Guidelines For Management Of

SPECIAL REPORT: Guidelines for Management of Incidental Pulmonary NodulesMacMahon et alFleischner Society 2017 Guidelines for Management of Incidentally Detected Pulmonary Nodules in AdultsA: Solid Nodules*SizeNodule Type36–8 mm (100–250 mm3),6 mm (,100 mm ).8 mm (.250 mm3)CommentsSingleLow risk†No routine follow-upCT at 6–12 months, thenconsider CT at18–24 monthsHigh risk†Optional CT at 12 monthsCT at 6–12 months, then CT Consider CT, PET/CT, or tissueat 18–24 monthssampling at 3 monthsNo routine follow-upCT at 3–6 months, thenconsider CT at 18–24monthsCT at 3–6 months, then at18–24 monthsMultipleLow risk†High risk†Optional CT at 12 monthsConsider CT, PET/CT, or tissuesampling at 3 monthsNodules ,6 mm do not require routine follow-up,but certain patients at high risk with suspiciousnodule morphology, upper lobe location,or both may warrant 12-month follow-up(recommendation 1A).Nodules ,6 mm do not require routine follow-up,but certain patients at high risk with suspiciousnodule morphology, upper lobe location,or both may warrant 12-month follow-up(recommendation 1A).CT at 3–6 months, thenconsider CT at 18–24 monthsUse most suspicious nodule as guide tomanagement. Follow-up intervals may varyaccording to size and risk (recommendation 2A).CT at 3–6 months, then at 18–24 Use most suspicious nodule as guide tomonthsmanagement. Follow-up intervals may varyaccording to size and risk (recommendation 2A).B: Subsolid Nodules*Nodule TypeSingleGround glass,6 mm (,100 mm3)Part solidNo routine follow-upMultipleSize 6 mm (.100 mm3)No routine follow-upCommentsCT at 6–12 months to confirm persistence, then CTevery 2 years until 5 yearsCT at 3–6 months. If stable,consider CT at 2 and 4years.In certain suspicious nodules , 6 mm, considerfollow-up at 2 and 4 years. If solid component(s)or growth develops, consider resection.(Recommendations 3A and 4A).CT at 3–6 months to confirm persistence. If unchanged and solid In practice, part-solid nodules cannot be definedcomponent remains ,6 mm, annual CTas such until 6 mm, and nodules ,6 mmshould be performed for 5 years.do not usually require follow-up. Persistentpart-solid nodules with solid components 6mm should be considered highly suspicious(recommendations 4A-4C)CT at 3–6 months. Subsequent management basedMultiple ,6 mm pure ground-glass noduleson the most suspicious nodule(s).are usually benign, but consider follow-up inselected patients at high risk at 2 and 4 years(recommendation 5A).Note.—These recommendations do not apply to lung cancer screening, patients with immunosuppression, or patients with known primary cancer.*Dimensions are average of long and short axes, rounded to the nearest millimeter.†Consider all relevant risk factors (see Risk Factors).Recommendations for ManagingIncidentally Discovered PulmonaryNodulesGeneral RecommendationsAll CT scans of the thorax in adultsshould be reconstructed and archivedwith contiguous thin sections ( 1.5Radiology: Volume 000: Number 0— 2017nmm, typically 1.0 mm) to enable accurate characterization and measurementof small pulmonary nodules, and routine acquisition and archiving of off-axis (coronal and sagittal) reconstructedseries is strongly recommended (grade1A; strong recommendation, high-quality evidence).radiology.rsna.orgUse of thick sections increasesvolume averaging, which effectivelyprecludes accurate nodule characterization of small nodules, with respect to part-solid morphology andfat or calcium content, which can affect management (Figs 1–3) (10–12).Coronal and sagittal series facilitate3

SPECIAL REPORT: Guidelines for Management of Incidental Pulmonary NodulesMacMahon et alFigure 1Figure 3Figure 1: (a) Lung window and (b) soft-tissue window 1-mm transverse CT sections show a smoothlymarginated solid nodule (arrow) with internal fat and calcification, consistent with a hamartoma. No furtherCT follow-up is recommended for such findings.Figure 2Figure 3: (a) Transverse 5-mm CT section showsan apparently pure ground-glass nodule in theleft lower lobe (arrow). (b) Transverse 1-mm CTsection at the same level as a reveals that this is asuspicious part-solid nodule with cystic components(arrow).Figure 2: (a) CT image shows a smoothly marginated solid nodule with central calcification, typical ofa healed granuloma. No further CT follow-up is recommended for such nodules. (b) CT image shows asmoothly marginated solid nodule with laminar calcification, typical of a healed granuloma. No further CTfollow-up is recommended for such findings.distinction between nodules and scars(Fig 4). This recommendation is notrestricted to examinations performedspecifically for nodule assessment orlung cancer screening, as lung nodulesmay be encountered incidentally in anyadult patient. If the initial examinationwas performed with thick sections,a short-term follow-up examinationwith contiguous thin sections should4be considered as a baseline for futurecomparison.CT examinations of the thorax performed to follow lung nodules shoulduse a low-radiation technique (grade 1A:strong recommendation, high quality evidence). Techniques to reduce radiationdose are of particular importance, giventhe frequency with which follow-upCT examinations are performed. Werecommend adjusting exposure factorsaccording to body habitus, with a goalof achieving a volumetric CT dose index(or CTDIvol) of no more than 3 mGy ina standard-size patient (height, 170 cm;weight, 70 kg), as per ACR recommendations for screening CT (8).A number of dose reduction techniques, including dose modulation anditerative reconstruction, may be used(13). It is important that a similar technique be used to perform the follow-upexamination to minimize interscan variability, with section thickness and reconstruction filter being the most important parameters in this respect.For these guidelines, manual nodule measurements should be based onthe average of long- and short-axis diameters, both of which should be obtained on the same transverse, coronal, or sagittal reconstructed images.Whichever image reveals the greatestradiology.rsna.orgnRadiology: Volume 000: Number 0— 2017

SPECIAL REPORT: Guidelines for Management of Incidental Pulmonary Nodulesshould include the earliest availablestudy and more recent studies. Notethat differences in scanning technique,such as use of thick sections for previous imaging, may make comparisonless accurate, especially for small nodules; therefore, routine use of contiguous thin-section reconstruction andarchiving is important (20).Figure 4Figure 4: (a) Transverse 1-mm CT section shows a nodular opacity adjacent to the minor fissure (arrow).(b) Coronal reconstructed CT image shows that the opacity is a benign linear scar or lymphoid tissue (arrow).dimensions is the image that should beused. Measurements should be madewith electronic calipers or semiautomated methods and should be recordedto the nearest whole millimeter (grade1C; strong recommendation, low- orvery-low-quality evidence).Although several screening trialshave used the maximum diameter ofnodules on transverse sections to estimate size, others (iELCAP) have usedthe average of long- and short-axis diameters measured by using lung windows (4,5,7,14,15). Prediction modelsused to estimate malignancy yield better results with the average diameterthan with the maximum transversediameter (16). The Fleischner Societyhas recommended use of the averagediameter since 2005, as the average oflong and short axes more accurately reflects three-dimensional tumor volume(1). For larger nodules and for masseslarger than 10 mm, it is generally appropriate to record both long- andshort-axis dimensions, with the longaxis dimension being used to determinethe T factor in lung cancer staging andbeing a criterion for tumor responseto treatment. Measurements shouldbe rounded to the nearest millimeter.Fractional millimeter measurementsRadiology: Volume 000: Number 0— 2017MacMahon et alnare not recommended, as their use implies a greater degree of accuracy thanthat which can be achieved in practice.Thus, the size threshold (,6 mm) corresponds to a rounded measurement of5 mm or less in these guidelines. As analternative to manual linear measurements, automated or semiautomatedvolumetric measurements can be used,and they have the advantage of beingmore reproducible than manual techniques (17). Volume thresholds of 100and 250 mm3 are used for volumetryinstead of the 6- and 8-mm thresholdsused for linear measurements. However, volumetry is substantially dependent on the specific software used(18,19). For this reason, volumetricmeasurements to assess nodule growthshould be performed with identicalsoftware versions. More comprehensive recommendations on nodule measurements, including a full discussion oftechnical and observer-related factors,will be provided in a separate WhitePaper from the Fleischner Society thatis currently in preparation.Prior imaging studies should alwaysbe reviewed whenever they are availableto determine possible growth or stability (grade 1A; strong recommendation,high-quality evidence). Comparisonsradiology.rsna.orgRecommendations for Solid Lung NodulesRecommendation 1: single solid noncalcified nodules.—Solid nodules smallerthan 6 mm (those 5 mm or smaller)do not require routine follow-up in patients at low risk (grade 1C; strong recommendation, low- or very-low-qualityevidence). There is a paucity of directevidence regarding cancer probabilityin small nodules in low-clinical-risk situations. However, there is abundant evidence for cancer risk in current smokersor those who recently quit smoking andwho have been studied in the contextof lung cancer screening programs.The risk of cancer in patients who havenever smoked and in younger patientsis known to be significantly lower, witha relative risk on the order of 0.15 inthe United States when compared withrisk in heavy smokers in the case of solidnodules (21). Given that the average riskof cancer in solid nodules smaller than 6mm in patients at high risk is less than1%, it is reasonable to assume an evenlower risk in a patient with low clinicalrisk (7,22). This recommendation isconsistent with our policy of excludingnodules with a less than 1% risk of cancer from routine CT follow-up.Solid nodules smaller than 6 mm donot require routine follow-up in all patients with high clinical risk; however,some nodules smaller than 6 mm withsuspicious morphology, upper lobe location, or both may warrant follow-up at12 months (grade 2A; weak recommendation, high-quality evidence.). Theserevised guidelines increase the sizethreshold for routine follow-up of solidnodules to 6 mm. This change is basedon supporting data from several screening trials that indicate the risk of cancerin nodules smaller than 6 mm is considerably less than 1%, even in patientsat high risk (6,7). On the other hand,5

SPECIAL REPORT: Guidelines for Management of Incidental Pulmonary Nodulessuspicious morphology, upper lobe location, or both can increase cancerrisk into the 1%–5% range; therefore,follow-up at 12 months may be considered, depending on comorbidity andpatient preferences. Earlier follow-up isnot recommended in such instances, asexperience has shown that such smallnodules, if malignant, rarely advance instage over 12 months, whereas a shortterm follow-up examination showing noapparent change may provide false reassurance. An exception may be madein some patients with technically suboptimal initial scanning results to obtain a high-quality baseline study for future comparison or in nervous patientswho may be reassured by evidence ofshort-term stability.Solitary noncalcified solid nodulesmeasuring 6–8 mm in patients with lowclinical risk are recommended to undergo initial follow-up at 6–12 monthsdepending on size, morphology, and patient preference (grade 1C: strong recommendation, low- or very-low-qualityevidence). One follow-up examinationshould suffice in many instances. Ifmorphology is suspicious or if stabilityis uncertain, an additional study may beobtained after a further 6–12 months.The risk of malignancy is very low inthis category, and not all solid nodulesrequire traditional 2-year follow-up. Therecommendation for 2-year follow-upwas based on earlier studies with thickCT sections or chest radiographs andwas made before the important differences between solid and subsolid nodules were recognized (6,7).Although some solid cancers havebeen reported to grow very slowly,with doubling times of more than 700days and failure to clearly demonstrategrowth for up to 2 years, these reportswere also based on analysis of thickersection CT images and evaluation onhard-copy images (23). More recentstudies have confirmed the reliability of2-year stability in the assessment of benignancy in solid nodules, and shorteror longer periods of follow-up may beappropriate in selected subjects, depending on risk factors, nodule morphology, and accuracy of measurements(24). Thus, we recommend optionally6MacMahon et aldiscontinuing follow-up of well-definedsolid nodules with benign morphologyat 12–18 months if the nodule is accurately measurable and unequivocallystable. For subsolid nodules, longerterm follow-up is recommended (2).For solitary solid noncalcified nodules measuring 6–8 mm in patients athigh risk, an initial follow-up examination is recommended at 6–12 monthsand again at 18–24 months (grade 1B:strong recommendation, moderate quality evidence). This recommendation isbased on an estimated average risk ofmalignancy of approximately 0.5%–2.0% for nodules in this size range andis derived from screening studies, mostnotably the PanCan, BCCA, and NELSON trials (6,7) Again, the precise intervals can be modified according to individual risk factors and preferences. Insome patients in whom nodule stabilityremains uncertain, further surveillancemay be required; however, two follow-upexaminations should be sufficient to exclude growth in most subjects.For solitary solid noncalcified nodules larger than 8 mm in diameter, consider 3-month follow-up, work-up withcombined positron emission tomography (PET) and CT (PET/CT), tissuesampling, or a combination thereof;any one of these options may be appropriate depending on size, morphology,comorbidity, and other factors. (grade1A; strong recommendation, high-quality evidence). Although the average riskof cancer in an 8-mm solitary noduleis only approximately 3% dependingon morphology and location, a considerably higher risk can be inferred incertain patients (25,26). As nodules become larger, their morphology becomesmore distinct, and management shouldbe strongly influenced by the appearance of the nodule rather than by sizealone (Figs 5–7). Thus, both invasiveand noninvasive management optionsare included in this article.Measurement of attenuation (inHounsfield units) in solid nodules canbe helpful to determine the presenceof calcification or fat, either of whichcan have major diagnostic implications. It is critical that such measurements be made on images without anFigure 5Figure 5: CT image shows a solid triangularsubpleural nodule (arrow) with a linear extensionto the pleural surface, typical of an intrapulmonarylymph node. No CT follow-up is recommended forsuch findings.Figure 6Figure 6: Transverse 1-mm CT section throughthe left upper lobe shows a suspicious solidspiculated nodule (arrow). Surgery revealed invasiveadenocarcinoma.edge-enhancing filter, such as the typethat is generally used on lung and boneimages. Measurements on a sharpenedimage may give erroneously high attenuation values, and other factors, suchas beam hardening, can affect the accuracy of the measurements. All attenuation measurements should be madeon the thinnest available nonsharpened(typically soft-tissue window) imageseries; the radiologist should use asmall region of interest (not a pointradiology.rsna.orgnRadiology: Volume 000: Number 0— 2017

SPECIAL REPORT: Guidelines for Management of Incidental Pulmonary NodulesFigure 7Figure 7: Transverse 1-mm CT sections obtained 10 months apart show a highly suspicious pattern of progressive thickening in the wall of a right lower lobe cyst (arrow). Resection revealed invasiveadenocarcinoma.Figure 8Figure 8: CT image shows multiple solid nodules of varying size with lowerzone predominance (arrows) secondary to metastatic thyroid carcinoma.value) and realize that substantial variations occur among different scanners,filters, and body locations, even withregular calibration (27).Recommendation 2: multiple solidnoncalcified nodules.—For multiplesolid noncalcified nodules smaller than6 mm in diameter, no routine follow-upis recommended (grade 2B; weakrecommendation,moderate-qualityRadiology: Volume 000: Number 0— 2017nevidence). Small nodules in this sizerange are frequently encountered inroutine clinical practice and are usuallybenign in origin. They most often represent either healed granulomata from aprevious infection (especially in regionswith endemic fungal infections) or intrapulmonary lymph nodes. In patientsat high risk, a 12-month follow-up examination may be considered.radiology.rsna.orgMacMahon et alNote that this recommendation assumes no known or suspected primaryneoplasm that might be a source ofmetastases. In patients with clinical evidence of infection and in those who areimmunocompromised, active infectionshould be considered, and short-termfollow-up may be appropriate.For multiple solid noncalcified nodules with at least one nodule 6 mm orlarger in diameter, follow-up is recommended at approximately 3–6 months,followed by an optional second scanat 18–24 months that will depend onestimated risk. (grade 1B; strong recommendation, moderate-quality evidence). If a larger or more suspiciousnodule is present, it should be used asa guide to management according tothe guidelines for solitary nodules, asstated previously. In such situations,metastases remain a leading consideration, particularly when the distributionof nodules has peripheral and/or lowerzone predominance and when the sizeof the nodules has a wide range (Fig 8)(28). In most instances, metastases willgrow perceptibly within 3 months. Ananalysis of subjects with multiple nodules in the NELSON trial showed anincrease in risk for primary cancer, asthe total nodule count increased from1 to 4, but a decrease in risk for thosewith five or more nodules, most ofwhich likely resulted from prior granulomatous infection (29).The dominant nodule should beused as a guide to management; however, additional nodules should also bemonitored on follow-up images. In thiscontext, the term dominant refers tothe most suspicious nodule, which maynot be the largest.Solitary Subsolid Lung NodulesRecommendation 3: solitary pureground-glassnodules.—Forpureground-glass nodules smaller than 6mm (ie, 5 mm and smaller) in diameter, no routine follow-up is recommended (grade 1B; strong recommendation,moderate-quality evidence). Becauseof the high prevalence of ground-glassnodules smaller than 6 mm, we do notrecommend follow-up scanning in everypatient with such findings. However,7

SPECIAL REPORT: Guidelines for Management of Incidental Pulmonary Nodulesthis does not preclude follow-up in selected patients with subsolid nodules(including those with pure ground-glassor part-solid types) close to 6 mm inMacMahon et alsize with suspicious morphology orother risk factors. This item has beenmodified slightly from the previousrecommendation, providing an optionFigure 9Figure 9: Transverse 1-mm CT sections through the right lower lobe. (a) A well-defined 6-mm groundglass nodule (arrow) can be seen. (b) Image obtained more than 2 years after a shows a subtle increase inthe size of the nodule (arrow). This finding was confirmed by noting the slightly altered relationship to adjacent vascular structures. Such subtle progression can be detected only by using 1-mm contiguous sections.Findings are consistent with adenocarcinoma in situ or minimally invasive adenocarcinoma, and continuedyearly follow-up is recommended.of 2- and 4-year follow-up in selectedsubjects at high risk. This reflects datafrom Asian populations, indicating thatup to 10% of such nodules may growand that nearly 1% may progress to adenocarcinoma over many years. However, the finding of malignant transformation in less than 1% of all patie

Radiology: Volume 000: Number 0— 2017 n radiology.rsna.org 1 1 From the Department of Radiology, University of Chicago, 5841 S Maryland Ave, MC 2026, Chicago, IL 60637 (H.M.); Department of Radiology, New York University Langone Medical Center, New York, NY (D.P.N.); Department of Radiology, Seoul National University College of Medicine,