Published online before print August 26, 2005, 10.1148/radiol.2371041686
(Radiology 2005;237:338-341.)
© RSNA, 2005
Postpneumonectomy Pulmonary Artery Stump Thrombosis: CT Features and Imaging Follow-up1
Boon Han Kwek, MD and
Conrad Wittram, MB, ChB
1 From the Department of Radiology, Division of Thoracic Radiology, FND 202, Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114. Received October 1, 2004; revision requested December 7; revision received December 22; accepted February 1, 2005.
Address correspondence to C.W. (e-mail: cwittram{at}partners.org).
 |
ABSTRACT
|
|---|
PURPOSE: To retrospectively evaluate the computed tomographic (CT) features of pulmonary artery stump thrombosis at initial and follow-up CT.
MATERIALS AND METHODS: The study was approved by institutional review board, which waived informed consent, and was HIPPA compliant. All patients who had undergone pneumonectomy and CT from January 2001 to August 2003, as identified with data search system, were included. Eighty-nine patients (49 men, 40 women; mean age, 60 years) were studied. Thrombus identification, categorization (concave or convex), and stump and thrombus measurements were made by two radiologists in consensus. The use of anticoagulation therapy was determined from patients' charts. The t test was used.
RESULTS: Initial CT scans were obtained 34 months ± 67 (standard deviation) after pneumonectomy; multiple CT scans were obtained in 58 patients during follow-up of 25.1 months ± 24.8. Eleven (12.4%) of 89 patients had stump thrombi with near equal frequency on either side. Five concave and six convex thrombi were initially identified. Anticoagulation was not commenced for stump thrombosis. The mean length of the right stump (31 mm ± 10) was greater than that of the left stump (13 mm ± 7) (P < .01). After a right and left pneumonectomy, there was a significant difference between the length of the stump in patients with (right, 40 mm ± 14; left, 21 mm ± 11) and patients without thrombosis (right, 30 mm ± 9; left, 12 mm ± 6) (P = .027 and P < .01, respectively). Follow-up CT scans were not available in four cases. CT findings demonstrated a reduction in thrombus size in four patients (one received anticoagulation therapy for concomitant pulmonary embolism). Two patients had stable concave thrombi, one with an initial concave thrombus developed convex thrombus, and one with an initial convex thrombus developed concave thrombus. No thrombi propagated outside of the stump.
CONCLUSION: There is a relationship between stump length and the development of in situ thrombosis. The data suggest a rather benign natural history.
© RSNA, 2005
|
INTRODUCTION
|
|---|
Thrombus formation within the pulmonary artery stump has been known to occur since the early days of pneumonectomy, albeit so recognized only at autopsy (1–4). In 1966, Chuang et al (4) reported two cases of fatal thromboembolism in a series of 200 pneumonectomies, which they attributed to stump thrombi seen during autopsy. In 1993, Takahashi et al (5) detected pulmonary artery stump thrombi in three patients at computed tomography (CT) performed 3–4 months after a right pneumonectomy. The outcome of these patients was however not known. Wechsler et al (6) described a patient with a pulmonary artery stump thrombus following right middle and lower lobe resection that remained unchanged on a CT scan obtained 6 months later, with no evidence of pulmonary embolism.
At present, little is known about the frequency of occurrence and the natural history of pulmonary artery stump thrombosis. To our knowledge, there is no study other than the aforementioned case reports and autopsy series on this subject. Thus, the purpose of our study was to retrospectively evaluate the CT features of pulmonary artery stump thrombosis at initial and follow-up CT imaging.
|
MATERIALS AND METHODS
|
|---|
Patients
This retrospective study was approved by our institutional review board, which waived informed consent, and the study was compliant with the Health Insurance Portability and Accountability Act. The study period included all patients who had undergone a pneumonectomy and chest CT at our institution from January 2001 to August 2003. Patients were identified with our radiology database search system (Folio, Woburn, Mass). When identified, all postpneumonectomy contrast material–enhanced chest CT scans were reviewed, and the time between pneumonectomy and subsequent CT was recorded. Patients who did not receive intravenous contrast material were excluded. Therefore, a total of 89 patients were included in our study, of whom 42 underwent right and 47 underwent left pneumonectomy. The indications for CT included follow-up of lung cancer imaging in 61 cases; suspected pulmonary embolism in 10 cases; chest pain, fever, or progressive dyspnea in each of five cases; pneumonia in two cases; and lower extremity edema in one case. There were 49 male and 40 female patients (mean age, 60 years; range, 42–87 years). The clinical indication for pneumonectomy was malignancy in 85 patients, infection in two patients, and gunshot injury and massive hemoptysis secondary to pulmonary hypertension in one patient each.
Imaging and Interpretation
Four–detector row CT scans (GE Medical Systems, Milwaukee, Wis) were obtained from the lung apices to the bases by using 5-mm section thickness, 15-mm table feed, 140 kVp, 220–280 mA, 0.8-second scan time, and 25–30-second scan delay. All patients received an injection of 100 mL of ioxilan (300 mg of iodine per milliliter, Isovue; Bracco Diagnostics, Princeton, NJ) at a rate of 2–3 mL/sec. The CT scans were retrospectively reviewed by two radiologists (B.H.K. and C.W. with 3 and 5 years of experience, respectively, in interpreting CT scans for pulmonary embolism) and were read in consensus by using mediastinal windows settings (window width, 350 HU; window level, 40 HU) on a picture archiving and communication system monitor.
On a CT scan, the distal end of the pulmonary artery stump was indicated by either surgical clips or was outlined by mediastinal fat. The length of the pulmonary artery stump was calculated as the maximal measurement along the longitudinal axis of the stump with reference to a line projected from the inner border of the remaining pulmonary artery (Fig 1). The patients' initial postoperative CT scan was used for all measurements. The presence and morphology of the soft-tissue attenuation within the pulmonary artery stump were noted, as well as the maximal length and attenuation measurement. Attenuation measurements were performed to explore the possible differences in the maturity of thrombi. During the initial part of the study, it became clear that there are two shapes of thrombi, which are similar to the configuration of acute or chronic thromboembolic disease. Therefore, we decided to use these descriptive terms during the study. Concave thrombus was recognized as intravascular soft tissue that has an inner concave margin (forms a cap at the tip of the stump) with an obtuse angle to the wall of the pulmonary artery. Convex thrombus was recognized as intravascular soft tissue with a sharp margin that forms a convex margin with acute angles to the wall of the pulmonary artery. For attenuation measurements, the region of interest was positioned on the middle image of three that demonstrated the thrombus. The diameter of the region of interest was selected to be half the diameter of the thrombus being measured and was positioned to avoid streak artifacts.
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Transverse contrast-enhanced CT scan at the level of main pulmonary artery depicts the method of measuring the right pulmonary artery stump. Length of the pulmonary artery stump was calculated as the maximal measurement along the longitudinal axis of the stump (double arrow) with reference to a line projected from the inner border of the remaining pulmonary artery (dotted line).
|
|
Statistical Analysis
Statistical samples were compared by using a two-sample t test by assuming unequal variances. P values were computed by using one-tailed distribution. A P value of less than .05 was considered to indicate a significant difference. Data were collected and analyzed by using software (Excel; Microsoft, Seattle, Wash).
|
RESULTS
|
|---|
Initial and Follow-up CT Scans
Thirty-one patients underwent a single CT examination, while 58 patients underwent multiple CT examinations (mean, 3.5 scans ± 3.2 [standard deviation]; range, 1–20 scans) following pneumonectomy. The period between pneumonectomy and the first CT scan in the group with stump thrombosis was 27 months ± 29; in the group without stump thrombosis, it was 35 months ± 72 (P = .5). Therefore, the time between the patient's operation and the first CT scan is not a confounding factor. The period of follow-up between the first and the last CT scan was 25.1 months ± 24.8 (range, 2 days to 139 months). Of the 89 patients in the study, 11 (12.4%) were found to have thrombi in the pulmonary artery stump during the follow-up. The follow-up period was similar for patients with (15 months ± 13) and patients without (26 months ± 26) (P = .19) stump thrombosis and was thus not a confounding factor.
Thrombi
Initial thrombi consisted of five concave and six convex thrombi. One of the patients had an initial concave thrombus and developed a convex thrombus (Fig 2), and another patient had an initial convex thrombus that organized into a concave thrombus (Fig 3). The maximal length of the convex thrombi (23 mm ± 8) was greater than that of the concave thrombi (8 mm ± 4) (P < .01), but the attenuation measurements (concave, 50 HU ± 17; convex, 36 HU ± 5) (P = .12) were similar.
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Pulmonary artery stump thrombosis in a 55-year-old woman who underwent right pneumonectomy for recurrent chest infection resulting from postirradiation fibrosis. (a) Transverse contrast-enhanced CT scan at the level of main pulmonary artery obtained 9 months after pneumonectomy shows intravascular soft tissue (arrow) that has a concave margin with respect to contrast material (forms a cap at the end of the stump). (b) Subsequent CT scan obtained 11 months after pneumonectomy shows intravascular soft tissue (arrow) with a sharp convex margin, which forms acute angles to the vessel wall. This convex thrombus has developed on top of an existing concave thrombus.
|
|
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Pulmonary artery stump thrombosis in a 55-year-old woman who underwent right pneumonectomy for recurrent chest infection resulting from postirradiation fibrosis. (a) Transverse contrast-enhanced CT scan at the level of main pulmonary artery obtained 9 months after pneumonectomy shows intravascular soft tissue (arrow) that has a concave margin with respect to contrast material (forms a cap at the end of the stump). (b) Subsequent CT scan obtained 11 months after pneumonectomy shows intravascular soft tissue (arrow) with a sharp convex margin, which forms acute angles to the vessel wall. This convex thrombus has developed on top of an existing concave thrombus.
|
|
View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Pulmonary artery stump thrombosis in a 63-year-old man who underwent left pneumonectomy for lung cancer. (a) Transverse contrast-enhanced CT scan at the level of left pulmonary artery obtained 2 months after pneumonectomy shows a convex thrombus (arrow) in the left pulmonary artery stump. The patient did not receive anticoagulation therapy. (b) Subsequent CT scan obtained 4 months after pneumonectomy demonstrates organization of the thrombus to form a concave shape (arrow).
|
|
View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Pulmonary artery stump thrombosis in a 63-year-old man who underwent left pneumonectomy for lung cancer. (a) Transverse contrast-enhanced CT scan at the level of left pulmonary artery obtained 2 months after pneumonectomy shows a convex thrombus (arrow) in the left pulmonary artery stump. The patient did not receive anticoagulation therapy. (b) Subsequent CT scan obtained 4 months after pneumonectomy demonstrates organization of the thrombus to form a concave shape (arrow).
|
|
Stump Length
The concave stump thrombi were present on the initial CT scan in four patients, ranging from 4 months to 7.5 years after pneumonectomy. One patient had multiple CT scans that showed a normal pulmonary stump, and the patient developed a concave thrombus 13 months after pneumonectomy. With regard to follow-up imaging, one patient did not undergo follow-up CT, one patient had complete spontaneous resolution of the concave thrombus, two patients had concave thrombi that did not show any change over a period ranging from 8 to 17 months, while the last patient developed a convex thrombus on top of the existing concave thrombus (Fig 2).
The convex stump thrombi were present on the initial CT scan obtained 2 months to 4.5 years after pneumonectomy in five of six patients. One patient underwent multiple CT examinations that showed a normal pulmonary stump, and the patient developed a convex thrombus 6 years after pneumonectomy. Three of six patients with convex thrombi did not undergo any subsequent CT examinations. Of the remaining three patients with follow-up imaging studies, one showed complete spontaneous resolution of the convex thrombus (at subsequent CT performed 10 months later), one showed spontaneous decrease in the size of the thrombus (Fig 3), while the last patient received anticoagulation therapy for a concomitant lower lobe pulmonary embolus, with a decrease in the size of the stump thrombus from 26 to 10 mm over a 5-week period.
A total of four patients with either convex or concave stump thrombi did not undergo follow-up CT. A 64-year-old patient with a convex thrombus died of pulmonary hypertension 2 months after CT. The 1.0-cm concave thrombus was not believed to be the cause of death since the patient was also treated with warfarin for preexisting pulmonary hypertension. An 87-year-old man with a 3.5-cm convex thrombus was noted to be well with no signs or symptoms of pulmonary embolism 4 months after CT; further clinical data were, however, not available. A 52-year-old woman with common variable immunodeficiency disease and a convex thrombus later died of pneumonia. The medical record of the last patient could not be obtained because he returned to his country for further treatment.
The length of the pulmonary artery stump was greater on the right (31 mm ± 10; range, 12–55 mm) than on the left (13 mm ± 7; range, 3–40 mm) (P < .01). However, thrombosis within the pulmonary artery stump occurred with near equal frequency between patients with left (n = 5) and patients with right (n = 6) pneumonectomy. The frequencies of thrombi within the left (three convex, two concave) and right (three convex, three concave) pulmonary artery stumps were also similar.
In patients who had undergone right pneumonectomy, there was a significant difference between the length of the pulmonary artery stump in patients with (40 mm ± 14) and patients without (30 mm ± 9) thrombosis (P = .027). Also, after left pneumonectomy, the length of the pulmonary artery stump in patients with thrombosis (21 mm ± 11) was significantly greater than that in patients without thrombosis (12 mm ± 6) (P < .01).
|
DISCUSSION
|
|---|
Pulmonary artery stump thrombosis was reported as early as 1938 by Crafoord (1), who described three patients in whom thrombi at the site of pulmonary artery ligation were found at autopsy. Further reports on this matter were based exclusively on autopsy findings (2–4). Among 200 consecutive pneumonectomies, Chuang et al (4) described two patients who died within a week of pneumonectomy; death was attributed to pulmonary artery stump thrombi discovered at autopsy. Chuang et al also surveyed 600 thoracic surgeons, and 46 of 332 respondents reported experience with 64 cases of autopsy-proved stump thrombosis (4). They concluded with "a plea for diligent search for this potentially lethal but curable postpneumonectomy complication." As of present, there is little information about the incidence and natural history of pulmonary artery stump thrombosis.
In our series of 89 patients, 11 (12.4%) developed pulmonary artery stump thrombosis that was detected on postpneumonectomy CT scans. No comparison can be made with other studies, since there are only two prior case reports about pulmonary artery stump thrombosis observed at CT (5,6).
In the majority (82%) of patients with pulmonary artery stump thrombosis, the thrombus was present on the initial CT scan. The remaining patients had an initial CT scan with a normal pulmonary artery stump that developed thrombus subsequently. Thus, a normal pulmonary artery stump at initial CT does not preclude the development of thrombus formation.
Follow-up CT scans were available in seven patients with stump thrombosis. Reduction in thrombus size was observed in four patients (three with convex and one with concave thrombus). Two patients had concave thrombus that remained stable at follow-up CT, while the last patient had a convex thrombus that developed on top of an existing concave thrombus.
One patient with a convex stump thrombus also had an embolus in the contralateral lower lobe pulmonary artery. It is impossible to determine whether the pulmonary embolus arose from the pulmonary artery stump or from deep veins, since the coincidence rate of pulmonary embolism is 1.5%–3.4% at CT performed for an indication other than pulmonary embolism (7,8).
None of the patients developed propagation of the thrombus outside of the stump. Interestingly, three of the four patients who showed reduction in thrombus size did not receive any anticoagulation therapy. Our limited experience with a small number of patients suggests a benign natural history of pulmonary artery stump thrombosis. However, more data with a larger group of patients with stump thrombosis would be needed to verify this initial finding.
Pulmonary artery stump thrombi were present with almost equal frequency following either left or right pneumonectomy, despite the greater length of the right artery stump (right, 31 mm ± 10; left, 13 mm ± 7; P < .01). This observation is likely a consequence of a change in the pulmonary artery flow dynamics, which does not affect each side equally. This finding is in keeping with the result of a survey conducted by Chuang et al (4) in which 26 right- and 12 left-sided autopsy-proved stump thrombi were reported. The site of pneumonectomy in a further 26 cases was, however, not indicated in the survey result. In patients who underwent left pneumonectomy, the pulmonary artery stump length for those with thrombus (21 mm ± 11) was longer than that for those without thrombus (12 mm ± 6) (P < .01). Similarly, in patients who underwent right pneumonectomy, there was a significant difference between stump length in patients with (40 mm ± 14) and patients without (30 mm ± 9) stump thrombus (P = .027). These results demonstrate the contribution of stump length to thrombus formation, which is a consequence of a change in flow dynamics. It is, therefore, prudent to leave the pulmonary artery stump as short as possible.
This study was limited by the small number of patients with pulmonary artery stump thrombosis, as well as incomplete clinical data in one of the patients with a stump thrombus. Any adverse outcome in the patient without follow-up data may well change the rather benign natural history of stump thrombosis suggested by our results. A study with more patients undergoing CT follow-up of stump thrombosis and correlation with clinical outcome are needed to verify our preliminary findings, since there are currently limited data on this issue.
In conclusion, pulmonary artery stump thrombosis occurs more frequently than suggested by the results from autopsy series. We found near equal frequency of stump thrombosis in patients following either left or right pneumonectomy, and there is a relationship between stump length and the development of in situ thrombosis. Our data suggest a rather benign natural history of postpneumonectomy artery stump thrombosis, since no propagation of the thrombus was seen outside the stump. However, more data are needed to verify this observation.
|
FOOTNOTES
|
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, B.H.K., C.W.; study concepts/study design or data acquisition or data analysis/interpretation, B.H.K., C.W.; manuscript drafting or manuscript revision for important intellectual content, B.H.K., C.W.; approval of final version of submitted manuscript, B.H.K., C.W.; literature research, B.H.K., C.W.; clinical studies, B.H.K.; statistical analysis, B.H.K., C.W.; and manuscript editing, B.H.K., C.W.
|
References
|
|---|
- Crafoord C. On the technique of pneumonectomy in man: critical survey of experimental and clinical development and report of authors material and techniques. Acta Chir Scand 1938;81(suppl):1–142.
- Wiklund T. Bronchogenic carcinoma: a clinical study of 259 cases, 100 of which were resected—follow-up study of the resected cases. Acta Chir Scand 1951; 162(suppl):1–152.
- Yeager GH, Mansberger AR. Postoperative vascular complications. Am Surg 1953;19:613–621.[Medline]
- Chuang TH, Dooling JA, Connolly JM, Shefts LM. Pulmonary embolization from vascular stump thrombosis following pneumonectomy. Ann Thorac Surg 1966;2:290–298.[Medline]
- Takahashi T, Yokoi K, Mori K, Miyazawa N. Clot in the pulmonary artery after pneumonectomy (letter). AJR Am J Roentgenol 1993;161:1110.[Medline]
- Wechsler RJ, Salazar AM, Gessner AJ, Spirn PW, Shah RM, Steiner RM. CT of in situ vascular stump thrombosis after pulmonary resection for cancer. AJR Am J Roentgenol 2001;176:1423–1425.[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]
This article has been cited by other articles:
|
|
|
|
 
C. Kotoulas and S. Lachanis
Embolism of the pulmonary artery stump after right pneumonectomy
Interactive CardioVascular and Thoracic Surgery,
May 1, 2009;
8(5):
563 - 564.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
