Abstract
Background:
To evaluate the significance of aspirin, as well as, other potential confounding risk factors, on the incidence and volume of pulmonary hemorrhage in patients undergoing percutaneous computed tomography-guided lung biopsy.
Methods:
This retrospective study was approved by the institutional review board. Between September 2013 and December 2014, 252 patients taking aspirin underwent transthoracic computed tomography-guided lung biopsy. Patient, technical, and lesion-related risk factors were evaluated. Univariate analysis was performed with a Student’s t test, chi-square test, or Fisher’s exact test, as appropriate followed by multivariate logistic regression.
Results:
Of 252 patients, 49 (19.4%) continued or stopped aspirin ≤4 days prior to biopsy and 203 (80.6%) patients stopped aspirin ≥5 days prior to biopsy. Pulmonary hemorrhage occurred in 174 cases (69.0%). The median volume of hemorrhage was 3.74 cm3 (range, 0-163.5 cm3). Multivariate analysis revealed that lesion size (P < 0.0001) and lesion depth (P < 0.0001) were independent risk factors for the incidence of pulmonary hemorrhage, while lesion size (P = 0.0035), transgression of intraparenchymal vessels (P < 0.0001), and lesion depth (P = 0.0047) were independent risk factors for severity of hemorrhage. Aspirin stopped ≤4 days from a percutaneous lung biopsy was not associated with pulmonary hemorrhage.
Conclusion:
Aspirin taken concurrently or stopped within 4 days of transthoracic lung biopsy is not an independent risk factor for pulmonary hemorrhage. The incidence of hemorrhage following lung biopsy is associated with lesion size and depth, while the severity of hemorrhage is associated with lesion size, depth, as well as traversal of intraparenchymal vessels.
Background
Percutaneous biopsy is a widely utilized technique for sampling of pulmonary nodules and masses. Many patients who undergo computed tomography (CT)-guided lung biopsy have c morbid conditions placing them at increased risk for a cardiac event or stroke, and, thus, require daily aspirin therapy. Aspirin is an irreversible cyclooxygenase-1 inhibitor that reduces platelet activation by thromboxane A2.1 Aspirin reduces the risk of coronary or cerebral artery thrombosis leading to myocardial infarction or stroke, respectively. As an irreversible inhibitor of platelet activation, aspirin also increases the risk of spontaneous and iatrogenic procedure-related hemorrhage.2-4
Prior evaluation of bleeding complications in patients receiving antiplatelet therapy who undergo transthoracic needle biopsy has shown that single agent aspirin does not increase the risk for hemorrhage.5This data, however, did not control for confounding risk factors such as lesion size and depth, as well as, a lack of quantification of parenchymal hemorrhage volume. A subsequent investigation evaluating the impact of antiplatelet therapy on bleeding complications evaluated hemoptysis as a primary outcome measure, which may be underestimated in a retrospective study, or used only a semiquantitative grading schema to assess the severity of hemorrhage.6,7 The purpose of our study was to evaluate the significance of aspirin, as well as, other potential confounding risk factors, on the incidence and volume of pulmonary hemorrhage in patients undergoing percutaneous CT-guided lung biopsy.
Methods
This retrospective case-control study was performed in accordance with the Declaration of Helsinki. It was approved by the Institutional Review Board with waiver of informed consent obtained. The protocol research number is PA17-0317.
Study Population
Between September 2013 and December 2014, 252 patients taking daily aspirin underwent transthoracic CT-guided lung biopsy at a single tertiary-referral center. In order to limit confounding effects of dual antiplatelet therapy or anticoagulation, patients taking concomitant antiplatelet agents or anticoagulants were excluded from this study.
Biopsy Technique
Biopsies were performed by 13 fellowship trained interventional radiologists with 8-32 years of experience. Electronic medical records and radiology images were available for all patients, including access to medication history, laboratory values, and intraprocedural chest CT images. CT-guided biopsies were performed on a SOMATOM Definition AS (Siemens Healthcare, Erlangen, Germany). Coaxial biopsy technique utilizing a 19-gauge Chiba needle (Cook Medical, Bloomington, IN) was used in all procedures. At the discretion of the interventional radiologist, fine needle aspiration (FNA) biopsy samples were performed with a 22-gauge Chiba needle (Cook Medical, Bloomington, IN) and/or core needle biopsy (CNB) samples were obtained with a 20-gauge Quick-Core needle (Cook Medical, Bloomington, IN). Following biopsy, CT images were obtained of the region biopsied to assess for hemorrhage and pneumothorax. Patients were monitored in the postanesthesia care unit for 3 hours. Follow-up chest radio-graphs were obtained 1 and 3 hours following the procedure to exclude pneumothorax.
Study Variables
All 252 patients included in this study were taking aspirin prior to the lung biopsy. Patient demographics and potential risk factors for pulmonary hemorrhage were recorded. These variables included age, sex, platelet count (K/μL), international normalized ratio (INR), CT imaging evidence of emphysema, lesion size (cm, longest axial dimension), lesion composition (solid, ground glass opacity, or mixed), largest vessel transgressed by the biopsy needle (mm, so long as the vessel diameter was at least 2 mm—vessel diameters below this threshold were deemed too small to accurately measure with CT), lesion location (right upper lobe, right middle lobe, right lower lobe, left upper lobe, and left lower lobe), type of tissue sample collected (FNA or CNB compared to FNA and CNB), lesion depth (cm, measured as distance from pleural entry site to the nearest edge of the target lesion along the needle tract), and date of last aspirin dose relative to the lung biopsy (patients were grouped into 2 cohorts based on date of last aspirin dose, ≤4 days and ≥5 days).
The incidence and volume of immediate post–biopsy-related hemorrhage were measured using CT images. Hemorrhage was quantified volumetrically from the final set of intraprocedural CT images following biopsy. A freehand region of interest was drawn around the area of hemorrhage on each axial CT image where hemorrhage was evident. The biopsied lesion was excluded from the region of interest. Taking into account slice thickness (range 4.8 mm-5.0 mm), the volume of hemorrhage on each individual slice was totaled to provide a single aggregate hemorrhagic volume (Fig 1A-C). The incidence of hemoptysis was also recorded after reviewing the intraprocedural and postanesthesia care notes.
FIG. 1.
Sixty-year-old female presents with a right upper lobe (RUL) nodule detected on surveillance neck CT for remote history of head and neck malignancy. (A) Image from a chest CT demonstrates a 1.1 cm nodule in the RUL (white arrow). (B) Image from an intraprocedural CT-guided lung biopsy demonstrates a coaxial core needle biopsy of the RUL nodule. (C) Image obtained immediately following removal of the coaxial biopsy needle demonstrates hemorrhage involving the RUL. The area of the hemorrhage (indicated by the black circles) measured 42.4 cm2.
Statistical Analysis
The relationship between potential risk factors and the incidence and volume of pulmonary hemorrhage was evaluated by a 2-sided Student’s t test for numeric values and the chi-square test or Fisher’s exact test, as appropriate, for categorical values. Subsequently, all variables were subjected to a multivariate logistic regression to identify independent risk factors for the incidence and severity of pulmonary hemorrhage. All risk factors in this study were offered as potential predictors in the statistical model. A P value of less than 0.05 was considered to be statistically significant. Statistical analysis was performed using statistics software SAS 9.3 (SAS Institute Inc., Cary, NC).
Results
There were 252 patients included in this study; 49 (19.4%) patients either continued (n = 36 patients) or stopped aspirin ≤4 days (n = 13 patients) prior to biopsy and 203 (80.6%) patients stopped aspirin ≥5 days prior to biopsy. Pulmonary hemorrhage, as identified by intraprocedural CT images, occurred in 174 cases (69.0%). Univariate analysis demonstrated the following risk factors to be associated with incidence of pulmonary hemorrhage following CT-guided lung biopsy: CT evidence of emphysema (P = 0.0307), lesion size (P < 0.0001), size of the largest vessel transgressed by the biopsy needle (P < 0.0001), and lesion depth (P < 0.0001). Of note, subgroup analysis was performed on the incidence of hemorrhage in patients who received FNA only (n = 13 of 15, 86.7%) compared to patients who received CNB only (n = 52 of 73, 71.2%), and no significant difference was identified (P = 0.3356). The multivariate regression results demonstrated that lesion size (odds ratio, 0.580, 95% confidence interval, 0.476-0.707, P < 0.0001) and lesion depth (odds ratio 1.957, 95% confidence interval, 1.527-2.508, P < 0.0001) were significant independent risk factors influencing the incidence of pulmonary hemorrhage following percutaneous lung biopsy, Table 1.
TABLE 1.
Risk factors influencing incidence of hemorrhage, n = 252 patients
| Risk factors | No hemorrhage, Mean [95% CI*], N = 78 patients |
Hemorrhage, Mean [95% CI*], N = 174 patients |
P value (univariate analysis) |
P value (multivariate analysis) |
|---|---|---|---|---|
| Age, years | 70.1 [68.3-71.9] | 68.0 [66.3-69.8] | 0.1083 | |
| Sex (%) | Female 33 (42.3%) Male 45 (57.7%) |
Female 71 (41.0%) Male 102 (59.0%) |
0.8504 | |
| Platelet count, K/μL | 240 [224-256] | 243 [229-257] | 0.8212 | |
| INR | 1.03 [1.01-1.05] | 1.01 [1.00-1.03] | 0.3111 | |
| Emphysema (%) | Yes 28 (35.9%) No 50 (64.1%) |
Yes 88 (50.6%) No 86 (49.4%) |
0.0307 | |
| Lesion size, cm | 4.1 [3.5-4.6] | 2.0 [1.8-2.2] | <0.0001 | <0.0001; OR 0.580 (95%CI* 0.476-0.707) |
| Lesion composition (%) | Solid 74 (94.8%) GGO 2 (2.6%) Mixed 2 (2.6%) |
Solid 152 (87.4%) GGO 11 (6.3%) Mixed 11 (6.3%) |
0.2477 | |
| Size of largest vessel transgressed during biopsy, median [range], mm | 0 [0-3] | 0 [0-4] | 0.0002 | |
| Lesion location (%) | RUL30 (38.5%) RML 5 (6.4%) RLL 14 (17.9%) LUL 14 (17.9%) LLL 15 (19.2%) |
RUL 57 (32.8%) RML 10 (5.7%) RLL 41 (23.6%) LUL 38 (21.8%) LLL 28 (16.1%) |
0.7301 | |
| Type of tissue collected (%) | FNA or CNB 23 (29.5%) FNA and CNB 55 (70.5%) |
FNA or CNB 65 (37.4%) FNA and CNB 109 (62.6%) |
0.2257 | |
| Lesion depth, cm | 1.1 [0.8-1.4] | 2.9 [2.5-3.4] | <0.0001 | <0.0001; OR 1.957 (95% CI* 1.527-2.508) |
| Last dose of aspirin | ≤4 days 14 (28.6%) ≥5 days 35 (71.4%) |
≤4 days 64 (31.5%) ≥5 days 139 (68.5%) |
0.7338 |
CNB, core needle biopsy; FNA, fine needle aspiration; GFR, glomerular filtration rate; GGO, ground glass opacity; INR, international normalized ratio; LUL, left upper lobe; LLL, left lower lobe; OR, odds ratio; RUL, right upper lobe; RML, right middle lobe; RLL, right lower lobe.
95% CI is provided, unless otherwise indicated.
95% CI: 95% Wald confidence limits for odds ratio.
The median volume of hemorrhage in this patient population was 3.74 cm3 (range 0-163.5 cm3). The distribution of hemorrhage per patient is provided in Figure 2. Univariate analysis demonstrated the following risk factors to be associated with volume of pulmonary hemorrhage following CT-guided lung biopsy: CT evidence of emphysema (P = 0.0133), lesion size (P = 0.0001), size of the largest vessel transgressed by the biopsy needle (P < 0.0001), lesion depth (P < 0.0001), and last dose of aspirin relative to the lung biopsy (P = 0.0214). Subgroup analysis was performed on the volume of hemorrhage in patients who received FNA only (n = 15, median 6.6 mL, range 0-23.2 mL) compared to patients who received CNB only (n = 73, median 4.2 mL, range 0-59.4 mL), and no significant difference was identified (P = 0.8531). The multivariate regression results demonstrated that lesion size (P = 0.0035), size of the largest vessel transgressed by the biopsy needle (P < 0.0001), and lesion depth (P = 0.0047) were significant independent risk factors influencing the volume of hemorrhage following percutaneous lung biopsy, Table 2.
FIG. 2.
Calculated volume of pulmonary hemorrhage based on immediate postbiopsy CT-images for each patient (n = 252 patients) included in the study. Patients in red also had documented, self-limited hemoptysis (n = 13,5.2%). (Colorversion of figure is available online.
TABLE 2.
Risk factors influencing volume of hemorrhage,* n = 252 patients
| Risk factor |
P value (univariate analysis) |
P value (multivariate regression analysis) |
|---|---|---|
| Age, years | 0.5931 | |
| Median 68.6, range 28.3-89.7 | ||
| Sex | 0.8912 | |
| 148 M (58.7%), 104F (41.3%) | ||
| Platelet count, K/μL Median 227, range 52-645 |
0.7916 | |
| INR | 0.4886 | |
| Median 1.0, range 0.8-1.5 | ||
| Emphysema | 0.0133 | |
| Yes 116 (46.0%), no 136 (54.0%) | ||
| Lesion size | 0.0001 | 0.0035 |
| median 2.0 cm, range 0.6-10.1 cm | ||
| Lesion composition | 0.0721 | |
| Solid, n = 226 (89.6%) | ||
| GGO, n = 13 (5.2%) | ||
| Mixed, n = 13 (5.2%) | ||
| Size of largest vessel transgressed during biopsy | <0.0001 | <0.0001 |
| median 0 mm, range 0-4 mm | ||
| Lesion location | 0.7680 | |
| RUL 87 (34.5%) | ||
| RML 15 (6.0%) | ||
| RLL 55 (21.8%) | ||
| LUL 52 (20.6%) | ||
| LLL 43 (17.1%) | ||
| Type of tissue collected | 0.1522 | |
| FNA or CNB 88 (34.9%) | ||
| FNA and CNB 164 (65.1%) | ||
| Lesion depth | <0.0001 | 0.0047 |
| median 2.0 cm, range 0-11 cm | ||
| Last dose of aspirin | 0.0214 | |
| ≤4 days, n = 49 (19.4%) | ||
| ≥5 days, n = 203 (80.6%) |
CNB, core needle biopsy; FNA, fine needle aspiration; GFR, glomerular filtration rate; GGO, ground glass opacity; INR, international normalized ratio; LUL, left upper lobe; LLL, left lower lobe; RUL, right upper lobe; RML, right middle lobe; RLL, right lower lobe.
Median volume of hemorrhage (3.74 cm3, range 0-163.5 cm3)
Hemoptysis occurred in 13 of 252 patients (5.2%). The mean volume of hemorrhage measured on the postbiopsy images for the patients with reported hemoptysis was 30.6 cm3 (range 0.8-106.4 cm3) compared to 8.9 cm3 (mean, range 0-163.5 cm3) for the patients without reported hemoptysis, P < 0.0001. Four of 13 (30.8%) cases of hemoptysis occurred in patients who had received aspirin within 4 days of biopsy. All cases of hemoptysis were self-limited and resolved with conservative management. One patient with self-limited hemoptysis required admission for reasons unrelated to bleeding.
Discussion
Our data indicate that aspirin which is either continued or suspended less than 4 days prior to a percutaneous lung biopsy is not a significant risk factor affecting the incidence or volume of parenchymal hemorrhage. Our observation is consistent with prior studies.5-8 Aspirin is an irreversible cyclooxygenase-1 inhibitor. This effect inhibits platelet generation of thromboxane A2, which results in an antithrombotic effect.9 In patients with normal bone marrow function, platelet lifespan is approximately 10 days. Thus, withholding aspirin for 5 days prior to an intervention will result in 30%-50% of platelets to be functional at the time of the procedure. Given the significance of pulmonary hemorrhage which may lead to difficulty with oxygenation as well as aspiration, interventionalists may choose to withhold aspirin for 5 days prior to proceeding with the procedure.10 Our study is the first to quantify the amount of hemorrhage following percutaneous lung biopsy followed by multivariate analysis to control for confounding risk factors. On the basis of our observations, the incidence and volume of hemorrhage following percutaneous CT-guided lung biopsy is not affected by aspirin therapy which is either continued or suspended within 4 days of the procedure.
Our study showed lesion size and depth are independent risk factors influencing the incidence and severity of hemorrhage following percutaneous lung biopsy. Smaller lung lesions and increased lesion depth have been shown to increase the risk for bleeding.6,7,11-14 The smaller and more deeply located lesions are more difficult to target percutaneously, requiring multiple needle passes for adequate sampling and/or frequent redirection of the biopsy needle. Thus, lung parenchymal hemorrhage due to biopsy can be greater in these situations. Furthermore, more peripheral lesions require fewer intraparenchymal vessels to be transgressed by the biopsy needle. Deeper lesions, in contrast, require a longer needle tract with greater potential for puncturing intraparenchymal vessels which increase in size toward the lung hilum.
A third risk factor for the severity of pulmonary hemorrhage is traversal of intraparenchymal vessels. Nour et al previously found that traversing pulmonary vessels in the needle biopsy track is a significant predictor of postbiopsy hemorrhage.13 Data from our study suggests that the size of the vessel traversed is an independent predictor of bleeding severity (P < 0.0001). This suggests that crossing larger parenchymal vessels with the biopsy needle increases risk of biopsy-associated hemorrhage. Furthermore, unlike solid organs, aerated lung likely offers little tamponade effect for injured vessels.
Aspirin does not appear to cause an increased incidence of hemoptysis (n = 13 of 252 patients, 5.2%), which is concordant with prior studies.11,13,14 Of the 13 patients with self-limited hemoptysis, 4 patients (30.8%) had taken aspirin within 4 days of biopsy, while the remaining 9 patients (69.2%) had stopped aspirin 5 or more days prior to biopsy. The difference between measured hemorrhage on the immediate postbiopsy CT images was significantly greater for patients with hemoptysis (mean 30.6 cm3) compared to patients without hemoptysis (mean 8.9 cm3), P < 0.0001. While this difference is limited by the retrospective nature of this study, it may be possible that volumetric assessment of hemorrhage following CT-guided lung biopsy may be a tool to predict the occurrence of hemoptysis. While patients included in our study who suffered from pulmonary hemorrhage (n = 174 of 252, 69.0%) and hemoptysis (n = 13 of 252, 5.2%) following percutaneous lung biopsy were successfully managed with conservatively, there is the potential for severe hemorrhage and hemoptysis to be life-threatening complications. Our results may indicate an association between the degree of hemorrhage seen on CT and occurrence of clinical hemoptysis.
Our study has limitations. Our study was a single-center retrospective cohort study. Our study focused on single agent antiplatelet therapy with aspirin. Many patients requiring antiplatelet therapy are maintained on a dual agent regimen composed of aspirin and another antiplatelet agent, such as clopidogrel. The safety of these other antiplatelet agents with respect to bleeding following transthoracic lung biopsy warrants further evaluation. Additionally, no distinction was made in this study between patients taking 81 mg aspirin and those taking 325 mg aspirin, though in a clinical study of 106 healthy subjects, Qayyum et al found near complete suppression of platelet aggregation to arachidonic acid at 81 mg/day, with no incremental suppression at 325 mg/day.15 Another potential limitation of the study is the method by which parenchymal hemorrhage volume was calculated. A freehand region of interest was drawn around the area of hemorrhage on each CT slice where hemorrhage was evident. Then, taking into account slice thickness, volume of hemorrhage on each individual slice was totaled to provide a single aggregated volume. This method, however, does not take into account heterogeneity due to variable density of hemorrhage. Despite this possible source of error, the described method of hemorrhage quantification used in this study provides a reasonable estimation of parenchymal hemorrhage. Finally, the authors used 20 gauge core needle biopsy devices. The results of this study, therefore, are not generalizable to larger gauge core needle biopsies used for percutaneous lung biopsy, which may be more prevalent as molecular testing on tumor tissue is increasingly performed.
Conclusions
In conclusion, pulmonary hemorrhage associated with percutaneous transthoracic lung biopsy is relatively common and generally self-limited. Aspirin taken concurrently or stopped within 4 days of transthoracic lung biopsy is not an independent risk factor for increased incidence or severity of pulmonary hemorrhage. The incidence of hemorrhage following lung biopsy is associated with lesion size and depth, while the severity of hemorrhage is associated with lesion size, depth, as well as traversal of intraparenchymal vessels.
Acknowledgments
Funding
Supported by the NIH/NCI under award number P30CA016672 and used services from the Biostatistics Resource Group (Cancer Center Support Grant shared resource).
Abbreviations:
- CT
computed tomography
- FNA
fine needle aspiration
- CNB
core needle biopsy
- INR
international normalized ratio
- GFR
glomerular filtration rate
- OR
odds ratio
- 95% CI
confidence interval
- GGO
ground glass opacity
- RUL
right upper lobe
- RML
right middle lobe
- RLL
right lower lobe
- LUL
left upper lobe
- LLL
left lower lobe
Footnotes
Declarations
Ethics approval and consent to participate: This retrospective case-control study was performed in accordance with the Declaration of Helsinki. It was approved by the Institutional Review Board with waiver of informed consent obtained. The protocol research number is PA17-0317.
Consent for Publication
N/A.
Availability of Data and Materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Declarations of Interests
None.
All authors confirm that they have read and approved the final manuscript.
References
- 1.Schror K Aspirin and platelets: the antiplatelet action of aspirin and its role in thrombosis treatment and prophylaxis. Semin Thromb Hemostasis 1997;23:349–56. [DOI] [PubMed] [Google Scholar]
- 2.Devereaux PJ, Mrkobrada M, Sessler DI, et al. Aspirin in patients undergoing noncardiac surgery. I Engl J Med 2014;370:1494–503. [DOI] [PubMed] [Google Scholar]
- 3.Sun JC, Whitlock R, Cheng J, et al. The effect of pre-operative aspirin on bleeding, transfusion, myocardial infarction, and mortality in coronary artery bypass surgery: a systematic review of randomized and observational studies. Eur Heart J 2008;29:1057–71. [DOI] [PubMed] [Google Scholar]
- 4.Narum S, Solhaug V, Myhr K, et al. Characterisation of non-warfarin-associated bleeding events reported to the Norwegian spontaneous reporting system. Eur J Clin Pharmacol 2013;69:1445–52. [DOI] [PubMed] [Google Scholar]
- 5.Atwell TD, Smith RL, Hesley GK, et al. Incidence of bleeding after 15,181 percutaneous biopsies and the role of aspirin. AJR Am J Roentgenol 2010;194:784–9. [DOI] [PubMed] [Google Scholar]
- 6.Song YS, Park CM, Park KW, et al. Does antiplatelet therapy increase the risk of hemoptysis during percutaneous transthoracic needle biopsy of a pulmonary lesion? AJR Am J Roentgenol 2013;200:1014–9. [DOI] [PubMed] [Google Scholar]
- 7.Tai R, Dunne RM, Trotman-Dickenson B, et al. Frequency and severity of pulmonary hemorrhage in patients undergoing percutaneous CT-guided transthoracic lung biopsy: single-institution experience of 1175 cases. Radiology 2016;279:287–96. [DOI] [PubMed] [Google Scholar]
- 8.Herth FJ, Becker HD, Ernst A. Aspirin does not increase bleeding complications after transbronchial biopsy. Chest 2002;122:1461–4. [DOI] [PubMed] [Google Scholar]
- 9.Patrono C Aspirin as an antiplatelet drug. N Engl J Med 1994;330:1287–94. [DOI] [PubMed] [Google Scholar]
- 10.Patel IJ, Davidson JC, Nikolic B, et al. Consensus guidelines for periprocedural management of coagulation status and hemostasis risk in percutaneous image-guided interventions. J Vasc Interv Radiol 2012;23:727–36. [DOI] [PubMed] [Google Scholar]
- 11.Yeow KM, See LC, Lui KW, et al. Risk factors for pneumothorax and bleeding after CT-guided percutaneous coaxial cutting needle biopsy of lung lesions. J Vasc Interv Radiol: JVIR 2001;12:1305–12. [DOI] [PubMed] [Google Scholar]
- 12.Anzidei M, Sacconi B, Fraioli F, et al. Development of a prediction model and risk score for procedure-related complications in patients undergoing percutaneous computed tomography-guided lung biopsy. Eur J Cardio-Thoracic Surg: Off J Eur Assoc Cardio-Thoracic Surg 2015;48:e1–6. [DOI] [PubMed] [Google Scholar]
- 13.Nour-Eldin NE, Alsubhi M, Naguib NN, et al. Risk factor analysis of pulmonary hemorrhage complicating CT-guided lung biopsy in coaxial and non-coaxial core biopsy techniques in 650 patients. Eur J Radiol 2014;83:1945–52. [DOI] [PubMed] [Google Scholar]
- 14.Yeow KM, Su IH, Pan KT, et al. Risk factors of pneumothorax and bleeding: multivariate analysis of 660 CT-guided coaxial cutting needle lung biopsies. Chest 2004;126:748–54. [DOI] [PubMed] [Google Scholar]
- 15.Qayyum R, Becker DM, Yanek LR, et al. Platelet inhibition by aspirin 81 and 325 mg/day in men versus women without clinically apparent cardiovascular disease. Am J Cardiol 2008;101:1359–63. [DOI] [PMC free article] [PubMed] [Google Scholar]