Coagulation and Transfusion Medicine / Heparin Assay in Newborns on ECMO

Evaluation of Heparin Assay for Coagulation Management

in Newborns Undergoing ECMO
Wassia A. Khaja, MD,1 Ozlem Bilen, MD,2 Ralf B. Lukner, PhD,1
Rachel Edwards, MBA, MT(ASCP),2 and Jun Teruya, MD, DSc3
Key Words: Extracorporeal membrane oxygenation; ECMO; Neonatal coagulation; Anticoagulation; Heparin assay

CME/SAM

DOI: 10.1309/AJCPGVD62LKKVDLH

Upon completion of this activity you will be able to:

define an overview of the extracorporeal membrane oxygenation
(ECMO) circuit.
compare the utility of heparin assay vs partial thromboplastin time
(PTT) and activated clotting time (ACT) in monitoring neonates on
ECMO.
define the caveats of using PTT, ACT, and anti-Xa assay.

Abstract
The objective was to identify the usefulness
of heparin level by antifactor Xa (anti-Xa) assay
vs activated partial thromboplastin time (PTT) or
activated clotting time (ACT) in neonates undergoing
extracorporeal membrane oxygenation (ECMO).
A retrospective record review of 21 patients in the
neonatal intensive care unit (mean ECMO initiation
age, 2 days; range, 0-4 days; male/female ratio, 1:1)
undergoing ECMO from 2006 to 2008 was performed.
Linear regression correlations between anti-Xa,
PTT, and ACT were determined by extrapolating PTT
and ACT therapeutic ranges that corresponded with the
ECMO heparin target range of 0.3 to 0.6 U/mL. Pearson
correlation coefficients between heparin levels and
PTT (0.903 to 0.984), PTT less than 40 seconds after
correction using PTT-heparinase (0.903 to 1.000),
and ACT (0.951 to 0.891) in this patient population
were widely variable.
Inconsistency of PTT and ACT therapeutic
ranges corresponding to heparin levels of 0.3 to 0.6
U/mL prompts a multifactorial approach to ECMO
management because no single laboratory test can
be used to determine appropriate anticoagulation
management.

950
950

Am J Clin Pathol 2010;134:950-954

DOI: 10.1309/AJCPGVD62LKKVDLH

The ASCP is accredited by the Accreditation Council for Continuing

Medical Education to provide continuing medical education for physicians.
The ASCP designates this educational activity for a maximum of 1 AMA PRA
Category 1 Credit per article. This activity qualifies as an American Board
of Pathology Maintenance of Certification Part II Self-Assessment Module.
The authors of this article and the planning committee members and staff
have no relevant financial relationships with commercial interests to disclose.
Questions appear on p 1010. Exam is located at www.ascp.org/ajcpcme.

Outcomes for newborns with severe cardiorespiratory

failure have substantially improved using extracorporeal
membrane oxygenation (ECMO).1-3 There are mainly 2 types
of ECMO: venoarterial and venovenous. They both provide pulmonary support, but only venoarterial ECMO provides cardiac support. During venovenous ECMO, blood is
extracted from a large central vein and returned to the venous
circulation. During venoarterial ECMO, blood is extracted
from a central vein and returned to the arterial system. During
ECMO, a large volume of blood is extracted from the native
vascular system and circulated outside the body by a mechanical pump. While outside the body, the blood passes through
an oxygenator and a heat exchanger. In the oxygenator,
hemoglobin becomes fully saturated with oxygen and carbon
dioxide is removed. Elimination of carbon dioxide can be
controlled by adjusting the rate of countercurrent gas flow
through the oxygenator. Finally, the blood is reinfused into the
native vascular system Figure 1.
The ECMO circuit needs continuous anticoagulation
because whenever blood is exposed to the foreign surfaces
such as the plastic tubing, the clotting process is triggered.
Clots tend to form initially before the membrane oxygenator.
As the circuit ages, some small clots are usually found within
the membrane oxygenator itself and can lead to major clots
in the patient. In ECMO, while unfractionated heparin (UFH)
is used as an anticoagulant to prevent thrombosis, intracranial bleeding is the most dangerous complication.4 Owing to
the immature hemostatic function in newborns, maintaining
appropriate coagulation is quite challenging. The anticoagulation effect of UFH can be monitored by use of different tests:
activated partial thromboplastin time (PTT), activated clotting
time (ACT), and antifactor Xa activity (anti-Xa).
American Society for Clinical Pathology

Coagulation and Transfusion Medicine / Original Article

Right
atrium

Aorta

Heat
exchanger
Postmembrane
pressure monitor

Patient
bridge

Membrane
oxygenator
Fluids Heparin

Premembrane
pressure monitor

Venous
reservoir

O2 in

Pump

Figure 1 The extracorporeal membrane oxygenation circuit.

PTT evaluates not only congenital and acquired factor

deficiencies, but also the efficacy of heparin therapy. In addition, by neutralizing heparin using heparinase, it can be determined whether PTT prolongation is due to UFH or coagulation factor deficiency. There are many reasons for a prolonged
PTT, such as factor deficiencies, vitamin K deficiency, liver
disease, and disseminated intravascular coagulation; thus, it is
not the best method to monitor the heparin level.5
ACT is another global coagulation test used to monitor
the heparin level.6 It is a convenient test because it uses fresh
whole blood and can be performed easily at the bedside to
monitor adequate anticoagulation therapy. One of the major
limitations of using ACT is its dependency on good user
technique. Variability in ACT results are due in part to poor
technique and preanalytic conditions such as hemolysis and
clotted samples.7
Quantitative measurement of UFH is done by chromogenic anti-Xa assay. This test also has its limitations such as
decreased sensitivity in the presence of increased amounts of
plasma free hemoglobin, bilirubin, and triglycerides.
This study attempted to identify the desired anticoagulation
levels that allow ECMO in neonates without significant risk for
clotting or hemorrhage. In particular, we aimed to identify the
usefulness of the anti-Xa assay vs PTT or ACT. We attempted
to find correlations among these coagulation values.

Materials and Methods

This was a retrospective review study of 21 patients
undergoing ECMO therapy. This study was approved by the

Baylor College of Medicine Internal Review Board (Houston,

TX). The patient population consisted of neonatal intensive
care unit patients at the Texas Childrens Hospital, Houston,
from 2006 to 2008. ECMO coagulation panels were performed every 4 to 24 hours for each patient.
The ECMO coagulation panel at the Texas Childrens
Hospital consists of the following measured parameters: prothrombin time, PTT, PTT with heparinase, fibrinogen, heparin
level, D-dimer, and functional antithrombin. Specimens were
collected in 3.2% citrated tubes, and all tests were performed
immediately without being frozen. The platelet count was
measured separately but at the same time. The heparin level
was measured by anti-Xa activity using the Rotachrom
(Diagnostica STAGO, Parsippany, NJ).
The individual data and the group data were tabulated,
and various coagulation indices were plotted to compare their
relationship. Based on the plotted empirical data, a line of
best fit and the correlation coefficient between the coagulation parameters were determined. Using the equation from
the linear regression line, therapeutic ranges for PTT and
ACT were extrapolated to be equivalent to the ECMO target
heparin range of 0.3 to 0.6 U/mL. At our institution, this range
corresponds to a PTT therapeutic range for ECMO of 73 to
90 seconds but was based on data using plasma from adults
receiving heparin therapy with an assumed normal baseline
PTT. The plotted data used to extrapolate current therapeutic
range yield a Pearson correlation coefficient (r) of 0.5386. In
these adult patients, this represents the desired degree of association between the heparin assay and PTT. We collected all
bilirubin and plasma free hemoglobin values for the patients.

American Society for Clinical Pathology

Am J Clin Pathol 2010;134:950-954

951

DOI: 10.1309/AJCPGVD62LKKVDLH

951
951

Khaja et al / Heparin Assay in Newborns on ECMO

Discussion
This study represents objective evidence for a commonly
held notion that adequate coagulation is difficult to maintain
for neonates undergoing ECMO. ACT and PTT are of interest
because they measure the effectiveness of heparin administration at different levels of the coagulation pathway. In addition,
they are relatively easy to measure and allow for concurrent
adjustments of the heparin assay levels. When comparisons
were made between heparin levels and PTT and/or ACT, it
was clear that neither ACT nor PTT was a perfect measurement
952
952

Am J Clin Pathol 2010;134:950-954

DOI: 10.1309/AJCPGVD62LKKVDLH

Characteristic

Result

M/F ratio
Mean (range) age at initiation of ECMO (d)
Mean (range) duration of ECMO (d)
Mean SD No. of ECMO coagulation
panels performed
Primary diagnosis
Congenital diaphragmatic hernia
Meconium aspiration
Pulmonary hypertension
Sepsis

1:1
2 (0-4)
10.8 (4-23)
2.35 1.08 per day
15
4
1
1

ECMO, extracorporeal membrane oxygenation.

300
250
ACT (s)

Of the 24 neonates who underwent ECMO from 2006 to

2008 in the neonatal intensive care unit, the data for 3 were
excluded from the study owing to incomplete coagulation
records. Patients started on ECMO day 1, and also patients who
required a circuit change during the course of ECMO were considered as starting on ECMO. Only 3 of the 21 patients required
circuit changes owing to some clot formation in the circuit. The
study included 21 patients, with a 1:1 male/female ratio Table
1. The common primary indications for ECMO therapy were
congenital diaphragmatic hernia, meconium aspiration syndrome, and pulmonary hypertension. The average age at initiation of ECMO was 2 days (range, 0-4 days) with mean length
of ECMO therapy of 10.8 days (range, 4-23 days).
Heparin level, PTT, and ACT were significant because
these are the values of interest to clinicians for the modification of treatment. Figure 2 compares the heparin level of all
ECMO panels of the patients with the PTT (corrected with
heparinase) and ACT. There is a wide variation in the values
of PTT and ACT within a given heparin level. For example,
a heparin level of 0.30 U/mL gives a range of PTT between
40 and 150 seconds and, similarly, with ACT yields a range
of 175 to 260 seconds. To better analyze these data, extrapolation of the line of best fit was performed. This analysis
yielded correlation coefficients for PTT and ACT of 0.364
and 0.125, respectively.
When the total data points for all patients are combined
Table 2, the PTT and ACT ranges that corresponded with
a heparin therapeutic range of 0.3 to 0.6 U/mL were narrow;
PTT had a range of 88 to 109 seconds, and the range was similar for a PTT less than 40 seconds with heparinase treatment.
ACT gave a narrow therapeutic range of 208 to 234 seconds.
Of the 3 cases requiring a circuit change, only 1 had a
negative correlation of the ACT with heparin level, and that
was not due to monitoring failure.
The monitoring details for a patient without clot formation within the circuit and for a patient who underwent circuit
change owing to clot formation within the circuit are given in
Figure 3 and Figure 4, respectively.

Table 1
Demographic Characteristics of 21 Neonatal Intensive Care
Unit Patients Undergoing ECMO During the 2006-2008 Study
Period

200
150

PTT (s)

Results

100
50
0
0.0

0.10

0.20

0.30

0.40

0.50

0.60

0.70

Heparin Level (U/mL)

Figure 2 Comparison of 316 heparin levels for all

extracorporeal membrane oxygenation (ECMO) panels of
patients with the activated partial thromboplastin time (PTT;
r = 0.364; corrected with heparinase) and/or activated clotting
time (ACT; r = 0.125).

of heparins anticoagulation effect. As previously mentioned,

the desired correlation, obtained from a control group of
adult patients receiving heparin therapy, was determined to
be 0.5386. The graph (Figure 2) shows a positive, although
not strong correlation between heparin level vs PTT and
ACT. According to Pearson correlation coefficients, PTT (r
= 0.364) showed and had a stronger relationship with heparin
levels than that of ACT (r = 0.125).
In addition, a positive correlation was expected between
the coagulation parameters for each patient data set. In Table
2, it is shown that for most patients, there was a positive correlation between the heparin levels and the extrapolated therapeutic ranges of PTT and ACT, although the correlation may
not be very strong for some cases. Multiple patients showed
an inverse relationship between the measured parameters as
seen with negative r values. In general, ACT and PTT are
American Society for Clinical Pathology

Coagulation and Transfusion Medicine / Original Article

Table 2
Comparison of Coagulation Panels for 21 Patients and for the Total Group
Pearson Product Correlation
Coefficient (r) With Heparin Level

Therapeutic Ranges for 0.3-0.6 U/mL of Heparin

Case No.

PTT (s)

PTT (<40 s)*

ACT (s)

PTT (s)

PTT (<40 s)*

ACT (s)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
All

101-150
86-188
74-115
96-131
71-95
108-155
98-141
97-156
91-63
100-127
72-105
94-160
91-110
93-110
106-188
108-203
82-103
97-115
100-92
82-121
112-137
88-109

95-157
86-188
70-113
96-131
71-95
108-155
98-141
97-156
91-63
123-218
72-105
88-161
81-66
93-115
106-188
108-203
81-105
97-115
100-97
82-121
185-354
90-108

208-234
228-226
235-257
205-215
205-215
213-245
212-211
199-211
213-219
208-191
224-254
219-216
227-247
233-253
246-329
226-308
226-308
216-227
221-212
204-277
169-81
208-234

0.547
0.686
0.644
0.688
0.435
0.908
0.671
0.932
0.903
0.272
0.445
0.29
0.326
0.217
0.364
0.984
0.373
0.397
0.127
0.695
0.368
0.295

0.8
0.686
0.808
0.686
0.435
0.908
0.671
0.932
0.903
0.279
0.445
0.362
0.108
0.291
0.364
0.99
0.395
0.397
0.06
0.695
1
0.364

0.234
0.058
0.512
0.142
0.361
0.517
0.01
0.292
0.163
0.092
0.464
0.466
0.361
0.289
0.477
0.783
0.126
0.272
0.123
0.891
0.951
0.095

ACT, activated clotting time; PTT, activated partial thromboplastin time.

* Following heparinase neutralization.

50
0
D

ay
D 1
ay
D 4
ay
D 6
ay
D 7
a
D y9
ay
D 11
ay
D 12
ay
D 13
ay
D 14
ay
D 14
ay
D 15
ay
D 16
ay
D 17
ay
D 17
ay
18

PTT (s)

100

400
300
200
100
0
ay
D 1
ay
D 1
ay
D 2
ay
D 2
ay
D 3
ay
D 3
ay
D 4
ay
D 4
ay
D 5
ay
D 5
ay
D 6
ay
D 6
ay
D 7
ay
D 7
ay
D 8
ay
D 8
ay
9

150

2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0

500
ACT (s)

200

2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0

PTT (s)

ACT (s)

250

Heparin

300

PTT

600

Heparin

Heparin (U/mL)

PTT

Heparin (U/mL)

ACT
ACT

Figure 3 Monitoring details for a patient without substantial

clot formation in the circuit. Activated clotting time (ACT),
activated partial thromboplastin time (PTT), and heparin level
were plotted.

Figure 4 Monitoring details for a patient with substantial

clot formation in the circuit. Activated clotting time (ACT),
activated partial thromboplastin time (PTT), and heparin level
were plotted.

beneficial for monitoring heparin anticoagulation; however,

if there is an inverse relationship, the information given is no
longer useful. It is impossible to determine whether a spot
check of ACT or PTT represents a direct or inverse relationship. Although it is stated in the Rotachrom package insert
that high bilirubin and plasma hemoglobin levels cause falsely
low heparin levels, it was not the main reason of the discrepancy between the heparin level and ACT and PTT. Overall,
ACT has more inverse correlations than PTT, making a less
reliable test. The correlation coefficients between the heparin
level and PTT and the heparin level and ACT are still positive,

but weak. Of note, the PTT has a stronger correlation with

heparin than does the ACT.
It is interesting that there is a direct relation between the
number of days of ECMO therapy and D-dimer levels Figure
5. An increasing D-dimer level suggests fibrin deposition and
is a sign of clot formation. It is possible that patients undergoing ECMO therapy require increasing amounts of heparin
therapy as the number of days on the circuit increases. Only
3 of the 21 patients required circuit changes, which is impressive considering the length of time on the circuit and the difficulty of maintaining adequate anticoagulation.

American Society for Clinical Pathology

Am J Clin Pathol 2010;134:950-954

953

DOI: 10.1309/AJCPGVD62LKKVDLH

953
953

Khaja et al / Heparin Assay in Newborns on ECMO

25.00
16.01
n = 27

D-Dimer (g/mL FEU)

20.00

17.63
n = 19

11.44
n = 42

20.00
n = 15

19.28
n = 13

19.94
n=8

15.00
5.46
n = 82

10.00

5.79
n = 53

5.00
0.00
0

5.00

10

12

14

Day

Figure 5 D-dimer levels plotted against the extracorporeal

membrane oxygenation day. D-dimer levels are given in
conventional units; to convert to Systme International units
(nmol/L), multiply by 5.476. FEU, fibrinogen equivalent units.
Error bars indicate standard deviations.

those for nonsurvivors. Regression analysis indicated that

increased heparin administration was predictive of survival (P
< .0001), independent of all other variables. It was concluded
that survival is improved by increased heparin administration
independent of the ACT.8 In addition, further investigation
of other methods to monitor anticoagulation such as anti-Xa
activity or thrombin generation is of interest.
PTT is, overall, a more reliable test than ACT but no
single laboratory test can be used alone to monitor heparin
anticoagulation.
From the 1Baylor College of Medicine; 2Department of Pathology,
Texas Childrens Hospital; and 3Departments of Pathology &
Immunology, Pediatrics, and Medicine, Texas Childrens Hospital
and Baylor College of Medicine; Houston.
Address reprint requests to Dr Teruya: Texas Childrens
Hospital, MC 2-2261, Baylor College of Medicine, 6621 Fannin
St, Houston, TX 77030.

References
PTT and ACT ranges corresponding to heparin levels
are quite variable, and no single laboratory test can be used
alone to monitor heparin anticoagulation. PTT is, overall,
a more reliable test than ACT; however, a multifactorial
approach must be used to provide an adequate coagulation
level for patients. In our institution, we adjust heparin dosing
to keep it within the target range; in addition, we try to keep
the prothrombin time and PTT in the desired ranges. If they
are prolonged, we transfuse appropriate blood products such
as plasma. We also aim to keep fibrinogen and antithrombin
levels within adequate ranges. If they are below the target
range, we give antithrombin concentrate to bring antithrombin
levels up and cryoprecipitate to bring fibrinogen levels up to
the target levels. We try to keep the platelet count at more than
100 103/L (100 109/L), and when the level goes down,
we administer platelets.
The significance of this study lies in whether there is a
substantial improvement in the outcomes of neonates receiving ECMO based on their coagulation during ECMO. Further
investigations of interest are whether proper anticoagulation
during ECMO has an impact on the long-term morbidity and
mortality of neonates. Baird et al8 recently conducted a retrospective review of 604 consecutive pediatric ECMO patients,
comparing the coagulation parameters for survivors with

954
954

Am J Clin Pathol 2010;134:950-954

DOI: 10.1309/AJCPGVD62LKKVDLH

1. Tan LH, Du LZ, He XJ, et al. Two successful neonatal

extracorporeal membrane oxygenation treatment for severe
heart failure after cardiac surgery. Chin Med J (Engl).
2009;122:2672-2674.
2. Betit P, Craig N. Extracorporeal membrane oxygenation for
neonatal respiratory failure. Respir Care. 2009;54:1244-1251.
3. Rajagopal SK, Almond CS, Laussen PC, et al. Extracorporeal
membrane oxygenation for the support of infants, children,
and young adults with acute myocarditis: a review of the
Extracorporeal Life Support Organization registry. Crit Care
Med. 2010;38:382-387.
4. Hirthler MA, Blackwell E, Abbe D, et al. Coagulation
parameter instability as an early predictor of intracranial
hemorrhage during extracorporeal membrane oxygenation.
J Pediatr Surg. 1992;27:40-43.
5. Hattersley PG, Hayse D. The effect of increased contact
activation time on the activated partial thromboplastin time.
Am J Clin Pathol. 1976;66:479-482.
6. Searles B, Nasrallah F, Darling E, et al. How does the
age of a blood sample affect its activated clotting time?
comparison of eight different devices. J Extra Corpor Technol.
2002;34:175-177.
7. Uden DL, Payne NR, Kriesmer P, et al. Procedural variables
which affect activated clotting time test results during
extracorporeal membrane oxygenation therapy. Crit Care
Med. 1989;17:1048-1051.
8. Baird CW, Zurakowski D, Robinson B, et al. Anticoagulation
and pediatric extracorporeal membrane oxygenation: impact
of activated clotting time and heparin dose on survival. Ann
Thorac Surg. 2007;83:912-920.

American Society for Clinical Pathology