Published online before print May 9, 2006, 10.1148/radiol.2401051129
(Radiology 2006;240:246-255.)
© RSNA, 2006
Incidental Pulmonary Emboli in Oncology Patients: Prevalence, CT Evaluation, and Natural History1
Gregory W. Gladish, MD,
Du Hwan Choe, MD,
Edith M. Marom, MD,
Bradley S. Sabloff, MD,
Lyle D. Broemeling, PhD and
Reginald F. Munden, DMD, MD
1 From the Departments of Diagnostic Radiology (G.W.G., E.M.M., B.S.S., R.F.M.) and Biostatistics (L.D.B.), University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 57, Houston, TX 77030; and Department of Radiology, Korea Cancer Center Hospital, Seoul, Korea (D.H.C.). Received July 7, 2005; revision requested September 8; revision received September 27; accepted October 18; final version accepted December 14.
Address correspondence to G.W.G. (e-mail: ggladish{at}mdanderson.org).
 |
ABSTRACT
|
|---|
Purpose: To retrospectively determine the prevalence and natural history of incidental pulmonary emboli in oncology patients, the number of such cases reported at initial thoracic computed tomographic (CT) image interpretation, and the factors that contribute to underdiagnosis.
Materials and Methods: Institutional review board approval, which included HIPAA-compliant access to protected health information and waived patient consent requirements, was obtained for this retrospective study. Four hundred three consecutive oncology patients (199 male, 204 female; age range, 14–87 years; mean age, 55 years) in whom adequate-quality multidetector thoracic CT was performed within a 10-day period for indications other than pulmonary emboli assessment were identified. There were 31 (7.7%) inpatients at the time of imaging. Each imaging case was reviewed by two independent radiologists, and all pulmonary emboli were confirmed by a panel of three thoracic radiologists. Clinical charts were reviewed for demographic data, embolus detection, and outcomes up to 2 years after the initial examination. Patient groups were compared by using 
2 and one-sided binomial tests.
Results: Sixteen (4.0%) of the 403 patients had pulmonary emboli. The highest prevalences were in patients with gynecologic malignancies (two of 13, 15%) and in those with melanoma (four of 41, 10%). Four (25%) of the 16 patients with emboli were identified at initial clinical CT image interpretation, and all had multiple emboli involving at least the lobar arteries. Missed emboli typically were solitary and involved smaller arteries; no other confounding factors were identified. Six (60%) of 10 patients with emboli who underwent any lower extremity imaging had deep vein thrombosis. With the exception of one patient, who was transferred back to the referring physician and lost to follow-up, all patients with reported pulmonary emboli were treated. Two patients had subsequent embolic events: one death despite treatment and one recurrent embolus in a nontreated patient.
Conclusion: Incidental pulmonary emboli were seen in 16 (4%) oncology patients but were initially reported in only four of them. The small size of involved arteries contributes to the failed detection at initial CT image interpretation, and patients with emboli in these small vessels may have deep vein thrombosis or recurrent emboli.
© RSNA, 2006
|
INTRODUCTION
|
|---|
Pulmonary embolism is a substantial cause of morbidity and mortality in the United States (1–4). Traditionally, the examination of a patient for pulmonary emboli has included nuclear medicine ventilation-perfusion scanning and conventional pulmonary angiography. Computed tomographic (CT) pulmonary angiography enables less invasive, direct demonstration of clots within the pulmonary artery system and has become the standard imaging method for the evaluation of pulmonary emboli in many institutions (5–7).
The technical advances that have made CT pulmonary angiography possible have been extended to routine thoracic CT. These advances include thinner collimation, breath-hold acquisitions, and optimized intravenous administration of contrast media. Although the technique used to perform routine thoracic CT differs from that used to perform CT pulmonary angiography, many routine thoracic CT examinations are adequate for the evaluation of pulmonary emboli.
Pulmonary emboli are detected incidentally in asymptomatic patients at a prevalence of approximately 1.0%–1.5% in the general patient population (8,9). Inpatients have a higher prevalence: around 4%–5% (9,10). Many patients found to have incidental pulmonary emboli have known risk factors for thromboembolic disease, such as clotting disorders, recent surgery, or malignancy (8–10). Oncology patients are at increased risk for pulmonary emboli (11,12), but the thoracic manifestations of their disease may distract the radiologist from complete assessment of the pulmonary arteries.
In addition, the clinical importance of these unsuspected emboli is unknown. To our knowledge, the natural history of asymptomatic, unsuspected pulmonary emboli that are not treated has been reported in no prior study. In two previous reports on incidentally detected pulmonary emboli, patient identification was based on the prospective examination of patients or the retrospective review of radiology reports (8,9). In both of these studies, the referring clinicians usually were alerted to the presence of incidental pulmonary emboli at the time the emboli were detected, and they may have altered their therapy on the basis of this information. The reporting of incidental emboli does alter therapy (8,13). Investigators in the most recent study of incidental emboli did not assess patient outcomes (10). Thus, the purpose of our study was to retrospectively determine the prevalence and natural history of incidental pulmonary emboli in oncology patients, the number of such cases reported at initial thoracic CT image interpretation, and the factors that contribute to underdiagnosis.
|
MATERIALS AND METHODS
|
|---|
Patients and CT Scanning Technique
Institutional review board approval, which included Health Insurance Portability and Accountability Act–compliant access to protected health information and waived informed patient consent requirements, was obtained for this retrospective study. Six hundred consecutive oncology patients who within a 10-day period beginning June 3, 2002, underwent intravenous contrast medium–enhanced thoracic CT for indications other than evaluation of pulmonary emboli were identified. In 482 of these patients, multidetector CT involving the acquisition of contiguous 3.75-mm images had been performed by using contrast timing (details to follow). The remaining 118 patients underwent either CT examinations that were performed by using single-detector scanners with thicker collimation or combined examinations that required different contrast timing; thus, they were excluded from further analyses.
For contrast medium administration, 100 mL of iohexol, 300 mg of iodine per milliliter (Omnipaque 300; Amersham Health, Princeton, NJ), was injected at 3.0 mL/sec with a 20-second scanning delay or at 2.5 mL/sec with a 30-second scanning delay. The rate of contrast medium injection was individually selected from among these two choices on the basis of the quality of peripheral intravenous access. For thoracic CT scanning with concomitant abdominal and/or pelvic CT scanning, the total intravenous contrast medium volume was increased to 150 mL, but the injection rate and scanning delay for the thoracic portion of the examination were not altered. Patients were excluded from the study if other contrast medium volumes, injection rates, or injection timing were used.
CT was performed by using a LightSpeed or LightSpeed Plus scanner (GE Medical Systems, Milwaukee, Wis) with 3.75-mm section collimation and a table speed of 11.25 mm per rotation. Typical exposure parameters were 320 mA at 120 kVp with a gantry speed of 0.5 second per rotation for the LightSpeed Plus scanner and 200 mA at 120 kVp with a gantry speed of 0.7 second per rotation for the LightSpeed scanner. For some patients, exposure parameters were adjusted according to body habitus. Contiguous 3.75-mm images were reconstructed with both standard and lung kernels.
Image Review and Final Patient Group
Each imaging case was independently reviewed by two chest radiologists from a group of four radiologists (G.W.G., D.H.C., E.M.M., B.S.S.) with 9–14 years experience. The images were evaluated at a locally developed picture archiving and communication system workstation using iSite (Stentor, Brisbane, Calif) image distribution software. The readers were free to use any window and level settings and both standard and lung kernel reconstructions. Multiplanar reformatting was available at a separate workstation (Vitrea; Vital Images, Plymouth, Minn). To markedly diminish recall bias, evaluations were performed at least 1 year after the initial clinical CT image interpretation.
Each of the two readers subjectively evaluated the imaging cases for the quality of vascular opacification on a three-point scale (ie, poor, adequate, excellent) and for the presence and severity of respiratory motion artifacts on a four-point scale (none, mild, moderate, severe). Patients were excluded if either reader rated the vascular opacification as poor or the respiratory motion artifacts as severe. Sixty-nine patients with poor pulmonary vascular opacification and two with severe respiratory motion artifacts were excluded from further evaluation. Eight of the remaining patients were determined not to have any confirmed malignancy and also were excluded.
The final study group consisted of 403 patients aged 14–87 years (mean, 55 years); 31 (7.7%) were inpatients at the time of imaging. There were 199 male patients, who were aged 15–85 years (mean, 56 years), and 204 female patients, who were aged 14–87 years (mean, 54 years). Twelve (5.9%) of the 204 female patients and 19 (9.5%) of the 199 male patients were inpatients at the time of imaging. The remaining 372 patients were outpatients.
The readers recorded the confounding imaging features—those intrathoracic abnormalities that could have limited the evaluation of the pulmonary arteries, such as lung masses, consolidation, effusion, or surgery. Standard criteria for identifying pulmonary emboli were used and included a sharply delineated pulmonary arterial filling defect present on at least two consecutive image sections and located centrally within the vessel or with acute angles at its interface with the vessel wall (Fig 1) (5). Each reader separately recorded the location of all emboli. The emboli were categorized according to the size of the affected pulmonary arterial vessels and the location of the embolus within the lobe of the lung. All disagreements were resolved and the presence of emboli was confirmed by a consensus panel of three fellowship-trained thoracic radiologists with 9–13 years experience: two members of the initial reading panel (G.W.G., B.S.S.) and an additional radiologist (R.F.M.) who did not perform any of the retrospective readings.
View larger version (89K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a: (a, b) Consecutive transverse contrast-enhanced CT images obtained at the level of the ascending aorta (Ao) in a 67-year-old man with metastatic melanoma show typical findings of pulmonary emboli (arrows) in both lower lobe pulmonary arteries, including sharply delineated pulmonary arterial filling defects present on consecutive sections and located centrally within the vessel or with acute angles at the interface with the vessel wall.
|
|
View larger version (93K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b: (a, b) Consecutive transverse contrast-enhanced CT images obtained at the level of the ascending aorta (Ao) in a 67-year-old man with metastatic melanoma show typical findings of pulmonary emboli (arrows) in both lower lobe pulmonary arteries, including sharply delineated pulmonary arterial filling defects present on consecutive sections and located centrally within the vessel or with acute angles at the interface with the vessel wall.
|
|
Clinical Review
Radiology reports and clinical notes were reviewed by two authors (D.H.C., G.W.G.) to determine if emboli were detected at the initial clinical CT image interpretation and if the reported emboli affected clinical management. Charts were reviewed to also determine the tumor type; presence of known metastasis; history of surgery, chemotherapy, or radiation therapy; ongoing anticoagulation therapy; and platelet count, prothrombin time, partial thromboplastin time, and international normalized ratio measured within a month before the CT examination. Patients who were judged to have emboli by the reading and consensus panels were followed up by one of the authors (G.W.G.) with evaluation of subsequent radiologic examination results and charts for the presence of emboli, deep vein thrombosis (DVT), and any subsequent embolic events up to 2 years after the initial examination.
Statistical Analyses
Statistical analyses were performed with SPSS, version 11.5, software (SPSS, Chicago, Ill). The prevalence of emboli in each tumor group was compared with the prevalence in the entire study group by using the one-sided binomial test. The frequencies of confounding imaging features, pulmonary embolus risk factors, and abnormal laboratory values in the patients with and in those without pulmonary emboli were compared by using the 2 test. The frequencies of confounding imaging features in the reported pulmonary emboli and unreported pulmonary emboli groups also were compared by using the 2 test. A P value of .05 was considered to indicate a statistically significant difference.
|
RESULTS
|
|---|
Prevalence of Pulmonary Emboli
In 14 patients, pulmonary emboli were identified by both readers. Twelve additional patients had possible emboli that were detected by only one reader; in two of these subjects, pulmonary emboli were detected by consensus. These detections resulted in a total of 16 patients with incidental pulmonary emboli—approximately 4.0% of the total study population. Emboli were present in nine (4.4%) of the 204 female patients and in seven (3.5%) of the 199 male patients. Emboli were present in two (6%) of the 31 inpatients and in 14 (3.8%) of the 372 outpatients. There was no significant difference in the prevalence of emboli between the inpatient and outpatient groups.
Of the 10 remaining patients with possible emboli that were detected by only one reader, eight had abnormalities that were deemed by the consensus panel to represent motion or volume-averaging artifacts. The other two patients had low-attenuating structures within the pulmonary arterial system that were judged not to represent acute pulmonary thromboemboli: One patient had a thrombus within an arterial stump following lobectomy (Fig 2). The other had intraarterial metastases that were demonstrated at subsequent examinations to be growing (Fig 3).
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a: (a) Transverse contrast-enhanced CT image obtained at the level of the left atrium (LA) in a 77-year-old man 2 years after right lower lobectomy for lung cancer shows a filling defect (arrow) in the right interlobar pulmonary artery. (b) Oblique coronal CT reformation along the right pulmonary artery (PA) in the same patient shows the thrombus (arrow) in the distal interlobar artery extending to the resection site (indicated by surgical clips [arrowhead]). Ao = aorta, SVC = superior vena cava.
|
|
View larger version (119K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b: (a) Transverse contrast-enhanced CT image obtained at the level of the left atrium (LA) in a 77-year-old man 2 years after right lower lobectomy for lung cancer shows a filling defect (arrow) in the right interlobar pulmonary artery. (b) Oblique coronal CT reformation along the right pulmonary artery (PA) in the same patient shows the thrombus (arrow) in the distal interlobar artery extending to the resection site (indicated by surgical clips [arrowhead]). Ao = aorta, SVC = superior vena cava.
|
|
View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a: (a) Initial transverse contrast-enhanced CT image obtained at the level of the aorta (Ao) and pulmonary artery (PA) in a 66-year-old woman with metastatic renal cell carcinoma shows a filling defect (arrow) in a segmental left upper lobe pulmonary artery. (b) Follow-up transverse contrast-enhanced CT image obtained 2 months later shows growth of the intraarterial lesion (arrow), which indicates that it is a tumor rather than a thrombus. In a and b, multiple pulmonary metastases (arrowheads) also are seen.
|
|
View larger version (70K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b: (a) Initial transverse contrast-enhanced CT image obtained at the level of the aorta (Ao) and pulmonary artery (PA) in a 66-year-old woman with metastatic renal cell carcinoma shows a filling defect (arrow) in a segmental left upper lobe pulmonary artery. (b) Follow-up transverse contrast-enhanced CT image obtained 2 months later shows growth of the intraarterial lesion (arrow), which indicates that it is a tumor rather than a thrombus. In a and b, multiple pulmonary metastases (arrowheads) also are seen.
|
|
The most common malignancies in our study group were lymphoma, lung cancer, and melanoma (Table 1). The largest numbers of emboli were seen in patients with lung cancer (n = 5) and in those with melanoma (n = 4). The highest prevalence of emboli was in the patients with gynecologic tumors (two of 13, 15%); the second highest prevalence was in the patients with melanoma (four of 41, 10%).
Risk Factors
A slight majority of patients had known metastatic disease at the time of CT (Table 2). About one-third of patients were currently undergoing chemotherapy or had completed their most recent course of chemotherapy within 1 month. Few of the examined patients had recently undergone surgery or radiation therapy. However, there was no significant difference in the frequency of any of these risk factors between the patients who did and those who did not have pulmonary emboli. There was also no substantial difference in the frequency of anticoagulation therapy at the time of imaging between the patients who did and those who did not have pulmonary emboli. Asymptomatic emboli were identified in two patients who were appropriately treated with anticoagulation therapy at the time of CT examination.
Laboratory coagulation values were normal in most of the patients for whom laboratory evaluation results were available (Table 3). There were no substantial differences in the distributions of laboratory values among the patients with emboli, although the number of these patients with available laboratory evaluation results was very small.
Extent and Distribution of Emboli
Fifty percent (n = 8) of the 16 patients with emboli had emboli in lobar or larger arteries (Table 4). Fifty percent (n = 8) of the patients with emboli also had multiple lobe involvement. Only one (6%) patient with emboli had them in only the subsegmental arteries (Fig 4). In two (12%) patients, the emboli involved the left lung only; in eight (50%), the emboli involved the right lung only; and in six (38%), the emboli involved both lungs. In four (25%) patients, the emboli were reported at initial CT image interpretation. All four of these clinically detected cases involved emboli in multiple lobes and included at least one embolus involving a lobar or larger artery. Six of the 12 emboli that were not documented in the original clinical report were solitary and involved only segmental or smaller arteries.
View larger version (68K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4: Transverse contrast-enhanced CT image obtained at the level of the aortic root (Ao) in a 59-year-old woman with metastatic uterine leiomyosarcoma shows a small pulmonary embolus (arrow) in a descending branch of the right lateral basal segmental pulmonary artery. No other emboli are present.
|
|
Clinical Follow-up
Four patients had clinically reported emboli. Three of these patients were started on anticoagulation therapy as a result of the report. The other patient had CT evidence of DVT also, but the decision regarding therapy was deferred to that patient's referring clinician at an outside institution. One of the patients with clinically reported emboli died of renal failure 17 days later. The other patients had no further evidence of embolic disease or complications related to anticoagulation therapy.
Most (n = 8, 67%) of the 12 patients with unreported emboli had no evidence of further embolic disease at clinical follow-up (range, 2 days to 24 months; average, 13 months). All four (33%) of the remaining patients had DVT that was demonstrated with other imaging modalities. One had known DVT that was being treated at the time of CT; this patient died of respiratory failure 20 days after imaging despite undergoing anticoagulation therapy and inferior vena cava filter placement. The death was attributed to a combination of pulmonary tumor progression and pulmonary emboli. One patient had symptoms of DVT at the time of imaging. She received an ultrasonography (US)-based diagnosis 3 days later (Fig 5) and was started on anticoagulation therapy. The other two patients had iliac DVT that was incidentally identified at pelvic CT (Fig 6). One of these patients received treatment for the DVT.
View larger version (88K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5: Lung cancer metastasized to the left hip in a 66-year-old woman with left hip pain and leg swelling but no dyspnea. Transverse contrast-enhanced CT image obtained at the level of the aortic root (Ao) shows an embolus (arrow) in the right posterior basal segmental pulmonary artery. Doppler US (not shown) performed 3 days after CT depicted an absence of blood flow in a noncompressible left superficial femoral vein, with normal flow in the superficial femoral artery.
|
|
View larger version (85K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a: Metastatic melanoma in a 68-year-old man. (a) Initial transverse contrast-enhanced CT image obtained at the level of the left atrium (LA) shows a small embolus (arrow) in the left anteromedial basal segmental pulmonary artery. (b, c) Transverse 3-month follow-up CT images show resolution of the left segmental pulmonary embolus (arrowhead in b) and a new embolus (arrow in c) in the right lower lobe pulmonary artery. (d) Transverse 3-month follow-up pelvic CT image obtained at the level of the femoral neck (F) shows a thrombus (arrows) at the confluence of the superficial and deep femoral veins.
|
|
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b: Metastatic melanoma in a 68-year-old man. (a) Initial transverse contrast-enhanced CT image obtained at the level of the left atrium (LA) shows a small embolus (arrow) in the left anteromedial basal segmental pulmonary artery. (b, c) Transverse 3-month follow-up CT images show resolution of the left segmental pulmonary embolus (arrowhead in b) and a new embolus (arrow in c) in the right lower lobe pulmonary artery. (d) Transverse 3-month follow-up pelvic CT image obtained at the level of the femoral neck (F) shows a thrombus (arrows) at the confluence of the superficial and deep femoral veins.
|
|
View larger version (71K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6c: Metastatic melanoma in a 68-year-old man. (a) Initial transverse contrast-enhanced CT image obtained at the level of the left atrium (LA) shows a small embolus (arrow) in the left anteromedial basal segmental pulmonary artery. (b, c) Transverse 3-month follow-up CT images show resolution of the left segmental pulmonary embolus (arrowhead in b) and a new embolus (arrow in c) in the right lower lobe pulmonary artery. (d) Transverse 3-month follow-up pelvic CT image obtained at the level of the femoral neck (F) shows a thrombus (arrows) at the confluence of the superficial and deep femoral veins.
|
|
View larger version (97K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6d: Metastatic melanoma in a 68-year-old man. (a) Initial transverse contrast-enhanced CT image obtained at the level of the left atrium (LA) shows a small embolus (arrow) in the left anteromedial basal segmental pulmonary artery. (b, c) Transverse 3-month follow-up CT images show resolution of the left segmental pulmonary embolus (arrowhead in b) and a new embolus (arrow in c) in the right lower lobe pulmonary artery. (d) Transverse 3-month follow-up pelvic CT image obtained at the level of the femoral neck (F) shows a thrombus (arrows) at the confluence of the superficial and deep femoral veins.
|
|
Imaging Follow-up
All patients with emboli who were followed up with imaging demonstrated resolution of the initially detected emboli at the first follow-up examination (range, 1–12 months; average, 4.7 months). One (8%) of the 12 patients with unreported emboli had developed new pulmonary emboli by the first follow-up CT examination at 3 months despite resolution of the original embolus (Fig 6). These new emboli also were not reported clinically. DVT was detected in this patient at 3-month follow-up pelvic CT, but no treatment was instituted. The new emboli had diminished at the next follow-up 5 months after the initial examination. The nuclear medicine venogram obtained at 5 months showed no evidence of residual lower extremity DVT. A segmental portion of the pulmonary embolus persisted to the final follow-up imaging examination at 8 months. The patient had no further reported evidence of recurrent emboli at the final clinical follow-up at 11 months.
No other pulmonary emboli were identified at completed imaging follow-up (range, 1–24 months; average, 10.4 months) of the patients with emboli. In all four patients with pulmonary emboli who underwent lower extremity venous Doppler US, DVT was demonstrated. Of eight patients with pulmonary emboli who underwent contrast-enhanced pelvic CT, three had DVT. Two patients underwent both lower extremity US and pelvic CT within 48 hours of chest CT. In one of these patients, DVT was demonstrated at both examinations; in the other, DVT was demonstrated at US only. The pelvic CT examinations were not optimized for the detection of DVT: CT images were obtained at the time of venous enhancement for evaluation of the urinary bladder and covered only short segments of the external iliac and common femoral veins. Overall, six (60%) of 10 patients who underwent some form of lower extremity imaging had DVT.
Confounding Factors
Thoracic masses and pulmonary consolidation were the most common potentially confounding factors (Table 5). There was no significant association between the presence of any confounding factor and the presence of pulmonary emboli. None of the confounding thoracic abnormalities was significantly more frequent in the patients whose emboli were reported at initial clinical CT image interpretation than in those whose emboli were not initially reported (Table 6). However, the patients whose emboli were not reported were significantly more likely to have only segmental and solitary emboli.
|
DISCUSSION
|
|---|
There have been preliminary assessments of the prevalence of asymptomatic pulmonary emboli in general outpatient and inpatient settings, with very limited assessment in other subgroups (8–10). The importance of these emboli is essentially unknown, as there have been no trials, to our knowledge, to assess the natural history of incidentally detected asymptomatic emboli. In addition, there has been little or no systematic effort to determine the extent of DVT associated with these emboli. These issues are especially difficult to address in patients with complex medical histories, such as oncology patients.
We observed an overall prevalence of incidental pulmonary emboli of 4.0% in this oncology patient population, with a prevalence of 3.8% in the outpatient group and 6% in the inpatient group. The difference in embolus prevalence between the outpatient and inpatient groups was not statistically significant. A variety of tumors—including pancreatic, ovarian, and brain tumors—are associated with an increased risk of pulmonary emboli and DVT (11,12). In this study, patients with gynecologic malignancies and melanoma showed a trend toward a higher prevalence of pulmonary emboli compared with the overall group. The lack of a difference in prevalence among the other tumors may have been due to a number of factors, including the exclusion of patients who were examined because of pulmonary embolus symptoms, the small sample size of some of the subgroups, and the overall low frequency of incidental emboli.
Results of previous studies (8–10) of incidental pulmonary emboli have suggested a prevalence of 1.0%–3.4% in the given patient populations, which included 20%–80% inpatients. The higher prevalence in the latest of these studies (10) can be attributed to a greater percentage of inpatients. In all of these studies, a large proportion of patients with emboli had cancer and many were inpatients. In all three studies, the prevalence of emboli in outpatients was lower than 1%. The present study included a higher prevalence of incidental emboli among both inpatients and outpatients. These findings presumably reflect the coagulopathic effects of malignancy. However, our use of thinner section collimation may have contributed to the higher prevalence of detected emboli in this study.
In 12 (75%) of 16 patients, the emboli identified by the reading and consensus panels were not detected at the initial clinical CT image interpretation. When we evaluated the confounding imaging factors in the detection of pulmonary emboli, only the size of the largest affected artery and the number of emboli were shown to be associated with the lack of CT-based clinical detection. The unreported emboli more frequently involved small arteries and were solitary, although emboli in lobar arteries and multiple emboli also were missed. Notably, all of the detected emboli were in lobar or larger arteries or were multiple in number. It appears that the small size of affected arteries and/or the limited extent of involvement may be a reason that pulmonary emboli are missed.
Review of the medical records and other imaging reports revealed that in the majority (n = 6, 60%) of patients with incidentally detected emboli who underwent any imaging of the lower extremity venous system, DVT was demonstrated. One of these patients subsequently died of respiratory failure. Three patients with DVT had emboli involving only segmental or smaller arteries. These findings suggest that asymptomatic and even small emboli may well be the harbinger of more important thromboembolic disease and support the current clinical practice of treating all patients with pulmonary emboli (6,13,14).
There were several possible limitations to our study. The prevalence of incidental emboli may have been underestimated because the imaging techniques used were similar to those that were commonly used for pulmonary emboli assessment about 5–7 years ago. The use of thinner section collimation and higher contrast medium injection rates might have enabled the detection of more emboli. Chronic emboli may have been excluded because of the requirement that some component of the embolus be free within the lumen or have angles acute to the vessel wall.
The exclusion of patients with poor contrast opacification and severe respiratory artifacts at CT also could have altered the apparent prevalence of incidental emboli. Patients with emboli may have altered contrast dynamics and greater respiratory difficulty that result in suboptimal examinations. However, the inclusion of poor-quality images probably would have resulted in a decrease in the apparent prevalence of emboli because of the difficulty in detecting small emboli at these examinations. The ratings for contrast opacification and respiratory motion artifact were subjective: Patients were excluded when either reader assigned the worst ratings; this ensured that all poor-quality images were excluded. In a previous study (10), there was a minor change, of 0.3%, in the prevalence of incidental emboli with use of wide window settings. The contribution of fixed window and level settings to the failed CT-based clinical detections of emboli could not be assessed in our study. At the time the examinations were performed, both hard- and soft-copy image interpretations were being used at our institution. The failed CT-based clinical detections of multiple emboli and of emboli involving lobar arteries suggest that the use of fixed window and level settings was not the only factor.
In conclusion, incidental pulmonary emboli were present in 4% of the general oncology population in this study, with higher prevalences likely among patients with gynecologic malignancies and those with melanoma. The detection of incidental pulmonary emboli remains a challenge. The majority of emboli go undetected at routine thoracic CT image interpretation, probably because of the small size of the affected arteries, but the presence of incidental emboli suggests the existence of other thromboembolic disease. These factors indicate a need to carefully evaluate pulmonary arteries at routine imaging examinations.
|
ADVANCES IN KNOWLEDGE
|
|---|
- Incidental pulmonary emboli were seen in 4% of oncology patients, more frequently in those with melanoma or gynecologic tumors.
- Only 25% of incidental emboli were reported at initial interpretation, partly because of their small size.
- Even patients with small emboli may have DVT or recurrent emboli.
|
FOOTNOTES
|
|---|
Abbreviations: DVT = deep vein thrombosis
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, G.W.G.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, G.W.G.; clinical studies, G.W.G., D.H.C., E.M.M., R.F.M.; statistical analysis, G.W.G., L.D.B.; and manuscript editing, all authors
|
References
|
|---|
- Stein PD, Kayali F, Olson RE. Estimated case fatality rate of pulmonary embolism, 1979 to 1998. Am J Cardiol 2004;93:1197–1199.[CrossRef][Medline]
- Laack TA, Goyal DG. Pulmonary embolism: an unsuspected killer. Emerg Med Clin North Am 2004;22:961–983.[CrossRef][Medline]
- Kearon C. Natural history of venous thromboembolism. Circulation 2003;107(23 suppl 1):I22–I30.
- Horlander KT, Mannino DM, Leeper KV. Pulmonary embolism mortality in the United States, 1979–1998: an analysis using multiple-cause mortality data. Arch Intern Med 2003;163:1711–1717.[Abstract/Free Full Text]
- Wittram C, Maher MM, Yoo AJ, Kalra MK, Shepard JA, McLoud TC. CT angiography of pulmonary embolism: diagnostic criteria and causes of misdiagnosis. RadioGraphics 2004;24:1219–1238.[Abstract/Free Full Text]
- Trowbridge RL, Araoz PA, Gotway MB, Bailey RA, Auerbach AD. The effect of helical computed tomography on diagnostic and treatment strategies in patients with suspected pulmonary embolism. Am J Med 2004;116:84–90.[CrossRef][Medline]
- Stein PD, Kayali F, Olson RE. Trends in the use of diagnostic imaging in patients hospitalized with acute pulmonary embolism. Am J Cardiol 2004;93:1316–1317.[CrossRef][Medline]
- Winston CB, Wechsler RJ, Salazar AM, Kurtz AB, Spirn PW. Incidental pulmonary emboli detected at helical CT: effect on patient care. Radiology 1996;201:23–27.[Abstract/Free Full Text]
- Gosselin MV, Rubin GD, Leung AN, Huang J, Rizk NW. Unsuspected pulmonary embolism: prospective detection on routine helical CT scans. Radiology 1998;208:209–215.[Abstract/Free Full Text]
- Storto ML, Di Credico A, Guido F, Larici AR, Bonomo L. Incidental detection of pulmonary emboli on routine MDCT of the chest. AJR Am J Roentgenol 2005;184:264–267.[Abstract/Free Full Text]
- Ziegler S, Sperr WR, Knobl P, et al. Symptomatic venous thromboembolism in acute leukemia: incidence, risk factors, and impact on prognosis. Thromb Res 2005;115:59–64.[CrossRef][Medline]
- Lee AY, Levine MN. Venous thromboembolism and cancer: risks and outcomes. Circulation 2003;107(23 suppl 1):I17–I21.
- Romano WM, Cascade PN, Korobkin MT, Quint LE, Francis IR. Implications of unsuspected pulmonary embolism detected by computed tomography. Can Assoc Radiol J 1995;46:363–367.[Medline]
- Eyer BA, Goodman LR, Washington L. Clinicians' response to radiologists' reports of isolated subsegmental pulmonary embolism or inconclusive interpretation of pulmonary embolism using MDCT. AJR Am J Roentgenol 2005;184:623–628.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
P U Dalal
Non-metastatic manifestations of cancer in the chest
Imaging,
September 1, 2008;
20(3):
194 - 204.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. R. Larici, L. Calandriello, F. Maggi, M. Torge, L. Bonomo, G. W. Gladish, D. H. Choe, E. M. Marom, B. S. Sabloff, and R. F. Munden
Prevalence of Incidental Pulmonary Emboli in Oncology Patients
Radiology,
December 1, 2007;
245(3):
921 - 922.
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Remy-Jardin, M. Pistolesi, L. R. Goodman, W. B. Gefter, A. Gottschalk, J. R. Mayo, and H. D. Sostman
Management of Suspected Acute Pulmonary Embolism in the Era of CT Angiography: A Statement from the Fleischner Society
Radiology,
November 1, 2007;
245(2):
315 - 329.
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. G. Cronin, D. G. Lohan, M. Keane, C. Roche, and J. M. Murphy
Prevalence and Significance of Asymptomatic Venous Thromboembolic Disease Found on Oncologic Staging CT
Am. J. Roentgenol.,
July 1, 2007;
189(1):
162 - 170.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. R Desai
Unsuspected pulmonary embolism on CT scanning: yet another headache for clinicians?
Thorax,
June 1, 2007;
62(6):
470 - 472.
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
