guideline

Guidelines on the assessment of bleeding risk prior to surgery

or invasive procedures

British Committee for Standards in Haematology

1 2
Y. L. Chee, J. C. Crawford, H. G. Watson1 and M. Greaves3
1
Department of Haematology, Aberdeen Royal Infirmary, Aberdeen, 2Department of Anaesthetics, Southern General Hospital, Glasgow,
and 3Department of Medicine and Therapeutics, School of Medicine, University of Aberdeen, Aberdeen, UK

surgery or invasive procedures to predict bleeding risk. The

Summary
aim is to evaluate the use of indiscriminate testing. Appro-
Unselected coagulation testing is widely practiced in the priate testing of patients with relevant clinical features on
process of assessing bleeding risk prior to surgery. This may history or examination is not the topic of this guideline. The
delay surgery inappropriately and cause unnecessary concern target population includes clinicians responsible for assess-
in patients who are found to have ‘abnormal’ tests. In addition ment of patients prior to surgery and other invasive
it is associated with a significant cost. This systematic review procedures.
was performed to determine whether patient bleeding history
and unselected coagulation testing predict abnormal periop-
Methods
erative bleeding. A literature search of Medline between 1966
and 2005 was performed to identify appropriate studies. The writing group was made up of UK haematologists with
Studies that contained enough data to allow the calculation of a special interest in bleeding disorders and an anaesthetist.
the predictive value and likelihood ratios of tests for periop- First, the commonly employed coagulation screening tests
erative bleeding were included. Nine observational studies were identified and their general and specific limitations
(three prospective) were identified. The positive predictive considered. Second, Medline was systematically searched for
value (0Æ03–0Æ22) and likelihood ratio (0Æ94–5Æ1) for coagula- English language publications from 1966 to September 2005.
tion tests indicate that they are poor predictors of bleeding. Relevant references generated from initial papers and
Patients undergoing surgery should have a bleeding history published guidelines/reviews were also examined. Meeting
taken. This should include detail of previous surgery and abstracts were not included. Key terms: routine, screening,
trauma, a family history, and detail of anti-thrombotic preoperative, surgery, coagulation testing, APTT, PT, bleeding,
medication. Patients with a negative bleeding history do not invasive procedures. Inclusion criteria: studies had to contain
require routine coagulation screening prior to surgery. enough data to enable the calculation of (i) the predictive value
(PV) and likelihood ratio (LR) of the coagulation test for
Keywords: surgery, coagulation screen, bleeding, clinical postoperative bleeding and/or (ii) the PV and LR of the
history. bleeding history for postoperative bleeding. The rationale and
methods for the calculations are described in Appendix 1. Nine
observational case series with usable data (Table I) and one
systematic review were identified (Table II).
Objective
Data elements extracted from these articles were study type,
The aim of this guideline is to provide a rational approach to surgical setting, number and age of patients and coagulation
the use of bleeding history and coagulation tests prior to tests performed. Outcome data extracted included abnormal
tests, positive bleeding history, postoperative bleeding and
Correspondence: BCSH Secretary, British Society for Haematology, change in management as a result of coagulation screening.
100 White Lion Street, London N1 9P, UK. E-mail: [email protected] Statistical analysis: standard methods were used to calculate the
Correspondence prior to publication: Dr Y. L. Chee, Department of PV and LR. 95% confidence intervals for proportions were
Haematology, Aberdeen Royal Infirmary, Foresterhill Road, Aberdeen calculated by the efficient-score method, corrected for conti-
AB25 2ZN, UK. E-mail: [email protected] nuity (Appendix 1) (Newcombe, 1998).

ª 2008 The Authors

doi:10.1111/j.1365-2141.2007.06968.x Journal Compilation ª 2008 Blackwell Publishing Ltd, British Journal of Haematology, 140, 496–504
 13652141, 2008, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2141.2007.06968.x by Nat Prov Indonesia, Wiley Online Library on [06/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Guideline

Table I. Characteristics of the trials.

Definition of an Patients with Preoperative

Surgical Number of Exclusion abnormal abnormal coagulation change in
Reference setting patients criteria coagulation test test results (%) management (%)

Gabriel et al (2000) Prospective 1479 No Not stated 4 0Æ3

Paediatric ENT
Houry et al (1995) Prospective 3242 Yes PT <70% control 17 1
General surgery APTT ratio >1Æ2 control
Burk et al (1992) Prospective 1603 No ‡3SD above mean 2 1
Paediatric ENT
Asaf et al (2001) Paediatric ENT
PT = 373 Yes* PT >13 APTT >39Æ9 s PT = 32 n/a
APTT = 346 (no normal range given) APTT = 18
Howells et al (1997) Paediatric ENT 261 Yes ‡2SD above mean 15 0
Myssiorek and Alvi (1996) ENT 1138 No Not stated 1 0
Kang et al (1994) Paediatric ENT 1061 No Not stated 3 0Æ1
Manning et al (1987) Paediatric ENT 994 Yes* ‡2SD above mean 6 0Æ1
Suchman and Mushlin (1986) General surgery APTT = 2134 Yes* APTT >26Æ5 s 16 n/a
(no normal range given)

Unless otherwise stated, all patients had PT and APTT performed.

ENT, ear, nose and throat; PT, prothrombin time; APTT, partial prothrombin time; s, seconds; SD, standard deviation; n/a, not applicable.
*Patients with known history of coagulopathy.
Patients on anti-thrombotic drugs.
92% of patients were less than 20 years old.

A draft guideline was produced by the writing group, revised

The principles of the coagulation screening tests
and agreed by consensus. Further comment was made by the
used most widely in an attempt to predict
members of the Haemostasis and Thrombosis Task Force of
bleeding and their limitations
the British Committee for Standards in Haematology (BCSH).
The guideline was reviewed by a sounding board of approx- When a blood vessel is injured the vascular, platelet, coagulation
imately 40 UK haematologists, the BCSH and the Committee and fibrinolytic systems react in a co-ordinated fashion to
of the British Society for Haematology and comments were prevent blood loss whilst localising thrombus to the site of injury.
incorporated where appropriate. Criteria used to quote levels Bleeding can arise from abnormalities in any one, or a combi-
and grades of evidence are as outlined in Appendix 7 of the nation, of the four components in the haemostatic system. The
Procedure for Guidelines commissioned by the BCSH (http:// physiology is complex and current widely used laboratory tests
www.bcshguidelines.com/process1.asp) (Appendix 2). cannot accurately reproduce the in vivo haemostatic processes.

Summary of key recommendations Coagulation tests

The first-line clotting tests commonly used are the activated
1 Indiscriminate coagulation screening prior to surgery or
partial thromboplastin time (APTT) and the prothrombin
other invasive procedures to predict postoperative bleed-
time (PT). These are both measured using automated analy-
ing in unselected patients is not recommended. (Grade B,
sers. The standardized skin bleeding time (BT) is occasionally
Level III).
performed. Thrombin clotting time and fibrinogen are not
2 A bleeding history including detail of family history,
generally considered to be first-line clotting tests and are not
previous excessive post-traumatic or postsurgical bleeding
discussed further.
and use of anti-thrombotic drugs should be taken in all
patients preoperatively and prior to invasive procedures.
APTT. The APTT is a test of the integrity of the intrinsic and
(Grade C, Level IV).
common pathways of coagulation. The in vitro clotting time is
3 If the bleeding history is negative, no further coagulation
measured after addition to plasma of calcium and the APTT
testing is indicated. (Grade C, Level IV).
reagent, which contains phospholipid (a platelet substitute,
4 If the bleeding history is positive or there is a clear clinical
also called ‘partial thromboplastin’ as it lacks tissue factor),
indication (e.g. liver disease), a comprehensive assessment,
and an intrinsic pathway activator e.g. kaolin. The APTT
guided by the clinical features is required. (Grade C,
should be designed to detect bleeding disorders due to
Level IV).

ª 2008 The Authors

Journal Compilation ª 2008 Blackwell Publishing Ltd, British Journal of Haematology, 140, 496–504 497
 13652141, 2008, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2141.2007.06968.x by Nat Prov Indonesia, Wiley Online Library on [06/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Guideline

Table II. Studies of patients with abnormal PT undergoing preinvasive procedures extracted from Segal and Dzik (2005) (see original paper for
references).

Patients with abnormal Patients with normal test

Procedure Reference test results with major bleeding results with major bleeding*

Bronchoscopy Kozak and Brath (1994) 3/28 = 0Æ11 28/218 = 0Æ13

Zahreddine et al (2003) 1/14 = 0Æ07 43/412 = 0Æ10
Central line Foster et al (1992) 0/122 0/57
Doerfler et al (1996) 0/33 NR
Fisher and Mutimer (1999) 1/580 = 0Æ002 NR
Femoral arteriogram Wilson et al (1990) 0/9 0/300
Darcy et al (1996) 1/85 = 0Æ012 15/915 = 0Æ016
MacDonald et al (2003) 1/10 = 0Æ1 NR
Liver biopsy Ewe (1981) 4/93 = 0Æ043 4/85 = 0Æ047
Denzer et al, 2001 0/29 1/50 = 0Æ02
Riley et al (1984) 1/20 = 0Æ05 NR
Tobin and Gilmore (1989) 1/100 = 0Æ01 NR
McVay and Toy (1990) 4/76 = 0Æ05 4/100 = 0Æ04
Caturelli et al (1993) 0/49 NR
Sawyerr et al (1993) 2/100 = 0Æ02 NR
Kamphuisen et al (2002) 0/27 0/9
Steadman et al (1988) 0/67 NR
Papatheodoridis et al (1999) 0/112 0/45
Choo et al (2000) 0/18 NR
Bruzzi et al (2002) 0/31 0/19
Smith et al (2003) 3/203 = 0Æ015 0/168
Paracentesis McVay and Toy (1991) 1/37 = 0Æ03 10/352 = 0Æ03
Renal biopsy Davis and Chandler (1995) 1/9 = 0Æ11 33/110 = 0Æ30
Thompson et al (2004) 2/10 = 0Æ2 0/15
Mixed Friedman and Sussman (1989) 0/51 NR

NR, not reported.

*Patients may have thrombocytopenia with normal PT/INR.

deficiencies of factors VIII, IX, and XI and inhibitors of the (PFA-100), the thrombelastogram and measures of endogenous
intrinsic and common pathway factors (including lupus thrombin potential. Presently, these methods are not used
anticoagulant and therapeutic anticoagulants). Inevitably, it routinely and have not been validated for use in a preoperative
also detects deficiency of factor XII. setting. For these reasons, they are not reviewed further here.

PT. The PT assesses the integrity of the extrinsic and common

Limitations
pathways. The in vitro clotting time is measured after addition
of the PT reagent, which contains thromboplastin Coagulation screening tests can be meaningfully interpreted
(phospholipids with tissue factor) and calcium to citrated only with knowledge of their limitations and the relevant
plasma. PT prolongation should detect important deficiencies clinical situation.
(or rarely inhibitors) of factors II, V, VII and X. Its main use is
for anticoagulant monitoring and detection of acquired
General limitations
bleeding disorders (especially disseminated intravascular
coagulation, liver disease and vitamin K deficiency). In vitro assays: Both the PT and APTT are in vitro laboratory
assays that measure the time to clot formation in a test tube
Skin bleeding time. This is the only in vivo haemostasis test and require the addition of exogenous reagents. Interpretation
available. It is used to test for defects of platelet-vessel wall requires caution, as they do not accurately reflect the in vivo
interaction and should detect inherited or acquired disorders haemostatic response.
of platelet function, von Willebrand disease (VWD) and
abnormalities of vessel wall integrity. Normal biological variation: In laboratory practice the ‘normal’
range is usually derived from disease-free subjects and defined
Other tests. A number of tests designed to better reflect primary as results falling within two standard deviations above and
haemostasis and global haemostatic mechanisms have been below the mean for the normal population. Therefore, by
developed. These include the platelet function analyser-100 definition, 2Æ5% of healthy subjects have a prolonged clotting

ª 2008 The Authors

498 Journal Compilation ª 2008 Blackwell Publishing Ltd, British Journal of Haematology, 140, 496–504
 13652141, 2008, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2141.2007.06968.x by Nat Prov Indonesia, Wiley Online Library on [06/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Guideline

time. In the absence of relevant clinical information, Detection of disorders that are not associated with a bleeding
unnecessary further investigations may be prompted, tendency: Two common causes of prolonged APTT in the
generating delay, anxiety, cost and potential harm. general population are factor XII deficiency and the lupus
The PT and APTT tests were designed as diagnostic tests to anticoagulant inhibitor. Neither is associated with bleeding.
confirm the clinical suspicion of a bleeding disorder. This is Test sensitivity to lupus anticoagulant and factor XII deficiency
different from their use as screening tests in healthy populations varies depending on choice of reagents.
where the prevalence of unsuspected bleeding disorders is low.
PT
Insensitivity to some clinically important bleeding disorders: For
reasons explained below, mild, but clinically significant Technical variability: Differences in the composition of the PT
haemophilia A or VWD may be missed, resulting in false reagent can result in variable sensitivity. The same constraints
reassurance. Although of much lower prevalence, factor XIII relating to the ability to detect factor deficiencies apply as in
deficiency and alpha2-antiplasmin deficiency may cause life- the APTT. Similarly, some reagent/instrument combinations
threatening surgical bleeding with normal APTT, PT and skin result in prolongation of the coagulation time only when a
bleeding time. However, most patients will have a positive relevant factor level drops to less than 30 iu/dl.
bleeding history.
Detection of disorders that are not associated with a bleeding
Artefact due to sample collection or pathological tendency: Prolongation of PT is an occasional manifestation of
conditions: Erroneous coagulation results can be caused by lupus anticoagulant.
prolonged tourniquet placement, difficult or traumatic
phlebotomy, inadequate sample volumes, heparin
Skin bleeding time
contamination, prolonged storage, sampling from a line and
failure to adjust the amount of citrate anticoagulant when the Technical variability: Despite attempts at standardization, the
haematocrit is significantly raised. Repeat testing with attention test remains poorly reproducible and subject to a large number
to technique, and ideally by direct venepuncture should exclude of variables. Technique-related factors include location and
most of these artefacts. Burk et al (1992) found 35 abnormal direction of the incision.
test results (either PT, APTT, BT or a combination) in 1603
prospectively screened preoperative patients. Only 15 (43%) of Poor sensitivity and specificity: The skin bleeding time does not
these abnormal results persisted on retesting 7–10 d later. necessarily reflect bleeding from any other site. A range of
commonly encountered patient-related factors can prolong
Specific limitations. APTT skin bleeding time without any clear relationship to bleeding
risk. These include medications (aspirin and other non-
Technical variability: The two main factors are the use of steroidal anti-inflammatory drugs), severe renal failure,
different APTT reagents and different end-point detection thrombocytopenia, paraproteinaemia and severe anaemia.
methods. APTT reagents vary enormously in their Similarly, the bleeding time may be within the normal range
phospholipid content and activator, resulting in significant in VWD, platelet storage pool disorder and in aspirin users, but
differences in the sensitivity of reagents to coagulation factor increased perioperative bleeding may still occur (Lind, 1991).
deficiencies and inhibitors, especially lupus anticoagulants, but
also heparin. Ideally, for screening purposes, the assay should From the above, it is clear that coagulation tests have
be set up to detect any clinically significant deficiency of factor considerable limitations due to technical factors, insensitivity
VIII (i.e. the lower limit of the normal range for the local to some significant bleeding disorders and sensitivity to some
population, usually around 45–50 iu/dl). However, some common abnormalities that carry no bleeding risk.
reagent/instrument combinations result in prolongation of
the APTT only when the factor VIII concentration is less than
Predictive value of coagulation screening tests
30 iu/dl. Mild, but clinically significant haemophilia A or
VWD may be missed, resulting in false reassurance. Further,
Predictive value of coagulation tests for postoperative
sensitivity of the APTT to common pathway factor
bleeding
deficiencies, especially fibrinogen and prothrombin, is low.
We calculated the positive predictive value (PPV) and negative
Disease and/or physiological variability: Clinically important predictive value (NPV) of a prolonged clotting time for
diseases may be modified or masked by physiological response. postoperative bleeding and the postoperative bleeding rates of
For example, factor VIII rises markedly in pregnancy and in patients with and without a prolonged clotting time (nine
response to physical stress and trauma. This results in a studies, Tables I and III) (Suchman & Mushlin, 1986; Manning
shortening of the APTT, which may mask the detection of mild et al,1987; Burk et al, 1992; Kang et al, 1994; Houry et al,
haemophilia A and VWD. 1995; Myssiorek & Alvi, 1996; Howells et al, 1997; Gabriel

ª 2008 The Authors

Journal Compilation ª 2008 Blackwell Publishing Ltd, British Journal of Haematology, 140, 496–504 499
 500
Guideline

Table III. Predictive value and likelihood ratios for the value of clotting tests or bleeding history in predicting postoperative bleeding and bleeding rate for patients with abnormal and normal coagulation
tests.

PPV and LR+ of coagulation test for PPV and LR+ of bleeding history
postoperative bleeding (95% CI) for postoperative bleeding (95% CI) Absolute risk
difference for
bleeding rate
Bleeding rate Bleeding rate between patients
for patients for patients with and without 95% CI of absolute risk
with abnormal with normal abnormal difference for bleeding rate*
Reference PPV LR+ PPV LR+ coagulation test coagulation test coagulation test (upper limit, lower limit)

Gabriel et al (2000) 0Æ16 (0Æ08–0Æ28) 1Æ65 (0Æ82–3Æ30) 0Æ23 (0Æ06–0Æ54) 2Æ64 (0Æ73–9Æ48) 0Æ174 0Æ100 0Æ074 )0Æ014, 0Æ210
Houry et al (1995) 0Æ04 (0Æ03–0Æ07) 1Æ33 (0Æ91–1Æ93) 0Æ04 (0Æ03–0Æ06) 1Æ27 (0Æ99–1Æ64) 0Æ045 0Æ032 0Æ013 )0Æ005, 0Æ036
Burk et al (1992) 0Æ06 (0Æ01–0Æ23) 2Æ84 (0Æ70–11Æ47) n/a n/a 0Æ065 0Æ023 0Æ042 )0Æ012, 0Æ206
Asaf et al (2001)
PT 0Æ09 (0Æ05–0Æ16) 0Æ94 (0Æ56–1Æ57) n/a n/a 0Æ091 0Æ099 )0Æ008 )0Æ070, 0Æ069
APTT 0Æ11 (0Æ05–0Æ23) 1Æ18 (0Æ59–2Æ40) n/a n/a 0Æ115 0Æ095 0Æ020 )0Æ056, 0Æ138
Howells et al (1997) 0Æ03 (0Æ00–0Æ15) 0Æ95 (0Æ15–6Æ00) 0Æ13 (0Æ01–0Æ53) 4Æ7 (0Æ64–34Æ68) 0Æ026 0Æ027 )0Æ001 )0Æ043, 0Æ125
Myssiorek and Alvi (1996) 0Æ14 (0Æ03–0Æ44) 5Æ10 (1Æ18–21Æ96) n/a n/a 0Æ143 0Æ030 0Æ113 )0Æ006, 0Æ408
Kang et al (1994) 0Æ22 (0Æ09–0Æ43) 4Æ45 (1Æ86–10Æ63) n/a n/a 0Æ222 0Æ056 0Æ166 0Æ037, 0Æ372
Manning et al (1987) 0Æ03 (0Æ01–0Æ13) 0Æ95 (0Æ24–3Æ74) n/a n/a 0Æ035 0Æ036 )0Æ001 )0Æ034, 0Æ094
Suchman and Mushlin (1986) 0Æ03 (0Æ01–0Æ05) 2Æ08 (1Æ21–3Æ57) 0Æ02 (0Æ01–0Æ03) 5Æ04 (3Æ48–7Æ31) 0Æ026 0Æ010 0Æ016 0Æ001, 0Æ041

PPV, positive predictive value; LR+, likelihood ratio for a positive test; n/a, not available; CI, confidence interval.
*Newcombe (1998) after EB Wilson, 1927 (with continuity correction).
Significant difference at alpha £0Æ05

ª 2008 The Authors

Journal Compilation ª 2008 Blackwell Publishing Ltd, British Journal of Haematology, 140, 496–504
13652141, 2008, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2141.2007.06968.x by Nat Prov Indonesia, Wiley Online Library on [06/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 13652141, 2008, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2141.2007.06968.x by Nat Prov Indonesia, Wiley Online Library on [06/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Guideline

et al, 2000; Asaf et al, 2001). In these studies, coagulation post hoc power calculation. Finally, publication bias is not
testing was performed routinely in all patients. Two studies excluded by the review methodology employed, although
excluded patients on antithrombotics (Houry et al, 1995; this is unlikely to have been a significant factor as all
Howells et al, 1997) and three studies excluded patients with the published studies are negative from the point of view of
a known history of coagulopathy (Suchman & Mushlin, 1986; the clinical utility of coagulation tests. In addition, the
Manning et al,1987; Asaf et al, 2001) The PPV of a prolonged patients included are probably representative of the general
clotting time for postoperative bleeding ranged from 0Æ03 to population and overall conclusions are in keeping with our
0Æ22. The likelihood ratio for a positive test (LR+) ranged from current understanding of the limitations of coagulation
0Æ94 to 5Æ10 and when limited to the three prospective studies, testing.
the LR+ was low (range 1Æ33–2Æ84) with 95% confidence Based on the evidence, the practice of indiscriminate
intervals crossing 1Æ0 in all three studies. Moreover, the coagulation testing is not justifiable, at least for the popula-
postoperative bleeding rates in patients with and without tion of preoperative patients included in this systematic
a prolonged clotting time were inconsistent but similar review. Although some defend it as a means of avoiding
(Table III). litigation, it has been demonstrated that 30–95% of unex-
pected laboratory results from screening tests are either not
Paediatric tonsillectomy. Coagulation testing has been documented or not pursued further (Muskett & McGreevy,
considered especially important in paediatric surgical practice 1986; Johnson & Mortimer, 2002). Therefore, random
where patients may not have been exposed to any prior screening could potentially increase rather than reduce the
haemostatic challenge. Six of nine studies identified consisted risk of litigation. There is also a perception that coagulation
only of paediatric tonsillectomy patients and, in all six, tests are inexpensive. While this may be true for individual
coagulation tests were performed routinely in all patients tests, the cumulative cost is considerable, especially when the
independent of bleeding history (Manning et al,1987; Burk financial cost of consequential additional unnecessary tests
et al,1992; Kang et al, 1994; Howells et al, 1997; Gabriel et al, and delays to treatment are considered. We did not include
2000; Asaf et al, 2001).The PPV of a prolonged clotting time patients undergoing surgery conventionally associated with a
for postoperative bleeding ranged from 0Æ03 to 0Æ22 (Table III). higher risk of morbidity or mortality from bleeding compli-
The LR+ ranged from 0Æ94 to 4Æ45 and when limited to the two cations e.g. intracranial, neurosurgical or ophthamological
prospective studies, the LR+ was low (1Æ65 and 2Æ8) with surgery, as the few studies that have been published did not
confidence intervals that crossed 1Æ0 in both studies. In fulfil the inclusion criteria. Although some will argue that
addition, the postoperative bleeding rates in paediatric patients these patients justify screening regardless of their clinical
with and without a prolonged clotting time were similar history, the same disadvantage of routine screening will apply
(Table III). i.e. poor sensitivity and specificity. In the absence of good
quality evidence to guide practice, the authors propose that if
Testing before invasive procedures. Recently Segal and Dzik routine screening is practiced, robust mechanisms to follow-
(2005) performed an evidence-based review of the ability of up on the results of the tests with appropriate management
a prolonged PT/raised International Normalized Ratio (INR) intervention should be in place.
to predict excessive bleeding resulting from an invasive
procedure. They identified 25 studies (one clinical trial and
Recommendation
24 observational studies) in a variety of settings (Table II). The
results show that the bleeding rates for patients with and Routine coagulation testing to predict postoperative
without abnormal coagulation test results were similar in bleeding risk in unselected patients prior to surgery or
groups of patients undergoing bronchoscopy, central vein other invasive procedures is not recommended (Grade B,
cannulation, angiography, liver and kidney biopsy and Level III).
paracentesis. Risk difference was calculated for 14 studies
and showed little absolute difference (although the confidence
Predictive value of the bleeding history for postoperative
intervals were wide).
bleeding
We calculated the value of a positive bleeding history for the
Limitations
prediction of bleeding (four studies, Table III) (Suchman &
All the studies included are case-series; most were retro- Mushlin, 1986; Houry et al, 1995; Howells et al, 1997; Gabriel
spective and may suffer from selection bias, imperfect recall et al, 2000). The PPV of the bleeding history for postoperative
and incomplete case record documentation. The studies are bleeding ranged from 0Æ02 to 0Æ23. The LR+ ranged from 1Æ27
also heterogeneous in terms of inclusion criteria, confound- to 5Æ04 and, when limited to the two prospective studies
ing factors, definition of abnormal clotting test, methods (Houry et al,1995; Gabriel et al, 2000), the LR+ was low (1Æ27
used to extract the bleeding history and definition of and 2Æ64) with 95% confidence intervals that crossed one in
postoperative bleeding. Also, no study has sample size or both studies.

ª 2008 The Authors

Journal Compilation ª 2008 Blackwell Publishing Ltd, British Journal of Haematology, 140, 496–504 501
 13652141, 2008, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2141.2007.06968.x by Nat Prov Indonesia, Wiley Online Library on [06/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Guideline

Limitations are indications that a structured approach may be predic-

tive. Therefore there is insufficient evidence to conclude that
Only four informative studies were identified. Of these only
the bleeding history has no PV for postoperative bleeding.
two were prospective and different methods were used in
A bleeding history, including family history, evidence of
each study to elicit the bleeding history. Currently, there is
excessive post-traumatic or postsurgical bleeding and use of
no standardised approach that has been validated for use as
antithrombotic drugs should be taken in all patients prior to
a screening tool. Bleeding symptoms are subjective and up
surgery or invasive procedures. (Grade C, Level IV).
to 25% of healthy subjects describe common symptoms such
as excessive epistaxis, gum bleeding and postpartum haem-
orrhage but have normal laboratory test results (Sadler, Disclaimer
2003).
While the advice and information in these guidelines is
Although the evidence indicates that a poorly structured
believed to be true and accurate at the time of going to press,
bleeding history does not predict postoperative bleeding, it has
neither the authors, the British Society for Haematology nor
been demonstrated that the predictive power of the history for
the publishers accept any legal responsibility for the content of
presence of a bleeding disorder is dependent on the precise
these guidelines.
questions asked (Sramek et al, 1995). In a case–control study
comparing patients with a proven bleeding disorder with
healthy volunteers and using a standardised questionnaire, it References
was shown that the presence of a positive family history and Asaf, T., Reuveni, H., Yermiahu, T., Leiberman, A., Gurman, G., Porat,
bleeding after traumatic events (except parturition) identified A., Schlaeffer, P., Shifra, S. & Kapelushnik, J. (2001) The need for
subjects with a bleeding disorder. In contrast, some reported routine pre-operative coagulation screening tests (prothrombin time
symptoms were non-discriminatory including gum bleeds, PT/partial thromboplastin time PTT) for healthy children under-
epistaxis and blood in the urine or stool. It was concluded that going elective tonsillectomy and/or adenoidectomy. International
a structured interview is useful as a screening tool. However, Journal of Pediatric Otorhinolaryngology, 61, 217–222.
this questionnaire has not been evaluated in a preoperative Bruzzi, J.F., O’Connell, M.J., Thakore, H., O’Keane, C., Crowe, J. &
setting and further studies should address this. Further, Murray, J.G. (2002) Transjugular liver biopsy: assessment of safety
and efficacy of the Quick-Core biopsy needle. Abdominal Imaging,
Tosetto et al (2006) looked retrospectively at the utility of
27, 711–715.
a mucocutaneous bleeding score in predicting bleeding after
Burk, C.D., Miller, L., Handler, S.D. & Cohen, A.R. (1992) Preoper-
surgery or tooth extraction in patients with VWD. The results
ative history and coagulation screening in children undergoing
showed that clinical assessment was at least as effective as tonsillectomy. Pediatrics, 89, 691–695.
laboratory testing in predicting bleeding after tooth extraction Caturelli, E., Squillante, M.M., Andriulli, A., Siena, D.A., Cellerino, C.,
and superior in postsurgical bleeding. The authors are not De Luca, F., Marzano, M.A., Pompili, M. & Rapaccini, G.L. (1993)
aware of any structured bleeding history that has been Fine-needle liver biopsy in patients with severely impaired coagu-
validated prospectively in a large number of routine lation. Liver, 13, 270–273.
preoperative surgical patients. We recommend that the Choo, S.W., Do, Y.S., Park, K.B., Kim, S.H., Han, Y.H. & Choo, I.
bleeding history should comprise of two sections. The first (2000) Transjugular liver biopsy: modified Ross transseptal
section should include brief questions about bleeding needle versus quick-core biopsy needle. Abdominal Imaging, 25,
483–485.
symptoms, prior haemostatic challenges, family history and
Darcy, M.D., Kanterman, R.Y., Kleinhoffer, M.A., Vesely, T.M., Picus,
drug history. If the first section is positive for bleeding
D., Hicks, M.E. & Pilgram, T.K. (1996) Evaluation of coagulation
symptoms, a quantitative description of the symptoms should
tests as predictors of angiographic bleeding complications. Radiol-
be obtained. ogy, 198, 741–744.
By targeting a subgroup of patients with a positive bleeding Davis, C.L. & Chandler, W.L. (1995) Thromboelastography for the
history for further assessment and coagulation testing, it is prediction of bleeding after transplant renal biopsy. Journal of the
plausible that the PV of the combination of an abnormal American Society of Nephrology, 6, 1250–1255.
bleeding history and abnormal coagulation test may be higher Denzer, U., Helmreich-Becker, I., Galle, P.R. & Lohse, A.W. (2001)
for postintervention bleeding than either alone. Importantly, Safety and value of minilaparotomy in high risk patients. Z Gas-
this strategy would enable testing to be focused on the troenterology, 39, 11–14.
minority of subjects in whom there is reasonable suspicion of Doerfler, M.E., Kaufman, B. & Goldenberg, A.S. (1996) Central venous
catheter placement in patients with disorders of hemostasis. Chest,
the presence of a bleeding disorder.
110, 185–188.
Ewe, K. (1981) Bleeding after liver biopsy does not correlate with
Recommendation indices of peripheral coagulation. Digestive Diseases and Sciences, 26,
388–393.
Although the published data considered in this guideline Fisher, N.C. & Mutimer, D.J. (1999) Central venous cannulation in
indicate that an unstructured bleeding history is not a good patients with liver disease and coagulopathy–a prospective audit.
predictor of postoperative bleeding (Grade B, level III) there Intensive Care Medicine, 25, 481–485.

ª 2008 The Authors

502 Journal Compilation ª 2008 Blackwell Publishing Ltd, British Journal of Haematology, 140, 496–504
 13652141, 2008, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2141.2007.06968.x by Nat Prov Indonesia, Wiley Online Library on [06/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Guideline

Foster, P.F., Moore, L.R., Sankary, H.N., Hart, M.E., Ashmann, M.K. & Papatheodoridis, G.V., Patch, D., Watkinson, A., Tibballs, J. & Burr-
Williams, J.W. (1992) Central venous catheterization in patients oughs, A.K. (1999) Transjugular liver biopsy in the 1990s: a 2-year
with coagulopathy. Archives of Surgery, 127, 273–275. audit. Alimentary Pharmacology and Therapeutics, 13, 603–608.
Friedman, E.W. & Sussman, I.I. (1989) Safety of invasive procedures in Riley, S.A., Ellis, W.R., Irving, H.C., Lintott, D.J., Axon, A.T. & Los-
patients with the coagulopathy of liver disease. Clinical and Labo- owsky, M.S. (1984) Percutaneous liver biopsy with plugging of
ratory Haematology, 11, 199–204. needle track: a safe method for use in patients with impaired
Gabriel, P., Mazoit, X. & Ecoffey, C. (2000) Relationship between clinical coagulation. Lancet, 2, 436.
history, coagulation tests, and perioperative bleeding during tonsil- Sadler, J.E. (2003) Von Willebrand disease type 1: a diagnosis in search
lectomies in pediatrics. Journal of Clinical Anesthesia, 12, 288–291. of a disease. Blood, 101, 2089–2093.
Houry, S., Georgeac, C., Hay, J.M., Fingerhut, A. & Boudet, M.J. Sawyerr, A.M., McCormick, P.A., Tennyson, G.S., Chin, J., Dick, R.,
(1995) A prospective multicenter evaluation of preoperative Scheuer, P.J., Burroughs, A.K. & McIntyre, N. (1993) A comparison
hemostatic screening tests. The French Associations for Surgical of transjugular and plugged-percutaneous liver biopsy in patients
Research. American Journal of Surgery, 170, 19–23. with impaired coagulation. Journal of Hepatology, 17, 81–85.
Howells, R.C., Wax, M.K. & Ramadan, H.H. (1997) Value of pre- Segal, J.B. & Dzik, W.H. (2005) Paucity of studies to support that
operative prothrombin time/partial thromboplastin time as a abnormal coagulation test results predict bleeding in the setting of
predictor of postoperative hemorrhage in pediatric patients under- invasive procedures: an evidence-based review. Transfusion, 45,
going tonsillectomy. Otolaryngology and Head and Neck Surgery, 1413–1425.
117, 628–632. Smith, T.P., Presson, T.L., Heneghan, M.A. & Ryan, J.M. (2003)
Johnson, R.K. & Mortimer, A.J. (2002) Routine pre-operative blood Transjugular biopsy of the liver in pediatric and adult patients using
testing: is it necessary? Anaesthesia, 57, 914–917. an 18-gauge automated core biopsy needle: a retrospective review of
Kamphuisen, P.W., Wiersma, T.G., Mulder, C.J. & de Vries, R.A. 410 consecutive procedures. AJR. American Journal of Roentgenology,
(2002) Plugged-percutaneous liver biopsy in patients with impaired 180, 167–172.
coagulation and ascites. Pathophysiology of Haemostasis and Sramek, A., Eikenboom, J.C., Briet, E., Vandenbroucke, J.P. &
Thrombosis, 32, 190–193. Rosendaal, F.R. (1995) Usefulness of patient interview in bleeding
Kang, J., Brodsky, L., Danziger, I., Volk, M. & Stanievich, J. (1994) disorders. Archives of Internal Medicine, 155, 1409–1415.
Coagulation profile as a predictor for post-tonsillectomy and ade- Steadman, C., Teague, C., Harper, J., Hayes, P., Nathan, N., Harris, O.
noidectomy (T + A) hemorrhage. International Journal of Pediatric & Kerlin, P. (1988) Transjugular liver biopsy–an Australian
Otorhinolaryngology, 28, 157–165. experience. Australian and New Zealand Journal of Medicine, 18,
Kozak, E.A. & Brath, L.K. (1994) Do ‘‘screening’’ coagulation tests 836–840.
predict bleeding in patients undergoing fiberoptic bronchoscopy Suchman, A.L. & Mushlin, A.I. (1986) How well does the activated
with biopsy? Chest, 106, 703–705. partial thromboplastin time predict postoperative hemorrhage?
Lind, S.E. (1991) The bleeding time does not predict surgical bleeding. JAMA, 256, 750–753.
Blood, 77, 2547–2552. Thompson, B.C., Kingdon, E., Johnston, M., Tibballs, J., Watkinson,
MacDonald, L.A., Beohar, N., Wang, N.C., Nee, L., Chandwaney, R., A., Jarmulowicz, M., Burns, A., Sweny, P. & Wheeler, D.C. (2004)
Ricciardi, M.J., Benzuly, K.H., Meyers, S.N., Gheorghiada, M. & Transjugular kidney biopsy. American Journal of Kidney Diseases, 43,
Davidson, C.J. (2003) A comparison of arterial closure devices to 651–662.
manual compression in liver transplantation candidates undergo- Tobin, M.V. & Gilmore, I.T. (1989) Plugged liver biopsy in patients
ing coronary angiography. Journal of Invasive Cardiology, 15, with impaired coagulation. Digestive Diseases and Sciences, 34,
68–70. 13–15.
Manning, S.C., Beste, D., McBride, T. & Goldberg, A. (1987) An Tosetto, A., Rodeghiero, F., Castaman, G., Goodeve, A., Federici, A.B.,
assessment of preoperative coagulation screening for tonsillectomy Batlle, J., Meyer, D., Fressinaud, E., Mazurier, C., Goudemand, J.,
and adenoidectomy. International Journal of Pediatric Otorhinolar- Eikenboom, J., Schneppenheim, R., Budde, U., Ingerslev, J., Vorlova,
yngology, 13, 237–244. Z., Habart, D., Holmberg, L., Lethagen, S., Pasi, J., Hill, F. & Peake,
McVay, P.A. & Toy, P.T. (1990) Lack of increased bleeding after liver I. (2006) A quantitative analysis of bleeding symptoms in type 1 von
biopsy in patients with mild hemostatic abnormalities. American Willebrand disease:results from a multicenter European study
Journal of Clinical Pathology, 94, 747–753. (MCMDM-1 VWD). Journal of Thrombosis and Haemostasis, 4, 766–
McVay, P.A. & Toy, P.T. (1991) Lack of increased bleeding after 773.
paracentesis and thoracentesis in patients with mild coagulation Wilson, E.B. (1927) Probable inference, the law of succession, and
abnormalities. Transfusion, 31, 164–171. statistical inference. Journal of the American Statistical Association,
Muskett, A.D. & McGreevy, J.M. (1986) Rational preoperative evalu- 22, 209–212.
ation. Postgraduate Medical Journal, 62, 925–928. Wilson, N.V., Corne, J.M. & Given-Wilson, R.M. (1990) A critical
Myssiorek, D. & Alvi, A. (1996) Post-tonsillectomy hemorrhage: an appraisal of coagulation studies prior to transfemoral angiography.
assessment of risk factors. International Journal of Pediatric Otorhi- British Journal of Radiology, 63, 147–148.
nolaryngology, 37, 35–43. Zahreddine, I., Atassi, K., Fuhrman, C., Febvre, M., Maitre, B. &
Newcombe, R.G. (1998) Interval estimation for the difference between Housset, B. (2003) [Impact of prior biological assessment of coag-
independent proportions: comparison of eleven methods. Statistics ulation on the hemorrhagic risk of fiberoptic bronchoscopy]. Revue
in Medicine, 17, 873–890. des Maladies Respiratoires, 20, 341–346.

ª 2008 The Authors

Journal Compilation ª 2008 Blackwell Publishing Ltd, British Journal of Haematology, 140, 496–504 503
 13652141, 2008, 5, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2141.2007.06968.x by Nat Prov Indonesia, Wiley Online Library on [06/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Guideline

Appendix 1 Appendix 2

Rationale and methods for calculation of the predictive Classification of evidence levels
value, likelihood ratio and absolute risk difference Ia
Evidence obtained from meta-analysis of randomized
Predictive value. When a test is used for screening, it is not controlled trials.
known who has and who does not have disease. It is therefore Ib Evidence obtained from at least one randomized
necessary to consider the probability of the presence or absence controlled trial.
of disease given a positive or negative test result. To do this, IIa Evidence obtained from at least one well-designed
the positive predictive value (PPV) and the negative predictive controlled study without randomization.
value (NPV) of the test are calculated (see figure below). The IIb Evidence obtained from at least one other type of
PPV and NPV are dependent on disease prevalence. When well-designed quasi-experimental study*.
III Evidence obtained from well-designed non-experimental
a disease has low prevalence, PPV will be low. In other words,
descriptive studies, such as comparative studies,
if the population is at low risk of having the disease, a positive
correlation studies and case studies.
result is likely to be a false positive, even when the specificity
IV Evidence obtained from expert committee reports or
and sensitivity of the test is close to 100%. opinions and/or clinical experiences of respected authorities.

Disease

Present Absent Classification of grades of recommendations

Test Abnormal a (true b (false Positive predictive
A Requires at least one randomized controlled trial as part of
result positive) positive) value (PPV) a/(a+b)
a body of literature of overall good quality and consistency
Normal c (false d (true Negative predictive
addressing specific recommendation. (Evidence levels Ia, Ib).
negative) negative) value (NPV) d/(c+d)
B Requires the availability of well conducted clinical studies but no
Sensitivity Specificity
randomized clinical trials on the topic of recommendation.
a/(a+c) d/(b+d)
(Evidence levels IIa, IIb, III).
C Requires evidence obtained from expert committee reports or
Likelihood ratio. Another important indicator of the diagnostic opinions and/or clinical experiences of respected authorities.
strength of a test is the likelihood ratio (LR). The likelihood ratio of Indicates an absence of directly applicable clinical studies
a positive test (LR+) tells you how many times more likely a positive of good quality. (Evidence level IV).
test result will occur in a patient with the disease, as compared to
a patient without the disease. A LR of 1Æ0 means the test provides no
additional information while ratios above or below this increase or Evidence obtained from the literature searches should be assessed by
decrease the probability of disease. The product of the LR and pre-test the drafting group and recommendations formulated from this evi-
odds determines the post-test odds of disease. In general, LRs of greater dence. As in the summary, the recommendations need to be graded
than 10 generate large shifts in pre to post-test probability, while LRs of according to the strength of supporting evidence using levels and
1Æ0 to 3Æ0 are very weak. Conversely, LRs of less than 0Æ1 generate large grades of evidence outlined in Appendix 7 of the Procedure for
shifts in pre to post-test probability, while LRs of between 0Æ3 and 1Æ0 Guidelines commissioned by the BCSH (http://www.bcshguide-
are very weak. lines.com/process1.asp). If there are several possible options for
management, these should be enumerated and also linked to sup-
Sensitivity porting evidence.
LRþ ¼
1  Specificity *refers to a situation in which implementation of an intervention is
outwith the control of the investigators, but an opportunity exists to
Absolute risk difference. The absolute risk difference is the evaluate its effect.
arithmetic difference of the bleeding rate between those with
abnormal coagulation and those with normal coagulation tests i.e. it
is the event rate between the two comparison groups. An absolute risk
difference of zero indicates no difference between the two groups.
A risk difference that is greater than zero indicates that the coagulation
testing was effective in reducing the risk of that outcome.

ª 2008 The Authors

504 Journal Compilation ª 2008 Blackwell Publishing Ltd, British Journal of Haematology, 140, 496–504