0887-21http://d

Grant Sof HBayThe

*DepartAnn

†DepartArb

AddressDepSystE-m

Connective Tissue Disease–AssociatedInterstitial Pneumonia and IdiopathicInterstitial Pneumonia: Similarity andDifferenceThomas Bryson, MD,* Baskaran Sundaram, MB, BS, MRCP, FRCR,*Dinesh Khanna, MD,† and Ella A. Kazerooni, MD, MS*

71/$-see frontx.doi.org/10.10

upport: Dr Khaealth and Sclerer, Celgene, DIGrapeutics.

ment of DiagnArbor, MI.ment of Internor, MI.

reprint requeartment of Dem, CVC 5481ail: sundbask@

Interstitial lungdiseases (ILDs) are increasingly recognized inpatientswith systemic diseases.Patients with early ILD changes may be asymptomatic. Features of ILD overlap amongsystemic diseases and with idiopathic variety. High-resolution computed tomography plays acentral role in diagnosing ILDs. Imaging features are often nonspecific. Therapy- andcomplication-related lung changes would pose difficulty in diagnosing and classifying anILD. Biology and prognosis of secondary ILDs may differ between different disease-relatedILDs and idiopathic variety. Combination of clinical features, serological tests, pulmonary andextrapulmonary imaging findings, and pathology findings may help to diagnose ILDs.Semin Ultrasound CT MRI 35:29-38 C 2014 Elsevier Inc. All rights reserved.

Connective tissue diseases (CTDs) represent a diversegroup of immune-mediated systemic inflammatory dis-

eases that frequently target the lungs, including the pulmonaryinterstitium, airway, pleura, and pulmonary vasculature.Interstitial lung disease (ILD) is common in many of thesedisorders, including systemic sclerosis (SSc) or scleroderma,rheumatoid arthritis (RA), systemic lupus erythematosus(SLE), and inflammatory myopathies such as polymyositis-dermatomyositis (PM-DM),mixed CTD (MCTD), and Sjögrensyndrome (SS). ILD may present as a forme fruste of systemicdisease, in some cases preceding extrapulmonary manifesta-tions of CTD by years.1,2

matter & 2014 Elsevier Inc. All rights reserved.53/j.sult.2013.10.010

nna receives Grant support from the National Instituteoderma foundation. He is also a consultant for Actelion,NA, Genentech, Gilead, Intermune, Pfizer, andUnited

ostic Radiology, University of Michigan Health System,

al Medicine, University ofMichiganHealth System, Ann

sts to Baskaran Sundaram, MB, BS, MRCP, FRCR,iagnostic Radiology, University of Michigan Health, 1500 E Medical Center Dr, Ann Arbor, MI 48104.umich.edu

ILD as the Initial Manifestation ofCTDUp to 15% of patients presenting with a diagnosis of ILD werediagnosed with CTD during their initial evaluation,3 and up to25% of ILDs occur in patients with an undiagnosed CTD,sometimes making the distinction between idiopathic pulmo-nary fibrosis (IPF) and CTD-related ILD difficult.4 There arepatients who present ILD and extrapulmonary features that donot meet the diagnostic criteria put forward for a systemicconnective disease. It has been suggested that these patientsmay be categorized separately as having “lung-dominantCTD,” as their mode of presentation, disease progression,and prognosis may differ compared with patients with featuresof clear-cut diagnosis of a CTD.2 The frequency of ILD,thoracic manifestations, and typical interstitial imaging find-ings of systemic diseases are mentioned in Tables 1-3.

Systemic SclerosisSSc is a systemic autoimmune disease of uncertain etiologycharacterized by cellular and humoral autoimmunity, vascularinjury, and tissue fibrosis. Its peak incidence is between 45 and64 years, and it is more common in women (3:1 to 8:1 femalepredominance) and African Americans.5 It is subdivided intoa limited cutaneous form, previously known as CREST

29

Table 1 Frequency of Interstitial Lung Diseases in Various Systemic Diseases80

Diseases Frequency of ILD (%) Comment

Rheumatoid arthritis 20-30 Increased risk with cigarette smokingPolymyositis-dermatomyositis 20-50 More common with antisynthetase antibodiesSystemic sclerosis 70 More common in diffuse disease; topoisomerase-1 antibodies

45 (clinically significant)Systemic lupus erythematosus 2-8 Usually in patients with multisystem diseaseMixed connective tissue disease 20-60Sjögren syndrome Up to 25

T. Bryson et al.30

(Calcinosis, Raynaud's syndrome, Esophageal dysmotility,Sclerodactyly, Telangiectasia) syndrome, and a diffuse cuta-neous form, with varying degrees of skin, esophageal, lung,cardiac, and vascular involvement. Both forms can be progr-essive in nature. Pulmonary manifestations include both ILDand vascular pathology manifesting as pulmonary arterialhypertension (PAH). These pulmonary complications arenow the leading cause of disease-related morbidity and mor-tality in patients with SSc.6 The African American populationwith SSc has greater disease severity, with 54% demonstrat-ing ILD and 32% having severe ILD in a recent study. AfricanAmerican patients with SSc autoantibodies had ILD and PAH(antitopoisomerase I and anti-U3-RNP (ribonucleoprotein),respectively) more frequently than patients without auto-antibodies.7

Changes of ILD on high-resolution computed tomography(HRCT) can be found in 55%-65% of unselected patientswith SSc.8,9 The most common pulmonary finding is apattern typical of nonspecific interstitial pneumonia (NSIP),with subpleural-predominant ground-glass opacities, retic-ulation, and traction bronchiectasis, without honeycombing(Fig. 1).10 Other patterns, including usual interstitial pneu-monia (UIP)-like fibrosis with peripheral honeycombing, areless common.11 The HRCT findings at initial presentationcarry prognostic significance; patients with extensive ILDoccupying 20% or more of lung volume experienced a rapid

Table 2 Thoracic Manifestations of CTD

SSc RA SLE DM-PM MCTD SS

ILD overall þ þ þ þ þ þ þ þ þ þ þ þ þNSIP þ þ þ þ þ þ þ þ þ þ þ þ þ þUIP þ þ þ þ þ þ þ þOP þ þ þ þ þ þ þ þ –

LIP – þ þ – – þ þDAD þ þ þ þ þ þ þ – þ

Airway – þ þ þ – þ þ þPleura – þ þ þ þ þ – þ þVascular þ þ þ þ þ þ þ þ þDAH – – þ þ þ – – –

Relative prevalence of thoracic findings in systemic sclerosis (SSc),rheumatoid arthritis (RA), systemic lupus erythematosus (SLE),dermatomyositis-polymyositis (DM/PM), mixed connective tissuedisease (MCTD) and primary Sjögren syndrome (SS). Overall ILDprevalence is reported, along with individual ILD histologies,including nonspecific interstitial pneumonitis (NSIP), usual inter-stitial pneumonitis (UIP), organizingpneumonitis (OP), lymphocyticinterstitial pneumonitis (LIP), and diffuse alveolar damage (DAD).Diffuse alveolar hemorrhage (DAH) is also considered.

decline in pulmonary function in 1 study,12 whereas patientswith less extensive lung abnormality occupying less than20% of lung volume or no pulmonary involvement haverelatively little progression of ILD at follow-up.8,12 Althoughsome have suggested that ground-glass opacities on HRCTrepresent reversible inflammatory alveolitis, ground-glassopacity associated with bronchiectasis is recognized torepresent irreversible fibrosis.13

Several extrapulmonary thoracic findings on HRCT helpsuggest the diagnosis of SSc. Esophageal involvement appe-ars as a dilated patulous esophagus often containing fluid ordebris (Fig. 1) and has been described in up to 97% ofpatients with progressive SSc.5 These patients may alsodemonstrate stigmata of aspiration because of esophagealdysmotility, including bronchiectasis, mucus plugging, andtree in bud nodularity.14 Enlargement of central pulmonaryarteries suggesting PAHmay also be seen in SSc. This may bea result of restrictive lung disease or CTD-related vasculop-athy in the absence of ILD.When lung disease is present, theseverity of PAH may not directly correlate with severity ofILD.15,16 In a study by Condliffe et al, computed tomog-raphy (CT) pulmonary angiography findings of 89 patientswith suspected SSc-related PAH were evaluated for diame-ters of pulmonary artery, qualitative grades of tricuspidregurgitation, and diameters of right and left ventricles. Theinvestigators concluded that the cardiac chamber and greatvessel measurements on CT correlate with invasive pulmo-nary hemodynamic measurements and have additive valuesobtained using echocardiography. The ratio between thediameters of cardiac ventricles may also have prognosticvalues in these patients.17 The severity and the extent of fib-rosis as detected by HRCT may correlate with peak pulmo-nary artery pressures and may be the most reliable inde-pendent predictor of PAH.18

HRCT also provides a sensitive method of detecting SSc-ILD. Analysis from the Scleroderma Lung Study–I, a largemulticenter randomized clinical trial of oral cyclophospha-mide vs placebo in 158 patients with symptomatic SSc-ILD,found that the most common HRCT findings at baselinewere fibrosis (reticular opacity, traction bronchiectasis, andbronchiolectasis, 93%), ground-glass opacities (90%), andhoneycombing (37%).19 A greater extent of pulmonaryfibrosis on HRCT correlated well with lower forced vitalcapacity (FVC) and DLCO (Transfer factor) values onpulmonary function tests (PFTs). The extent of pulmonaryfibrosis seen on HRCT scans was significantly negativelycorrelated with FVC (r ¼ �0.22), diffusing capacity of the

Table 3 Typical HRCT Findings Suggestive of ILD Histology

Typical HRCT Findings

NSIP Bilateral symmetrical ground-glass opacities in a basal and peripheral distribution. Septal thickening and reticulation in 50%with associated traction bronchiectasis� subpleural sparing. Honeycombing and consolidation are rare.

UIP Basal and peripheral reticulation� ground-glass opacities. Frequent traction bronchiectasis. Architectural distortion,reflecting lung fibrosis, with frequent honeycombing and lobar volume loss in advanced disease.

OP Areas of airspace consolidation with associated air bronchograms and mild cylindrical bronchiectasis. Adjacent ground-glass opacities are common. Basal distribution with peripheral or peribronchovascular predominance. Nodular foci ofconsolidation are common, including small bronchovascular nodules.

LIP Ground-glass opacities with scattered thin-walled cysts, primarily in a perivascular distribution. Reticulation in 50%. Lungnodules and patchy consolidation may occur.

DAD Bilateral patchy ground-glassopacitieswith geographic involvement and randomdistribution. Extent of involvementworsenswith disease duration. Consolidations, typically basal within dependent portions of the lung. May organize witharchitectural distortion and traction bronchiectasis.

DAD, diffuse alveolar damage.

Secondary versus primary interstitial pneumonia 31

lung for carbon monoxide (r ¼ �0.44), and total lungcapacity (r ¼ �0.36).20 In the Scleroderma Lung Study–I,the extent of fibrosis seen on baseline HRCT scan waspredictive of the progressive rate of decline in FVC in theplacebo group.21 Conversely, patients with the most exten-sive fibrosis seen on baseline HRCT scans showed thegreatest response to cyclophosphamide (CYC) treatment.In addition, the baseline degree of reticular disease on HRCTand extent of disease are independent predictors of mortalityin SSc-ILD.12

A baseline evaluation for ILD is critical in all patients withSSc and should be done with a combination of chest HRCTsand PFTs. In an analysis of 215 patients with SSc followed upin a clinical practice for 10 years, baseline PFTs and chestHRCTswere predictive of mortality risk.12 An increased extentof disease (as defined by the extent and coarseness ofreticulation [fibrosis] and proportion of ground-glass opacity),420% on HRCT, correlated with an increase in mortality(hazard ratio [HR] ¼ 2.48, P o 0.0005). Patients with adecreased baseline FVC o70% also had increased mortalityrisk (HR ¼ 2.11, P ¼ 0.001). When the 2 modalities werecombined in the risk assessment, patients withHRCT extent ofZ20% in addition to a FVCo70% had the highest mortalityrisk (HR ¼ 3.46, Po 0.0005).

Figure 1 Two axial HRCT sections from a patient with systemicreticulation, ground-glass opacity, and traction bronchiectasisrelatively spared (A) but demonstrate a patulous, debris-filled e

Rheumatoid ArthritisRA is a relatively common collagen vascular disease, occurringin 1%-2% of the population, more frequently in women(3:1 female predominance), and typically between 25 and 50years of age.22,23 Although RA is primarily characterized bysynovial inflammation, extra-articular manifestations are seenin approximately half of patients with RA, and evidence of lungdisease is seen in up to 80% of patients who are clinicallysuspected to have RA-associated ILD.24 A recent HRCT studydemonstrated a 10%-12% incidence of ILD and 8% incidenceof bronchiolitis in 1 RA cohort.25,26 Patients with RA-associated ILD have a higher mortality than those withoutRA-associated ILD, with approximately 80% dying of pulmo-nary complications.27-29 Risk factors for developing RA-associated ILD include advanced age, high titers of rheumatoidfactor, carriage of the HLA-DRB1n1502 allele, and high tiers ofanticyclic citrullinated peptide antibodies, which is associatedwith both airway disease and ILD.30

Unlike most other CTDs, the predominant ILD histology inRA-associated ILD is UIP, found in up to 56% of patients.31,32

NSIP is also commonly seen, followed by organizing pneumo-nia (OP). The major patterns on HRCT in RA are subpleu-ral reticulation with or without honeycombing (Fig. 2A),

sclerosis demonstrate lower lung–predominant peripheraltypical of NSIP-like fibrosis (B). The upper lobes are

sophagus typical of this disease (arrow).

Figure 2 Four different patients with rheumatoid arthritis–associated lung disease: (A) peripheral reticulation and extensivehoneycombing typical of UIP; (B) extensive ground-glass opacities and septal thickening typical of NSIP, with peripheraltraction bronchiectasis that canmimic honeycombing; (C) areas of consolidation in a patientwith organizing pneumonia atbiopsy; and (D) expiratory images with bronchial wall thickening andmosaic attenuation suggesting air trapping related toairway involvement.

T. Bryson et al.32

ground-glass attenuation opacities (Fig. 2B), and areas of con-solidation (Fig. 2C).33 There is also an airway-predominantpattern in RA, which manifests as bronchial wall thickening,bronchiectasis, and mosaic attenuation with air trapping(Fig. 2D). Several studies have demonstrated a good correla-tion between the imaging findings and lung biopsy results,particularly when the histology is UIP.31-33 The primary utilityof HRCT in patients with RA is the reliable identification ofpatients with UIP-type ILD who carry a worse prognosis,34

similar to patients with IPF.Patients with RA may also have pulmonary macronodules,

which are typically asymptomatic.14 They are found morefrequently in men, commonly in the upper and midperipherallungs. The nodules may cavitate, enlarge, stay stable, sponta-neously resolve, or newly appear on serial imaging. Unfortu-nately, the appearance of these nodules is nonspecific, usuallynecessitating follow-up imaging at a minimum, and broncho-scopyor evenbiopsy to excludemalignancyor infection (Fig. 3Aand B). In essence, these nodules are a diagnosis of exclusion.Extrapulmonary thoracic findings are common in RA, with

pleural involvement more common than lung disease. Pleuralthickening or effusion or both are seen in 38%-73% of patientson HRCT and in 40%-75% of patients with RA on auto-psy.35,36 Large airway involvement is a less common finding in

RA and includes cricoarytenoid arthritis, with abnormalitiesrarely identified on CT.24

Systemic Lupus ErythematosusSLE is a complex rheumatologic condition with a broadspectrum of systemic involvement. It has an overall incidenceof 2-11 persons per 100,000 a year; it is higher in women,who make up approximately 90% of patients with SLE,multiracial populations. In the African American population,the diagnosis is common among women of childbearing age;however, in other populations, it occurs more commonly inwomen older than 40 years.37 Respiratory symptoms andabnormal lung function are relatively common in patientswith SLE, occurring in 63% and 66%, respectively, in a recentstudy.38 Thoracic manifestations are varied and includepleural disease, acute pneumonitis, alveolar hemorrhage,chronic ILD, and shrinking lung syndrome. Seropositivityfor specific antinuclear antibodies including anti-RNP andanti-dsDNA has been linked with pulmonary and pleuralinvolvement in SLE.39,40

Patients with SLE may present emergently with suddenonset of fever, dyspnea, cough, pleuritic chest pain, and

Figure 3 CT images of a patient with rheumatoid arthritis and multiple cavitary nodules within the upper lobes. Thedominant nodule within the left upper lobe (A) was proven to be squamous cell carcinoma at biopsy, whereas the smallerright upper lobe nodule (B) demonstrated granulomatous change suggestive of a rheumatoid nodule at pathology.

Secondary versus primary interstitial pneumonia 33

hypoxia, resulting in a confusing clinical picture that mimicspulmonary infection, acute pulmonary embolus, acute lupuspneumonitis, or alveolar hemorrhage. Infections represent thesecond most frequent cause of death in patients with SLE andrange from community-acquired pneumonia to those byatypical or opportunistic pathogens.41,42 The prevalence ofacute lupus pneumonitis is difficult to estimate, and by bestestimates ranges from2%-10% inpatientswith SLE,with up to50% of these patients developing acute lupus pneumonitis astheir presenting manifestation of SLE. Chronic ILD is relativelyrare, occurring in only 3% of patients with SLE. Some chronicILD cases may develop after recovery from acute lupuspneumonitis or develop insidiously, with an increasing inci-dence of ILD with age and disease duration.43,44 Fewer than1% of patients develop progressive loss of lung volume withassociated dyspnea, a process known as shrinking lungsyndrome. The pathogenesis of this condition is still unclear.The current thinking is that this condition is associated withdiaphragmatic dysfunction related to diaphragmatic myopa-thy, phrenic neuropathy, or pleural adhesions. In the rare caseof shrinking lung syndrome, there is frequently little radio-graphic abnormality other than reduced lung volumes with

Figure 4 Prone HRCT image demonstrating peripheral septal thica patient with SLE (A). Contrast-enhanced CT image in a pademonstrates serositis, with small pleural effusions and a mode

elevated diaphragms in patients with significant restrictive PFTabnormalities and with no significant pulmonary interstitialchanges.45

Imaging patterns in SLE are frequently nonspecific and maynot correlatewith patient symptoms. Several small studies haveshown a relatively high rate of abnormal findings on HRCT, asfound in 70% of a mixed population of asymptomatic andsymptomatic patients with SLE46,47 and up to 93% ofsymptomatic patients with SLE.48 HRCT typically demon-strates lower lobe–predominant septal thickening, architec-tural distortion, and ground-glass opacities typical of NSIPwith or without pleural effusions (Fig. 4A).43 In the acutesetting,HRCTfindings can range fromminimal abnormality todiffuse ground-glass opacities or patchy consolidation or both,suggestive of diffuse alveolar damage. Patients with SLE mayalso present emergently with diffuse alveolar hemorrhage,which manifests as nonspecific diffuse alveolar opacities, oftenaccompanied by hemoptysis.Pleural disease is the most common extrapulmonary

thoracic manifestation of SLE, with pleural effusions occurringin 30%-50% of patients. Lupus pleuritis frequently presentswith the same confusing clinical symptoms as acute pulmonary

kening and reticulation suggesting early NSIP-type ILD intient with SLE who presented with pleuritic chest painrate pericardial effusion (B).

T. Bryson et al.34

manifestations describedpreviously, particularly pleuritic chestpain. In patients with SLE and pleural effusions of uncertainetiology, thoracentesis with pleuralfluid analysis, including thepresence of antinuclear antibodies, may be helpful in differ-entiating lupus pleuritis from other disease processes.49,50

Other manifestations of lupus serositis include pericarditis,occasionally presenting with significant pericardial effusion(Fig. 4B). Patients with SLE may also demonstrate airwayinvolvement such as bronchiectasis in up to 20% of patients orvascular involvement including stigmata of PAH in up to 11%of patients.14

Inflammatory MyopathiesIdiopathic inflammatorymyopathies are a heterogeneous groupof disorders characterizedby subacute or chronic skeletalmuscleinflammation and weakness. This group includes inclusion-body myositis, PM, and DM. PM and DM are frequentlyconsidered together as their signs and symptoms are similar,with the exception of cutaneous manifestations such as helio-trope rash on the upper eyelids and raised Gottron papules onthe dorsal surface of the hands seen in DM. Pathologically, theseare distinct entities. Because these diseases frequently coexistwith other CTD and can be difficult to distinguish from eachother, their exact prevalence is unknown,with estimates rangingfrom 6-10 per 1,000,000.51 PM-DM primarily affects women,with DM seen in both children and adults and PM seen after thesecond decade, and inclusion-body myositis primarily affectsmen older than 50 years.52 A subset of patients manifest aclinical syndrome known as antisynthetase syndrome, which ischaracterized by the presence of anti–aminoacyl-tRNA synthe-tase antibodies (such as anti-PL-12, anti-PL-7, and anti-Jo-1),stigmata of PM-DM, ILD, arthritis, Raynaud phenomenon, andmechanic's hands.53,54 Anti-Jo-1 is the best studied of theantisynthetase antibodies, with a prevalence of 1.2-2.5 per1,000,000, at least 2:1 female predominance, and a strongcorrelation with ILD in patients with PM-DM.54,55

In the general PM-DMpopulation, the rate of ILD is variable,ranging from 5%-30%14 to as high as 65% or more in HRCTstudies, with many asymptomatic patients having imagingfindings of ILD.56

HRCT most commonly demonstrates subpleural septalthickening, reticulation, and ground-glass opacities typical ofan NSIP pattern. The presence of additional areas of consol-idation suggesting OP is common and frequently coexists withthe NSIP pattern on histopathology.57 Foci of OP onHRCT aretypically seen as peripheral airspace opacities in the subpleuralor peribronchovascular regions that vary in attenuation fromground glass to dense consolidation and may be bandlike,polygonal, or nodular in shape. Over time, these opacities maybe migratory, clearing in one area and involving new areas.Crescentic opacities surrounding a focus of groundglass knownas the atoll or reverse halo sign have been described as specificfor OP.58 A UIP-type pattern is seen infrequently. Antisynthe-tase syndromemay have a distinct appearance, including basal-predominant septal thickening, ground glass, peribronchovas-cular consolidation, and associated loss of lung volume.54,59

Additionally, the patients with PM-DM may develop respi-ratory complications such as hypoventilation from respiratorymuscle weakness and aspiration from pharyngeal muscleweakness, potentially the most common respiratory compli-cation in this population. Relatively few extrapulmonarythoracic manifestations of PM-DM are seen, with little asso-ciated airway or pleural involvement. Inflammatory myopa-thies can affect the heart in up to 75% of patients; however,much of this represents subclinical disease without obviousstructural abnormality. Cardiovascular involvement is a rela-tively common cause of morbidity and mortality, with up to45% of patients developing congestive heart failure. Therefore,imaging stigmata of heart failure or coronary artery diseasemaybe related to the underlying CTD.60 These patients are also atan increased risk for malignancy, most commonly lungcancers.61

MixedConnective TissueDiseaseMCTD is characterized by antiribonucleoprotein autoanti-bodies (specifically anti-U1-RNP) and a spectrum of clinicalfeatures such as Raynaud phenomenon, swollen hands,arthritis, serositis (with pleural and pericardial involvement),myositis, esophageal dysmotility, pulmonary hypertension,and ILD. It is debated whether this constellation of symptomsdefines a unique CTD or represents an overlap between SLE,SSc, and PM-DM. In a recent study, the prevalence of MCTDwas 3.8 per 100,000, with a strong female predominance ofapproximately 3:1 and age at onset in the third to fourthdecade.62 This and other studies have demonstrated thatpulmonary involvement is a common feature of MCTD,occurring in 52%-85% of patients on imaging. ILD is the mostcommon pulmonary manifestation, with active ILD occurringin up to 66% of patients with MCTD.63 These patients form aspectrum of disease, from asymptomatic to advanced fibrosis,with significant mortality implications including a 20.8%mortality rate over 4.2 years for those with severe fibrosis.62

PAH occurs in 10%-45% of patients with MCTD and is alsopredictive of a poor prognosis, representing themost commoncause of death in 1 study.64

Much as the clinical signs or symptoms of MCTD representan overlap of several other CTDs, the results of HRCT alsoreflect a combination of findings seen in other conditions. Themost common findings follow the NSIP pattern, with lowerlobe–predominant ground-glass opacities, nonseptal linearopacities, and peripheral reticulation (Fig. 5).63,65 Findingssuch as traction bronchiectasis, UIP-type honeycombing, andconsolidation suggesting OP are less common. Other reportshave suggested that ground-glass opacities are less commonand that septal thickening or reticulation is the predominantfinding; however, this may reflect differences in study pop-ulation and possibly response to corticosteroid therapy assome studies have shown reversibility of abnormalities onHRCT finding following treatment.63,66 Other pulmonaryfindings may include low lung volumes related to respiratorymuscle weakness, stigmata of aspiration or pneumonia orboth, and rarely alveolar hemorrhage.14

Figure 6 Patchy ground-glass opacities and thin-walled cysts through-out both lungs in a patient with Sjögren syndrome and likely LIP.

Figure 5 Lower lobe–predominant reticulationwith subpleural sparingsuggesting NSIP-type ILD in a patient with MCTD. Note the patulousesophagus (arrow).

Secondary versus primary interstitial pneumonia 35

As seen in SLE, SSc, and PM-DM, patients with MCTDmayalso have extrapulmonary thoracic findings. Signs of PAH,including enlarged central pulmonary arteries and right heartenlargement, should be sought given the mortality implica-tions. Aswith SSc, signs of esophageal dysmotilitymay be seen,including a patulous esophagus and retained debris. Pleuraland pericardial effusions or thickening also occur, particularlyin patients with SLE-dominant symptomatology,36 with pleu-ral disease occurring in 12%-66% of patients with MCTD.63,66

Sjögren SyndromeSS is a relatively common autoimmune condition occurring in0.1% of the general population and up to 3% of older adults,almost exclusively in women (up to 13:1 female:male ratio). It ischaracterized by lymphocytic infiltration of exocrine glandsresulting in typical sicca symptoms such as dry eyes andmouth,with focal lymphocytic sialoadenitis on salivary gland pathologyor positive results for the presence of antinuclear antibodiesagainst ribonucleoproteins, Ro/SSA, or La/SSB.4,67,68 SS canoccur either alone or in association with other CTDs, termedprimary SS (pSS) and secondary SS, respectively. Thoracicinvolvement, including airway disease and ILD, is common inpSS; however, the prevalence depends on the patient populationstudied and method of diagnosis. Imaging evidence of thoracicinvolvement was found in 79% of patients on HRCT in onepatient population,69whereas only11%ofpatients had clinicallysignificant lung involvement in a separate study.68 Patients withprimary SS have an increased rate of respiratory infections owingto disease involvement ofmucosal glandswithin the airways andresultant impairedmicrobial clearance. It canbedifficult in somecases to differentiate scarring sequel to prior infection fromparenchymal abnormalities related to ILD on imaging.69

The most common histologic pattern observed in patientswith pSS-related ILD is NSIP, with OP, lymphocytic interstitialpneumonia (LIP) and less frequently, UIP.70,71 The characteristicHRCT findings are ground-glass opacities, reticular opacities,and consolidation, generally in a lower lobe–predominant

distribution. This constellation of findings suggests NSIP or OPif consolidations predominate.69-71 Thin-walled parenchymalcysts and small peripheral nodulesmay also be identified (Fig. 6)and can raise suspicion for LIP, particularly when seen in con-junction with ground-glass opacities.69 Unfortunately, althougha CT diagnosis of NSIP positively correlates with histopathologyin patients with pSS, other CT patterns show poor correlationwith histology.70 Honeycombing is rare. Airway involvement iscommon in patients with pSS, with HRCT demonstratingbronchiectasis in up to 46% of patients and potentially resultingin an obstructive pattern at spirometry.4,69

SS is associatedwith an increased risk of lymphoma, with riskofmalignant non-Hodgkin's lymphoma estimated to be up to 44times greater in SS than the general population.72 These patientsfrequently develop extranodal disease thatmost commonly invo-lves the parotid gland, but can present with pulmonary involve-ment.72,73 On HRCT, pulmonary lymphoma has a variableappearance, including ill-defined consolidation, ground-glassopacities, well-defined nodules, and cavitary or cystic lesions.Pulmonary lymphoma should be suspected in large nodulesor masses.73,74 However, there are no significant differencesbetween patients with lymphoma and LIP for the prevalence offindings such as lung masses, lung nodules, ground-glass opa-city, cavity, pericardial effusion, and hilar lymphadenopathy.74

Idiopathic ILD vs ILD in CTDSome clinical features can be used to discriminate between theILD of CTD and idiopathic ILD. These generally include thedemographics of the affected population, presence of serumantibodies, and to somedegree, the findings onHRCT.75 IPF isdefined as a chronic progressive fibrosing ILD of unknownorigin characterized by UIP histology on surgical lung biopsyor findings typical of UIP onHRCT of the chest. The pattern oflower lung subpleural-predominant reticulation with honey-combing that may be accompanied by traction bronchiectasisis more than 95% specific for UIP. This UIP pattern typicallyhas relatively little associated ground-glass opacity. IPF is

T. Bryson et al.36

limited to the lung with serological evaluation unrevealing,including a nonspecific elevation of inflammatory markers.The precise prevalence and incidence of IPF are unknown,

but estimates range from 2-29 cases per 100,000 in the generalpopulation, with approximately 7-16 new cases per 100,000 ayear. Patients with IPF are more frequently male and aretypically in their sixth or seventh decade of life, with anincreasing incidencewith advancing age. Patients younger than50 years are rarely diagnosed with IPF, and there is a strongassociationwith cigarette smoking. Clinically, patientswith IPFmanifestwith as the insidious onset of dyspnea on exertion andbibasilar inspiratory crackles on auscultation. Importantly,other causes for these findings, including environmentalexposure, drug toxicity, or systemic disease such as CTDmustbe excluded before a diagnosis of IPF can be made.4,76

In contrast to IPF, the demographics of CTDs are typically ayounger and more frequently female patient population. Pati-ents with CTD often present with multisystem disease, includ-ing a host of extrathoracic manifestations such as Raynaudphenomenon or articular involvement or both. Similarly, whenpatients with CTD develop pulmonary manifestations, they aretypically multicompartmental, with varying degrees of lung,pleura, airway, and pulmonary vascular involvement (Table 1).Along with the variable clinical manifestations of CTD, theassociated ILD may produce any of the established histologicpatterns of ILD, including NSIP, UIP, LIP, cryptogenic OP, ordiffuse alveolar damage.77 Although these patterns are seenwith varying frequency in different disorders, the histologiccalling card of CTD-related ILD is NSIP (Table 1).Importantly, patients with CTD-related ILD have a better

prognosis than IPF, with an approximately 70% 5-yearsurvival compared with 20%-35% for IPF.78 Prognosis hasbeen best studied in SSc and RA, controlled for age, sex,smoking status, ILD histology, and pulmonary function testingresults among other factors.Mortality ismost closely associatedwith measures of disease severity at the time of diagnosis,including extent of fibrosis and pulmonary function abnor-malities, with relatively lessmortality difference observed basedon ILD histology alone.10,12 RA is the one exception to thisgeneralization, with several studies documenting mortalitysimilar to IPF in patients with RA with UIP histology or typicalimaging findings of UIP on HRCT. In 1 study, these patientshad a median survival of 3.2 years.31,34,79 This has led someauthors to suggest that treatment of RA-associated ILD shouldbe determined by the underlying histopathology, with patientswith NSIP receiving immunosuppressive therapy and patientswith UIP referred for possible lung transplantation.34 Thediagnosis and treatment of these patients are multidisciplinaryin nature,with radiologists andHRCTplaying a significant rolein screening andmonitoring patients with suspected or knownCTD-related ILD.We have reviewed CTDs and their commonthoracic manifestations, with an emphasis on HRCT findingsthat can help suggest the diagnosis of CTD-related ILD.

ConclusionsCTDs represent a heterogeneous group of disorders that canproduce a broad spectrum of thoracic pathologies on imaging.

In patients with CTD with subacute or chronic thoracicsymptoms, HRCT is an important diagnostic tool to performa thorough evaluation of ILD. These patients may show avariety of findings, from stigmata of ILD to pleural disease,airway disease, or pulmonary hypertension. Although mostpatients with CTD-related ILD have NSIP pathology andassociated nonspecific imaging findings, radiologists shouldseek out findings suggestive of a more specific histologicdiagnosis whenever possible. Similarly, when faced with a newdiagnosis of ILD, the radiologist should seek to differentiatebetween UIP and NSIP imaging patterns when possible andidentify secondary imaging signs of undiagnosed CTD, such asa patulous esophagus, or pleural disease, as these findingswould significantly alter the prognosis and clinical coursecompared with IPF. Acute symptoms frequently present adiagnostic challenge as the imaging findings associatedwith anacute exacerbation of ILD may overlap with those of pulmo-nary infections, aspiration, drug reactions, or other patholo-gies. Imaging is a key tool in the appropriate diagnosis andmanagement of these patients and can provide referringclinicians with important information in the management ofthis difficult patient population.

References1. Cottin V: Interstitial lung disease: Are we missing formes frustes of

connective tissue disease? Eur Respir J 28(5):893-896, 20062. Fischer A, West SG, Swigris JJ, et al: Connective tissue disease-associated

interstitial lung disease: A call for clarification. Chest 138(2):251-256,2010

3. Mittoo S, Gelber AC, Christopher-Stine L, et al: Ascertainment of collagenvascular disease in patients presenting with interstitial lung disease. RespirMed 103(8):1152-1158, 2009

4. de Lauretis A, Veeraraghavan S, Renzoni E: Review series: Aspects ofinterstitial lung disease: Connective tissue disease-associated interstitiallung disease: How does it differ from IPF? How should the clinicalapproach differ? Chron Respir Dis 8(1):53-82, 2011

5. Coral-Alvarado P, Pardo AL, Castano-Rodriguez N, et al: Systemicsclerosis: A world wide global analysis. Clin Rheumatol 28(7):757-765,2009

6. Steen VD, Medsger TA: Changes in causes of death in systemic sclerosis,1972-2002. Ann Rheum Dis 66(7):940-944, 2007

7. Steen V, Domsic RT, LucasM, et al: A clinical and serologic comparison ofAfrican American and Caucasian patients with systemic sclerosis. ArthritisRheum 64(9):2986-2994, 2012

8. Launay D, Remy-Jardin M, Michon-Pasturel U, et al: High resolutioncomputed tomography in fibrosing alveolitis associated with systemicsclerosis. J Rheumatol 33(9):1789-1801, 2006

9. Devenyi K, Czirjak L: High resolution computed tomography for theevaluation of lung involvement in 101 patients with scleroderma. ClinRheumatol 14(6):633-640, 1995

10. Bouros D, Wells AU, Nicholson AG, et al: Histopathologic subsets offibrosing alveolitis in patientswith systemic sclerosis and their relationshipto outcome. Am J Respir Crit Care Med 165(12):1581-1586, 2002

11. Silva CI, Muller NL: Interstitial lung disease in the setting of collagenvascular disease. Semin Roentgenol 45(1):22-28, 2010

12. Goh NS, Desai SR, Veeraraghavan S, et al: Interstitial lung disease insystemic sclerosis: A simple staging system. Am J Respir Crit Care Med177(11):1248-1254, 2008

13. Shah RM, Jimenez S,Wechsler R: Significance of ground-glass opacity onHRCT in long-term follow-up of patients with systemic sclerosis. J ThoracImaging 22(2):120-124, 2007

14. Capobianco J, Grimberg A, Thompson BM, et al: Thoracic manifestationsof collagen vascular diseases. Radiographics 32(1):33-50, 2012

Secondary versus primary interstitial pneumonia 37

15. Grosse C, Grosse ACT: findings in diseases associated with pulmonaryhypertension: A current review. Radiographics 30(7):1753-1777, 2010

16. Wells AU, Steen V, Valentini G: Pulmonary complications: One of themost challenging complications of systemic sclerosis. Rheumatology(Oxford) 48(suppl 3):iii40-iii44, 2009

17. Condliffe R, Radon M, Hurdman J, et al: CT pulmonary angiographycombined with echocardiography in suspected systemic sclerosis-associated pulmonary arterial hypertension. Rheumatology (Oxford) 50(8):1480-1486, 2011

18. Pandey AK, Wilcox P, Mayo JR, et al: Predictors of pulmonary hyper-tension on high-resolution computed tomography of the chest in systemicsclerosis: A retrospective analysis. CanAssocRadiol J 61(5):291-296, 2010

19. Goldin JG, Lynch DA, Strollo DC, et al: High-resolution CT scan findingsin patients with symptomatic scleroderma-related interstitial lung disease.Chest 134(2):358-367, 2008

20. Goldin J, Elashoff R, Kim HJ, et al: Treatment of scleroderma-interstitiallung disease with cyclophosphamide is associated with less progressivefibrosis on serial thoracic high-resolution CT scan than placebo: Findingsfrom the scleroderma lung study. Chest 136(5):1333-1340, 2009

21. KhannaD, Tseng CH, Farmani N, et al: Clinical course of lung physiologyin patients with scleroderma and interstitial lung disease: Analysis of theScleroderma Lung Study Placebo Group. Arthritis Rheum 63(10):3078-3085, 2011

22. Kim DS: Interstitial lung disease in rheumatoid arthritis: Recent advances.Curr Opin Pulm Med 12(5):346-353, 2006

23. Gabriel SE: The epidemiology of rheumatoid arthritis. Rheum Dis ClinNorth Am 27(2):269-281, 2001

24. Nannini C, Ryu JH, Matteson EL: Lung disease in rheumatoid arthritis.Curr Opin Rheumatol 20(3):340-346, 2008

25. Mori S, Cho I, Koga Y, et al: Comparison of pulmonary abnormalities onhigh-resolution computed tomography in patients with early versuslongstanding rheumatoid arthritis. J Rheumatol 35(8):1513-1521, 2008

26. Mori S, Koga Y, Sugimoto M: Small airway obstruction in patients withrheumatoid arthritis. Mod Rheumatol 21(2):164-173, 2011

27. Bongartz T, Nannini C, Medina-Velasquez YF, et al: Incidence andmortality of interstitial lung disease in rheumatoid arthritis: A population-based study. Arthritis Rheum 62(6):1583-1591, 2010

28. Olson AL, Swigris JJ, Sprunger DB, et al: Rheumatoid arthritis-interstitiallung disease-associated mortality. Am J Respir Crit Care Med 183(3):372-378, 2011

29. Tsuchiya Y, Takayanagi N, Sugiura H, et al: Lung diseases directlyassociated with rheumatoid arthritis and their relationship to outcome.Eur Respir J 37(6):1411-1417, 2011

30. Mori S, Koga Y, Sugimoto M: Different risk factors between interstitiallung disease and airway disease in rheumatoid arthritis. Respir Med 106(11):1591-1599, 2012

31. LeeHK, KimDS, Yoo B, et al: Histopathologic pattern and clinical featuresof rheumatoid arthritis-associated interstitial lung disease. Chest 127(6):2019-2027, 2005

32. Kim EJ, Collard HR, King TE Jr: Rheumatoid arthritis-associated inters-titial lung disease: The relevance of histopathologic and radiographicpattern. Chest 136(5):1397-1405, 2009

33. Tanaka N, Kim JS, Newell JD, et al: Rheumatoid arthritis-related lungdiseases: CT findings. Radiology 232(1):81-91, 2004

34. Kim EJ, Elicker BM, Maldonado F, et al: Usual interstitial pneumonia inrheumatoid arthritis-associated interstitial lung disease. Eur Respir J 35(6):1322-1328, 2010

35. Tanoue LT: Pulmonarymanifestations of rheumatoid arthritis. Clin ChestMed 19(4):667-685, 1998; [viii]

36. Franquet T: High-resolution CT of lung disease related to collagenvascular disease. Radiol Clin North Am 39(6):1171-1187, 2001

37. Pons-Estel GJ, Alarcon GS, Scofield L, et al: Understanding the epidemi-ology and progression of systemic lupus erythematosus. Semin ArthritisRheum 39(4):257-268, 2010

38. Allen D, Fischer A, Bshouty Z, et al: Evaluating systemic lupus erythe-matosus patients for lung involvement. Lupus 21(12):1316-1325, 2012

39. JurencakR, FritzlerM, Tyrrell P, et al: Autoantibodies in pediatric systemiclupus erythematosus: Ethnic grouping, cluster analysis, and clinicalcorrelations. J Rheumatol 36(2):416-421, 2009

40. Mittoo S, Gelber AC, Hitchon CA, et al: Clinical and serologic factorsassociated with lupus pleuritis. J Rheumatol 37(4):747-753, 2010

41. Kinder BW, Freemer MM, King TE Jr., et al: Clinical and genetic riskfactors for pneumonia in systemic lupus erythematosus. Arthritis Rheum56(8):2679-2686, 2007

42. Bernatsky S, Boivin JF, Joseph L, et al: Mortality in systemic lupuserythematosus. Arthritis Rheum 54(8):2550-2557, 2006

43. Cheema GS, Quismorio FP Jr.: Interstitial lung disease in systemic lupuserythematosus. Curr Opin Pulm Med 6(5):424-429, 2000

44. Matthay RA, Schwarz MI, Petty TL, et al: Pulmonary manifestations ofsystemic lupus erythematosus: Review of twelve cases of acute lupuspneumonitis. Medicine (Baltimore) 54(5):397-409, 1975

45. Karim MY, Miranda LC, Tench CM, et al: Presentation and prognosis ofthe shrinking lung syndrome in systemic lupus erythematosus. SeminArthritis Rheum 31(5):289-298, 2002

46. Fenlon HM, Doran M, Sant SM, et al: High-resolution chest CT insystemic lupus erythematosus. Am J Roentgenol 166(2):301-307, 1996

47. Sant SM, DoranM, Fenelon HM, et al: Pleuropulmonary abnormalities inpatients with systemic lupus erythematosus: Assessment with highresolution computed tomography, chest radiography and pulmonaryfunction tests. Clin Exp Rheumatol 15(5):507-513, 1997

48. Ooi GC, Ngan H, Peh WC, et al: Systemic lupus erythematosus patientswith respiratory symptoms: The value of HRCT. Clin Radiol 52(10):775-781, 1997

49. Good JT Jr, King TE, Antony VB, et al: Lupus pleuritis. Clinical featuresand pleural fluid characteristics with special reference to pleural fluidantinuclear antibodies. Chest 84(6):714-718, 1983

50. Porcel JM, Ordi-Ros J, Esquerda A, et al: Antinuclear antibody testingin pleural fluid for the diagnosis of lupus pleuritis. Lupus 16(1):25-27,2007

51. FathiM,Vikgren J, BoijsenM, et al: Interstitial lungdisease in polymyositisand dermatomyositis: Longitudinal evaluation by pulmonary functionand radiology. Arthritis Rheum 59(5):677-685, 2008

52. Dalakas MC, Hohlfeld R: Polymyositis and dermatomyositis. Lancet 362(9388):971-982, 2003

53. Marguerie C, Bunn CC, Beynon HL, et al: Polymyositis, pulmonaryfibrosis and autoantibodies to aminoacyl-tRNA synthetase enzymes. Q JMed 77(282):1019-1038, 1990

54. Solomon J, Swigris JJ, Brown KK: Myositis-related interstitial lung diseaseand antisynthetase syndrome. J Bras Pneumol 37(1):100-109, 2011

55. Koreeda Y, Higashimoto I, Yamamoto M, et al: Clinical and pathologicalfindings of interstitial lung disease patients with anti-aminoacyl-tRNAsynthetase autoantibodies. Intern Med 49(5):361-369, 2010

56. Fathi M, Dastmalchi M, Rasmussen E, et al: Interstitial lung disease, acommon manifestation of newly diagnosed polymyositis and dermato-myositis. Ann Rheum Dis 63(3):297-301, 2004

57. Tansey D,Wells AU, Colby TV, et al: Variations in histological patterns ofinterstitial pneumonia between connective tissue disorders and theirrelationship to prognosis. Histopathology 44(6):585-596, 2004

58. KimSJ, Lee KS, Ryu YH, et al: Reversed halo sign on high-resolutionCT ofcryptogenic organizing pneumonia: Diagnostic implications. Am J Roent-genol 180(5):1251-1254, 2003

59. Fischer A, Swigris JJ, du Bois RM, et al: Anti-synthetase syndrome in ANAand anti-Jo-1 negative patients presenting with idiopathic interstitialpneumonia. Respir Med 103(11):1719-1724, 2009

60. Lundberg IE: The heart in dermatomyositis and polymyositis. Rheuma-tology (Oxford) 45(suppl 4), 2006; iv18-21

61. Hill CL, Zhang Y, Sigurgeirsson B, et al: Frequency of specific cancer typesin dermatomyositis and polymyositis: A population-based study. Lancet357(9250):96-100, 2001

62. Gunnarsson R, Aalokken TM,Molberg O, et al: Prevalence and severity ofinterstitial lung disease in mixed connective tissue disease: A nationwide,cross-sectional study. Ann Rheum Dis 71(12):1966-1972, 2012

63. Bodolay E, Szekanecz Z, Devenyi K, et al: Evaluation of interstitial lungdisease in mixed connective tissue disease (MCTD). Rheumatology(Oxford) 44(5):656-661, 2005

64. Burdt MA, Hoffman RW, Deutscher SL, et al: Long-term outcome inmixed connective tissue disease: Longitudinal clinical and serologicfindings. Arthritis Rheum 42(5):899-909, 1999

T. Bryson et al.38

65. Kozuka T, Johkoh T, Honda O, et al: Pulmonary involvement in mixedconnective tissue disease: High-resolution CT findings in 41 patients.J Thorac Imaging 16(2):94-98, 2001

66. Saito Y, Terada M, Takada T, et al: Pulmonary involvement in mixedconnective tissue disease: Comparison with other collagen vascular diseasesusing high resolution CT. J Comput Assist Tomogr 26(3):349-357, 2002

67. Dalvi V, Gonzalez EB, Lovett L: Lymphocytic interstitial pneumonitis(LIP) in Sjogren's syndrome: A case report and a review of the literature.Clin Rheumatol 26(8):1339-1343, 2007

68. Ramos-Casals M, Solans R, Rosas J, et al: Primary Sjogren syndrome inSpain: Clinical and immunologic expression in 1010 patients. Medicine(Baltimore) 87(4):210-219, 2008

69. Lohrmann C, UhlM,Warnatz K, et al: High-resolution CT imaging of thelung for patients with primary Sjogren's syndrome. Eur J Radiol 52(2):137-143, 2004

70. Ito I, Nagai S, Kitaichi M, et al: Pulmonary manifestations of primarySjogren's syndrome: A clinical, radiologic, and pathologic study. AmJ Respir Crit Care Med 171(6):632-638, 2005

71. Parambil JG, Myers JL, Lindell RM, et al: Interstitial lung disease inprimary Sjogren syndrome. Chest 130(5):1489-1495, 2006

72. Voulgarelis M, Dafni UG, Isenberg DA, et al: Malignant lymphoma inprimary Sjogren's syndrome: A multicenter, retrospective, clinical studyby the European Concerted Action on Sjogren's Syndrome. ArthritisRheum 42(8):1765-1772, 1999

73. Tonami H, Matoba M, Kuginuki Y, et al: Clinical and imaging findings oflymphoma in patients with Sjogren syndrome. J Comput Assist Tomogr27(4):517-524, 2003

74. Honda O, Johkoh T, Ichikado K, et al: Differential diagnosis oflymphocytic interstitial pneumonia and malignant lymphoma on high-resolution CT. Am J Roentgenol 173(1):71-74, 1999

75. Song JW, Do KH, Kim MY, et al: Pathologic and radiologic differencesbetween idiopathic and collagen vascular disease-related usual interstitialpneumonia. Chest 136(1):23-30, 2009

76. Raghu G, Collard HR, Egan JJ, et al: An official ATS/ERS/JRS/ALATstatement: Idiopathic pulmonary fibrosis: Evidence-based guidelines fordiagnosis and management. Am J Respir Crit Care Med 183(6):788-824,2011

77. American Thoracic Society/European Respiratory Society Inter-national Multidisciplinary Consensus Classification of the IdiopathicInterstitial Pneumonias: This joint statement of the American ThoracicSociety (ATS), and the European Respiratory Society (ERS) wasadopted by the ATS board of directors, June 2001 and by the ERSExecutive Committee, June 2001. Am J Respir Crit Care Med 165(2):277-304, 2002

78. ThomeerMJ, Vansteenkiste J, Verbeken EK, et al: Interstitial lung diseases:Characteristics at diagnosis and mortality risk assessment. Respir Med 98(6):567-573, 2004

79. Park JH, Kim DS, Park IN, et al: Prognosis of fibrotic inter-stitial pneumonia: Idiopathic versus collagen vascular disease-related subtypes. Am J Respir Crit Care Med 175(7):705-711,2007

80. Castelino FV, Varga J: Interstitial lung disease in connective tissue diseases:Evolving concepts of pathogenesis and management. Arthritis Res Ther12(4):213, 2010