An audit of the utility of the D-dimer test in the diagnosis of pulmonary embolism in a private emergency unit in Johannesburg
Date
2014-08-25
Authors
Schur, Amanda J
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Background: The D-Dimer test has a high negative predictive value used primarily to
exclude clinically suspected possible thrombo-embolic disease. In Emergency Unit (EU)
practice, this test is often done not only for suspected Pulmonary Emboli (PE) but also to
rule out atypical PE. In South Africa, diagnostic usefulness of this test has not been
evaluated in a private hospital EU. The health profile of patients presenting in public and
private EUs is different and therefore, it was hypothesized that the usefulness of the DDimer
test in these two settings may be different. Results of this study may inform private
hospital EU best practice in the optimal utilization of this test.
Objective: To evaluate the usefulness of the D-Dimer test in the diagnosis of PE at the
Morningside MediClinic (MMC) private hospital EU in Johannesburg, South Africa.
Patients and Methods: After approval by the University of the Witwatersrand Human
Research Ethics Committee, audit of clinical records was done at the MMC EU from 1
March to 1 June 2009. Informed consent was not required from study subjects as the study
was done retrospectively with data extracted from clinical records in an anonymous and
delinked fashion. The study population included all patients who had a D-Dimer test done
in the MMC EU as part of their diagnostic workup. Extracted data included demographic
information, diagnoses and confirmatory tests done. Continuous and categorical variables
of data collected were summarized using Stastistica version 9.0 statistical package. A
Wells Score was calculated according to the Wells Criteria.
Results: In the study period, 189 of 2948 (5%) patients seen at MMC EU had D-Dimers
measured. Their population mean age was 57 years (range 38 – 84 years) and 51% were
males. Positive D-Dimers were present in 40 (21%) of the total patient population sample
group (189 patients). Within the diagnostic categories, the following percentages were the
results found per category of the positive D-Dimers within each category: PE (5)(100%),
Chest Infection (5)(56%), AMI (2)(33%), Arrhythmia (2)(33%), Hypertension (2)(25%),
Chest Pain (6)(14%), Anxiety (3)(23%), Headache (1)(14%), Syncope (1)(14%) and
Others (13)(32%). The mean Wells Score in PE was 3.6 (3.0-4.5.) indicating medium
probability of PE. All other diagnostic groups had low probability Wells Scores. It was
impossible to comment on findings in public hospitals, as there is no known literature
found to date on an audit performed concerning the usefulness of the D-Dimer test in a
public hospital or any of the public sector, in Johannesburg or elsewhere in South Africa,
regarding the diagnosis of PE. However, data has been published by other countries
regarding the D-Dimer in various hospital and EU settings (public and private).
Conclusion: In the cohort, the D-Dimer was done in only a fifth of patients seen at the
private MMC EU and it was positive in less than half of cases. The test yield was highest
in PE and had high negative predictive value in more than half of non-PE diagnoses.
Therefore, the results suggested that a positive D-Dimer is highly predictive of a diagnosis
of PE in this private EU. A negative D-Dimer result appears to be largely associated with
any of the non PE wide differential of diagnoses.