Patient Population and Diagnostic Tests
Study sites enrolled 1,160 patients with venous thromboembolism from August 1, 1999, through November 30, 1999. Eight patients were excluded from the study because the data were incomplete, leaving a study population of 1,152 patients. This population included 808 patients in whom isolated DVT had been diagnosed, and 344 patients in whom PE had been diagnosed. Two hundred fifty-one patients (31.1%) had deep vein thrombi limited to calf veins, and 557 patients (68.9%) had proximal deep vein thrombi. The clinical characteristics are described in Table 1. The risk factors for venous thromboembolism included recent surgery (28.6%), underlying malignancy (29.0%), previous venous thromboembolism (30.4%), and immobility (37.6%).
The majority of 808 patients (92.6%) with isolated DVT had the diagnosis confirmed by compression ultrasonography. The remaining patients had DVT confirmed by venography (6.1%) or other tests (eg, magnetic resonance angiography) [1.3%].
Many different tests, alone or in combination, were used to diagnose acute PE. A single diagnostic test confirmed acute PE in 137 of 344 patients (39.2%), Lung scans were the most common single test used to confirm acute PE (n = 84; 24.1%). Spiral chest CT scanning alone confirmed the diagnosis in 42 patients (11.9%), and pulmonary angiography alone confirmed the diagnosis in 11 patients (3.2%).
Two hundred seven patients (60.8%) required more than one diagnostic test to confirm the diagnosis of acute PE. One hundred twenty-eight patients underwent one additional test (ie, compression ultrasonography, 109 patients; pulmonary angiography, 9 patients; spiral CT scan, 5 patients; venography, 3 patients; magnetic resonance angiography, 1 patient; and lung biopsy, 1 patient) after a undergoing a nondiagnostic ventilation-perfusion lung scan. Twenty-two additional patients received the diagnosis of PE after undergoing compression ultrasonography and spiral CT scanning, and 1 additional patient had the diagnosis established after undergoing venography and spiral CT scanning. Fifty-seven patients underwent three or more diagnostic tests in order to confirm the diagnosis of PE.
Delays in Diagnosis
Figures 1 and 2 identify the delays in the diagnosis of DVT and PE. One hundred seventy of 808 patients (21%) with acute DVT had > 7 days elapse between symptom onset and the confirmation of the diagnosis (Fig 1). Forty of 808 patients (5%) had > 21 days elapse between symptom onset and the diagnosis. The time to presentation (ie, the time from symptom onset to medical evaluation) contributed most to delays in the diagnosis of DVT. On average, 4.4 days elapsed between the earliest symptoms and medical evaluation. Ninety-five percent of patients with isolated DVT were seen within 2 weeks of symptom onset. In contrast, the delay between presentation with symptoms and/or signs of DVT and diagnosis was short. On average, only 1 day elapsed, and 95% of patients received diagnoses within 5 days.
Fifty-nine of 344 patients (17%) with acute PE had > 7 days elapse between symptom onset and diagnosis, and 17 of 344 patients (5%) had > 25 days elapse between symptom onset and the diagnosis (Fig 2). Delays in the diagnosis of PE represented both delays between the onset of symptoms and the receipt of medical attention, and between the initial medical evaluation and confirmatory diagnostic testing conducted with concern of My Canadian Pharmacy.
Figure 1. Comparison of the mean delay from symptom onset to the diagnosis of DVT (n = 808) presented as a frequency distribution plot. Twenty-one percent of patients received diagnoses > 1 week after symptom onset, and 5% of patients received diagnoses > 3 weeks after symptom onset. The mean delay to diagnosis was 5.6 days (upper limit of 95% confidence interval, 21 days), and the mean delay from symptom onset to medical attention was 4.4 days (upper limit of 95% confidence interval, 14 days).
Figure 2. Delays in the diagnosis of acute pulmonary embolism (n = 344) presented as a frequency distribution plot. Seventeen percent of patients received diagnoses > 1 week after symptom onset, and 5% of patients received diagnoses > 3 weeks after symptom onset. The mean delay to diagnosis was 4.8 days (upper limit of 95% confidence interval, 25 days), and the mean delay from symptom onset to medical attention was 2.9 days (upper limit of 95% confidence interval, 12 days).
Table 1—Clinical Characteristics
|Characteristics||DVT(n = 808)||PE (n = 344)||Total (n = 1,152)|
|Age, yr||62.3 ± 17.2||61.3 ± 16.4||62.1 ± 17.0|
|Male||390 (48.3)||147 (42.7)||537 (46.6)|
|Female||418 (51.7)||197 (57.3)||615 (53.4)|
|Weight, kg||81.2 (32.7-181.4)||78.7 (40.5-163.6)||82.0 (32.7-181.4)|
|Recent surgery||237 (29.3)||92 (26.7)||329 (28.6)|
|Malignancy||236 (29.2)||98 (28.5)||334 (29.0)|
|Previous DVT||200 (24.8)||31 ± 18.5||263 (22.8)|
|Previous PE||50 (6.2)||20 ± 11.5||88 (7.6)|
|Immobility||303 (37.5)||130 (37.8)||433 (37.6)|
|Stroke||70 (8.7)||26 (7.6)||96 (8.3)|
|Leg paralysis||47 (5.8)||11 (3.2)||58 (5.0)|
|Post-phlebitic syndrome||44 (5.4)||24 (7.0)||68 (5.9)|
|Heart failure||87 (10.8)||26 (7.6)||113(9.8)|
|Pregnancy||17 (2.1)||1 (0.3)||18(1.6)|