Upload
others
View
33
Download
0
Embed Size (px)
Citation preview
11
The Solitary Pulmonary The Solitary Pulmonary Nodule RevisitedNodule RevisitedNodule RevisitedNodule Revisited
Eric Bensadoun MDEric Bensadoun MDDivision of Pulmonary, Critical Care, and Sleep MedicineDivision of Pulmonary, Critical Care, and Sleep Medicine
Multidisciplinary Lung Cancer ClinicMultidisciplinary Lung Cancer Clinicp y gp y gUniversity of KentuckyUniversity of Kentucky
DefinitionsDefinitions
• Solitary Pulmonary Nodule (SPN)A discrete more or less rounded
SPNSPN
– A discrete, more or less rounded opacity < 3 cm in diameter, completely surrounded by lung parenchyma without associated adenopathy, atelectasis or pneumonia
Lung MassLung Mass
• Lung Mass– A discrete more or less rounded
opacity > 3 cm in diameter
22
Common Causes of SPNsCommon Causes of SPNs
MalignantMalignant BenignBenign
Primary lung carcinoma Primary lung carcinoma Infectious Granulomas Infectious Granulomas
Solitary metastasis Solitary metastasis HamartomaHamartoma
Carcinoid tumorCarcinoid tumor Round pneumoniaRound pneumonia
Primary lung lymphomaPrimary lung lymphoma AA--V malformationV malformation
Rheumatoid noduleRheumatoid nodule
Wegener’s granulomatosisWegener’s granulomatosis
PseudonodulePseudonodule
SPN: Prevalence of MalignancySPN: Prevalence of Malignancy
StudyStudy Total # Total # of nodulesof nodules
MalignantMalignantN d l (%)N d l (%)
yy of nodulesof nodules Nodules (%)Nodules (%)
Steele et al (1963)Steele et al (1963) 887 36
VA Cooperative Study VA Cooperative Study (1975)(1975) 1134 32
Toomes et al (1983)Toomes et al (1983) 955 49
Rubins et al (1996)Rubins et al (1996) 370 79
Early Lung Cancer Action Project (ELCAP) (1999)
233 8.6
Mayo Clinic Study (2002) 782 1.8
33
SPN Management: An Approach Based SPN Management: An Approach Based on the Probability of Malignancyon the Probability of Malignancy
SPNSPN
Probability assessmentProbability assessmentbased on clinical based on clinical and CT findingsand CT findings
VATS orVATS orThoracotomyThoracotomy
Observation with Observation with serial imagingserial imaging
“Watchful waiting”“Watchful waiting”
More testing suchMore testing suchas PET, FNA toas PET, FNA to
further define riskfurther define risk
Calculating Probability: Art or Science?Calculating Probability: Art or Science?
• Informal estimate by clinician or radiologist based on risk factors test results and experienceon risk factors, test results and experience
• Bayesian analysis– Uses likelihood ratios of various risk factors and imaging
results to calculate the probability of cancer– Bayesian models assume variables are independent i.e., no
interaction– Probability can be determined using Bayesian model at y g y
this free website: www. chestx-ray.com• In general, assessments by prediction models are
similar to those of experienced clinicians
44
SPN: Risk AssessmentSPN: Risk Assessment
• Assess the risk of malignancy in each individual ti t b s d H P d si l i si t sts:patient based on H+P and simple non-invasive tests:
– Clinical risk factors– Radiographic finding on CXR and/or CT
• Use additional testing if necessary– Contrast enhanced CT and/or PET– Transthoracic needle biopsy or bronchoscopy
• Based on the above risk assessment decide on the next step:
– VATS or thoracotomy– Close observation with serial radiographic imaging
Clinical Risk Factors for CaClinical Risk Factors for Ca
• AgeRisk of malignancy increases with age– Risk of malignancy increases with age
– Malignancy very uncommon in patients < 35 yrs old
• Smoking history– Risk increases with # pack-years– Ex-smokers still at risk
• Prior history of malignancyy g y– 70-80% of SPNs are malignant (solitary metastasis or
lung primary)
• Symptoms– Most patients are asymptomatic
55
SPN: Imaging SPN: Imaging
• CXR/CT characteristics– Size– Edge/margin characteristics– Presence of calcification and/or fat– Growth rate
Prevalence of Malignancy Based on SizePrevalence of Malignancy Based on Size
100
94%94%
Lung MassLung Mass
20
40
60
80
Pe
rce
nt
Ma
lig
na
nt
31%31%
48%48%
82%82%
0
20
<1 1-2 2-3 >3
Nodule Diameter (cm)
n=58 n=197 n=149 n=100
Pooled data from Zerhouni et al, Radiology 1986; 160: 319Pooled data from Zerhouni et al, Radiology 1986; 160: 319--327 327 and Siegelman et al, Radiology 1986; 160: 307and Siegelman et al, Radiology 1986; 160: 307--312. (total n=886)312. (total n=886)
66
Calcifications in SPNsCalcifications in SPNs
• The presence of a benign pattern of calcification is i di ti f b i it (LR 0 01)indicative of benignity (LR=0.01)
• Detection of calcifications on CXR (Berger et al. AJR 2001)(Berger et al. AJR 2001)
• Sensitivity: 50% • Specificity: 81%
• Thin-cut CT is gold standard for detection of and characterization of calcification within SPNsf f w
• 6% of malignant nodules have Ca++ detected on CT– 85% of these tumors are >3 cm– Often punctate or eccentric pattern
Patterns of Calcifications Patterns of Calcifications
DIFFUSEDIFFUSE CENTRALCENTRAL POPCORNPOPCORNLAMINARLAMINAR
ECCENTRICECCENTRIC STIPPLEDSTIPPLED
77
Nodule Margins in SPNsNodule Margins in SPNs
SmoothSmooth LobulatedLobulated
IrregularIrregular SpiculatedSpiculated
DiagnosisDiagnosis SpiculatedSpiculated LobulatedLobulated SmoothSmooth
Margin Characteristics of SPNsMargin Characteristics of SPNs
DiagnosisDiagnosis SpiculatedSpiculated LobulatedLobulated SmoothSmooth
Primary lung carcinoma 257 (84%) 112 (28%) 21 (11%)
Solitary metastasis 13 (4%) 62 (16%) 34 (18%)
Benign nodule 36 (12%) 222 (56%) 132 (71%)
Pooled data from Zerhouni et al, Radiology 1986; 160: 319Pooled data from Zerhouni et al, Radiology 1986; 160: 319--327 327 and Siegelman et al, Radiology 1986; 160: 307and Siegelman et al, Radiology 1986; 160: 307--312312
88
SPN: Other Radiological SignsSPN: Other Radiological Signs
• Intranodular fat on CT is a reliable indicator of a Intranodular fat on CT is a reliable indicator of a hamartoma– 20/20 nodules with fat or fat and Ca++ seen on CT scan
were hamartomas (Siegelman et al, Radiology 1986)
• Cavitation can be seen with both benign and malignant nodule (Woodring et al, AJR 1983)g– 95% of lesions with wall thickness < 5mm were benign– 84% of lesions with wall thickness > 15mm were malignant
SPN: Growth RatesSPN: Growth Rates
• The growth rate is expressed as the doubling time hi h f s t th d bli f lwhich refers to the doubling of volume
– 4/3 r3 ie, 25% increase in the diameter equals a doubling of the volume
• Most malignant tumors have doubling times between 30-450 days
• No growth over a 2 year period (730 days) usually No growth over a 2 year period (730 days) usually indicates benignity (LR=0.01)– Can be established retrospectively (get old CXRs or CTs!)– Can be established prospectively in low risk patients
(“watch and wait approach” with serial imaging)
99
“Watchful Waiting”“Watchful Waiting”
• To establish nodule stability/benignity prospectively in low risk patients– Uses 2 year rule to confirm benignity– Serial CXR or CT (CT measurements more accurate) X 2 yrs
• At 3, 6, 12 , 18, and 24 months– No growth over 2 year period = benign– Any growth during this period is an indication for VATS or
bi psbiopsy
NonNon--Invasive TestingInvasive Testing
• Contrast enhanced CT• PET scan
1010
Contrast Enhanced CTContrast Enhanced CT
• Benign and malignant nodules differ in vascularity d b h diff tl ft t t and behave differently after contrast
administration • Contrast-enhanced CT uses nodule enhancement to
differentiate benign from malignant lesions• CT nodule enhancement protocol:
– Contrast: 2 ml/sec, 300 mg iodine/ml, 420 mg/kg dose– Nodule is scanned pre-contrast and at 1, 2, 3, 4 min after
contrast injection– Nodule enhancement (HU) = peak post-contrast
measurement – pre-contrast measurement
Swenson SJ et al. Radiology 2000; 214: 73Swenson SJ et al. Radiology 2000; 214: 73--8080
Pre-contrast 1 min 2 min
= Peak Post= Peak Post--contrast contrast ––PrePre--contrastcontrastMax Max =88=88--46= 42 46= 42
4 min3 min
Enhancement= 42 HUEnhancement= 42 HU
Test Interpretation:Test Interpretation:≤ 15 HU = Benign≤ 15 HU = Benign> 15 HU = Malignant> 15 HU = Malignant
1111
Contrast Enhanced CTContrast Enhanced CT
StudyStudy SensitivitySensitivity SpecificitySpecificity PPVPPV NPVNPV
Swenson et al. 2000 * 98% 58% 68% 96%
Meta-analysisCronin et al. 2008
93% 76% 80% 95%
* C* C ff d 15 HU B i 15 HU M liff d 15 HU B i 15 HU M li
Swenson SJ et al. Radiology 2000; 214: 73Swenson SJ et al. Radiology 2000; 214: 73--8080Cronin et al. Radiology 2008; 246: 772Cronin et al. Radiology 2008; 246: 772--782782
* Cut* Cut--off used: ≤ 15 HU = Benign; > 15 HU = Malignantoff used: ≤ 15 HU = Benign; > 15 HU = Malignant
Contrast Enhanced CT: SummaryContrast Enhanced CT: Summary
• Useful test in the evaluation of SPNsUseful test in the evaluation of SPNs
• The absence of enhancement (≤ 15 HU) is strongly predictive of benignity (NPV=95%)
• The presence of enhancement (> 15 HU) can be seen with malignancy or active granulomas (PPV=70-80%)
• Limited data for nodules < 1 cm• Limited data for nodules < 1 cm
1212
Positron Emission Tomography (PET)Positron Emission Tomography (PET)
• 2-deoxy-2-18fluoro-D-glucose (18FDG) is the radiotracer most commonly used in clinical PETradiotracer most commonly used in clinical PET
• Why FDG?– Cancer cells have an increased glucose metabolism – FDG is an analogue of glucose and there is an increased
uptake of FDG within cancer cells
PET: MethodologyPET: Methodology
• Patients should be fasting for at least 4 hours to minimize the physiologic glucose utilizationminimize the physiologic glucose utilization
• Blood glucose may be measured prior to the test– Glucose < 200 mg/dl; proceed with the test– Glucose > 200 mg/dl; test is delayed or rescheduled
• IV injection of 10-15 mCi of FDG is given
Im i b i s b t 45 mi t s ft i j ti• Imaging begins about 45 minutes after injection– Whole body scans (ears-knees) take about 45-60 minutes
to complete
1313
SPN: PET ImagingSPN: PET Imaging
CT scanCT scan PET scanPET scan PET/CT fusion imagePET/CT fusion image
• PET interpretationQ lit ti i l t– Qualitative visual assessment
• Area of abnormality is detected by comparison with mediastinal activity
– Semi-quantitative assessment• Standardized uptake value (SUV)• SUV 2.5 cut-off for malignancy is often used in clinical practice
SPN: PET ImagingSPN: PET Imaging
• Two meta-analyses of PET imaging with results y g gapplicable to SPNs > 1 and < 3 cm
MetaMeta--analysisanalysis SensitivitySensitivity SpecificitySpecificity PPVPPV NPVNPV
Gould et al. 2001 94% 83% - -
C i t l
Gould MK et al. JAMA 2001; 285: 914Gould MK et al. JAMA 2001; 285: 914--924924
Cronin et al. Radiology 2008; 246: 772Cronin et al. Radiology 2008; 246: 772--782782
Cronin et al.2008
95% 82% 91% 90%
1414
PET: SPN ImagingPET: SPN Imaging
Causes of FalseCauses of False--negativesnegativesBronchoalveolar carcinomaBronchoalveolar carcinomaCarcinoidCarcinoidTumors < 1 cmTumors < 1 cmLow grade adenocarcinomaLow grade adenocarcinomaH l iH l i
Causes of FalseCauses of False--positivespositives
Granulomatous inflammation or Granulomatous inflammation or infection (ie.,Tuberculosis, infection (ie.,Tuberculosis, Histoplasmosis, Sarcoidosis)Histoplasmosis, Sarcoidosis)CWP/SilicosisCWP/Silicosis
HyperglycemiaHyperglycemia Rheumatoid nodulesRheumatoid nodules
PET: SummaryPET: Summary
• Useful test in the evaluation of SPNs– High negative and positive predictive value > 90%
• Can also provide staging information for malignant nodules
• Be aware of the clinical settings where false positives or false negatives are more likely
• Limited use in lesions < 1cmLimited use in lesions 1cm– PET spatial resolution 5-7 mm – Limited data on accuracy for lesions < 1cm
1515
Invasive TestingInvasive Testing
• Bronchoscopy• Trans-thoracic needle aspiration• Trans-thoracic needle aspiration• VATS• Thoracotomy
SPN: BronchoscopySPN: Bronchoscopy
• Sensitivity (20-80%) varies according toSi f d l– Size of nodule
• < 1.5 cm: 10%• 2-3 cm: 40-60%
– Location of lesion– Skill of the bronchoscopist– Positive bronchus sign on CT (increased 30%60%)
• A specific benign diagnosis is rarely madep g g y• Bronchoscopy is of limited use in the evaluation of
an SPN– Useful in the reluctant surgical patient, poor surgical
candidate or inoperable patient with > 2cm nodule
1616
SPN: Transthoracic Needle AspirateSPN: Transthoracic Needle Aspirate
• Performed under CT guidance or fluoroS siti it : 60 95% S ifi it : 98 100%• Sensitivity: 60-95% Specificity: 98-100%– Factors that affect yield:
• Needle size (core vs.aspirate)• Number of passes• Size and location of lesion• Experience of radiologist• Presence of an experienced cytopathologist on-site
• Specific benign diagnosis rarely made
SPN: Transthoracic Needle AspirateSPN: Transthoracic Needle Aspirate
• Utility of TTNA in the work-up of an SPN– In an inoperable patient for a tissue diagnosis– Patient who is reluctant to have surgery without a
diagnosis or who is a high risk operative candidate – In an operable patient the decision for a TTNA should be
based on:• The likelihood of making a specific benign diagnosis
– Low pre-test probability for malignancy
• Will a “negative” result alter your management?– What is the pretest probability of cancer?
• Is knowing the patient has cancer prior to surgery helpful?– Does it shorten OR time?
1717
VideoVideo--Assisted Thoracic Surgery (VATS)Assisted Thoracic Surgery (VATS)
• Most useful for small peripheral lesionsAl t 100% iti it d ifi it• Almost 100% sensitivity and specificity– If positive for malignancy on frozen section may need to
convert to thoracotomy for optimal management – If benign then no further intervention is required
• Mortality rate: very rare• Complication rate: 5-8%p
– Atelectasis– Pneumonia– Prolonged air leak
SPN: ThoracotomySPN: Thoracotomy
• The “gold standard”g• 100% sensitivity and specificity• Diagnostic and therapeutic• 30 day mortality: 1-4%
– pneumonectomy > lobectomy– Age of patient– Age of patient
1818
SPNSPN>1 and <3 cm>1 and <3 cm
Obtain oldObtain oldCXR/CTCXR/CT
NonNon--operableoperablepatientpatient
Observation Observation or biopsy and/or nonor biopsy and/or non--
invasive testinginvasive testing
> 3 cm> 3 cm
VATS/ThoracotomyVATS/Thoracotomyunless benign Ca++ unless benign Ca++ or no growth x 2yror no growth x 2yr
YESYESNONO
Growth withGrowth withDT c/w CancerDT c/w CancerNo growthNo growth
or no growth x 2yror no growth x 2yr
CT scan/ HRCTCT scan/ HRCT
VATS/ThoracotomyVATS/Thoracotomy> 2 yr interval> 2 yr interval< 2 yr interval< 2 yr interval
Continued observationContinued observation No further workNo further work--upup
CT/HRCTCT/HRCT
IndeterminateIndeterminateNodule Nodule Risk factors: clinicalRisk factors: clinical
Benign Ca++Benign Ca++and/or fatand/or fat
No further W/UNo further W/URisk factors: clinicalRisk factors: clinicaland imaging featuresand imaging features
on CT scanon CT scanProbabilityProbability
of malignancyof malignancy
I t di tI t di t
Patient Patient preferencepreferenceSurgical riskSurgical risk
Low riskLow risk
Observation withObservation withserial imaging at serial imaging at 3,6,12,24 mos3,6,12,24 mos
High riskHigh risk
VATS/VATS/ThoracotomyThoracotomy
IntermediateIntermediateriskrisk
Contrast enhanced CT, PETContrast enhanced CT, PEToror
TTNA or BronchoscopyTTNA or Bronchoscopy
Positive Negative
1919
Small Pulmonary Nodule < 1cmSmall Pulmonary Nodule < 1cm
• Small pulmonary nodules (<1 cm) are an increasingly common in clinical practiceincreasingly common in clinical practice– The routine use of multi-row detector CT increases
the ability to detect small pulmonary nodules
The Prevalence of Malignancy inThe Prevalence of Malignancy inNodules < 1 cm: Lessons Learned From Nodules < 1 cm: Lessons Learned From
CT Screening Studies CT Screening Studies
Mayo Clinic (2003)Mayo Clinic (2003) < 4 mm< 4 mm 44--7 mm7 mm ≥ 8 mm≥ 8 mm
ELCAP (2004)ELCAP (2004) < 5 mm< 5 mm 55--9 mm9 mm ≥ 10 mm≥ 10 mm
Nodules detected on baseline screening CT
307 391 84
Diagnosis of lung cancer
0 2 (<1%) 24 (29%)
Swenson et al. Swenson et al. RadiologyRadiology 2003; 226: 7562003; 226: 756--6161
ELCAP (2004)ELCAP (2004) < 5 mm< 5 mm 55--9 mm9 mm ≥ 10 mm≥ 10 mm
Nodules detected on baseline screening CT
378 238 109
Diagnosis of cancer at 1 year
0 14 (6%) 56 (51%)
Henschke et al. Henschke et al. RadiologyRadiology 2004; 231: 1642004; 231: 164--168168
2020
The Evaluation of the SPN < 1 cm The Evaluation of the SPN < 1 cm
• CT imaging– Calcifications– Size– Nodule margins– Growth
• Always obtain prior imaging for comparison, if available
• Contrast enhanced CT– Of limited use in SPN < 8-10 mm
The Evaluation of the SPN < 1 cm The Evaluation of the SPN < 1 cm
• PET– Spatial resolution of most PET scanners is 5-7 mm– Limited data on PET for SPN < 10 mm
• 8 malignant nodules ≤ 10mm were all PET negative (Nomori et al. Lung Cancer 2004; 45: 19-27)
• 36 nodules ≤ 10mm (only 8 nodules were ≤ 8mm) (Herder et al. Eur J Nucl Med Mol Imaging 2004;31:1231-6)
– Sensitivity: 93% Specificity: 77%
– How should PET be interpreted if a nodule < 1cm?• Should a cut-off SUV of 2.5 be used?• More research is needed
2121
Number of Number of
• CT-guided TTNA
The Evaluation of the Small SPNThe Evaluation of the Small SPN
StudyStudy
Number of Number of patients with patients with nodules nodules ≤ 1 ≤ 1 cmcm
SensitivitySensitivity AccuracyAccuracy Pneumothorax Pneumothorax raterate Chest tubeChest tube
Wallace et al. 2002
61 total51= 8-10 mm10= 5-7 mm
82%
50%
88%
70%62% 31%
Ng et al. 2008
55 total50= 8-10 mm5= 5-7 mm
68% 79% 52% 9%5= 5-7 mm
• VATS– Can be a challenge finding nodules
• Special techniques may be required to localize small nodules
The Evaluation of the Small SPNThe Evaluation of the Small SPN
• Follow-up of nodules– Increase in nodule diameter needs to exceed 1 5-2 0 mm to Increase in nodule diameter needs to exceed 1.5 2.0 mm to
be 95% sure that the nodule has increased in size– 3-D volumetric assessment may be the answer in the future
1 month later1 month laterBaselineBaseline
2222
Rationale for the Fleischner Society Rationale for the Fleischner Society Guidelines for Small Nodules < 1 cmGuidelines for Small Nodules < 1 cm
• Low prevalence of malignancy in small nodules in p g ypatients without a history of cancer– < 1% in nodules < 5mm in diameter– < 10% in nodules 5-9mm in diameter– Even lower in low risk individuals (eg,. non-smokers)
• Experience gained from the management of small nodules detected during CT screening studies– Short term f/u exams @ 3-6 mos for nodules < 5 mm are Short term f/u exams @ 3 6 mos for nodules < 5 mm are
unnecessary– Frequent f/u exams @ 3, 6, 12, and 24 mos are
unnecessary for nodules < 9 mm• Limited utility of PET and TTNA for small nodules
Fleischner Society Guidelines for the Fleischner Society Guidelines for the Management of Small SPNs Management of Small SPNs
Nodule size Nodule size (mm)(mm) Low risk patient*Low risk patient* High risk patient**High risk patient**
≤ 4 No follow-up needed Follow-up CT at 12 months; if unchanged, no further follow-up
>4-6 Follow-up CT at 12 months; if unchanged, no further follow-up
Initial follow-up CT at 6-12 months then at 18-24 months if no change
>6-8 Initial follow-up CT at 6-12 months then at 18-24 months if no change
Initial follow-up CT at 3-6 months then at 9-12 and 24 months if no changemonths if no change
>8
Follow-up CT at around 3, 9, and 24 months; may consider dynamic contrast enhanced CT, PET, and/or biopsy
Follow-up CT at around 3, 9, and 24 months; may consider dynamic contrast enhanced CT, PET, and/or biopsy
MacMahon H, et al. MacMahon H, et al. RadiologyRadiology 2005; 237:3952005; 237:395--400400
*minimal or absent smoking history and other known risk factors*minimal or absent smoking history and other known risk factors**history of smoking or other known risk factors**history of smoking or other known risk factors
2323
The Fleischner Society Guidelines for The Fleischner Society Guidelines for the Management of Small Nodulesthe Management of Small Nodules
• Guidelines do not apply to the following groups:Guidelines do not apply to the following groups:– Patients with active or previous extrapulmonary
malignancy • Higher likelihood of malignancy (60-80%)• Biopsy or VATS may be indicated depending on
– Cell type, stage of primary tumor, and prognosis after metastectomy
– Young patients < 35 years oldg p y– Patients with unexplained fever
Nodule AttenuationNodule Attenuation
SolidSolid PartPart--SolidSolid NonNon--Solid or Solid or Ground Glass OpacityGround Glass Opacity
2424
PartPart--Solid and NonSolid and Non--Solid NodulesSolid Nodules
• ELCAP CT screening study
N d l N d l N b f N d lN b f N d l N b f M l N b f M l Nodule TypeNodule Type Number of NodulesNumber of Nodules(Total n=233)(Total n=233)
Number of Malignant Number of Malignant Nodules (% prevalence)Nodules (% prevalence)
Solid 189 (81%) 14 (7%)
Part-solid 16 (7%) 10 (63%)
Non-solid or GGO 28 (12%) 5 (18%)
– Higher prevalence of cancer in part-solid and GGO than solid opacities (34% vs. 7%)
– Malignant part-solid lesions or GGO often represent bronchioloalveolar carcinoma (BAC) or adenocarcinoma with bronchioloalveolar features
Henschke et al. AJR 2002;178; 1053Henschke et al. AJR 2002;178; 1053--77
The Evaluation of the PartThe Evaluation of the Part--Solid or Solid or Ground Glass Opacity (GGO)Ground Glass Opacity (GGO)
• PET scan for part-solid and GGO– 15 GGO nodules (Nomori et al. Lung Cancer 2004; 45: 19-27)– Sensitivity 10% and specificity 20%
• CT-guided TTNA and VATS may be problematic• Follow-up with serial thin cut CT
– Usually best course of action for small lesions < 1 cm– Overall size may not change, but the lesion may become y g y
more solid• Change in size or change in density is an indication for biopsy
– Follow-up duration may need to be longer because of slower growing tumors such as BAC
2525
SPN Management: SummarySPN Management: Summary
• All SPNs must be regarded as potentially All SPNs must be regarded as potentially malignant and require prompt evaluation
• The management of SPNs should be guided by the probability of malignancy in an individual patient
• The management of small nodules < 1 cm should be based on risk and size as delineated in the Fleischner Society guidelines