From Our Print Archives

Defining Critical Values in Coagulation

Critical values depend on several variables. Strategies to optimize coagulation results are explored.

View Comments (0)Print ArticleEmail Article
Vol. 21 • Issue 5 • Page 22


Coagulation results are not only dependent on the status of the patient, but the anticoagulant drugs that have been given to the patient as well as pre-analy-tical and analytical variables. So, what makes a coagulation value critical?

A critical value in the laboratory is defined as a result that requires a physician to be notified immediately.1 Critical values were first introduced to the laboratory by George Lundberg 40 years ago. The laboratory's approach to reporting critical values is required by CLIA, the Joint Commission and the College of American Pathologists (CAP).2

The outcome of a critical value in coagulation may result in the transfusion of blood or blood products or the adjustment of anticoagulant therapy. However, it has been well established that a coagulation result is also instrument/reagent dependent. Even though your method has been performed properly, the inter-laboratory variability makes for uncertainty in the interpretation of the results and adds to the difficulty in standardizing critical values in coagulation. As a result, defining critical values in coagulation varies among laboratories.

How to Define Critical Values

It is up to individual laboratories to assess their patient population and evaluate what values are critical by asking two questions:

1. Was the test critical to preserve life?

2. What low- and high-value cutoffs would require immediate intervention?

At present there are no coagulation guidelines that aid in determining critical values.

This leaves the task up to individual hospitals with input from laboratories and clinicians.

It is important for clinicians to define their patient population and for laboratories to understand how their reagents and instruments perform. For example, do they use heparin sensitive reagents for APTT testing, which will produce longer APTTs on patients with heparin? Is their reagent lupus sensitive, and will APTTs be prolonged in patients that have lupus or lupus-like inhibitors? What is the ISI of the PT reagent? Will the PT in seconds be sensitive to small adjustments in warfarin therapy? Also, how sensitive or insensitive are the laboratory's reagents to factor deficiencies? Does a normal PT and APTT represent a minimum of 30-35% of all factor levels, or are the reagents insensitive to some factors, and patients with normal PT and APTTs may bleed? A good understanding of these issues will help in defining critical values.

Consensus Among Laboratories

A hospital system comprised of six different sites undertook a project to standardize critical value reporting. They were able to agree on the classification of 85 tests; however, only two tests, the INR and APTT, stayed as hospital-specific critical values. Pathologists and clinicians could not agree. Community hospital clinicians felt the critical values were set too high; physicians at the teaching hospitals felt the values were too low. The patient population, severity of disease, anticoagulation and residency support appeared to be major contributors to this discordance.3

Values may be set differently depending on patient populations. An emergency room may set values lower than inpatients, since these patients are supposed to be normal, whereas surgery patients may have a different set of critical values based on the type of surgery, while inpatient results may vary due to anticoagulant therapy.4

A survey was conducted by the North American Specialized Coagulation Laboratory Association (NASCOLA) in which they found that critical value policies for coagulation testing were in place for 97% of the laboratories surveyed. Seventy percent of laboratories had critical values in place for the PT/INR, aPTT and fibrinogen. Median critical values for the PT was >37 seconds; aPTT > 100 seconds and fibrinogen levels <100mg/dL.5

Critical values were set at >1 U/mL of anti-Xa activity for both unfractionated and low molecular levels. Some laboratories had critical values for coagulation factor assays while the median critical limit was 10% to 15% of normal.5

The INR value seems to be the most commonly established critical value in coagulation. A CAP Q-PROBE was conducted to look at laboratories' utilization of the INR. Q-PROBE is an external peer-comparison program that addresses process, outcome and structure-oriented quality assurance issues to establish benchmarks to improve performance.

In a CAP Q-probe it was demonstrated that most laboratories use an INR in the range of four and five as critical values, with other laboratories using ranges from 2.6 all the way up to 10. In addition, for point-of-care (POC) testing, 23% of laboratory participants set an INR of 6.1 or higher for checking whether INR values produced by a POC device should be confirmed by a value determined by the central lab testing method.6

As well, in a CAP survey in which almost 5,000 laboratories participated INR critical values were found to be between 4-6 for 77%, but 23% used values above or below that range, with some using an INR value of 10 as critical. In addition, up to 13% reported a maximum INR of 20 or higher.6

Critical Values and Reducing TAT

It is important for laboratories to evaluate and track the process of notifying physicians of critical value results. For example, many laboratories require critical values to be repeated. If they fall within established repeat criteria the clinician and/or nurse is notified. The process of repeating a sample, finding the tube, repeating the test and comparing results can take over 20 minutes.3

The results of the NASCOLA survey revealed that the reporting of a critical value took a median (range) time of 7 minutes (2-60 minutes) for inpatients, 10 minutes (2-75 minutes) for outpatients, and 20 minutes (5-150 minutes) for referred-in samples and 89% of the laboratories repeated the test prior to reporting the critical value.5

Having different critical values for different patient populations may help to decrease turnaround time by knowing what results might be expected by a particular patient population. Emergency room patients may have a larger normal population so their values may be different than a cardiac care unit in which most patients may be on anticoagulants that will prolong tests. Laboratories may also choose to report a preliminary result since physicians may act on an initial result if clinically indicated.3

Donna Castellone is clinical science manager, Roche Diagnostics.


1. Critical Values Reporting QT10 Dec 2008, Q track

2. Lundberg GD. When to panic over an abnormal values. MLO 1972:4; 47-54.

3. Kranc MC, Hartwell G, Kiely SC, Silverman JF. Standardization, integration of critical tests, values. ADVANCE for the Administrators of the Laboratory 2010;19(10): 22.

4. Jenkins JJ II, Crawford JM, Bissell MG. Studying critical values: Adverse event identification following a critical laboratory values study at the Ohio State University Medical Center. American Journal of Clinical Pathology 2007;128(4):604-609.

5. Pai M, Moffat KA, Plumhoff E, Hayward CP. Results of a survey of the North American specialized coagulation laboratory association. American Journal of Clinical Pathology 2011;136(6):836-41.

6. Lusky K. INR practice gaps found in Q-Probes, March 2011, CAP Today.

7. McGilvray ID, Rotstein OD. Assessment of coagulation in surgical critical care patients In: Surgical Treatment: Evidence-Based and Problem-Oriented. Holzheimer RG, Mannick JA, editors. Munich: Zuckschwerdt; 2001.


Email: *

Email, first name, comment and security code are required fields; all other fields are optional. With the exception of email, any information you provide will be displayed with your comment.

First * Last
Title Field Facility
City State

Comments: *
To prevent comment spam, please type the code you see below into the code field before submitting your comment. If you cannot read the numbers in the below image, reload the page to generate a new one.

Enter the security code below: *

Fields marked with an * are required.