Date: April 29th, 2015

Guest Skeptic: Dr. Kerstin deWit did an Internal medicine and EM training in the UK. Since then she has worked in Thrombosis research and received a doctorate in the UK and a Masters in Epidemiology from the University of Ottawa. Kerstin currently works as a Thrombosis physician and Emergency Physician at McMaster University and member of Best Evidence in Emergency Medicine (BEEM).

Case: An 84-year-old woman presents to the ED with vague symptoms of shortness of breath and mild chest discomfort for the last two days. She does not say “pain” but rather just an ache when taking a deep breath. Her past medical history includes hypertension, dyslipidemia, type-2 diabetes and osteoarthritis. Her vital signs are normal. She is Well’s low but PERC positive due to her age. You do not want to order a D-dimer because of the high false positive rate but pulmonary embolism (PE) is on your differential.

Dr. Kerstin deWit

Dr. Kerstin deWit

Background: A negative D-dimer has been shown to safely rule out pulmonary embolism in patients who are not high risk. It has a high sensitivity but low specificity. This means that there is a high false positive rate.

The false positive rate increases with age. This is because normal D-dimer levels increase with age. The result is older patients are subjected to more diagnostic image testing to rule-out PE.

There is a worldwide epidemic of over-testing with the downstream consequences of over-diagnosis and over-treatment. The problem of over-testing is complex. It stems from an imperfect mix of incomplete and flawed diagnostic evidence, constant malpractice threats (worse in some parts of the world than others), and clinicians struggling to balance clinical care (our day job) with maintaining awareness of the ever-evolving research landscape.

PE is a prime target to reduce over-testing and there have there have been a number of studies looking at age adjusting the D-dimer. The derivation study was done by Douma et al 2010. They suggested using a new cut off for patients over the age of 50 years of age.

The formula was patient’s age x 10. So in our case it would be 84×10=840 μg/L

  • Conclusion: “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.”

The concept of age-adjusting the D-dimer to rule-out PE was validated in two trials (van Es et al 2011 and Penaloza et al 2012). Both of these studies demonstrated the clinical usefulness of this approach.

In 2013, Schouten et al in the BMJ published a systematic review and meta-analysis. It had over 12,000 patients and compared conventional D-dimer cut off of 500 μg/L vs. age-adjusted D-dimer.

  • Conclusion: “The application of age adjusted cut-off values for D-dimer tests substantially increases specificity without modifying sensitivity, thereby improving the clinical utility of D-dimer testing in patients aged 50 or more with a non-high clinical probability.”

There were some concerns about this meta analysis expressed by some including Dr. Brent Thoma from Boring EM. His concerns seemed to focus on the heterogeneity of the studies and combining DVT studies with PE studies in the meta analysis.

Theme Music: Good Riddance by Green Day performed by Ethan Milne

Another turning point, a fork stuck in the road. Nurse grabs you by the hand, directs you where to go.
So make the best of this test, and don’t ask why. It’s not a question, but a lesson learned in time.
It’s something unpredictable, but in the end is right. I hope you had a negative D-Dimer.

Clinical Question: Can an age-adjusted D-dimer threshold safely exclude PE in non-high risk ED patients with suspected PE?

Reference: Righini et al. Age-adjusted D-dimer Cuttoff Levels to Rule Out Pulmonary Embolism: The ADJUST-PE Study. JAMA 2014

  • Population: Consecutive patients from 19 hospitals in 4 European countries with a clinical suspicion of PE. This was defined as acute onset or worsening shortness of breath or chest pain without another obvious etiology.
    • Exclusion: Clinical suspicion for PE more than 24 hours after hospitalization, anticoagulated for another reason, creatinine clearance less than 30 mL/min, contrast dye allergy, or life-expectancy less than 3 months.
  • Intervention: Age adjusted D-dimer (age x 10)
  • Comparison: D-dimer as >500 μg/L
  • Outcome:
    • Primary Outcome: Failure rate of each diagnostic strategy computed as number of adjudicated DVT + non-subsegmental PE divided by the number of patients with a negative D-dimer result that were left without any anticoagulant therapy.
    • Secondary Outcome: Additional diagnostic yield of the age-adjusted cutoff value strategy.

Authors’ Conclusions:Compared with a fixed D-dimer cutoff of 500 μg/L, the combination of pretest clinical probability assessment with age-adjusted D-dimer cutoff was associated with a larger number of patients in whom PE could be considered ruled out with a low likelihood of subsequent clinical venous thromboembolism.”

Quality Checklist:

  1. checklist-cartoonThe clinical problem is well defined. Yes
  2. The study population represents the target population that would normally be tested for the condition including (i.e. no spectrum bias). Yes
  3. The study population included or focused on those in the ED. Yes

  4. The study patients were recruited consecutively (i.e. no selection bias). Yes
  5. The diagnostic evaluation was sufficiently comprehensive and applied equally to all patients (i.e. no evidence of verification bias). No
  6. All diagnostic criteria were explicit, valid and reproducible (i.e. no incorporation bias). Yes
  7. The reference standard was appropriate (i.e. no imperfect gold-standard bias). Yes
  8. All undiagnosed patients underwent sufficiently long and comprehensive follow-up (i.e. no double gold-standard bias). Yes
  9. The likelihood ratio(s) of the test(s) in question is presented or can be calculated from the information provided. No
  10. The precision of the measure of diagnostic performance is satisfactory. Yes

Results: There were 3,346 patients with suspected PE included in the study of which 337 (10%) were over the age of 50 with a normal age-adjusted D-dimer and non-high clinical probability. The overall prevalence of PE was 19.0% (95% CI 17.7%-20.4%).

  • 2898 patients were risk stratified as non-high risk or unlikely clinically to have a PE.
    • 1154/2898 (39.8%) had a negative D-dimer according to the age-adjusted cutoff
    • 817/2898 (28.2%) had a negative D-dimer using the <500 μg/L cutoff
    • This gives an absolute difference of 11.6%
  • Three-month thromboembolic risk for patients <500 μg/L
    • 1/810 or 0.1% (95% CI 0%-0.7%)
  • Three-month thromboembolic risk for patients with D-dimer >500 μg/L but < age-adjusted cutoff
    • 1/331 or 0.3% (95% CI 0.1%-1.7%)
  • None of the elderly (age >75) with non-elevated age-adjusted D-dimer had confirmed VTE during follow-up
    • 0/195 or 0% (95% CI 0%-1.9%)

Screen Shot 2015-04-25 at 3.11.12 PM

Geriatric patients are assessed for possible PE worldwide every day. However, US and Canadian studies consistently report PE rates lower than European sites. Therefore, the European data likely represent spectrum bias skewed towards a “sicker” population relative to the U.S. that skews estimates of sensitivity upward, but does not affect specificity.

PE was a prime target in EM to quickly and safely reduce over-testing and this prospective, outcomes-based research provides strong evidence up which to do so for older adults. Their 3-month adjudicated outcomes provide compelling proof that age-adjusted D-dimers provide a rationale to use risk stratification plus appropriate D-dimer testing to safely reduce unnecessary ancillary PE testing – even in older adults.

They used the revised Genava score or the 2-level Well’s score to risk stratified patients who they clinically suspected of having a PE. However, the authors do not explain who performed this risk stratification (research team vs. clinicians).

Based upon risk stratification, patients either proceeded to CTPA (high risk Wells or likely clinical probability Geneva), whereas non-high risk had D-dimer testing.

Neither the CTPA method or interpreting radiologist experience was detailed. Failure to report their CT instruments or radiologists’ experience is technically a flaw (risk of bias) from a purist perspective. In addition, there were six different quantitative high sensitivity D-dimer assays used.

However, these “flaws” also increase the external validity and pragmatic applicability of this research since each of us is stuck with D-dimer assay used in our lab and radiologist that we have. Furthermore, their results did not seem to vary based on D-dimer assay.

Although only 1 of 7 deaths had autopsy to confirm whether PE caused death, the longitudinal 3-month follow-up using structured questionnaire provides strong patient-centric evidence of effectiveness. In addition, 3-month follow-up is standard in PE literature and at the outer range of timeframe within control of ED systems.

This was a management study and not a diagnostic study. Therefore, the authors did not report sensitivity, specificity, likelihood ratios, or interval likelihood ratios in their manuscript. Future diagnostic research should evaluate the sensitivity, specificity, likelihood ratios, and interval likelihood ratios for age-adjusted D-dimer, as well as implementation strategies to incorporate these new levels of abnormal into geriatric emergency care. This prospective, outcomes based research adds to an expanding volume of retrospective investigations indicating that age-adjusted D-dimer is ready for widespread use and ought to be incorporated into guidelines, textbooks, and routine bedside care.

The use of an age-adjusted D-dimer safely increases the proportion of non-high risk for PE patients from 28.2% (<500 μg/L) to 39.8% (age adjusted) with a non- elevated D-dimer and no further work-up required. This represents an 11.6% increase in D-dimer diagnostic efficiency.

Comment on Author’s Conclusion Compared to SGEM Conclusion: The authors’ conclusions seem very reasonable.

The SGEM Bottom Line: Using an age-adjusted D-dimer cutoff in ED patients with suspected PE increases the diagnostic yield of D-dimer testing by 11.6%

Clinical Application: This is ready for prime time. We should incorporate age-adjusted D-dimer into medical education, continuing medical education, electronic medical record protocols, and patient shared decision-making instruments.

What Do I Tell My Patient? Blood clots in the lung are called pulmonary embolisms. They can be hard to diagnose sometimes. We have a blood test for people like you who are not at high risk. We adjust the test based on your age. If the test is below your age-adjusted level you most likely do not have a blood clot in your lung and no further testing is needed. If the test is above your age-adjusted level we will need to do a CT scan to check for a blood clot.

FOAMed Reviews of ADJUST PE Study:Screen-Shot-2012-12-29-at-6.08.14-PM

Keener Kontest: Last weeks winner was Jim Priano a PGY2 Emergency Medicine Pharmacy Resident. Jim knew that Mettronidazole is often used to treat C.diff infections but is not actually FDA approved for this indication.

Listen to the podcast for this weeks keener question. If you know the answer send an email to TheSGEM@gmail.com with “keener” in the subject line. The first correct response will win a cool skeptical prize.

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.

EMA Conferences


Screen Shot 2014-11-15 at 7.07.05 PM