Critical values are vital to safe clinical and laboratory practice. To address the lack of information on critical values in coagulation, pattern-of-practice surveys were distributed to members of the North American Specialized Coagulation Laboratory Association. More than 70% of respondents had critical values for commonly performed tests. Median values were as follows: prothrombin time, more than 37 seconds; international normalized ratio, more than 5; activated partial thromboplastin time, more than 100 seconds; and fibrinogen level, less than 100 mg/dL. Critical value reporting generated a significant workload, with up to 15% of these tests yielding critical results. The median time to report critical values was 7 minutes for inpatients. Despite the lack of guidelines surrounding critical values in coagulation, this survey confirms that laboratories have reasonable and uniform practices. It also provides critical value medians and ranges for a wide range of tests. Laboratories without critical values or in the process of reviewing their values may find this survey of their peers useful.
Critical values were first defined by Lundberg1 in 1972 as “pathophysiologic states at such variance with normal as to be life-threatening unless something is done promptly and for which some corrective action can be taken.” Since then, the recognition, documentation, and communication of critical values by clinical laboratories have been promoted as vital to patient safety and good laboratory practice. Critical value reporting has been included in external peer-comparison programs.2–7 It has also become a widely recognized patient safety goal. In its National Patient Safety Goals, the Joint Commission recommended that persons reporting critical values “…verify the complete order or test result by having the person receiving the information record and ‘read back’ the complete order or test result.” The Joint Commission also recommends that hospitals “… measure, assess, and, if needed, take action to improve the timeliness of reporting and the timeliness of receipt of critical tests and critical results and values by the responsible licensed caregiver.”2
Laboratory staff who perform coagulation tests are uniquely positioned to recognize and communicate critical values quickly and effectively to health care professionals in inpatient and outpatient settings at any time of day or night. At present, there are no published guidelines and policies on the appropriate critical values to use for coagulation parameters or the reporting of these values. Patterns-of-practice surveys have revealed that the critical values for activated partial thromboplastin time (aPTT), prothrombin time (PT), and fibrinogen are variable, and practices for other assays have not been published. The primary goal of our study was to determine if coagulation laboratories have similar critical values and how critical values for coagulation parameters are handled and reported by conducting 2 surveys of members of the North American Specialized Coagulation Laboratory Association (NASCOLA).
Materials and Methods
This study was conducted in accordance with NASCOLA requirements to anonymize participant identities for patterns-of-practice surveys. Two surveys were distributed electronically to coagulation laboratory members of NASCOLA. The first survey (containing 37 questions) was distributed in February 2009 to gather data on critical values (if any) for different coagulation parameters and to determine how laboratories had established these critical values, whether different critical values were used for special populations (eg, patients undergoing anticoagulation, neonates and children), laboratory practices for reporting critical values, and opinions on the potential for improvement and standardization of critical values for coagulation parameters. The second survey (containing 10 items) was distributed in July 2009 to gather more information on the critical values used for special populations and the laboratory workload associated with communicating critical values.
Data were analyzed in SPSS version 15.00 (SPSS, Chicago, IL), after anonymizing participant identities. Skipped responses were treated as missing data, and percentage responses were calculated based on the number of laboratories that answered individual questions.
Of the 69 NASCOLA member laboratories, 37 (54%) and 34 (49%), respectively, participated in the first and second surveys. The majority of participating laboratories (29/37 [78%]) indicated that they were accredited by federal and/or state or provincial licensing authorities. Most performed coagulation tests on inpatient (34/37 [92%]), outpatient (35/37 [95%]), and referred-in (27/37 [73%]) samples.
Critical Value Policies
Most participants (36/37 [97%]) reported that they had critical value policies. The sources of their critical value policies were as follows: local clinical opinions or consensus, 85% (28/33); published literature, 64% (21/33); survey of laboratory personnel, 24% (8/33); and unknown, 8% (3/33). Of the laboratories, 80% (28/35) had a formal system to review critical value policies, and most did this annually.
Critical Values for Coagulation Tests
Table 1 summarizes the critical values for the coagulation tests that laboratories performed. Most laboratories (>70%) had critical values for the common coagulation tests, such as PT/international normalized ratio (INR), aPTT, and fibrinogen level. The median critical values for these assays were as follows: PT, more than 37 seconds; INR, more than 5; aPTT, more than 100 seconds; and fibrinogen level, less than 100 mg/dL (Table 1). A minority of laboratories (3% to 30%) had critical values for other coagulation assays. The median critical value limit for heparin levels was more than 1 U/mL of anti-Xa activity for low-molecular-weight heparin (LMWH) and unfractionated heparin (UFH) levels (Table 1). For coagulation factor assays, the median critical limit was 10% to 15% of normal. Some laboratories (7/37 [19%]) had critical values for heparin-induced thrombocytopenia antibodies, mainly tested by enzyme-linked immunoassays. Few had critical values for thrombin times (n = 3), bleeding time (n = 1), the prothrombin gene mutation (n = 1), or the factor V Leiden mutation (n = 1). Few laboratories had critical values for special populations, including neonates (n = 5), pediatric patients (n = 3), obstetric patients (n = 1), and cardiac patients (n = 1).
The majority of laboratories (30/37 [81%]) had similar policies for reporting critical values for inpatients and outpatients, although some (5/37 [14%]) had different policies for each population. The majority of laboratories (28/37 [76%]) authorized the technologist who performed the test to report the critical value. Smaller numbers authorized reporting by a senior technologist or technical specialist (11/37 [30%]), laboratory physician (7/37 [19%]), or clerical staff (5/37 [14%]). The majority of laboratories required that the person notified of a critical value be the physician (30/37 [81%]) or nurse (27/37 [73%]) directly involved in the patient’s care or the physician who ordered the test (27/37 [73%]). A minority of laboratories allowed reporting of critical values to any other staff nurse (n = 14) or physician (eg, hematologist, laboratory director on call; n = 18), or unlicensed persons, such as clerical staff (n = 4) or medical students (n = 1).
All laboratories used telephone communication to report a critical value, with a minority supplementing notification by facsimile or by laboratory information system electronic broadcast/forced print functions. None of the laboratories relied solely on their laboratory information system to communicate a critical value. One site commented that all coagulation critical values were screened by a physician who decided whether a clinical consult should be performed. Almost all laboratories (35/36 [97%]) documented the receipt of critical value notifications, recording the identity of the person who communicated the result, the person who received the notification, and the time and date of communication. Most laboratories (34/36 [94%]) also documented the communication on the laboratory report. Five sites (5/36 [14%]) specifically mentioned that they required recipients of critical value communication to “read back” the information.
Participants were asked how they handled repeatedly critical values from the same patient. The majority (22/31 [71%]) communicated the initial critical value and all subsequent critical values, regardless of prior results, whereas 2 laboratories (6%) communicated the initial critical value but not repeated critical values. Three laboratories (10%) had other strategies such as reporting worsening values, values that were “grossly different” from previous values, and values that moved in and out of the critical range. Two laboratories (6%) commented that they reported repeated critical values after a set time had elapsed (eg, critical values occurring more than 12 hours after the last critical value was communicated). Two laboratories (6%) commented that they reported repeated results differently for different patient populations (eg, critical factor VIII activity levels were not reported for patients with known severe hemophilia A; critical levels for patients with cancer were reported every 7 days).
For laboratories serving areas that did not operate on a 24-hour cycle, critical value reporting after regular business hours was handled in a variety of ways. All 10 laboratories that responded to this question indicated that they attempted to contact the on-call physician or clinical service caring for the patient. If this person or service could not be contacted, most laboratories (7 [70%]) left a phone message for the ordering physician, while some communicated the results the next day (3 [30%]), faxed a critical value report (2 [20%]), required a physician to contact the patient directly (3 [30%]), or contacted the laboratory chief of service or designate (1 [10%]). No laboratories required laboratory staff to contact the patient directly.
Workload Associated With Critical Values
Participants were asked to estimate the percentage of reported tests that yielded a critical value for the 5 most commonly reported coagulation tests with critical values. The median (range) percentage of tests that were critical values were as follows: 1% (1%–15%) for INRs for anticoagulated patients; 1% (0%–15%) for aPTT tests for anticoagulated patients; 1% (0.3%–5%) for fibrinogen levels; and 1% (0%–3%) for UFH and LMWH anti-Xa heparin levels.
Participants stated 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. Fourteen laboratories provided information on their target or “benchmark” time to complete a critical value report. This ranged from immediate (facsimile/electronic notification) to 60 minutes for inpatients, outpatients, and referred-in samples. All respondents stated that if they could not complete the critical value report within the target time (eg, because they could not reach the most responsible caregiver), they continued trying until they could complete the report. The majority of participants (16/18 laboratories [89%]) indicated that they confirmed a critical value by repeated testing before reporting the value. The majority (20/29 [69%]) indicated that NASCOLA should form a working group to further address critical values for coagulation tests.
The NASCOLA surveys confirm that laboratories have reasonable and uniform practices surrounding critical values for coagulation tests. We found good consensus on critical values for the most commonly performed coagulation tests: PT/INR, aPTT, fibrinogen level, and UFH and LMWH levels. The reporting of critical values for commonly performed coagulation tests generates a significant workload, as laboratories noted that about 1% to 15% of all such tests yield critical values. Furthermore, considerable time is spent reporting critical values, ranging from a median of 7 minutes for inpatients to 20 minutes for referred-in samples. Laboratories are generally following Joint Commission guidelines surrounding critical value reporting, including communicating critical values to appropriate people who can respond in a timely manner and ensuring that critical value communication is prompt and unambiguous.2
Not surprisingly, in the absence of widely accepted guidelines, special coagulation laboratories have relied on local clinical opinions and consensus to develop critical value limits and policies. Some authors have suggested that when guidelines are lacking, laboratories should consider the patient population and solicit the input of clinicians to establish safe and clinically relevant critical values.2,8 Information shared by peer laboratories via the NASCOLA survey may be of additional value for laboratories setting critical value cutoffs or reviewing existing critical value policies.
The surveys provide critical value medians and ranges for a wide range of tests performed in the coagulation laboratory. More than 70% of laboratories have predefined critical values for commonly performed tests, such as the PT/INR, aPTT, and fibrinogen; there is reasonable consensus for these tests. Median critical values are as follows: PT, more than 37 seconds; INR, more than 5; aPTT, more than 100 seconds; and fibrinogen level, less than 100 mg/dL. These critical values are clinically relevant because they indicate an elevated bleeding risk from excessive anticoagulation or factor deficiency.9–21 Recent guidelines on anticoagulation indicate that the target INR for therapeutic warfarin should be between 2 and 3.5, depending on the patient population, with INR values greater than 5 considered supratherapeutic (the increased bleeding risk outweighs clinical benefit) in all patient populations.14 Such supratherapeutic values require urgent attention, such as holding warfarin doses for an interval or giving agents such as vitamin K, fresh frozen plasma, or prothrombin complex concentrates to reverse the effect of warfarin if the patient is bleeding.14 Similarly, an aPTT of more than 100 seconds in many centers would indicate an increased bleeding risk that might merit urgent action (eg, readjustment of intravenous heparin dosing).15 When setting high critical value limits for the aPTT, laboratories need to consider the aPTT results specific for their reagent and instrument combination. Each laboratory should establish its own therapeutic range for UFH treatment, ideally by establishing the relationship between the aPTT and an objective measure such as an UFH anti-Xa level.22,23
Few laboratories had set critical value limits for anti-Xa activity levels for LMWH and UFH. However, among the laboratories with critical values, the median value considered critical for both types of heparin anti-Xa levels was more than 1 U/mL. This is also clinically informative; there have been studies confirming that bleeding risk doubles in patients with anti-Xa activity of more than 1 U/mL vs anti-Xa activity of less than 1 U/mL.15,21
In our survey, few laboratories had predefined critical values for less commonly performed tests, such as coagulation factor levels, coagulation factor inhibitors, and protein C and S levels. However, abnormalities in these test results can point to clinically relevant disorders (eg, life-threatening purpura fulminans in a neonate with low protein C, factor VIII deficiency in hemophilia A). Therefore, laboratories may want to consider setting critical values for these tests, taking into account age-related changes in reference intervals.
It is important for laboratories to compare their own critical values with consensus values to ensure that they are choosing limits that meet the needs of their clinicians and patients. Laboratories dealing with unique populations should make every effort to consult with peer laboratories who perform testing on similar patient populations and with clinical and laboratory experts in the field.
The survey confirms that some laboratories tailor the handling of repeatedly critical test results to the patient population. The practice of notifying clinicians of each repeatedly critical value on the same patient can be beneficial in some circumstances, and it ensures that high acuity results “remain on the clinical radar.” However, repeatedly critical values, such as a factor VIII level in a person with known hemophilia, may not be a surprise to the clinician and may not warrant taking action. Reporting every instance of these critical values can be a poor use of laboratory time and resources, and it results in frequent interruptions for clinicians. The Joint Commission states that it is permissible for organizations to define critical results differently for repeated tests. Creating safe, clearly documented guidelines surrounding the reporting of repeatedly critical values requires communication between the laboratory and clinicians.
It is clear that laboratories want guidance and discussion on critical values in coagulation; the majority of survey participants thought that one of NASCOLA’s priorities should be to form a working group to address this issue. Coagulation laboratories serve diverse populations and have unique needs; therefore, designing a uniform critical value policy for all coagulation laboratories would be challenging. It would require consensus on which coagulation tests should have critical values, what these values should be, and how they should be communicated. However, consensus guidelines would be a valuable resource for laboratories. NASCOLA has had success observing laboratory patterns-of-practice and translating them into consensus guidelines.24–26 These surveys are an important preliminary step toward this goal.
Upon completion of this activity you will be able to:
list the coagulation tests that have widely accepted critical value limits.
describe strategies that can be used to develop critical value limits and policies.
describe strategies to ensure that critical value reporting is timely and unambiguous.
The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. 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.
The participation of NASCOLA members in this study is gratefully acknowledged. We acknowledge Elizabeth Van Cott, MD, Dorothy Adcock Funk, MD, FASCP, and Marlies Ledford-Kraemer, MBA, MT, for providing helpful feedback during this project.
Dr Pai is the recipient of a Hemostasis Research Fellowship from the Canadian Hemophilia Society (Montreal, Canada), and funding from the Hamilton Health Sciences New Investigator Fund (Hamilton). Dr Hayward is the recipient of a Heart and Stroke Career Investigator Award (Ottawa, Canada) and a Canada Research Chair in Molecular Hemostasis (Ottawa).
College of American Pathologists Conference XXXI on laboratory monitoring of anticoagulant therapy: the clinical use and laboratory monitoring of low-molecular-weight heparin, danaparoid, hirudin and related compounds, and argatroban. Arch Pathol Lab Med. 1998;122:799–807.
Development of North American consensus guidelines for medical laboratories that perform and interpret platelet function testing using light transmission aggregometry. Am J Clin Pathol. 2010;134:955–963.
Variability in clinical laboratory practice in testing for disorders of platelet function: results of two surveys of the North American Specialized Coagulation Laboratory Association. Thromb Haemost. 2005;93:549–553.