Sensitivity and Specificity of Age adjusted d-dimer Cut off Values in a Single Hospital Site: A

Retrospective Analysis

Ryan Cracknell, BMBS, MSc, BMedSci1 and Ehsan E. Salim, MBChB, BMedSci2
1
Golden Jubilee National Hospital, Clydebank, NHS Scotland
2
Monklands Hospital, Airdrie, NHS Scotland

Corresponding Author:

Ehsan E. Salim, MBChB, BSc,

NHS Greater Glasgow and Clyde, NHS Scotland,

Monklands Hospital, Monkscourt Avenue, Airdrie ML6 0JS

Email: [email protected]
Abstract

The d-dimer is a commonly utilised test in the assessment of Pulmonary Embolism (PE) in acute settings.

With a high sensitivity and low specificity, a significant number of false positive outcomes occur and lead to

unnecessary medical intervention.

A retrospective audit was carried out in a district general hospital by obtaining all acute CTPAs carried out

between December 2019 and August 2020. The age-adjusted d-dimer was calculated for each patient and

sensitivity and specificity reassessed.

After exclusion, 133 patients under 50 years of age with low pre-test probability scores were included in the

analysis. Age-adjusted d-dimer was found to increase specificity from 2% to 28% whilst maintaining a

sensitivity of 94%.

Utilisation of the age-adjusted d-dimer results of the increased specificity with the potential to reduce the

number of unnecessary admissions, radiation and medication. A prospective study would be of use in the

further investigation of age-adjusted d-dimer.

Keywords: d-dimer, pulmonary embolism, CTPA, chest pain

Chest pain is a common presentation to emergency and acute medical departments throughout the world 1.

Despite its commonplace at the front door of the hospital, chest pain continues to be a symptom resulting in

diagnostic uncertainty. Around 15% of all chest pain presentations to acute medicine are screened for
2
pulmonary embolism .The process of diagnosis combines skilled history taking alongside routine and

symptom specific investigations, including pre-test probability scoring and the high sensitivity, low

specificity d-dimer test. Following a positive d-dimer, the patient then most commonly undergoes a

Computed Tomography Pulmonary Angiogram (CTPA). CTPA is the gold-standard diagnostic test for PE
3
with a high positive and high negative sensitivity and specificity diagnostic value . Despite its high

diagnostic value, the CTPA has its disadvantages, such as radiation and contrast medium exposure, expense,

and time consumption/use of already constrained resources. Considering the negative factors associated with

CTPAs, their use should be avoided where possible. The d-dimer test acts as a gatekeeper to CT scanning. If

the d-dimer is negative and pre-test scoring identifies low/medium risk, PE can be excluded, without the

need for a scan. Despite its high negative predictive value, as a byproduct of cross-linked fibrin degradation,

a raised d-dimer can be attributed to a number of pro-thrombotic states, including those which are
4
physiological such as advanced age . The physiological increase in d-dimer value in older adults leads to an

overall reduced ability to exclude PE in this population.

The use of a poorly specific test in the Emergency Department and medical receiving wards results in

unnecessary admissions, increased risk of hospital acquired infections, cost to the health service, exposure of

patients to unnecessary radiation and anticoagulation.

A validated modification to account for physiological increases in d-dimer with age should result in

increased specificity with preserved sensitivity. In this study we add to the data which suggests that adjusting

the d-dimer for age is both validated and effective in the screening of PE.

Age adjusted d-dimers have been shown to have an increased specificity to PEs without any reduction in
5
sensitivity . By utilising this as a front door test, the expectation would be that the unnecessary risks afore

mentioned are reduced.

UK hospital site with the hope that age -adjusted d-dimers may be implemented across sites within the NHS

trust.

Methods

As part of a quality improvement project in a district general hospital, data was collected retrospectively for

a cohort of patients who had undergone a CTPA scan as part of the diagnostic work up for Pulmonary

Embolism between December 2019 and August 2020 (n=163). The local radiology department provided a

list of all patients with a CTPA requested directly from the Emergency department, or within the first 24

hours of admission to the medical receiving unit. Patients were excluded if they were under the age of 50

years (n=16), if they had a high Wells score (n=5), or if no-d-dimer was taken on admission (n=6).

Our local laboratory reports using d-dimer units as opposed to fibrinogen equivalent units and so the d-dimer

result was considered positive using our locally accepted cut-off of 230ng/L. The age adjustment process

considered the d-dimer to be negative if it was lower than the patient’s age in years multiplied by five. This

negative cut-off was chosen in light of previously published data which considered age adjustment in the
6
context of d-dimer units with a similar positive threshold

Data was analysed using SPSS version 28.0 and Microsoft Excel version 16.61. As a result of sample size,

the Shapiro-Wilk test was used to test normal distribution. Normally distributed values are represented as

mean and standard deviation, whilst those not normally distributed as median and interquartile range. The

Welch’s t-test was used to test normally distributed parametric variables for significance. Where not

normally distributed, the Mann-Whitney U test was used. The Chi-squared test was used to analyse

frequencies. A 95% confidence interval was used when referencing sensitivities and specificities.
 163 patients who had undergone a CTPA (at ED or within 24hrs of
admission) between Dec 2019 - Aug 2020

Excluded patients:
16 under the age of 50
5 with a high Well's score
6 with no D-dimer test on admission

32 patients had a PE on CTPA 101 patients had no PE on CTPA

Figure 1: Patient recruitment, exclusion criteria and outcome of CTPA scan

Results

From December 2020 - August 2021, a total of 163 patient underwent CTPA’s in the diagnostic work up of

PE. Following exclusion for not meeting inclusion criteria, 133 patients over the age of 50, with low to

moderate Well’s scores underwent CPTAs for suspected PE. Of the included patient, thirty-two patients

(24.0%) were diagnosed with PE. The confirmed PE group had significantly higher D-dimer levels than

those who were negative, however there were no significant differences in age or sex distribution.

PE No PE Significance

Number 32 101

Age 70.6 [95%CI 66.7-74.5] 70.7 [95%CI 68.6-72.8] p=0.482

Male 51.3% 40% p=0.138

D-dimer 1491 [646-3291] 487 [333-772] p<0.00001

Table 1: Basic demographic data of PE screening outcomes and statistical significance

and true negative results of 32 and 2, respectively. This is relayed as a sensitivity of 1.00 (95% CI 0.89-1.00)

and a specificity of 0.02 (95% CI 0.002-0.069). The positive predictive value (PPV) and the negative

predictive value (NPV) of the conventional d-dimer were 0.244 (95%CI 0.237-0.247) and 1.00 respectively.

Age adjusted d-dimer demonstrated a reduction in false positive results (n=73) and an increase in true

negative results (n=28). The age-adjusted d-dimer resulted in a false negative total of two. The sensitivity

was 0.94 (95%CI 0.79-0.99) and the specificity 0.28 (95%CI 0.19-0.38). The PPV and NPV were

0.291(95%CI 0.26-0.32) and 0.93(95%CI 0.78-0.98), respectively.

Age adjustment was also calculated with an age adjusted cut-off of 6x the patient’s age. This resulted in the

following results:

The sensitivity was 0.90(95%CI 0.75-0.98) with a specificity of 0.44 (95%CI 0.34-0.54). The PPV was 0.34

(95%CI 0.29-0.38) and a NPV of 0.94 (95%CI 0.83-0.98).

Discussion

Most published studies looking at the use of age adjusted d-dimers in the assessment of PE have

considered the conventional assay threshold of 500nm/ml, age adjusting to 10X the age of the patient.

These studies have demonstrated that when compared to conventional d-dimer thresholds, the age-

adjusted d-dimer has a higher specificity without reduction in sensitivity. There are limited studies in the

literature which have considered age adjusted d-dimer in the context of the Instrumentation Lab D-dimer

assay, which has a normal value threshold of 230nm/mL. This study has aimed to fill a gap in the literature

looked at age adjusted d-dimer in the context of the Instrumentation Lab D-dimer assay. Considering the

age-adjusted cut of is roughly half that of the conventional d-dimer assay, a cut off of 5X the patient’s age

was used. Whilst we demonstrated an improvement in specificity of over ten times when age adjusting the

d-dimer, the sensitivity was only minimally reduced by 6% to 94%. Both false negative results pertained to

CTPA confirmed sub-segmental PE’s with no segmental or lobar involvement. `These findings are similar to
7,8
studies by Douma, et al and Dutton et al . A recent study by Dutton et al. considered alternative age

adjustment thresholds showing an improved specificity with only a mild reduction in sensitivity when age
adjusting by 6X the patient’s age8. We have therefore also considered this is our data evaluation. In this

study, multiplication of the patient’s age by six rather than five increased the specificity from 27% to 44%

but reduced sensitivity from 94% to 90% due to an additional false negative result which was a segmental

PE.

Many argue that sub segmental PEs are not pathological and are simply part of the impact of physiological

aging on the lungs, therefore disputing the need for harmful diagnostics and treatments.

We conducted ad-hoc analysis where we considered sub-segmental PEs as negative findings. The sensitivity

of 5X age adjusted d-dimer in our cohort increased from 94% to 100% whilst the sensitivity of 6X age

adjustment increased from 90% to 94%.

Limitations

This is self-limiting retrospective study which carries its own inherent limitations. The findings however are

similar to those obtained in prospective studies using the traditional 500ng/mL threshold. The study

excludes patients with high risk scores and does not assess age adjusted D-dimer in varying age ranges.

Despite the selected population, the NPV is reassuringly high. It would be useful to perform a prospective

study using the age adjusted threshold with this d-dimer assay.

Conclusion

Age adjusting the d-dimer cut-off by 5X the patient’s age is safe when utilised in non-high risk patients,

under 50 years old. It an effective approach to increasing the specificity whilst maintaining a high level of

sensitivity in the clinical setting of suspected PE. The number of False positive results are therefore reduced

which is turns improve service utilization by minimizing further futile investigations and pharmacological

interventions. In our cohort 5X age adjustment would have resulted in 30 (23%) patient avoiding

unnecessary CTPAs and interim anticoagulation.

We would like to acknowledge the support from the radiology department at our affiliated district general

hospital, Inverclyde Royal Hospital, for providing the data required to carry out this study. Dr Shiva

Kotesvaren contributed to this study and is a named collaborator.

Conflict of interest disclosure

The Authors declare that there is no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit

sectors.
 References

1. Ruigómez A, Rodríguez LAG, Wallander MA, Johansson S, Jones R. Chest pain in general

practice: incidence, comorbidity and mortality. Fam Pract. 2006;23(2):167-174.

doi:10.1093/FAMPRA/CMI124

2. Diagnosing the Cause of Chest Pain. Accessed June 27, 2022.

https://www.aafp.org/pubs/afp/issues/2005/1115/p2012.html

3. Stein PD, Fowler SE, Goodman LR, et al. Multidetector Computed Tomography for Acute

Pulmonary Embolism. New England Journal of Medicine. 2006;354(22):2317-2327.

doi:10.1056/NEJMOA052367/SUPPL_FILE/NEJM_STEIN_2317SA1.PDF

4. Moresco RN, Vargas LCR, Voegeli CF, Vianna Santos RC. D-dimer and its relationship to

fibrinogen/fibrin degradation products (FDPs) in disorders associated with activation of coagulation

or fibrinolytic systems. J Clin Lab Anal. 2003;17(3):77-79. doi:10.1002/JCLA.10072

5. Robert-Ebadi H, Robin P, Hugli O, et al. Impact of the Age-Adjusted D-Dimer Cutoff to Exclude

Pulmonary Embolism: A Multinational Prospective Real-Life Study (the RELAX-PE Study).

Circulation. 2021;143:1828-1830. doi:10.1161/CIRCULATIONAHA.120.052780

6. Dutton J, Dachsel M, Crane R. Can the use of an age-adjusted D-dimer cut-off value help in our

diagnosis of suspected pulmonary embolism? Clinical Medicine. 2018;18(4):293-296.

doi:10.7861/CLINMEDICINE.18-4-293

7. Carrier M, Klok FA. Symptomatic subsegmental pulmonary embolism: to treat or not to treat?

Hematology. 2017;2017(1):237-241. doi:10.1182/ASHEDUCATION-2017.1.237

8. Douma RA, le Gal G, Söhne 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(7753):962. doi:10.1136/BMJ.C1475