside leukocyte filtration of red cell concentrates. Transfus Sci.1992; 13 :467 –472[CrossRef][Web of Science][Medline]
- Formanek G, Frech R, Amplatz K. Arterial thrombus formation during clinical percutaneous catheterization.
Circulation.1970; 41
:833
–839
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2004 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
