Potential of an age adjusted D-dimer cut-off value to improve the exclusion of pulmonary embolism in older patients: a retrospective analysis of three large cohorts
BMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c1475 (Published 30 March 2010) Cite this as: BMJ 2010;340:c1475
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dear Editor,
We read Douma et al.‘s paper with interest, since its publication
coincides with our ongoing retrospective analysis of Dade Behring
Innovance/Sysmex CA-1500 D-dimer data from all (800+) tests performed for
patients between the ages of 60 and 80 during 2009 in our General Hospital
setting.
We can report that, while our results do suggest a substantive
increase in the D-dimer cut-off from the 500µg/L value routinely employed
would better serve the needs of this patient group, it appears a static
cut-off affords greater improvement in specificity for VTE than that
generated by a ‘(10 x age) = cut-off’ algorithm, with identical
sensitivity and NPV.
Yours faithfully
Kelly Nightingale Esq.
kelly.nightingale@buckshosp.nhs.uk
Competing interests:
None declared
Competing interests: No competing interests
Dear Editor,
Douma et al highlight the often differing test characteristics for
the same test applied to different populations, in this case according to
age. However, they have made a fundamental error in the interpretation of
these characteristics which appears to have led to a potentially unsafe
conclusion.
The specificity of a test is dependent on the true negative and the
false positive rates and becomes better as the false positive rate
approaches zero, i.e. if the test is positive, it is unlikely to be
falsely positive. Hence, better specificity allows the clinician to rule
in disease. Conversely sensitivity is dependent on the true positive and
false negative rates and improves as the false negative rate approaches
zero, i.e. if the test is negative, it is unlikely to be falsely negative.
Hence, it is primarily the sensitivity that determines a test’s ability to
rule out disease. There is often a balance to be struck between the
sensitivity and specificity of a test such that as one improves, the other
worsens according to the cut-off value set. The specificity takes no
account of the false negative rate so, even if the true negative rate is
high, if the false negative rate is equally high, the clinician cannot
tell if a negative test is true or false and cannot therefore exclude
disease. This rule is often referred to by the mnemonic Spin and Snout:
specificity to rule in, sensitivity to rule out.
The other consideration is the actual role of the D-dimer test in the
investigative pathway for suspected pulmonary embolism. Thus far, it has
been used as a screening test to exclude those who definitely do not have
disease. The remainder then undergo more diagnostic studies. It is
therefore much more important to consider the sensitivity of D-dimers in
this screening role than the specificity. In each of the three cohorts in
Douma et al’s study, the conventional cut-off had fewer false negatives
than the age-related cut-off. This means that the age-related cut-off is
less safe as it misses some patients with the disease who would otherwise
have been identified. The benefit of Douma et al’s proposal is a reduction
in need for further, more expensive investigation. Hence, such a proposed
change in the management of suspected pulmonary embolism needs not a
further clinical evaluation but an ethical one to balance non-maleficence
(doing no harm by not missing the diagnosis with false negatives) against
equity of access (ensuring optimum use of resources).
References
1. Renée A Douma, Grégoire le Gal, Maaike Söhne, Marc Righini, Pieter W
Kamphuisen, Arnaud Perrier, Marieke J H A Kruip, Henri Bounameaux, Harry R
Büller, and Pierre-Marie Roy. Potential of an age adjusted D-dimer cut-off
value to improve the exclusion of pulmonary embolism in older patients: a
retrospective analysis of three large cohorts BMJ 2010; 340: c1475
Competing interests:
None declared
Competing interests: No competing interests
Douma et al found in their validation cohorts that using a D-dimer
age-based cut-off resulted in a false negative rate for pulmonary embolism
of 0.3-0.6%. They concluded that age-based D-dimer cut-offs greatly
increased the proportion of older patients in whom pulmonary embolism
could be safely excluded. This is misleading and could be dangerous to
patients.
Their finding is based on the entire cohort of patients with D-
dimer values below the age-based cutoff but the critical element is the
false negative rate in the patients with a D-dimer result above 500 and
less than the age-based cut-off. According to the data presented in table
3, there were an additional 110 patients with a negative D-dimer value
using an age-based cut-off and this resulted in an additional 5 tests with
false negative results. This would give a false negative rate of 4.5%.
Likewise according to the information in table 5, the false negative rate
in the second validation cohort would be 1.9%. By combining the patients
in the proposed new range with the larger group of patients who already
test negative under the standard cutoff, they have underestimated the risk
of a false negative result for the patients who would purportedly benefit
from the newer cut-off by about a factor of 8.
Based on the data
presented, we think it would be prudent to not change the cut-off
threshold for a negative D-dimer test at this time.
Competing interests:
None declared
Competing interests: No competing interests
Editor, I read the report by Douma et al with a great interest [1].
Duoma et al. concluded that "The age adjusted D-dimer cut-off point,
combined with clinical probability, greatly increased the proportion of
older patients in whom pulmonary embolism could be safely excluded [1]." I
have some questions on this report. Based on the nature of a retrospective
analysis, there are several considerations. First, how can Douma et al.
confirmed on the homogeniticity of the subjects in the three cohorts?.
Second, how can Douma et al. confirm on the QC of the laboratory analysis
in all included cohorts? Do the analyzers in all cohorts share the
complete common diagnostic properties? These questions can be the points
for further discussion.
References
1. Renée A Douma, Grégoire le Gal, Maaike Söhne, Marc Righini, Pieter W
Kamphuisen, Arnaud Perrier, Marieke J H A Kruip, Henri Bounameaux, Harry R
Büller, and Pierre-Marie Roy. Potential of an age adjusted D-dimer cut-off
value to improve the exclusion of pulmonary embolism in older patients: a
retrospective analysis of three large cohorts
BMJ 2010; 340: c1475
Competing interests:
None declared
Competing interests: No competing interests
Dear Editor,
Renee et al highlight the usefulness of an age adjusted cutoff
value for D-dimer to exclude pulmonary embolism for
elderly(1). However, many of the elderly patients suspected of
having pulmonary embolism have multiple risk factors in the same
individual. Hence raising the cut off for such patients only
based on age adjusted value may not be appropriate. Hence, a
decision on a case-by-case basis by clinicians with awareness of
the characteristics of d-dimer in subgroups appears to be more
reasonable.
Reference:
(1)Renée A Douma, Grégoire le Gal, Maaike Söhne, Marc Righini,
Pieter W Kamphuisen, Arnaud Perrier, Marieke J H A Kruip,
Henri Bounameaux, Harry R Büller, and Pierre-Marie
Roy.Potential of an age adjusted D-dimer cut-off value to
improve the exclusion of pulmonary embolism in older patients:
a retrospective analysis of three large cohorts.BMJ 2010; 340:
c1475
Competing interests:
None declared
Competing interests: No competing interests
Accuracy of D-dimer in suspected pulmonary embolism by physicians
Pulmonary embolism is a major health problem, with an estimated
575,000 persons presenting with a suspected pulmonary embolism each year
in the United States (1), and for the international community of
clinicians.
Recently a review by Agnelli and colleagues (2) evaluated the clinical
assessment in suspected acute pulmonary embolism.
On the basis of our clinical experience we disagree with Agnelli and
colleagues in the diagnostic workup, and we have concern about some points
of general interest by physicians that should be allocated to evidence-based strategies in daily practice:
First, D-dimer testing is of limited clinical utility for inpatient
with clinically suspected pulmonary embolism for the poor accuracy. The
frequency of false positive with high level of the test is associated with
many other diseases is patients (Table 1)(1).
In addition meta analysis by Brown and colleagues (3) found a
reasonably high sensitivity (93%), but only a moderate specificity (51%)
of D-dimer testing with ratio positive of 1.9 and a ratio negative of 0.14
for the detection of pulmonary embolism in outpatients. The cutoff for a
normal D-dimer was 500 ? /L. The diagnostic indexes compared favorably
with those found in a meta analysis of the D-dimer ELISA (4).
Second, in our diagnostic investigation of older patients (? 70 yrs)
with suspected pulmonary embolism, and according to literature D - dimer (
? 500 ? /L) not excluded a pulmonary embolism (5).
Third, we suppose that having a new D - dimer cut-off value is
impractical in a clinical setting. However, at present there is not a
validation of this cut-off value prospectively in an outcome study with
patient follow-up.
Finally, in the clinical approach of acute pulmonary embolism we
suggest the algorithm showed (Figure 1), and the findings of the study by
Agnelli and colleagues should be reconsidered.
References
1.Rathbun S W, Whitsett TL, Vesely SK, and Raskob GE Clinical
Utility of D-dimer in Patients With Suspected Pulmonary Embolism and Non
diagnostic Lung Scans or Negative CT Findings. Chest 2004; 125(3): 851-55
2.Agnelli G, Becattini C. Acute pulmonary embolism. N Eng J Med; 363:
-266-74
3.Brown MD, Lau J, Nelson DR, and Kline JA. Turbidimetric D-Dimer
Test in the Diagnosis of Pulmonary Embolism: A Metaanalysis. Clin
Chemistry 2003; 49:11 1846-1853
4.Brown MD, Rowe BH, Reeves MJ, Bermingham JM, Goldhaber SZ. The
accuracy of the enzyme-linked immunosorbent assay D-dimer test in the
diagnosis of pulmonary embolism: a meta-analysis. Ann Emerg Med 2002;
40:133-44.
5.Douma RA, le Gal G, Sohne M et al. Potential of an age adjusted D-
dimer cut-off value to improve the exclusion of pulmonary embolism in
older patients: a retrospective analysis of three large cohorts. BMJ
2010;340:c1475
Table 1
Figure 1
Competing interests: No competing interests