Pathology Consultation / S����󰁣����󰁬 A󰁴��󰁣��󰁥

Pathology Consultation on Reporting of Critical Values

Jonathan R. Genzen, MD, PhD,1 and Christopher A. Tormey, MD,2 for the Education Committee
of the Academy of Clinical Laboratory Physicians and Scientists

Key Words: Pathology consultation; Critical values; Critical results; Panic values; Laboratory results; Results reporting

DOI: 10.1309/AJCP9IZT7BMBCJRS

C
Upon completion of this activity you will be able to: The ASCP is accredited by the Accreditation Council for Continuing M
• define the terms “critical value” and “critical diagnosis.”
• examine obstacles associated with reporting critical values and
Medical Education to provide continuing medical education for physicians.
The ASCP designates this educational activity for a maximum of 1 AMA
E
diagnoses in a laboratory setting. PRA Category 1 Credit ™ per article. This activity qualifies as an American
• discuss practical solutions for problems associated with critical value Board of Pathology Maintenance of Certification Part II Self-Assessment
reporting. Module. A
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 643. Exam is located at www.ascp.org/ajcpcme.
S
Ab s t r a c t

Among the most important functions of a Case Scenario S

pathology or laboratory medicine service is the clear, A patient with atrial fibrillation who is receiving
accurate,
/
warfa- rin has an afternoon cardiology appointment for
and rapid communication of critical test results oD
routine care
(critical
and anticoagulant monitoring. A basic metabolic profile
w
n
o
M
values) to patient care providers. Pathologists and and l
o
laboratory professionals are often confronted with l
prothrombin time are ordered. The specimen is transported a
d
e
many obstacles in the reporting of such critical to the laboratory by courier, and laboratory testing is d
f
values, including establishing clinically relevant completed at 7:30 PM. All values are within normal limits r
criteria for critical values, resolving difficulties in except for an elevated potassium level (K+) of 6.9 mEq/L r
f
locating an ordering provider when a critical value (6.9 mmol/L; reference range, 3.2-5.2 mEq/L [3.2-5.2 mo
is obtained, h
mmol/L]) and a prothrombin time of 64.7 seconds (reference t
and ensuring that the provider understands the range, 11.1-13.2 seconds), corresponding to an international p
:
severity and implications of a critical result when he normalized ratio of 7.4. These results qualify as critical /
or she values by your clinical laboratory policy, and the laboratory /
a
has questions. This article presents a hypothetical (yet technologist attempts to contact the ordering clinician by jj
fairly common) clinical case scenario regarding c
telephone. Calls to the phy- sician’s office are not c
critical values and then provides an up-to-date forwarded to an answering service or
.
po
discussion and
covering clinician, but rather directed to an office answering o
x
review of the literature on the reporting of machine. The ordering physician does not respond to o
f
critical results. pages
d
r
or telephone calls made to the contact numbers listed in the j
o
hospital telephone directory or laboratory information o
u
system (LIS). The laboratory technologist contacts the on- n
r
call pathol- ogy resident and asks for assistance.
n
a
sl
o.
o
g
r
Questions /
b
y
g
u
1. What are laboratory critical values? e
s
ts
2. What are the requirements for critical value reporting? o
n
3. How should clinical laborattories establish critical D
e
c
e
value lists and determine appropriate thresholds? m
b
e
4. Who should make and receive critical r
3
0
value notifications? ,
2
0
5. How might critical value reporting be improved?
61

© American Society for Clinical Am J Clin Pathol 2011;135:505-513 505

Pathology 505 DOI: 10.1309/AJCP9IZT7BMBCJRS 505
Genzen and Tormey / C󰁲��󰁴��󰁣��󰁬 V��������󰁳

6. What are the responsibilities of pathologists available online.4 Of note, the read-back requirement for be
and laboratory directors in the critical value results conveyed by telephone is ca
process? being changed to a formal standard. Read-back is us
imperative e
sig
nif
Background ica
Lundberg 1 first outlined the fundamental components nt
of critical value reporting in a Medical Laboratory Observer err
article, describing critical laboratory values as “values or
which reflect pathophysiological derangements at such rat
variance with normal as to be life threatening if therapy es
is not instituted immediately.” They have more recently ha
been described as “laboratory results that indicate a life- ve
threaten- ing situation for the patient. Because of their be
en
critical nature, urgent notification of a critical value to the
de
appropriate healthcare professional is necessary.”2
tec
Alternative terms for critical values include critical results,
te
panic values, and alert values. The term panic values
d
carries a suggestion of emo- tional stress and runs against
in
the thoughtful and organized process of communicating
th
important information clearly. Its use is therefore
e
discouraged.
pr
Laboratories are required by numerous regulatory agen- oc
cies to develop and put into practice critical value policies. 3- es
6
s
Although the content of these policies varies according of
to
institutional needs, the core components are often quite tel
simi- lar. This article begins by describing the current ep
regulatory requirements for critical value reporting. This ho
ne
information will be followed by a detailed analysis of the
res
fundamental components of critical value notification.
ult
Aspects of critical value reporting that have been evaluated
co
in the literature are emphasized, as are current technological
m
advances that may change the way in which critical value
m
reporting takes place.
un
ica
tio
Regulations ns.
7,8
Critical value reporting is required by a variety of laws,
A
regulations, and accreditation programs. In the United
dd
States, the Clinical Laboratory Improvement Amendments iti
of 1988 (CLIA ’88) include requirements on critical value on
reporting. Current regulations specify that the laboratory al
manual must address critical values (when applicable to test sta
procedures), along with protocols for reporting critical te,
value results [see pr
§493.1251 (b; 11 and 13); and §493.1291 (g)].6 The ov
Joint
in
Commission (TJC) National Patient Safety Goals also cia
address critical value reporting. Specifically, Goal 2— l,
improve the effectiveness of communication among an
caregivers—includes “report critical results of tests and d
diagnostic procedures in a timely basis.” This goal and its loc
specific performance elements were updated in 2010 and are
al regulations regarding critical value reporting may also exist determination is the responsibility of the laboratory director,
and can be relevant to an individual laboratory’s performance it
requirements. should be made in communication with the clinicians who
Critical value reporting is addressed by the College use
of American Pathologists (CAP) Laboratory Accreditation
Program as part of several checklist components.3 These
items delineate the specific requirements for critical value pro-
cedures, including documentation of reporting and read-back
of verbally communicated results. Finally, the International
Organization for Standardization also includes critical value
reporting in its clinical laboratory standard ISO 15189:2007. 5

Establishing a Critical Values List

Although there are many regulations specifying that labo- D
o
ratories must define and communicate critical values, it may w
n
o
seem surprising that regulations do not state which laboratory l
o
l
tests require critical value limits and notification. Indeed, indi- a
d
e
vidual clinical laboratories face unique challenges that reflect d
f
institutional organization, clinical demand, patient population, r
instrumentation, and staffing. Such variations have hindered r
f
the development of universal standards for critical value mo
h
reporting across laboratories. t
The idea of a universal critical value list is appealing p
:
to many laboratorians and clinicians. For example, many /
clinicians would likely consider a sodium (Na+) level of /
a
168 mEq/L (168 mmol/L) a “critical” value regardless of jj
c
which laboratory performs the test. Indeed, the practice of c
assigning the laboratory director responsibility for creating .
po
and refining the critical value list has led to similar overall
o
x
inclusion of tests between laboratories without there being a o
f
universal mandate or requirement. As an example, virtually d
r
j
all laboratories include Na+ on their critical value list pre- o
cisely because it is important for patient care. Furthermore, o
u
not communicating a critically elevated Na+ level could r
n

have medicolegal ramifications if an adverse clinical out- n

a
come occurred. Defining (and then mandating) a universal sl

set of thresholds for tests, however, would be a daunting task o.

o
g
given the scarcity of outcomes-based data on critical value r
/
b
thresholds. Inherent variability in assay-specific reference y
g
u
intervals between institutions is also a complicating factor. e
s
ts
o
An individual laboratory director can account for this vari- n
D
ability by defining critical ranges consistent with his or her e
c
e
own assays and instrumentation. m
b
e
How should a laboratory determine which tests to include r
3
0
,
on a critical value list? Moreover, how should the critical high 2
0
and low thresholds be established? While ultimately, this 6
1

506 Am J Clin Pathol 2011;135:505-513 © American Society for Clinical Pathology

506 DOI: 10.1309/AJCP9IZT7BMBCJRS
 Pathology Consultation / S����󰁣����󰁬 A󰁴��󰁣��󰁥

laboratory services, as well as with a medical review board or LIS will notify the laboratory staff (usually the perform-
3,9
of the institution, if applicable. This task may include ing technologist) of the critical value. Laboratory policies
meeting with relevant physicians, medical and surgical must clearly indicate whether the assay should be verified
section chiefs, hospital administrators, and/or nurse and/or repeated before reporting and, if so, within what time
managers to discuss crit- ical value policies and to determine frame. Repeat testing is not feasible in many circumstances
if there are any tests that should be included (or omitted) (eg, blood culture results), and ongoing improvements in
and whether any thresholds should be adjusted according to laboratory assays may decrease the clinical usefulness of
clinical needs. routine repeat testing before reporting. This is a topic of
Not every laboratory test should have critical values continued clinical interest and debate.
associated with it. Critical value lists are, by nature, limited The National Patient Safety Goals state that laboratory
to not hinder the clinical effectiveness of notification.1 procedures must indicate “by whom and to whom” critical
Critical lists that are too inclusive (or that have critical value results are reported, as well as “the acceptable length of time
thresh- olds that require excessive notification) place an between the availability and reporting of critical results.”4
unnecessary burden on laboratory staff. Such lists annoy Documentation is required. The CAP checklist (component
clinicians, foster a negative attitude toward important GEN.41330) specifies what information must be document-
laboratory services, and, most important, provide uncertain ed during critical value notifications, including “date, time, D
o
additional benefit to patient care. At the other extreme, lists 3 w
responsible laboratory individual, [and] person notified.” n
o
that are too exclusive (or with thresholds that are too high or Laboratory personnel who perform the actual tests are l
o
l
low) might not prevent adverse clinical outcomes, as a delay currently responsible for making the vast majority of critical a
d
e
in the recognition of life- threatening laboratory results by value notifications. A 2008 survey of 121 institutions found d
f
clinicians can be disastrous. A balance must be achieved. that approximately 90% of calls are made by medical tech- r
The best place to start when establishing or modifying nologists/technicians, 1% are made by client services or call r
f
critical value lists is by comparison with previously pub- center staff, and 9% are made by a combination of the two. mo
h
lished lists, practice parameters, and consensus 26 t
documents A 2002 survey of 623 institutions showed similar results, p
:
because these sources have been refined with the benefit of with some differences between inpatient and outpatient /
2,9- 13 /
time, institutional comparison, and clinical performance. notification. That study recommended that critical value a
18
notifications should be made by one of the “team members” jj
Several published studies from CAP (Q-Probes and Q- c
involved in performing the procedure.13 A separate 2008 c
Tracks) have compared critical value reporting across survey of laboratory professionals and pathologists (at >350 .
po
hundreds of institutions and are a valuable resource for hospitals) revealed that nearly 18% of respondents were o
critical value policy assessment. The 1997 American using a call center for critical value notifications. 27
x
o
f
Society for Clinical Pathology “Critical Values Practice The workflow benefits of centralized call centers (and d
r
Parameter” (published in the Journal) is another having a laboratory technologist or someone with laboratory j
o
outstanding resource.9 ❚Table 1❚ and expertise involved in the call center mechanism) are well o
u
2,3,9,13,14,16-21
❚Table 2❚ present lists of common tests that described in the aforementioned survey and a separate 2008 r
n
fre- CAP Today n
quently have critical values defined. Although most 27,28 a
sl
published critical value lists do not include blood bank o.
o
g
testing, we have included a number of transfusion r
/
medicine–related scenarios b
y
g
u
e
s
ts
o
n
D
e
c
e
m
b
e
r
3
0
,
2
0
61
that may benefit from rapid communication and feature article. Laboratories face an ever-increasing
discussion
with a responsible clinician. lists) online, facilitating comparison of lists between peer
It should be noted that most published reports focus laboratories.22-25

on critical value notifications in general laboratory testing,

although several studies of critical diagnoses in surgical
pathology and cytology have recently been published and
will be discussed later in this article. Finally, many institu- Critical Value Notification Procedures
tions place their laboratory policies (including critical value
 The initial step in the critical value communication dilemma in critical value notification—the overall
volume
process involves identification of an abnormal result by
of laboratory testing is increasing, but a continued shortage
someone in the laboratory.9 For automated assays, the
in the number of laboratory professionals means that fewer
instrument, middle ware,
people are expected to do more. Shifting the task of critical
value notification away from laboratory technologists may
be inevitable at many institutions.
Indeed, several hospitals have implemented the use of
automated notification systems for critical value reporting.
At one institution, critical values transmitted from the LIS
to a hospital clinical information system trigger the genera-
tion of text messages directed to the responsible clinician’s
mobile phone and computer.15 If the clinician does not
confirm receipt in the clinical information system
within
60 minutes, results are communicated by telephone. This
approach improved the speed of communication and
allowed
for full electronic documentation of critical value
reporting.

© American Society for Clinical Am J Clin Pathol 2011;135:505-513 507

Pathology 507 DOI: 10.1309/AJCP9IZT7BMBCJRS 507
Genzen and Tormey / C󰁲��󰁴��󰁣��󰁬 V��������󰁳

❚Table
1❚
Examples of Possible Critical Laboratory Values for Chemistry, Hematology/Coagulation, and Pediatric-Specific Ranges*

Low Threshold High Threshold

Chemistry
Blood urea nitrogen, mg/dL (mmol/L) — 100 (35.7)
Calcium, mg/dL (mmol/L)
Total 6.5 (1.6) 13 (3.3)
Ionized 3.2 (0.8) 6.2 (1.6)
Carbon dioxide, total, mEq/L (mmol/L) 10 (10) 40 (40)
Chloride, mEq/L (mmol/L) 75 (75) 125 (125)
Creatinine, mg/dL (µmol/L) — 6.0 (530.4)
Glucose, mg/dL (mmol/L) 45 (2.5) 450 (25.0)
Glucose, CSF, mg/dL (mmol/L) 40 (2.2) 200 (11.1)
Lactate, mg/dL (mmol/L) — 30.6 (3.4)
Magnesium, mg/dL (mmol/L) 1.0 (0.4) 4.9 (2.0)
Osmolality, mOsm/kg (mmol/kg) 250 (250) 325 (325)
Phosphate, mg/dL (mmol/L) 1.0 (0.3) 9.0 (2.9)
PO2, mm Hg (kPa) 40 (5.3) —
PCO2, mm Hg (kPa) 20 (2.7) 70 (9.3) D
o
w
pH 7.2 7.6 n
Potassium, mEq/L (mmol/L) 2.8 (2.8) 6.2 (6.2) o
l
Sodium, mEq/L (mmol/L) 120 (120) 160 (160) o
l
Troponin I or T — See comment† a
d
Uric acid, mg/dL (µmol/L) — 13 (773) e
d
Hematology/coagulation f
Prothrombin time, s — 30‡ r
Partial thromboplastin time, s — 80 r
f
Fibrinogen, mg/dL (µmol/L) 90 (2.6) 800 (23.5)
mo
Hemoglobin, g/dL (g/L) 7 (70) 20 (200) h
Hematocrit, % (proportion of 1) 20 (0.2) 60 (0.6) t
Platelet count, × 103/µL (× 109/L) 40 (40) 1,000 (1,000) p
:
WBC count, × 103/µL (× 109/L) 2 (2) 40 (40)
Blasts — First /
or sterile body fluid) observation /
Organisms/parasites detected on smear review (CSF, blood, — First
observation a
anemia (schistocytes and low platelet jj
Smear
count)review suggestive of microangiopathic hemolytic — First c
observation c
Urinalysis (pathologic crystals) — First .
po
WBC count, CSF (cells/µL) — observation
5
Pediatric-specific ranges o
x
Ammonia, µg/dL (µmol/L) — 154 (110.0) o
Bilirubin, neonatal, mg/dL (µmol/L) — 15 (256.5) f
Blood urea nitrogen, mg/dL (mmol/L) — 55 (19.6) d
r
Creatinine, mg/dL (µmol/L) — 3.8 (335.9) j
o
Glucose, neonatal, mg/dL (mmol/L) 30 (1.7) 325 (18.0) o
u
PO2, neonatal, mm Hg (kPa) 40 (5.3) 100 (13.3)
n
PCO2, neonatal, mm Hg (kPa) 20 (2.7) 63 (8.4) r
Potassium, neonatal, mEq/L (mmol/L) 2.8 (2.8) 7.8 (7.8)
n
Sodium, pediatric, mEq/L (mmol/L) 121 (121) 156 (156) a
sl
Hemoglobin, neonatal, g/dL (g/L) 10 (100) 22 (220)
o.
Hematocrit, neonatal, % (proportion of 1) 33 (0.33) 71 (0.71) o
g
Platelet count, × 103/µL (× 109/L) 50 (50) 900 (900) r
/
CSF b
y
Glucose, mg/dL (mmol/L) 30 (1.7) 200 (11.1) g
u
Protein, mg/dL (g/L) — 190 (1.9) e
s
ts
WBC count, cells/µL o
n
0-1 y — 30 D
1-4 y — 20 e
c
5-17 y — 10 e
m
b
e
r
3
0
,
2
0
6
1
CSF, cerebrospinal fluid.
*
Information in this table is meant as a starting point for the evaluation of a laboratory’s critical value list and not as a strict guideline applicable to all clinical scenarios.
The table incorporates critical value data from the literature, particularly the outstanding reports from Kost,16-18,20,21 the College of American Pathologists Q-Probes
studies, 13,14 the ASCP Practice Parameter,9 a pediatric study by Gong and Adeli, 12 and critical values included in the Tietz Textbook of Clinical Chemistry and Molecular
Diagnostics.2
Système International (SI) conversion factors can be found in textbooks.2 In many cases, SI values have been rounded to 1 decimal point for the purposes of clarity. Because
the table integrates information from multiple studies, tests (and ranges) may vary from individual sources and may not be representative of the original authors’ opinions.
The table is also not inclusive of all tests that may be considered critical. For example, it does not include toxicology and therapeutic drug monitoring owing to the diversity
of tests

† p e rf o r m e d b e tw e e n i n d i id u al i n s t itu t io n s .
F i rs t r e su l t o b s er v e d a b o ve a n a s s a y o r la b o ratory-defined cutoff for myocardial infarction. A discussion of one laboratory’s
experience in establishing a troponin critical value cutoff (in collaboration with the emergency department) can be found in a 2008 CAP Today feature article.19
‡
An international normalized ratio equivalent (eg, >5) may also be used.

508 Am J Clin Pathol 2011;135:505-513 © American Society for Clinical Pathology

508 DOI: 10.1309/AJCP9IZT7BMBCJRS
 Pathology Consultation / S����󰁣����󰁬 A󰁴��󰁣��󰁥

❚Table
2❚
Examples of Possible Critical Laboratory Values for Microbiology and Transfusion Medicine*

Microbiology
Positive Gram, acid-fast bacillus, or mycology stains of smears (CSF, blood, or sterile body
fluid) Positive blood cultures
Positive CSF cultures
Positive sterile body fluid cultures
Positive stool culture for select organisms
Positive bacterial antigen tests
Transfusion medicine†
Gross, visible hemolysis in a post–transfusion reaction specimen
Positive direct antiglobulin test (IgG and/or C3) in a post–transfusion reaction specimen
Evidence of crossmatch incompatibility with a post–transfusion reaction specimen
Discrepancy in identifiers noted on a blood product label, tag, or container during transfusion reaction evaluation
Positive blood cultures from a unit implicated in a transfusion reaction
Discovery of a new alloantibody in a patient undergoing surgery

CSF, cerebrospinal fluid.

*
Information in this table is meant as a starting point for the evaluation of a laboratory’s critical value list and not as a strict guideline applicable to all clinical scenarios.
†
We have included possible transfusion medicine scenarios that would benefit from prompt communication with a responsible clinician, although most published critical
value lists do not specifically include blood bank testing. D
o
w
n
o
l
o
l
a
d
e
d
f
r
r
f
mo
h
t
At another institution, an automated paging system was To whom should critical values be reported? The cli
developed for critical value notification.11 In that program, rationale stated in the National Patient Safety Goals states ni
critical values transmitted from the LIS generate a page con- that results are to be conveyed to the “responsible cal
taining the patient name, medical record number, collection licensed caregiver.” 4 The pe
time, critical result, and reference range. The clinician must CAP checklists describe notification to a “physician (or rs
confirm receipt of the critical value by dialing a phone other on
number listed in the message. If the clinician does not ne
respond within l
10 minutes (or rejects the notification), the call is escalated res
to a trained group of operators who proceed with telephone po
noti- fication. Implementation of that system increased nsi
documenta- tion of critical value receipt by physicians and bl
decreased the median time for notification. Several other e
studies have also evaluated the role of automated paging for
systems in critical value reporting.29,30 pa
tie
It should be emphasized that automated solutions
should allow for an escalation policy (see the next section) nt
to ensure
car
communication of critical results when clinicians do not
e)”
10
acknowledge receipt. Laboratory contact information an
should also be available so that clinicians with additional d
questions can ask a laboratory professional or medical the
director as appro- priate. Patient privacy requirements “a
should also be considered with automated solutions
pp
because data conceivably might be transmitted and stored
ro-
on nonencrypted devices. Finally, device compatibility with
pri
alphanumeric characters (par- ticularly units) and character
ate
limits should also be evaluated because an inaccurate or
cli
incomplete notification could lead to medical error and
ni
adverse clinical outcome.
cal
 :
individual.”3 CLIA refers to “the individual or entity to ordering and/or covering physicians. According to most p
requesting the test and, if applicable, the individual regulatory agencies, this would also be an acceptable /
/
responsible for using the test results” [§493.1291 practice as long as there is documentation that the critical a
(g)].6 value was then conveyed by the nurse to the ordering
jj
c
A 2007 survey of 163 clinical laboratories asked the ques- physician and/or licensed caregiver. c
.
tion “Who can receive critical values?” for the inpatient and As an alternative, some hospital networks have adopted po

outpatient settings.14 As expected, answers from virtually all a policy of reporting all critical values generated from o
x
facilities included any licensed caregiver, ordering physician, the o
f
on-call physician, or resident.14 Many laboratories, however, d
r
j
also permitted administrative personnel (ward clerks or recep- o
o
tionists) to accept critical values for outpatients (48%) and u
n
inpatients (27%). A separate 2008 study showed that almost r
18% of institutions authorized release to other staff (eg, ward n
a
26 sl
clerks and/or unit secretaries). In that study, the calls made
o.
o
g
to “other providers” occurred slightly more quickly than those r
made to licensed caregivers. Any timesaving was lost, how- /
b
y
ever, when factoring in the subsequent time it took this other g
u
e
s
provider to then contact a licensed caregiver. The “authorized ts
o
n
agent” approach to critical value notification (calling someone D
e
whom a licensed caregiver specifies can receive critical value c
e
m
notifications but is not necessarily capable or authorized to act b
e
r
on them independently) should be discouraged. Many facili- 3
0
,
2
ties allow for reporting of critical values directly to licensed 0
nurses, who are then responsible for conveying these results 6
1

© American Society for Clinical Am J Clin Pathol 2011;135:505-513 509

Pathology 509 DOI: 10.1309/AJCP9IZT7BMBCJRS 509
Genzen and Tormey / C󰁲��󰁴��󰁣��󰁬 V��������󰁳

outpatient setting during “off” hours to a hospital emergency about a patient’s medical history and the potential
department (ED) or triage center. For example, the standard ramifica- tions of the critical value.
operating procedure at many Veterans Administration hospi- We ultimately advocate for active involvement of a
tals is to report critical results to an ED attending pathology resident, an attending pathologist, and/or a
physician, who can then decide whether to act on these medical director in difficult-to-convey critical value calls. It
results. This approach works well for Veterans has been our experience that this involvement usually opens
Administration hospitals, particularly because of the avenues (eg, investigations via the EMR) that are not
extensive electronic medical record (EMR) available to all readily avail- able to bench technologists. Such interventions
clinicians. Such a call reporting system may be of limited can ultimately result in more rapid communication to a
benefit, however, at institutions with less robust EMRs. clinician familiar with the patient involved.
Results would still need to be conveyed to the
responsible clinician for long-term
management.

Repeat Critical Values

Escalation Policies
Another common problem in critical value reporting is
What should a laboratory do when a technologist is not how a laboratory should handle repeat critical values, or D
o
able to reach a responsible clinician with the critical value? sub- sequent critical values for a given assay on the same w
n
o
In these circumstances, abandoning the call entirely is patient (but subsequent specimen). Approximately 70% of l
o
l
almost never an acceptable solution. An escalation policy surveyed laboratories have a policy on repeat critical a
d
e
or a fail- safe mechanism can be beneficial in such values.27 For those that do not, one is strongly d
f
circumstances. An escalation policy would direct the recommended because it will clarify laboratory r
laboratory technologist to contact a supervisor, pathology technologist responsibility and estab- lish consistency in r
f
resident, and/or medical director to assist in critical value performance. There are only 3 options: (1) Call only the first mo
h
notification.28 This policy allows the technologist to critical value. (2) Call each critical value. (3) Call critical t
refocus on the important task of laboratory testing, and it values once per interval of time. p
:
transfers the responsibility for notifi- cation to people who As clinicians become quickly annoyed by repetitive /
may have greater access to an inpatient or outpatient EMR calls for critical values, and as such calls may have dimin- /
a
and who can put the finding in a broader clinical context. In ishing value over time, some advocate using interval criteria jj
c
our experience, the pathology and labora- tory medicine (for example, calling once every 24 hours).13 Determining c
residents are usually able to contact a covering physician and whether critical results meet interval criteria might add .
po
convey these critical results. Verification of cli- nician additional tasks to laboratory technologists, although LIS o
notification should be subsequently conveyed back to the x
or middleware-based rules can be used to perform com- o
f
laboratory technologist (and entered into the LIS) to com- parisons automatically. Others have suggested that interval d
r
ply with TJC and CAP requirements. A fail-safe mechanism calling is appropriate for only select analytes. 9 Of note, j
o
(or safety net) can also be used in cases in which notification one study demonstrated that lower rates of undocumented o
u
continues to be unsuccessful.13,31 For example, if a “physi- critical value results in the medical record were r
n

associated n
a
33 sl
o.
o
g
r
/
b
y
g
u
e
s
ts
o
n
D
e
c
e
m
b
e
r
3
0
,
2
0
61
cian representing the laboratory” determines that immediate with policies that require calling all critical results. Such
care of the patient may be required, the laboratory result ing to contact the ordering provider, the algorithms include
and
attempting to identify and contact the patient’s primary care
patient information might be conveyed to a physician in the
provider. If still unsuccessful, a chief of service or chief of staff
ED to contact the patient directly.13
would ultimately be notified. This system has an added benefit of
An approach to dealing with unreachable clinicians was
bringing the issue of critical value reporting to the attention of a
recently proposed in a 2010 article.32 The report included hospital or departmental administrator who might not otherwise
algorithms for the inpatient and outpatient settings. After try- be aware of problems associated with the process. However, as
with other systems in which the ordering provider is not the an approach in high-volume laboratories, however, can
be
physician ultimately receiving the
exceedingly burdensome to technologists and clinical staff,
call, this process involves clinicians who may know very
little and there are minimal data to argue the clinical benefit
of one approach vs the other.
Critical value lists and procedures should include not
just critical ranges but also the frequency of when to call for
each given test. A laboratory may determine that some tests
should be called with each critical value, while others (such
as a mark- edly elevated blood urine nitrogen level) may be
called using an interval approach. The laboratory policy
should also clarify how to handle critical value notification
after a subsequent normal result during the same interval
(eg, 8 AM, critical high;
9 AM, normal; then 10 AM, critical high). In our policies, if
a normal test result occurs after a critical result, a
subsequent critical result is considered new and would be
called again. The
laboratory’s policy should be clear for such
scenarios.

510 Am J Clin Pathol 2011;135:505-513 © American Society for Clinical Pathology

510 DOI: 10.1309/AJCP9IZT7BMBCJRS
 Pathology Consultation / S����󰁣����󰁬 A󰁴��󰁣��󰁥

Critical Value Audits gr

Anatomic pathologists evaluate organs, tissues, and
The importance of using critical value data to better cellular specimens and provide diagnoses when os
under- stand laboratory process and preanalytic error cannot abnormal findings are observed. Preanalytic and analytic s
be over- emphasized. For example, one program identified a processes (eg, ex
specimen transport issue that led to falsely elevated K+ a
results in some patients.27 Changing the transport mi
requirements decreased the number of critical high K+ na
results. This change not only enhanced the quality of the tio
laboratory’s performance but also eased the burden of n,
unnecessary critical value calls. Analysis of critical value tis
limits can also be used to estimate the impact on call su
frequency that would result from changing threshold e
requirements. 10 Analysis can reveal differences in sli
critical value patterns by patient location (eg, falling ci
hematocrit values on surgical services vs low K+ values on ng
medical services).30 ,
This information could be used, for example, in the st
evaluation of new point-of-care programs. Others have used ai
critical val- ues analysis in studies of adverse events and ni
ng
clinical activity at discrete hospital locations.34 Critical
,
value audits provide tremendous information on laboratory
an
processes, and they are a great starting point for quality
d
improvement initiatives.
im
m
un
Opt-Out and Specific oh
Lists is-
Physician “opt-out” of critical value notifications (or
to
“no call hours”) were prohibited by more than 80% of
ch
programs sur- veyed in one study.14 Opt-out is also strongly
e
discouraged in the CAP checklist.3 The laboratory has a
mi
clear mandate to convey critical values, and permitting
ca
physician opt-out runs counter to the overall patient care
l
objective of the notification process.
an
Physician- and/or location-specific critical value lists
al
are another issue of controversy. For example, clinicians ys
at a dialysis clinic may be concerned about a different range is)
of electrolyte results than clinicians at an orthopedics oft
rehabilita- tion unit. Maintaining multiple, separate lists in en
the laboratory can be challenging (if not impossible) at most m
institutions and ea
would be prone to technologist error in underreporting and n
overreporting of critical results. Physician- and/or th
location-
at
specific lists are not in widespread
use. ho
Many laboratories, however, include some population- ur
specific critical values. Critical value lists often include s
unique limits for neonates, but these values are easily incor- to
porated onto a single laboratory-wide critical value list da
and would not be maintained separately. Tables including ys
m
pedi- atrics critical value limits are available,2,14,16 and one
ay
study on interlaboratory variability in pediatric critical
pa
values was recently published.12
ss
be
Critical Diagnoses in Anatomic Pathology fo
re a diagnosis can be made. In certain circumstances, the F N A s p ec im en s
Fun g i in a n F NA s pecimen from an
diagnosis suggests that “immediate treatment or prompt immunocompromised patient
The finding of certain microorganisms in any patient
evaluation of the patient” may be indicated.35 This concept of
critical diagnoses in surgical pathology and cytology (analo- FNA, fine-needle aspiration.
gous to critical values in the clinical laboratory) has received
significant attention in recent years.35-43
For example, several retrospective reviews and multi-
institutional surveys led to the creation of a list of possible
critical diagnoses in anatomic pathology and cytology ❚Table
3❚.35-39,42,44 Even though these conditions were suggested as
being important for immediate communication, the authors
found an overall lack of consensus from participants on what
might actually be included on such lists. 35-37
In fact, there are no specific national guidelines as to what
types of diagnoses in surgical pathology should qualify as
critical. The concept of critical diagnoses, however, has been D
o
endorsed by the Association of Directors of Anatomic and w
n
Surgical Pathology, and a generic list of possible diagnoses o
l
o
was published in 2006 based on the aforementioned studies, l
a
d
with the caveat that any such list “needs to be customized e
d
f
to each individual hospital based on specific requests from
r
clinicians and institutional factors such as the scope of ser- r
f
vices provided, case mix, acuity level, and protocols.” 38,42,44 mo
Finally, there is no consensus on how to actually report these h
t
diagnoses nor on the appropriate time frame for communica- p
:
38,40,45,46
tion. In the absence of specific guidelines, it will /
remain necessary for institutions to establish local, custom- /
a
ized protocols for handling critical diagnoses. jj
c
c
.
po

o
❚Table
3❚ x
o
f
Examples of Possible Critical Diagnoses in Anatomic
d
Pathology35-39,42,44 r
j
o
o
u
Surgical pathology
Crescents in >50% of glomeruli in a kidney biopsy specimen n
r
Vasculitis
Bacteria in a heart or bone marrow specimen n
a
Select organisms in immunocompromised patients sl
Uterine contents without villi or trophoblast o.
o
Fat in an endometrial curettage g
r
Mesothelial cells in a cardiac biopsy specimen /
b
Fat in colonic endoscopic polypectomy specimens y
Transplant rejection g
u
e
s
Malignancy in superior vena cava syndrome ts
Neoplasms causing paralysis o
n
Significan t disagreeme nt between frozen section D
e
and final diagnoses c
e
Cytology m
b
e
Unexpected malignancy r
Malignancy in critical places that can cause spinal cord injury 3
0
,
Disagreemen t between immediate and final interpretations of 2
0
6
1

© American Society for Clinical Am J Clin Pathol 2011;135:505-513 511

Pathology 511 DOI: 10.1309/AJCP9IZT7BMBCJRS 511
Genzen and Tormey / C󰁲��󰁴��󰁣��󰁬 V��������󰁳

Conclusion 2.
Along with the entire team of personnel involved in B
References u
criti- cal value notifications, pathologists and/or laboratory r
direc- tors have an important role in many aspects of 1. Lundberg G. When to panic over an abnormal value. t
Med i
critical value reporting. They are involved in establishing Lab Obs. 1972;4:47-54. s
and updating the critical value lists and policies (in
consultation with clinicians and institutional medical C
boards), they assist in critical value escalation and ,
interpretations, they are responsible for ensur- A
s
ianngd ctohmeyplciaannceusweitchriTtiJcCa-l h
w
avnadluCesAPau- o
o
rdeiqtsuitroedimdopcruomveenotvaetiroanll, laboratory d
performance. Technological advancements will certainly
E
alter the way in which critical value notifications are made, ,
but these will not change the overall responsibility of
pathologists and laboratory directors for ensuring B
r
compliance with critical value notification requirements. A u
recent CAP Today Q&A discussion on critical value n
notification con- cluded by stating that “a focus on building s
a patient-centric system requires strong pathologist D
31 ,
leadership to ensure a safe and reliable system.” As such, a
strong commitment to the critical value notification process
e
can enhance overall patient care and should be a focus of d
ongoing quality improvement. s
.

T
Case Summary i
After being unable to contact a covering physician, e
t
the laboratory technologist paged the on-call pathology z
resident. The resident had access to the outpatient EMR and
noticed that a cardiology fellow (and not the patient’s T
e
primary attending physician) wrote the clinic notes. The on- x
call cardiology fel- low was paged by the resident and stated t
that she was familiar with the patient. She accepted the b
o
critical results, performed appropriate read-back, and o
contacted the patient for clinical follow-up. The resident k
called back the laboratory technologist and provided
o
necessary information for documentation of the critical f
value notification. On investigation the next day, it was
C
determined that a new member of the outpatient office staff l
did not know how to set the telephone system appropriately i
to for-
n
ward calls to the answering service. The telephone (and i
pager) c
directory was updated for all physicians at this clinic. a
l

C
From the 1Department of Pathology and Laboratory Medicine,
h
Weill Cornell Medical College, Cornell University, and New
e
York-Presbyterian Hospital, New York, NY; and 2Department m
of Laboratory Medicine, Yale University School of Medicine, i
New s
Haven, CT, and the Pathology and Laboratory Medicine t
Service, VA Connecticut Healthcare System, West Haven. r
y
Address reprint requests to Dr Genzen: Weill Cornell
Medical College, 525 E 68th St, F-705, New York, NY 10065. a
 nd Molecular Diagnostics. 4th ed. St Louis, MO: Elsevier 21. Kost GJ. The significance of ionized calcium in cardiac and
Saunders; 2006. critical care: availability and critical limits at US medical
centers and children’s hospitals. Arch Pathol Lab Med.
3. College of American Pathologists. Laboratory General
1993;117:890-896.
Checklist [components GEN.41320, GEN.41330, and
GEN.41340]. Available at www.cap.org. Accessed August 6,
2010.
4. The Joint Commission. Accreditation Program: Laboratory,
National Patient Safety Goals (NPSG.02.03.01). http://www.
jointcommission.org/. Accessed July 12, 2010.

5. International Organization for Standardization. ISO

15189:2007: medical laboratories: particular requirements
for quality and competence [items 5.5.3n, 5.8.7, and 5.8.8].
Available at www.iso.org. Accessed August 7, 2009.
6. Medicare, Medicaid, and CLIA Programs: laboratory
requirements relating to quality systems and certain personnel
qualifications. Fed Regist. 2003;68:3639-3714. Codified at 42
CFR §493.
7. Barenfanger J, Sautter RL, Lang DL, et al. Improving patient
safety by repeating (read-back) telephone reports of critical
information. Am J Clin Pathol. 2004;121:801-803. D
o
8. Rensburg MA, Nutt L, Zemlin AE, et al. An audit on the w
n
reporting of critical results in a tertiary institute. Ann Clin o
l
Biochem. 2009;46:162-164. o
l
a
9. Emancipator K. Critical values: ASCP practice parameter. d
e
Am J Clin Pathol. 1997;108:247-253. d
f
r
10. Dighe AS, Rao A, Coakley AB, et al. Analysis of laboratory
critical value reporting at a large academic medical center. r
f
Am J Clin Pathol. 2006;125:758-764.
mo
11. Parl FF, O’Leary MF, Kaiser AB, et al. Implementation of a h
t
closed-loop reporting system for critical values and clinical
communication in compliance with goals of the Joint p
:
Commission. Clin Chem. 2010;56:417-423. /
/
12. Gong Y, Adeli K. A national survey on pediatric critical /
values used in clinical laboratories across Canada. Clin a
a
Biochem. 2009;42:1610-1615. jj
13. Howanitz PJ, Steindel SJ, Heard NV. Laboratory critical c
values policies and procedures: a college of American c
.
Pathologists Q-Probes study in 623 institutions. Arch Pathol po
Lab Med. 2002;126:663-669.
o
x
14. Wagar EA, Friedberg RC, Souers R, et al. Critical values
o
comparison: a College of American Pathologists Q-Probes f
d
r
s22su0r0v7e;y1o3f11:16736c9l-i1n7ic7a5l. laboratories. Arch j
o
Pathol Lab Med. o
u
15. Piva E, Sciacovelli L, Zaninotto M, et al. Evaluation of n
r
effectiveness of a computerized notification system for
reporting critical values. Am J Clin Pathol. 2009;131:432-441. n
a
sl
16. Kost GJ. Table of critical limits. MLO Med Lab Obs.
o.
2004;36(13 suppl):6-7. o
g
r
17. Kost GJ. Critical limits for urgent clinician notification at US /
b
medical centers. JAMA. 1990;263:704-707. y
g
u
18. Kost GJ. Clinical and professional reference guides: table of e
s
ts
critical limits. MLO Med Lab Obs. 2001;33:12-13. o
n
19. Lusky K. Critical values: looking more closely at calls, cutoffs. D
e
c
CAP Today. November 2008:80. e
m
20. Kost GJ. Critical limits for emergency clinician notification b
e
r
at United States children’s hospitals. Pediatrics. 3
0
1991;88:597-603. ,
2
0
6
1

512 Am J Clin Pathol 2011;135:505-513 © American Society for Clinical Pathology

512 DOI: 10.1309/AJCP9IZT7BMBCJRS
 Pathology Consultation / S����󰁣����󰁬 A󰁴��󰁣��󰁥

22. Massachusetts General Hospital Pathology Service. 35. Pereira TC, Liu Y, and Silverman JF. Critical values in
Laboratory Handbook. http://mghlabtest.partners.org/ surgical pathology. Am J Clin Pathol. 2004;122:201-
CriticalValues.htm. Accessed September 27, 2010. 205.
23. Brigham and Women’s Hospital. Clinical Laboratory 36. Pereira TC, Clayton AC, Tazelaar HD, et al. Critical
Manual. values in cytology. Diagn Cytopathol. 2006;34:447-451.
http://www.brighamandwomens.org/clinicallabs/medical/
37. Pereira TC, Silverman JF, LiVolsi V, et al. A multi-
resources.aspx. Accessed October 3, 2010.
institutional survey of critical diagnoses (critical values)
24. Mayo Medical Laboratories. Critical Values and Semi- in surgical pathology and cytology. Am J Clin Pathol.
Urgent Results. 2008;130:731-735.
http://www.mayomedicallaboratories.com/articles/
criticalvalues/index.html. Accessed October 2, 2010. 38. Silverman JF. Critical diagnoses (critical values) in
anatomic pathology. Am J Clin Pathol. 2006;125:815-817.
25. Yale-New Haven Hospital. Laboratory Manual. https://
39. Silverman JF, Pereira TC. Critical values in
labmanual.ynhh.org/. Accessed October 1, 2010. anatomic pathology. Arch Pathol Lab Med.
26. Valenstei n PN, Wagar EA, Stankovic AK, et al. 2006;130:638-640.
Notification
of critical results: a College of American Pathologists 40. Visscher DW. What values are critical [editorial]? Am J Clin
Pathol. 2008;130:681-682.
Q-Probes study of 121 institutions. Arch Pathol Lab Med.
2008;132:1862-1867. 41. Huang EC, Kuo FC, Fletcher CD, et al. Critical diagnoses in
27. Dighe AS, Jones JB, Parham S, et al. Survey of critical value surgical pathology: a retrospective single-institution study to
reporting and reduction of false-positive critical value monitor guidelines for communication of urgent results. Am
J Surg Pathol. 2009;33:1098-1102. D
results. Arch Pathol Lab Med. 2008;132:1666-1671. o
w
28. Dighe AS. Making the right calls on critical values, tests. 42. Silverman JF, LiVolsi VA, Fletcher CD, et al. Critical n
o
diagnoses (critical values) in anatomic pathology. Am J l
CAP Today. August 2008:12. o
Surg Pathol. 2006;30:897-899. l
a
29. Etchells E, Adhikari NK, Cheung C, et al. Real-time d
43. Allen TC. Critical values in anatomic pathology? Arch e
clinical alerting: effect of an automated paging system on d
response time to critical laboratory values: a randomised Pathol f
or
Lab Med. 2007;131:684-687. o
controlled trial. Qual Saf Health Care. 2010;19:99-102. m
44. Silverman JF, Fletcher CD, Frable WJ, et al. Critical h
30. Kuperman GJ, Teich JM, Tanasijevic MJ, et al. Improving t
response to critical laboratory results with automation: diagnoses (critical values) in anatomic pathology.
p
Hum Pathol. 2006;37:982-984. :
results /
45. LiVolsi VA, Leung S. Communicating critical values in /
o f a r a n d o m iz e d controlled trial. J Am Med anatomic pathology. Arch Pathol Lab Med. 2006;130:641- /
119 9 9 ; 6 :5 1 2 -5 2 2 . a
Inform Assoc. 644. a
31. Hernandez JS. Q & A. CAP Today. May 2010:94. 46. LiVolsi VA. Critical values in anatomic pathology: jj
how do we communicate [editorial]? Am J Clin Pathol. c
32. Singh H, Vij MS. Eight recommendations for policies for c
2004;122:171-172.
communicating abnormal test results. Jt Comm J Qual .
po
Patient Saf. 2010;36:226-232.
33. Wagar EA, Stankovic AK, Wilkinson DS, et al. o
x
Assessment monitoring of laboratory critical values: a o
f
College of American Pathologists Q-Tracks study of 180 d
institutions. Arch Pathol Lab Med. 2007;131:44-49. r
j
34. Jenkins JJ, Crawford JM, Bissell MG. Studying critical o
o
values: adverse event identification following a critical u
laboratory values study at the Ohio State University Medical n
r
Center.
n
Am J Clin Pathol. 2007;128:604-609. a
sl
o.
o
g
r
/
b
y
g
u
e
s
ts
o
n
D
e
c
e
m
b
e
r
3
0
,
2
0
6
1
© American Society for Clinical Am J Clin Pathol 2011;135:505-513 513
Pathology 513 DOI: 10.1309/AJCP9IZT7BMBCJRS 513