Keywords fresh frozen plasma - rebleeding - screening coagulation tests - therapeutic endoscopic
procedures - thromboelastometry
Introduction
Screening coagulation tests which include platelet count, prothrombin time (PT), and
activated partial thromboplastin time (aPTT) are routinely performed in some centers
prior to surgery and therapeutic endoscopy to identify patients with bleeding disorders,
assuming that testing will predict perioperative bleeding which can be prevented by
giving necessary treatment.[1 ]
[2 ]
[3 ] There is sometimes also a medicolegal angle for the need of performing these tests.[4 ] At the same time, there are society guidelines which do not support doing coagulation
tests unless there is prior history of bleeding, a clinical condition that predisposes
to bleeding or a procedure fraught with high bleeding risk.[5 ]
[6 ] This has led to confusion among physicians as whether to do these tests prior to
procedures, leading to varying policies.[4 ]
Screening coagulation tests are done prior to therapeutic endoscopic procedures in
our department based on British guidelines.[5 ] Abnormal results in these tests often warrant the administration of blood products
prior to endoscopy. These can lead to increased cost of care and are fraught with
the low but certain risk of transfusion reactions. Viscoelastic tests such as thromboelastography
(TEG) and rotational thromboelastometry (ROTEM) are now used to overcome these shortcomings
in patients with cirrhosis due to rebalanced hemostasis in liver disease.[7 ] TEG is a point-of-care test which provides a more global overview of the hemostasis.
If TEG is abnormal, then appropriate products are administered.
The parameters in TEG are R-time, K-time, angle, maximal amplitude (MA), and LY30.
R-time is the time taken to initiate clot formation, and it reflects clotting factor
levels. K-time represents the time taken to reach certain level of clot strength.
Angle denotes kinetics of clot strength and clot build up. K-time and angle reflect
fibrinogen contribution up to 80% and platelets up to 20%. MA indicates maximum clot
strength contributed by platelets up to 80% and fibrinogen up to 20%. LY30 is a measure
of fibrinolysis, and it refers to the percentage in amplitude reduction after 30 minutes
of MA.[8 ]
ROTEM is also a point-of-care viscoelastic test similar to TEG. In TEG, the pin is
stationary and the cup oscillates, whereas in ROTEM, the pin oscillates and the cup
is stationary. The parameters in ROTEM are clotting time (CT), clot formation time
(CFT), α angle, maximum clot firmness (MCF), and maximum clot lysis and their counterparts
in TEG are R-time, K-time, angle, MA, and LY30, respectively.[9 ]
A previous study attempting to understand the relationship between TEG and ROTEM showed
linear association with strong correlation for CFT with K-time and MA with MCF. Rest
of the parameters showed either moderate or poor correlation.[10 ]
The use of TEG in nonliver patients to guide transfusion requirements is a recent
concept which can be useful in patients with deranged screening coagulation parameters
based on its success in other clinical situations.[11 ] With this background, our study is aimed to compare the amount of blood products
transfused and bleeding rates in patients with and without hypocoagulable ROTEM who
underwent therapeutic endoscopic procedures with deranged screening coagulation tests
(PT and aPTT) without liver disease.
Materials and Methods
A retrospective study of all patients with deranged PT or aPTT, without liver disease,
who underwent therapeutic endoscopic interventions under the department of medical
gastroenterology at our hospital, a tertiary care center in South India, between January
1, 2020, and May 31, 2022, was conducted. ROTEM was done only in those patients who
had deranged PT or aPTT. Deranged PT was defined as PT more than 12.5 seconds or international
normalized ratio (INR) ≥ 1.4.[12 ]
[13 ] Deranged aPTT was defined as aPTT more than 40.4 seconds.
Viscoelastic testing was performed with citrated whole blood on ROTEM (Tem International,
Munich, Germany) using modified EXTEM mode. EXTEM reagent with low tissue factor (TF)
concentration was used to reflect physiological conditions.[14 ] The test was performed by trained personnel and activated using TF. The TF was prepared
by dilution of PT reagent, Innovin (Dade Behring, United States) at 1:2,000 dilution,
modified from the method described by Sørensen et al.[15 ] The variables assessed were CT, CFT, α angle (α), MCF, and maximal lysis. The reference
ranges for these parameters were established using samples from more than 300 blood
donors.
A hypercoagulable state was defined as two or more of the following: short CT and/or
CFT time, increased α-angle, and increased MA. Hypocoagulable state was defined as
two or more of the following: prolonged CT and/or CFT, decreased α-angle, and decreased
MA. Normal state was defined as all indices being within the normal ranges. In case,
if ROTEM that did not fall into one of the three categories above (i.e., only a single
abnormality or mixed hyper- and hypoindices in the same ROTEM), the predominant abnormality
was used to categorize the ROTEM into the most appropriate category.[16 ]
The amount of blood products transfused was decided by clinician by the advice of
hematologist and/or transfusion medicine specialist. Patient details were collected
through the hospital computerized data system. Data consisting of demographics, etiology
of primary gastrointestinal disease, screening coagulation parameters, therapeutic
intervention performed, amount of blood products transfused, and important outcomes
such as 30-day bleeding and 30-day mortality were recorded and compared between those
with and without hypocoagulable ROTEM. Vitamin K was not given for patients with cholestasis
prior to therapeutic intervention as utility of ROTEM was being studied in deranged
coagulation.
Statistical analysis was performed using the chi-square test or Fisher's exact test,
as applicable for categorical data. Mann–Whitney's U test was used for continuous data. The data were analyzed with SPSS v. 21.0 data
(Statistical Package for the Social Sciences, SPSS Inc., Chicago, Illinois, United
States). A p -value of < 0.05 was taken to be statistically significant.
This study was approved by our Institutional Review Board (Ref: IRB: 14844 [Retro]
dated October 26, 2022), and being a retrospective study, there was waiver of consent.
Results
A total of 204 patients who underwent therapeutic endoscopic procedures had deranged
screening coagulation parameters during the study period. After excluding 180 patients
with liver disease, 24 patients were included for the study ([Fig. 1 ]).
Fig. 1 Patient flow diagram. FFP, fresh frozen plasma; ROTEM, rotational thromboelastometry.
Six patients (M:F = 5:1; median age: 37, 20–54 years) had hypocoagulable ROTEM and
18 patients M:F = 11:7; median age: 56, 20–71 years) with either normo-/hypercoagulable
ROTEM. Both groups were comparable in baseline characteristics except platelet count
(p = 0.04) and aPTT level (p = 0.04) ([Table 1 ])
Table 1
Comparison of baseline characteristics between hypocoagulable and normo-/hypercoagulable
ROTEM groups
Characteristics
Hypocoagulable ROTEM (n = 6)
Normo-/hypercoagulable ROTEM (n = 18)
p -Value
Age (y)
37 (20–54)
56 (20–71)
0.083
Male sex
5 (83.3%)
11 (61.1%)
0.31
Indications for therapeutic endoscopy (n )
EHBO with cholangitis
1
4
EHBO without cholangitis
0
5
Gastric ulcer
1
2
Cholangiocarcinoma
0
2
Mallory–Weiss tear
1
2
Walled-off necrosis pancreas
0
2
Others[a ]
3
1
Procedures (n )
Combination endotherapy for ulcer bleed
4
5
ERCP + biliary sphincterotomy
1
9
ERCP + biliary sphincterotomy + EUS + FNA
0
1
EUS + FNA
0
1
EUS + cystogastrostomy
0
2
POEM
1
0
Laboratory results
Hemoglobin (g/L)
75.5 (27–176)
101.5 (55–136)
0.15
Platelet count (109 /L)
120 (26–200)
220 (11.6–580)
0.04[b ]
Thrombocytopenia (platelet count < 50 × 109 /L)
1 (16.7%)
1 (5.55%)
0.39
INR
2.16 (1.18–10)
1.535 (0.93–2.72)
0.27
INR > 1.4
4 (66.7%)
12 (66.7%)
1.0
aPTT (s)
53.7 (41.2–120)
42.4 (33.9–62.7)
0.04[b ]
aPTT > 40.4 (s)
6 (100%)
11 (61.1%)
0.13
Fibrinogen (g/L)
1.96 (0.44–5.13)
3.21 (1.53–6.63)
0.11
Fibrinogen < 1.5 g/L
2 (33.4%)
0
0.05
Abbreviations: aPPT, activated partial thromboplastin time; EHBO, extrahepatic biliary
obstruction; ERCP, endoscopic retrograde cholangiopancreatography; EUS, endoscopic
ultrasound; FNA, fine-needle aspiration; INR, international normalized ratio; POEM,
peroral endoscopic myotomy; ROTEM, rotational thromboelastometry.
Note: Data are presented as n (%) or median (range).
a Others: Achalasia cardia, rectal stercoral ulcer bleed, colonic diverticular bleed,
postsphincterotomy bleed.
b Statistically significant.
[Table 2 ] depicts comparison of blood products transfused and outcomes between both the groups.
There was significant difference in total amount of fresh frozen plasma (FFP) infused
and FFP infused per patient between the groups (9,000 vs. 4,500 mL and 2,000 vs. 1,000 mL;
p = 0.04), respectively.
Table 2
Comparison of blood products transfused and outcomes between hypocoagulable ROTEM
group and normo-/hypercoagulable ROTEM group
Variable
Hypocoagulable ROTEM (n = 6)
Normo-/hypercoagulable ROTEM (n = 18)
p -Value
No. of patients transfused with FFP
4 (66.7%)
5 (27.8%)
0.08
Total amount of FFP infused (L)
9
4.5
Median FFP (L) infused/patient
2 (1.75–3.25)
1 (0.75–1)
0.04[a ]
No. of patients transfused with platelets
2 (33.4%)
1 (5.56%)
0.07
Total amount of random donor platelet pools infused (U)
15
4
Median pooled platelets (U) infused/patient
7.5 (2–13)
4
0.33
No. of patients transfused with cryoprecipitate
2
0
Total amount of cryoprecipitate infused (U)
17
0
Median cryoprecipitate (U) infused/patient
8.5 (7–10)
0
No. of patients transfused with packed cells
4 (66.7%)
3 (16.67%)
0.019[a ]
Total amounts of packed cells transfused
23
6
Median RBCs transfused (packs) per patient
6 (3–8)
2
0.03[a ]
FFP transfusion only
2 (33.4%)
5 (27.8%)
0.79
Platelet transfusion only
0
1
FFP, platelets, and cryoprecipitate transfusions
2 (33.4%)
0
Outcome
30-d bleeding
2 (33.3%)
0
0.05
30-d mortality
1 (16.7%)
0 (0)
0.25
Lost to follow-up
0
4
Abbreviations: FFP, fresh frozen plasma; RBC, red blood cell; ROTEM, rotational thromboelastometry.
Note: Data are presented as n (%) or median (range).
a Statistically significant.
Two patients with hypocoagulable ROTEM bled within 30 days and none in the other group
(p = 0.05). Only one patient in the hypocoagulable ROTEM group died.
Discussion
TEG-guided transfusion strategy has been shown to significantly decrease the blood
component requirement in patients with advanced cirrhosis, coagulopathy, and nonvariceal
upper gastrointestinal bleeding.[17 ] It is already known that PT and INR do not predict bleeding risk in liver disease
because it relies on thromboplastins and measures only the activity of procoagulants
and not of the anticoagulants, both of which may be depressed in patients with advanced
liver disease.
Our study is a first of its kind trying to analyze the benefit of a similar strategy
in those patients without liver disease but with derangement of the screening coagulation
tests for whom therapeutic endoscopy is being considered.
Giles et al[18 ] reported that abnormal PT and aPTT tests do not predict bleeding risk in children
undergoing endoscopic procedures. Deranged aPTT may indicate underlying factor deficiency
which may or may not influence the overall coagulation cascade (e.g., factor XII deficiency).
So, ROTEM is an excellent point-of-care test which will inform about the overall coagulation
status.
In our study, we noted that significantly less amount of FFP was needed when a ROTEM-based
transfusion strategy was followed. The 30-day bleeding and mortality rates did not
increase in those who had restricted transfusion based on this strategy. Such an approach
could reduce blood product utilization, time, cost, and potential transfusion-related
adverse reactions without increasing the risk of bleeding or death.
Limitations of this study include its retrospective nature, small sample size, and
being confined to single tertiary care center. Clinician discretion in deciding blood
product transfusion based on ROTEM results is another limitation. Some patients with
normo-/hypercoagulable ROTEM also received transfusions as they belonged to some departments
where ROTEM-based transfusion strategy was not implemented. Instead, transfusions
were administered on the basis of the abnormal screening coagulation test results.
There is a need for large multicenter prospective studies to validate the results
of ROTEM-based transfusion strategy which we have utilized in our study.
Conclusion
For centers pursing the policy of coagulation-based screening prior to interventional
endoscopic procedures, the use of ROTEM as point-of-care test can help reduce blood
component usage in patients without liver disease who have deranged screening coagulation
parameters without any increased risk of 30-day bleeding and mortality.
