Pathology Patterns ReviewsA b s tra c tCystic fibrosis (CF) remains the most common life-limiting inherited disease in America. Making anaccurate, early diagnosis is essential to themanagement of the disease. The diagnostic criteria forCF require the presence of 1 or more typical clinicalfeatures, a family history of CF, or a positive newbornscreening test, plus laboratory evidence of the CFtransmembrane conductance regulator (CFTR)dysfunction. In the past, the laboratory test of abnormalCFTR function was based largely on an elevated sweatchloride test result. The recent development of agenotypic CFTR mutation screen has greatly improveddiagnostic accuracy. Increased screening of the CFTRlocus has led to the recognition of a number of atypicalCF disorders. Recently, a 2-tiered newborn screeningprotocol including CFTR genotyping has becomepopular, increasing the likelihood of early diagnosis.Cysticfibrosis(CF)isthemostcommonlife-limitingautosomalrecessivedisorderinwhites,withanincidenceof1 in 3,200 live births in the United States.1More than 25,000Americans have CF, with approximately 850 new cases diag-nosedeachyear.Thediseasealsohasarelativelyhighinci-denceamongHispanics(1in9,500)butisrareinnativeAfricans and native Asians.2TheresponsibleCFgeneproduct,theCFtransmem-braneconductanceregulator(CFTR),isaproteinof1,480aminoacidsthatfunctionsasacyclicadenosinemonophos-phate(cAMP)-regulatedchloridechannelattheapicalsurface of epithelial cells.3The defective electrolyte transportacrosstheepitheliainpatientswithCFleadstoincreasedmucusviscosityandrecurrentepisodesofobstruction,inflammation, and progressive destruction of affected organs.TheCFphenotypeismostconspicuousinthelungs,pancreas, sweat glands, and, in men, the vas deferens. TypicalclinicalfeaturesaresummarizedinTable1.4Theclinicalmanifestationsarehighlyvariableamongunrelatedindivid-ualsandwithinfamilies,withtheclinicalcourseofthediseaserangingfromearlydeathasaresultofpulmonarycomplications to relatively mild manifestations in the secondand third decades, and a rare normal length of life. The majorpredictorsofmortalityincludepulmonaryfunction,airwaymicrobiology,andnutritionstatus.Withadvancesinthera-peutic technology and aggressive antimicrobial management,survivalhasimprovedsignificantly.IntheUnitedStates,mediansurvivalincreasedfrom18yearsin1976to31.1yearsformenand28.3yearsforwomenin1995.2Despitethese advances, CF remains an incurable disease.SinceCFhasseriouslifelongmedicalandsocialimplica-tions for affected individuals and their families, making an accu-ratediagnosisasearlyaspossibleisclearlyessentialsothatAm J Clin Pathol 2002;117(Suppl 1):S109-S115 S109 American Society for Clinical PathologyLaboratory Tests for the Diagnosis of Cystic FibrosisLan Wang, MD,1and Steven D. Freedman, MD, PhD2Key Words: Laboratory test; Cystic fibrosis; DiagnosisWang and Freedman / LABORATORY TESTS FOR THE DIAGNOSIS OF CYSTIC FIBROSISappropriate treatment, genetic counseling, and access to special-izedmedicalservicescanbeensured.Themajorityofpatientswith CF are diagnosed during the first decade of life. The diag-nosisofCFinthepastwasbasedlargelyonapositivefamilyhistory,suggestiveclinicalmanifestations,andapositivesweattest result. However, many false-positive and false-negative caseshave been documented. Most misdiagnoses have been attributedto the vagaries of the sweat test technique. In addition, the criteriafor diagnosing atypical CF with a borderline or negative sweattest result in the face of convincing clinical evidence is debated.TherecentidentificationofthegeneforCFTRandthereadilyavailable laboratory techniques to detect CF mutations have clar-ifiedsomediagnosticuncertaintiesandimproveddiagnosticaccuracy.Ontheotherhand,genotypinganalysesofthegenemutationsledtotherecognitionofanexpandingphenotypicspectrumofatypicaldisease.ThediagnosisofCFcontinuestobechallenging,andcliniciansareincreasinglyfacingthedilemma: What makes the CF diagnosis?This review summarizes currently available major diag-nosticandancillarylaboratorytests,andprovidessomeinsight into the uses of these tests in the diagnosis of CF.Major Diagnostic Laboratory TestsThree major tests are used to detect CFTR abnormality.Sweat TestThe sweat test, first described in 1959, remains the goldstandard for the diagnosis of CF.3It is composed of 3 steps:sweatstimulationbypilocarpineiontophoresis;collectionofthesweatontogauze,filterpaper,coil,patch,orcapillarytube; and qualitative and quantitative analysis of the sweat forchloride concentration, sodium concentration, conductivity, orosmolality. A qualitative sweat test is a screening test for CFandmaynotrequiremeasurementoftheamountofsweatcollected.Thestandardsweattestorquantitativepilocarpineiontophoresistestisthediagnosticprocedureofchoiceandshould be carried out in accordance with the guidelines of theNationalCommitteeforClinicalLaboratoryStandardsinanexperienced laboratory or center.5Thereferencevaluesforsweatchlorideconcentrationare as follows: Greaterthan60mmol/L:consistentwithCF(meanvalue around 100 mmol/L) A range between 40 and 60 mmol/L: borderline Less than 40 mmol/L: negative (mean value around 30mmol/L)Assweatelectrolyteconcentrationincreaseswithage,interpretationoftheresultshouldbeadjustedaccordingtothepatientsage.Somedatahavesuggestedthatininfantsyoungerthan3months,asweatchlorideconcentrationofmore than 40 mmol/L is highly suggestive of a diagnosis ofCF;about10%ofhealthyadolescentsandadultscanhavesweatchlorideconcentrationofmorethan60mmol/L.6,7Inthesecases,afludrocortisonesuppressiontestmaybeconsidered.Theadministrationoforal9-alphafludrocorti-sone(3mg/m2perdayfor2days)increasessodiumreab-sorptioninthenormalsweatduct,andthesweatelectrolyteconcentrationdecreasesinunaffectedindividualsbutnotinpatients with CF.8Measurement of the sweat sodium concen-trationaloneisnotrecommendedbecauseofthehighinci-dence of false-positive results. However, in some cases withborderlinesweatchlorideresults,measurementofbothsodiumandchlorideconcentrationscanbehelpful.Indirectmeasurementofosmolalityorconductivityisconsideredunreliable for the purpose of diagnostic testing and NationalConsensusGuidelinesadvisethatthesemeasurementsbeconfinedtouseasscreeningtests.Thesweattestshouldberepeated at least twice on separate occasions before a conclu-sion is made.Thesweatelectrolyteconcentrationisknowntodecreaseatlowsweatrates.Therefore,aminimumaccept-ablevolume(15mfortheWescorMacroductcoilsystem,Wescor, Logan, UT) or weight (75 mg for the Gibson-Cookeprocedure)ofsweatmustbecollectedfromasinglesiteduring a 30-minute period of sampling to ensure an averagesweat rate of more than 1 g/m2per minute.5Thesweattestisatechnicallylaboriousprocedureandrequiresexperiencedpersonnelusingstandardizedmethodsin facilities that perform adequate numbers of tests to main-tainlaboratoryproficiencyandqualitycontrol.DependingS110 Am J Clin Pathol 2002;117(Suppl 1):S109-S115 American Society for Clinical PathologyTable 1Clinical Symptoms and Findings of Cystic Fibrosis4RespiratoryRecurrent pneumoniaRecurrent bronchiolitisPersistent pneumonia or productive coughPansinusitis or nasal polyposisPersistent colonization or infection with Pseudomonas aeruginosa, Staphylococcus aureusGastroenterologicFailure to thrive in infancy and childhood (protein-calorie malnutrition)Voluminous, foul-smelling, or fatty stoolsMucoid appendiceal impaction (pathologic finding)Distal intestinal obstruction in children and young adultsRectal prolapseIntussusception in childhood or adolescenceUnexplained hepatic cirrhosis, cholelithiasis, or pancreatitis before the age of 30 yearsSymptoms secondary to deficiency of vitamin A, D, E, or KMiscellaneousSalty-tasting skinHyponatremic dehydration or metabolic alkalosis in infancy and recurrent hyponatremic heat prostration in adultsDigital clubbing in children and young adultsAzoospermia or infertility; congenital absence of the vas deferensHypoproteinemia or anasarca (in infancy)Parental request for testing for cystic fibrosisPathology Patterns Reviewsonthemethodsused,sweattestscanhaveafalse-positiverateashighas15%orafalse-negativerateashighas12%.9,10Methodologicandtechnicalerrorshaveaccountedforalargeproportionofthefalseresults.Inaddition,false-positivesweattestresultscanariseinsituationssuchassampleevaporation,patientdehydration,skineczema,orrash.False-negativeresultscanbeseenwhenthepatientisedematousortakingmineralcorticoidsorwhenaninade-quate amount of sweat is collected.ElevatedsweatelectrolyteconcentrationsalsocanbeseeninmanyotherclinicaldisordersTable2.Fortunately,most of these disorders have clinically distinct manifestations,and a differential diagnosis is not very difficult to exclude.Thesweattestisnotarecommendedmethodforscreening. It also is not very useful in the early diagnosis ofyounginfants,fromwhomitusuallyismoredifficulttoobtainasufficientquantityofsweatsample.Inaddition,atypical cases have been reported in patients with normal orborderlinesweattestresultswhohaveorsubsequentlydevelop clinical manifestations of CF. Therefore, an alterna-tivereliablediagnosticandscreeningtoolisdesirable.Thediscovery of the CFTR gene and the identification of disease-producing mutations have raised considerable enthusiasm.Genotyping of CFTR MutationsTheCFphenotypeiscausedbymutationsinasinglegene, the CFTR gene, which was mapped to chromosome 7and cloned in 1989.11Genotyping of the gene has identifiedmorethan900mutationsinpatientswithCF,andthelistcontinues to grow. (For an up-to-date registry, view the Websiteathttp://www.genet.sickkids.on.ca/cftr/.)Themostfrequent mutation is a 3-base-pair deletion of the codon for aphenylalanineresidueataminoacidposition508(deltaF508),accountingfornearly70%oftheCFallelesinthewhite population. Some 15 to 20 other common mutationsaccount for 2% to 15% of the CF alleles. By using a panel ofcommerciallyavailableprobesfor70commonmutations,genetic testing for CF has approached a sensitivity of 90% inthegeneralpopulationoftheUnitedStates.2SinceaDNAsamplecouldbeeasilycollectedfromapersonofanyageand the DNA analysis technique is readily available and reli-ableintodayslaboratory,genotypingofCFTRmutationshasbeenusedasoneoftheconfirmativetestsforearlyCFdiagnosisandasascreeningmethodinantenatalandnewborndiagnosisforCFandgeneticcounselingforCFcarriers in the United States and other countries. It has beenpostulatedthatgeneticcounselingandantenataldiagnosismay have contributed to the decrease in the incidence of CFfrom 1 in 2,500 to 1 in 3,200 live births.12Identificationof2mutatedallelesinanindividualisaveryusefuladjunctinthediagnosisofCF.However,intheabsence of typical clinical manifestations, a family history, oranothertestresultpositiveforaCFTRabnormality,geno-typing alone can neither establish the diagnosis nor rule it out.The interpretation of genotyping results may be compli-cated by the following factors:1. Thesensitivityofmutationanalysisisdecreasedbecause of the large number of CFTR mutations. Since mostmutationsarerare,expandingthepanelofscreenedmuta-tionsachievesonlymarginalgainsinsensitivity.Themuta-tiondetectionratevariesindifferentethnicgroupsdependingonfrequenciesofthecommonmutations.Forexample,thedeltaF508 mutationsaccountfor70%ofCFallelesinwhites,48%inAfricanAmericans,46%inHispanics, and only 30% in Asian Americans and AshkenaziJews.2ThedetectionratesindifferentethnicgroupsusingthecurrentlyavailablescreeningpanelaresummarizedinTable3.DelayeddiagnosisofCFinchildrenfromethnicminoritieshasbeenreported.Customizingmutationpanelspotentiallycouldenhancethesensitivityofgenotypingindifferentethnicbackgrounds,althoughthisisnotroutineclinical practice.2. MutationanalysisoftheCFTRgenehasrevealedanexpandingspectrumofphenotype,rangingfromatypicalmanifestations(suchascongenitalbilateralabsenceofthevas deferens only in men) to more typical CF manifestations(suchaspancreaticinsufficiency,sinopulmonarydisease).Exceptforpancreaticinsufficiency,genotype-phenotypecorrelationofotherCFmanifestationswithdifferentCFTRmutations is poor. Some mutations (eg, R117H) are reportedto be associated with widely variable clinical manifestations,fromentirelyhealthyindividualstothetypicalmultiorganmanifestations of CF. Therefore, the diagnosis of CF cannotbebasedsolelyontheidentificationofCFTRmutations.ItusuallyneedstobeconfirmedbyobservingtypicalclinicalAm J Clin Pathol 2002;117(Suppl 1):S109-S115 S111 American Society for Clinical PathologyTable 2Differential Diagnosis of Elevated Electrolyte Levels in Sweat4Metabolic causes (true-positive results)Cystic fibrosisFucosidosisGlucose-6-phosphate dehydrogenase deficiencyGlycogen storage disease type 1MucopolysaccharidosisUntreated hypothyroidismVasopressin-resistant diabetes insipidusAdrenal insufficiencyFamilial cholestasisFamilial hypoparathyroidismHypogammaglobulinemiaSkin and sweat-gland disease (false-positive results)MalnutritionEctodermal dysplasiaAtopic dermatitisIatrogenic cause: long-term infusion of prostaglandin E1Methodologic errors (examples)Partial evaporation of sampleError in calculationWang and Freedman / LABORATORY TESTS FOR THE DIAGNOSIS OF CYSTIC FIBROSISmanifestationsorapositivesweattestresult.Controversialopinionshavebeenraisedonwhetheraclinicallyhealthymanwithcongenitalbilateralabsenceofthevasdeferenswho carries only 2 CFTR mutations should be diagnosed ashavingCF.Asaresult,aconsensusstatementforthediag-nosis of CF has been developed.13When the same mutationispassedontothemanschildren,itcouldmanifestwithamore typical CF phenotype.3. Worth mentioning are that cases have been reported inwhichindividualsdonothavetypicalCFsymptomsuntiltheyareolderthan60years.14Carefulsurveillanceandfollow-upofindividualswithatypicaldiseasestilliswarranted.4. Inrarecases,2ormoremutationsoccurringonthesame chromosome (in cis) may interact with each other andmodifytheCFphenotype.Forexample,thecomplexalleleR553Q-deltaF508 hasbeenshowntorevertpartiallyoramelioratethephenotypeofdeltaF508 mutation.15Otherrevertants(deltaF508-V1212I andR334W-R1158X)associ-ated with mild or atypical CF have also been described.16,17Nasal Potential-Difference MeasurementsThis is a functional test for the CFTR gene product. Therespiratoryepithelia,includingnasalepithelia,maintainatransepithelialelectricalpotentialdifference(PD)byregu-latingactivetransportofionssuchassodium(Na+)andchloride(Cl)acrossthemembrane.ThenasalPDcanbemeasuredinvivobyusingasaline-filledbutterflyneedleinserted subcutaneously in the forearm as the reference elec-trodeandasaline-filledpolyethylenetubethatgentlytouches the mucosa near the inferior turbinate as the mucosalelectrode.TheCFTRmutationscauseareducedchloridesecretionthroughthecAMP-regulatedClchannelandincreasedNa+resorptionthroughtheepithelialsodiumchannel,resultinginadifferentpatternofnasalPDinepitheliaofpatientswithCFcomparedwithnormalepithelia. The 3 following features characterize the nasal PDassociated with CF Figure 118:1. HigherbasalPD(meanSD:normal,24.70.9mV;abnormal,531.8mV),whichreflectsthehyperpo-larizationofthemembranegeneratedbyenhancedNa+transport relative to reduced Cl transport.2. Greater inhibition of PD after nasal perfusion with theNa+channelinhibitor,amiloride(normal,53%;abnormal,73%), which reflects inhibition of accelerated Na+ transport.3. Little response or absence of response in PD to perfu-sion of the nasal epithelial surface with a solution free of Clinconjunctionwithabeta-agonist,isoproterenol,whichreflects a significant reduction or absence of CFTR-mediatedCl secretion.InterpretationofthePDmeasurementsrequiresaclearunderstandingofphysiologyofiontransportofthenasalepithelium and the PD responses to perfusion with differentprobesofiontransport.False-negativeresultsmayoccurinthe presence of inflamed epithelium, previous nasal surgery,allergicrhinitis,ornasalpolyps.Inaddition,thePDmeasurementsareinfluencedbythepreciseanatomicloca-tion in the nose and the genotype.19Therefore, absence of araised basal PD does not rule out CF. As with the sweat test,nasal PD measurements must be repeated on at least 2 sepa-rate occasions.Nasal PD can be measured in patients as young as a fewhoursandmaydemonstrateabnormalCFTRfunctionmorereliably than the sweat test.18-20However, the measurements areS112 Am J Clin Pathol 2002;117(Suppl 1):S109-S115 American Society for Clinical PathologyTable 3Carrier Incidence and Detection Rate of CFTR Mutations in Different Ethnic Backgrounds*CF Carrier Carrier TestEthnic Background Incidence Detection Rate (%)Northern European 1/25-1/29 90Southern European 1/25-1/29 70Ashkenazi Jewish 1/26-1/29 97Hispanic 1/46 57African American 1/60-1/65 72Asian American 1/90 30*From Cystic fibrosis DNA testing [educational service brochure]. Framingham,MA: Genzyme Genetics; 1999.6050403020100Amiloride0 CIAmiloride Iso0 CIAmiloride 0 1 2 3 4 5 6 7 8 9PD (mV)706050403020100Amiloride0 CIAmiloride Iso0 CIAmiloride 0 1 2 3 4 5 6 7 8 9PD (mV)Time (Minutes)Figure 1 Nasal potential-difference (PD) tracing in a healthysubject (top) and a patient with cystic fibrosis (bottom).Tracings illustrate the response of PD to perfusion withamiloride (104mol/L concentration), addition of a solutionfree of Cl (gluconate buffer) to amiloride, and addition ofisoproterenol (Iso; 105mol/L concentration) to the solutionfree of Cl containing amiloride (see text). Reprinted withpermission from Rosenstein and Cutting.13Pathology Patterns Reviewstime-consumingandrequiresubstantialexpertisetoperformand, therefore, are best undertaken only by laboratories experi-enced in the technique. It is currently more a research tool forclinical studies and rarely used in the diagnosis of CF.Newborn Screening for Cystic FibrosisIt is widely believed that early detection would improveoutcome for patients with CF. Newborn screening programsfor CF were developed in Europe, Australia, and some partsoftheUnitedStatesafter1979whenCrossleyandcoworkers21showedthatthelevelofimmunoreactivetrypsinogen (IRT) is 2 to 5 times higher in neonates with CF.ThescreeningtestsareperformedondriedbloodspotsfromtheGuthriecards(usuallytakenduringthefirstweekof life) by 2 protocols. In the 1-tiered IRT protocol, neonateswith an IRT level of 180 ng/mL or more are considered posi-tive.Anotherspecimenobtainedatweek4usuallyismeasured to reduce the number of false-positive results. Thismethodyieldsasensitivityof87%,aspecificityof99.9%,andapositivepredictivevalueof6.4%.The2-tieredIRT/DNA protocol uses an IRT level of 110 ng/mL or moreasthecutofffollowedbyDNAanalysisforthedeltaF508CFTR mutation from the original dried blood sample on theGuthriecard.Thepresenceofatleast1mutatedalleleisconsideredpositive.Neonateswithapositivescreeningtestresult then are evaluated by the sweat test for the diagnosis ofCF. Greater sensitivity (90.9%) and positive predictive value(15.2%) can be achieved using the latter protocol, and morerapidresultscanbeobtainedsincebothtiersofthescreenusesamplesfromtheinitialGuthriecards.22Apositivepredictivevalueupto37.4%hasbeenshowninanotherstudyusingthesimilarscreeningstrategy.23IncludingmoreDNAmutationsinthescreeningprotocolwillslightlyincreasethesensitivityofthetestbutwilladdsubstantiallyto the cost of screening.The level of IRT decreases after 1-2 months of age andthereafter becomes unreliable.Ancillary Testing for Cystic FibrosisInpatientswhohaveanatypicalphenotypeofCFwithanequivocalsweattestresultornasalPDmeasurement,additional evidence for or against CF is very important. Thiscan be accomplished by using the following ancillary tests.Assessment of Exocrine Pancreatic FunctionMostpatientswithCFhavesignsandsymptomsofmalabsorption and abnormal exocrine pancreatic function. Anumberofdirectandindirecttestsareavailable,suchasmeasurementsofp-aminobenzoicacidabsorption,serumtrypsinogenandcarotenelevels,stoolenzymelevels,andfecal fat analysis with a minimum of a 72-hour pooled stoolcollection.24Moresubtlepancreaticfunctionalabnormalitycanbedetectedbydirectanalysisofpancreaticfluidandanion secretion before and after stimulation with secretin andcholecystokinin.25Theexocrinepancreashasahighfunctionalcapacity,andmorethan90%ofthepancreaticenzymesecretioncapacity must be lost before clinical signs and symptoms ofmalabsorptionareevident.Ontheotherhand,patientswithatypical CF often have mild clinical symptoms of pancreaticinsufficiency. For this reason, tests for pancreatic function todatehavebeenundertakentoassesstheneedforandadequacy of treatment rather than to prove the diagnosis.Sinus RadiographPansinusitisisvirtuallyuniversalinpatientswithCFanduncommoninotherwisehealthychildrenoryoungadults.26Thepresenceofpansinusitisisstronglysuggestiveof but not specific for the diagnosis of CF.Evaluation of Airway Function and Respiratory TractMicrobiologySmallairwayabnormalitiescouldbedemonstratedbyradiologic imaging and functional assessment using spirometry.CharacterizationoftherespiratorytractmicrobialfloraisanotherusefuladjunctforthediagnosisofCF.PersistentandprogressivepulmonaryinfectionbymucoidPseudomonasaeruginosa isstronglysuggestiveofCF.Persistent colonization with other organisms such as Staphy-lococcus aureus, Haemophilus influenzae, and Burkholderiacepacia alsomaysupportadiagnosisofCF,althoughthesepathogensarenotspecific.27Theorganismsusuallycanbeculturedfrombronchoalveolarlavagefluid,sputum,anoropharyngealswab,orasinusaspirate.Insomepatientswith clinically suspected CF, indirect tests for rising levels ofserum antibodies against P aeruginosa may document prob-able infection despite negative culture results.Inaddition,substantialinflammationoftheairwayisvirtually universal in patients with CF, even in the absence ofanyotherevidenceofinfection.28InpatientswithatypicalCF manifestations and no overt pulmonary disease, the pres-enceofahighpercentageofneutrophils(50%ormoreinpatientswithCFcomparedwithabout3%inhealthysubjects)andahighabsoluteneutrophilcountinbron-choalveolar lavage fluid, even in the absence of pathogens, isstrong evidence of CF.29Urogenital EvaluationObstructiveazoospermiaisafindingpresentin98%to99%ofindividualsaffectedwithCF.12AcarefulevaluationAm J Clin Pathol 2002;117(Suppl 1):S109-S115 S113 American Society for Clinical PathologyWang and Freedman / LABORATORY TESTS FOR THE DIAGNOSIS OF CYSTIC FIBROSISof postpubertal male subjects includes urologic examination,semenanalysis,ultrasonographicstudyoftheurogenitalstructures, and, in rare cases, scrotal exploration.Embryologic development of the vas deferens seems torequireahigherlevelofCFTRactivitythanisrequiredforthe normal function of other organs affected by CF.30There-fore,congenitalbilateralabsenceofthevasdeferenscanrepresentamildervariantofCFTRdysfunctionwithoutapparentabnormalityinpancreasandlungs.However,adiagnosis of CF is assigned to an individual who is first seenwith obstructive azoospermia only if there is evidence of anelevated sweat chloride concentration, identification of 2 CFmutations, or an abnormal nasal PD.ConclusionThediagnosisofCFneedstobemadeasearlyandasconfidentlyaspossible.ThediagnosticcriteriaforCFinclude the presence of 1 or more typical clinical features, afamilyhistoryofCF,orapositivenewbornscreeningtestresult,pluslaboratoryevidenceofCFTRdysfunction.AbnormalCFTRfunctionusuallywillbedocumentedby2separate elevated sweat chloride test results, 2 CF mutationsbygenotyping,oranabnormalnasalPDmeasurementrecorded on 2 separate days.For the group of patients who do not have the classic CFphenotypes,perhapsthetermCFTR-associateddiseaseshouldbeassigned.Theydonotnecessarilydeservethesamerigoroustreatment,butfollow-upisrequiredsincecases of delayed CF have been reported.From the 1Division of Laboratory Medicine, Department ofPathology, Massachusetts General Hospital and Harvard MedicalSchool, and 2Department of Medicine, Beth Israel DeaconessMedical Center and Harvard Medical School, Boston.Address reprint requests to Dr Wang: c/o Michael Laposata,MD, PhD, MGH Division of Laboratory Medicine, 55 Fruit St,Gray 235, Boston, MA 02114.References1. Hamosh A, Fitz-Simmons SC, Macek M Jr, et al. Comparisonof the clinical manifestation of cystic fibrosis in black andwhite patients. J Pediatr. 1998;132:255-259.2. Genetic testing for cystic fibrosis: National Institutes ofHealth Consensus Development Conference statement ongenetic testing for cystic fibrosis. Arch Intern Med.1999;159:1529-1539.3. Gibson LE, Cooke RE. Test for the concentration ofelectrolytes in sweat in cystic fibrosis of the pancreas utilizingpilocarpine by iontophoresis. Pediatrics. 1959;24:545-549.4. Stern RC. The diagnosis of cystic fibrosis. N Engl J Med.1997;336:487-491.S114 Am J Clin Pathol 2002;117(Suppl 1):S109-S115 American Society for Clinical Pathology5. LeGrys VA, Burritt MF, Gibson LE, et al. Sweat Testing SampleCollection and Quantitative Analysis; Approved Guideline. C34-A. Villanova, PA: National Committee for ClinicalLaboratory Standards; 1994.6. Farrell PM, Koscik RE. Sweat chloride concentrations ininfants homozygous or heterozygous for F508 cystic fibrosis.Pediatrics. 1996;97:524-528.7. Gibson LE, di Sant Agnese PA, Schwachman H. Procedure forthe Quantitative Iontophoresis Sweat Test for Cystic Fibrosis.Bethesda, MD: Cystic Fibrosis Foundation; 1985.8. Hodsen ME, Beldon I, Power R, et al. Sweat tests to diagnosecystic fibrosis in adults. Br Med J (Clin Res Ed).1983;286:1381-1383.9. Denning CF, Huang NN, Cuasay LR. Cooperative studycomparing three methods of performing sweat test to diagnosecystic fibrosis. Pediatrics. 1980;66:752-757.10. Legrys VA, Wood RE. Incidence and implications of falsenegative sweat test reports in patients with cystic fibrosis.Pediatr Pulmonol. 1988;4:169-172.11. Riordan JR, Rommens JM, Kerem B, et al. Identification ofthe cystic fibrosis gene: cloning and characterization ofcomplementary DNA. Science. 1989;45:1066-1073.12. Welsh MJ, Tsui L-C, Boat TF, et al. Cystic fibrosis. In: SciverCF, Beaudet AL, Sly WS, et al, eds. The Metabolic andMolecular Bases of Inherited Disease. New York, NY: McGraw-Hill; 1995:3799-3876.13. Rosenstein BJ, Cutting GR, for the Cystic FibrosisFoundation Consensus Panel. The diagnosis of cystic fibrosis:a consensus statement. J Pediatr. 1998;132:589-595.14. Spencer DA, Venkataraman M, Weller PH. Delayed diagnosisof cystic fibrosis in children from ethnic minorities. Lancet.1993;342:238-239.15. Teem JL, Berger HA, Ostedgaard LS. Identification ofrevertants for the cystic fibrosis delta F508 mutation usingSTE6-CFTR chimeras in yeasts. Cell. 1993;73:335-346.16. Duarte A, Amaral M, Barreto C. Complex cystic fibrosis alleleR334W-R1158X results in reduced levels of correctlyprocessed mRNA in a pancreatic sufficient patient. HumMutat. 1996;8:134-139.17. Macek M Jr, Mickle J, Vavrova V. The identification of apossible revertant mutation (V1212I) in two Czech deltaF508 homozygous siblings with cystic fibrosis and delayedonset of pancreatic insufficiency. Isr J Med Sci. 1996;suppl32:S182.18. Knowles MR, Paradiso AM, Boucher RC. In vivo nasalpotential difference: techniques and protocols for assessingefficacy of gene transfer in cystic fibrosis. Hum Gene Ther.1995;6:445-455.19. Sauder RA, Chesrown SE, Loughlin GM. Clinical applicationof transepithelial potential difference measurements in cysticfibrosis. J Pediatr. 1987;111:353-358.20. Glowen CW, Lawson EE, Gingras-Leatherman J, et al.Increased nasal potential difference and amiloride sensitivityin neonates with cystic fibrosis. J Pediatr. 1986;108:517-521.21. Crossley JR, Elliott RB, Smith PA. Dried blood screening forcystic fibrosis in the newborn. Lancet. 1979;i:472-474.22. Gregg RG, Simantel A, Farrell PM, et al. Newborn screeningfor cystic fibrosis in Wisconsin: comparison of biochemicaland molecular methods. Pediatrics. 1997;99:819-824.23. Wilcken B, Wiley V, Sherry G, et al. Newborn screening forcystic fibrosis: a comparison of two strategies for casedetection in 1.2 million babies. J Pediatr. 1995;127:965-970.Pathology Patterns Reviews24. Nousia-Arvanitakis S, Arvanitakis C, Desai N, et al.Diagnosis of exocrine pancreatic insufficiency in cystic fibrosisby the synthetic peptide N-benzoyl-L-tyrosyl-p-aminobenzoicacid. J Pediatr. 1978;92:734-737.25. Lankisch PG. Exocrine pancreatic function tests. Gut.1982;23:777-798.26. Gharib R, Allen RP, Joos HA, et al. Paranasal sinuses in cysticfibrosis: incidence of roentgen abnormalities. Am J Dis Child.1964;108:499-502.27. Thomassen MJ, Demko CA, Doershuk CF. Cystic fibrosis: areview of pulmonary infections and interventions. PediatrPulmonol. 1987;3:334-351.28. Khan TZ, Wargener JS, Bost T, et al. Early pulmonaryinflammation in infants with cystic fibrosis. Am J Respir CritCare Med. 1995;151:1075-1082.29. Konstan MW, Hilliard KA, Norvell TM, et al.Bronchoalveolar lavage findings in cystic fibrosis patients withstable, clinically mild lung disease suggest ongoing infectionand inflammation. Am J Respir Crit Care Med. 1994;150:448-454.30. Chillon M, Casals T, Mercier B, et al. Mutations in the cysticfibrosis gene in patients with congenital absence of the vasdeferens. N Engl J Med. 1995;332:1475-1480.Am J Clin Pathol 2002;117(Suppl 1):S109-S115 S115 American Society for Clinical Pathology